Citation
Anticholinergic Medication Use and Risk of Cognitive Impairment and Dementia in Older Adults with Preclinical Alzheimer Disease

Material Information

Title:
Anticholinergic Medication Use and Risk of Cognitive Impairment and Dementia in Older Adults with Preclinical Alzheimer Disease
Creator:
Swami, Sunil
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (19 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Epidemiology
Committee Chair:
MANINI,TODD M
Committee Co-Chair:
STRILEY,CATHERINE L
Committee Members:
COHEN,RONALD A
KAIRALLA,JOHN ANDREW
Graduation Date:
12/19/2014

Subjects

Subjects / Keywords:
Alzheimers disease ( jstor )
Anticholinergics ( jstor )
Cholinergics ( jstor )
Cognition ( jstor )
Cognitive impairment ( jstor )
Dementia ( jstor )
Diseases ( jstor )
Medications ( jstor )
Memory ( jstor )
Older adults ( jstor )
Epidemiology -- Dissertations, Academic -- UF
anticholinergics -- cognition -- dementia
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Epidemiology thesis, Ph.D.

Notes

Abstract:
Age dependent decline in learning and memory is associated with atrophic or degenerative changes in the brain characterized by cortical cholinergic deficits and decreased cholinergic neuronal activity. Alzheimer disease (AD) cholinergic dysfunction is also associated with cognitive impairment, deposition of neuritic amyloid plaques and neurofibrillary tangles in brain. Furthermore, dementia severity is associated with decrease in cholinergic neurons, cholinergic signaling, and cholinergic receptors. Older adults have higher prevalence of chronic disease and polypharmacy. In addition, age associated decline in central cholinergic activity makes older adults susceptible to harmful effects of anticholinergic (AC) agents. However, there is an inadequate understanding of effects of AC drug use on progression of cognition. Recognizing individuals with very subtle clinical, cognitive and pathological alterations years before they develop noticeable clinical signs and symptoms of preclinical AD (PAD) and quantifying their anticholinergic burden might help us understand effects of anticholinergics on cognition. It is important to comprehend clinical and epidemiological effects of AC drug use on PAD and progression of cognitive decline. These cognitive impairments are very difficult to decipher clinically and the neuropsychological tests are required to identify these questionable cognitive impairment (QCI).The objectives of the proposed study was to assess the effects of chronic AC use on global cognition and specific cognitive domains among older adults with QCI. We assess anticholinergic load by Anticholinergic Drug Scale (ADS) using temporal medication records in the National Alzheimer's Coordinating Center (NACC) database. The Central objective of this study was to determine the association between AC drug use and global and domain specific cognition among older adults with QCI. We hypothesize that older adults with QCI who have high exposure to AC drugs are more likely to have lower global and domain specific cognitive scores, have a higher incidence rate of mild cognitive impairment (MCI)/AD, and have more rapid declines in global and domain specific cognition as compared to older adults who have no or low exposure to AC drugs. The proposed study provided information about temporal associations of AC drug use and risk of MCI/dementia in older adults with QCI. This study provided greater understanding of the role of AC drug use in cognitive impairment, which might help in reducing public health burden of cognitive impairment and AD. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2014.
Local:
Adviser: MANINI,TODD M.
Local:
Co-adviser: STRILEY,CATHERINE L.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2016-12-31
Statement of Responsibility:
by Sunil Swami.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
12/31/2016
Classification:
LD1780 2014 ( lcc )

Downloads

This item has the following downloads:


Full Text



PAGE 1

http://jah.sagepub.com/Journal of Aging and Health http://jah.sagepub.com/content/19/4/612 The online version of this article can be found at: DOI: 10.1177/0898264307300975 2007 19: 612 J Aging Health and James A. Mortimer Christopher B. Rosnick, Brent J. Small, Cathy L. McEvoy, Amy R. BorensteinOlder Adults Negative Life Events and Cognitive Performance in a Population of Published by: http://www.sagepublications.com can be found at: Journal of Aging and Health Additional services and information for http://jah.sagepub.com/cgi/alerts Email Alerts: http://jah.sagepub.com/subscriptions Subscriptions: http://www.sagepub.com/journalsReprints.nav Reprints: http://www.sagepub.com/journalsPermissions.nav Permissions: http://jah.sagepub.com/content/19/4/612.refs.html Citations: What is This? Aug 6, 2007 Version of Record >> at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 2

612 Negative Life Events and Cognitive Performance in a Population of Older AdultsChristopher B. Rosnick,PhD Brent J. Small,PhD Cathy L. McEvoy,PhD Amy R. Borenstein,PhD James A. Mortimer,PhDUniversity of South Florida,TampaObjectives: This study examined the association between negative life events in the past year and cognitive performance in a population of older adults. Methods: Secondary data analysis was conducted on 428 participants from the Charlotte County Healthy Aging Study. Participants completed tests of episodic memory,attention,and psychomotor speed and endorsed the presence and severity of 24 life events. Life events were examined in the aggregate as well as individually. Results: Hierarchical multiple regression results suggest no significant relationship between the aggregate frequency and severity measures of negative life events and cognitive performance. At the individual-event level, individuals who experienced the injury or illness of a friend during the past year and rated it as having more of an effect on their lives performed better on all three cognitive tasks. However,individuals who reported having less money to live on over the past year and rated the event as having more of an effect on their lives performed more poorly on the psychomotor speed tasks. Discussion: The findings support previous research indicating that using estimates of individual stressors rather than aggregate stress measures increases the predictive validity of stress measurement. Furthermore,the individual negative life events can have both a positive and a negative effect,which nullify one another when using the sum score of events. Keywords: cognition,aging,acute stress,negative life eventsOver the past few decades,a substantial amount of research has focused on age-related differences in cognitive functioning (for review,see Bckman,Small,& Wahlin,2001). Based on this research,it is well knownJournal of Aging and Health Volume 19 Number 4 August 2007 612-629 2007 Sage Publications 10.1177/0898264307300975 http://jah.sagepub.com hosted at http://online.sagepub.com at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 3

Rosnick et al. / Negative Life Events and Cognition613 Authors’Note: We wish to thank the Charlotte County Foundation for its generous support of this study. We also wish to thank the Healthy Aging Study staff in Charlotte County,including Dr. Barbara Sherman,Dr. Adrianna Austin,Sue Campbell,Kathy Neu,Donna Ryan,and Joyce Stathopolous. Finally,we wish to acknowledge the Charlotte County Healthy Aging Study’s community advisory council,as well as all of the participants who gave their time. Please address correspondence to Dr. Christopher B. Rosnick,Brandeis University, Department of Psychology MS 062,415 South Street,Waltham,MA 02454; e-mail:crosnick@brandeis.edu. Correspondence concerning the Charlotte County Healthy Aging Study should be addressed to Dr. James A. Mortimer,USF Department of Epidemiology and Biostatistics-MDC 56,13201 Bruce B. Downs Blvd.,Tampa,FL 33612. that deficits in multiple domains of cognitive functioning are associated with the normal aging process (Bckman et al.,2001; Zacks,Hasher,& Li, 2000). Although mean-level deficits are presented,considerable variability exists in terms of the magnitude of age-related differences in cognitive performance. Given such findings,many researchers have adopted an individualdifferences perspective in an attempt to better understand these age-related differences. For example,associations between age-related differences in cognition and leisure activities (Albert et al.,1995; Yaffe,Barnes,Nevitt,Lui,& Covinsky,2001),genetics (Bretsky,Guralnik,Launer,Albert,& Seeman, 2003; Farrer et al.,1997; Small,Rosnick,Fratiglioni,& Bckman,2004), health (Anstey,Lord,& Williams,1997; Rosnick,Small,Borenstein,& Mortimer,2004),and demographics (Herlitz,Nilsson,& Bckman,1997; Zelinski & Burnight,1997) have been reported. In the current study,another class of individual-differences variables,negative life events,was examined in relation to cognitive performance in the elderly. Acute stressors are typically measured by examining daily hassles,negative life events,or major life events; past research suggests that several background characteristics are associated with each type of acute stressor. Nacoste and Wise (1991) reported that younger and older adults are more affected by negative life events as compared to middle-aged adults. In addition,personality characteristics such as neuroticism appear to be related to the frequency and the severity rating of daily hassles. Using a sample of 358 participants ranging in age from 20 to 62,De Jong,van Sonderen,and Emmelkamp (1999) found that individuals scoring higher in neuroticism reported experiencing more stress. Furthermore,Russell and Davey (1993) observed that in a sample of college students,those who scored higher in trait anxiety and worrying reported more daily hassles and rated the hassles as being more severe. Also,acute stressors affect males and females differently. Women are more likely to report major life crises (Willis,Thomas,Garry,& Goodwin, 1987) and more frequent hassles as compared to men (Flannery,1986). Last, at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 4

research suggests that reporting more acute stressors is related to poor psychological well-being (Beasley,Thompson,& Davidson,2003; Brilman & Johan,2001; Carmack,Boudreaux,Amaral-Melendez,Brantley,& de Moor, 1999; De Jong et al.,1999) and physical health (Brand,Hanson,& Godaert, 2000; Leserman,Zhiming,Yuming,& Drossman,1998; Lutgendorf,Reimer, Schlechte,& Rubenstein,2001) and that this relationship appears to be moderated by religiosity (for review,see Ano & Vasconcelles,2005) and an individual’s social support system (i.e.,Krause,2004). Relatively little information is available on how negative life events may affect cognitive performance in the elderly. The literature that has examined self-reported stressors and cognitive performance reveals mixed results. One study examined whether the amount of stress as measured by the Schedule of Recent Events predicts intellectual decline (Sands,1981). Differences between the sum score of stressors and the individual stressors were also examined. The sample comprised 112 women older than 65 (mean age = 76.8; range = 65) living independently in the community. Intelligence was measured by four subtests from the Weschler Adult Intelligence Scale (WAIS): vocabulary,comprehension,block design,and object assembly. A ratio was developed by dividing the sum of vocabulary and comprehension (crystallized intelligence) by the sum of block design and object assembly (fluid intelligence) to estimate intellectual decline. The results revealed that the stress sum score that was created with values assigned to the events (e.g.,turning 65 was assigned a stress value of 400) was not associated with the WAIS subtests or the intelligence ratio. However,the stress sum score that was based on selfreported stressfulness of the events was negatively associated with block design and positively associated with the intelligence ratio (indicating more decline). At the individual stressor level,individuals who reported positive events (e.g.,vacations) during the last 2 years showed increased intellectual performance,whereas those who reported negative changes (e.g.,changes in the health of a family member or personal health) over the last 2 years experienced greater intellectual decline. Similarly,Amster and Krauss (1974) found that women who declined mentally over a 5-year period experienced many more crises and higher levels of stress,as indexed by the Geriatric Social Readjustment Questionnaire. However,Grimby and Berg (1995) did not find an association between the number of stressful events reported and cognitive decline,although they did find an association between men who were bereaved during the previous 6 years and cognitive decline as compared to men who did not experience any major life events. In recent work from the Baltimore Epidemiologic Catchment Area study (Saczynski,Rebok,& Holtzman,2002),researchers found that participants 614Journal of Aging and Health at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 5

Rosnick et al. / Negative Life Events and Cognition615 who reported more stressful life events performed better on a task of delayed recall. There was no association between stressful life events and Mini-Mental State Examination scores or between immediate and recognition memory. The authors also examined the relation between individual life events and cognitive performance and reported that retirement and having experienced an injury or illness over the past year were related to better memory performance. However,experiencing the loss of a loved one in the past year was associated with poorer memory performance. The previous findings suggest the utility of examining not only the sum of stressors but also the individual stressors themselves and their associations with cognitive functioning. Total scores may obscure the relationship between individual stressors and cognitive performance (Sands,1981). In summary,life events appear to have both a positive and a negative effect on cognitive performance. In addition,the effect of stressful events on cognitive functioning appears to depend on the length of time that the stressor is endured. Transient or acute stress may be beneficial (Kim & Diamond,2002), whereas chronic stress can have detrimental effects on cognitive performance (for reviews,see McEwen & Sapolsky,1995; Sapolsky,2000a,2000b). In the present study,we examined the association between negative life events and cognitive performance in a population-based sample of older adults. The literature examining the effects of negative life events on cognitive performance is limited,especially within elderly populations. The current literature focuses on the effects of negative life events on physical and emotional health. Furthermore,when assessing negative life events,researchers typically use an aggregate measure as the predictor variable. In the current analyses, we examined the occurrence and severity rating of negative life events in the aggregate and individually. Because previous research has indicated that life events may have positive or negative effects on cognitive performance,the use of the aggregate measures may obscure potential relationships to cognitive performance because of opposing effects. Finally,the analysis was derived from a population-based study of older adults,increasing the generalizability of the results.MethodParticipantsData from the Charlotte County Healthy Aging Study,a population-based sample of older adults,were utilized (for more information on the data collection,see Small et al.,2000). Two census tracts were selected for study. The goal was to recruit 504 persons aged 60 to 84 years. In each census tract,126 at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 6

616Journal of Aging and Health persons were between the ages of 60 and 74 and the other 126 between the ages of 75 and 84. Congregate living and long-term care facilities were not included in the sampling frame. The total number of persons completing the study was 466. For the current analysis,428 persons (212 men and 216 women) were examined for whom complete data on the measures of relevance were available.MeasuresCognitive Performance The measures of cognitive performance were chosen to examine several broad domains of functioning,including episodic memory,psychomotor speed,and attention. Episodic memory. The episodic memory domain was indexed by a modified Hopkins Verbal Learning Test (Benedict,Schretlen,Groninger,& Brandt,1998; Brandt,1991). There were four measures of memory performance derived from this test:immediate recall from the first three learning trials,delayed free recall,cued recall,and a discrimination score corrected for guessing. Psychomotor speed. The psychomotor speed domain was assessed by the Trailmaking Test,Parts A and B (Reitan & Wolfson,1985). This test was administered according to standard procedures. The primary outcome measure was time taken to complete each part. Higher scores indicate poorer performance. Attention. The attention domain was indexed with the Stroop Test (Stroop, 1935),including color,word,and discrimination trials. This task was administered according to standard procedures. The primary outcome measure was the number of items correctly identified for each task. Given the lack of information regarding the association between negative life events and cognitive performance in the elderly,we standardized all of the cognitive tasks and combined them to create three standardized cognitive variables:episodic memory,psychomotor speed,and attention. Negative Life Events Negative life events were measured by a subset of items from the Louisville Older Persons Events Schedule (LOPES; Murrell & Norris,1984; Murrell, at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 7

Rosnick et al. / Negative Life Events and Cognition617 Norris,& Hutchins,1984). The full measure assesses 54 negative life events, and participants were asked if each event was positive or negative. The individual events utilized in the present study were rated negative by 90% of the participants. This is a similar method to that of Owen and colleagues (2002), although they used an 80% threshold. Using a 90% threshold,we utilized 24 items from the overall measure. The LOPES was designed to be used with an older population,and it includes such questions as “Have you had a good friend die in the past 12 months?”“Other than the loss of a job,has something happened in the last 12 months to give you much less money to live on?”and “Did any of your children have a new problem with money in the past 12 months?” (Murrell & Norris,1984; Murrell et al.,1984). Participants are asked to indicate which of the 24 items occurred during the past 12 months (0 = no,1 = yes) and to rate the effect that the event had on their lives (1 = no effect,2 = slight effect ,3 = moderate effect ,4 = strong effect ). Two aggregate variables were created from the LOPES:frequency,number of negative life events experienced; and cumulative severity,created by summing the severity of the events and dividing by 24. If the event did not occur,it was coded as having no effect. Background Characteristics Demographic information included age (in years),gender (0 = male,1 = female),education (in years),marital status (1 = not married,2 = married), and neuroticism. Neuroticism was measured with the NEO Five Factor Inventory (Costa & McCrae,1989). The neuroticism scale was utilized because it is associated with the frequency and severity of daily hassles and stress (De Jong et al.,1999; Russell & Davey,1993).Statistical AnalysisCorrelation analyses were performed to examine the bivariate associations among demographic characteristics,the aggregate frequency and severity of life event measures,and cognitive performance. These analyses were followed by hierarchical multiple regressions to control for possible covariates.1The first block comprised background characteristics,and the second block consisted of the aggregate life event measures. Separate models were run for the aggregate frequency and severity measures and each of the cognitive outcomes. at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 8

618Journal of Aging and Health For the individual negative life event analyses,correlation analyses were performed among the 24 LOPES items and the cognitive variables to determine associations among the independent and cognitive variables. The results of these analyses were used for the selection of predictor variables in the individual regression analyses (see Dixon & Hultsch,1983; Rosnick & Reynolds, 2003). Similar to the analyses with the aggregate measures,hierarchical multiple regressions were performed with the background characteristics in the first block and with the significant individual life events entered in the second block. Separate models were run for the individual frequency and severity measures.ResultsBackground CharacteristicsTable 1 provides the means and standard deviations for the demographic characteristics and cognitive performance. On average,respondents were in their early 70s,had almost 2 years of college education,and experienced approximately four negative life events over the past year with an average severity rating of 1.26,which indicates that the majority of the events had little to no effect on the participants’lives.Frequency and Cumulative EffectThe correlation analyses between the aggregate frequency score and cognitive performance revealed that individuals who reported more negative life events recalled a greater number of words on the episodic memory tasks. Similarly,individuals who had higher average severity ratings performed better on the attention tasks. These associations were no longer significant in the regression analyses,after controlling for age,gender,education,marital status,and neuroticism (results not shown).Individual Events and EffectsCorrelation analyses were performed with the three cognitive outcome measures and the 24 individual LOPES items to determine whether individual items were associated with cognitive performance. Because of the small sample,those LOPES items that were significantly correlated with cognitive performance were the ones included in the regression models (see Table 2). at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 9

Rosnick et al. / Negative Life Events and Cognition619 Table 1 Means and Standard Deviations for the Background Characteristics, Cognitive Performance,and Negative Life Events MS D Range Background characteristics Age 72.96 6.21 60 Female 50.50% Education 13.98 2.88 4 Married 77.80% Neuroticism 15.15 6.93 0 Cognitive performance Immediate recall 6.78 1.75 2.00.67 Delayed recall 7.60 2.73 0 Cued recall 8.53 2.39 1 Discrimination index 9.60 1.97 2 Trailmaking Test–Part A43.25 16.93 16.16.00 Trailmaking Test–Part B117.58 63.37 36 Stroop–Color 58.39 12.72 20 Stroop–Word 87.91 15.34 20 Stroop–Discrimination 28.34 9.30 5 Negative life events Frequency 3.69 2.35 0 Severitya1.26 0.23 1.0.3 a. 1 = no effect,2 = slight effect,3 = moderate effect ,4 = strong effect . This allowed us to maintain an acceptable predictor-variable-to-subject ratio. Hierarchical multiple regressions were independently conducted for the three cognitive variables,with the demographic and personality covariates (age, gender,education,marital status,and neuroticism) entered in the first step and the individual frequency measures entered as the second block. The same analyses were performed with the severity measures. Episodic Memory The results of the regression for the associations between individual events and severity measures and episodic memory are presented in Table 3. The demographic characteristics contributed a statistically significant amount of variance to cognitive performance in both models (16.7%). At the individual variable level,higher age,being male,having fewer years of education, and higher neuroticism scores were associated with recalling fewer words. at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 10

620Journal of Aging and Health Table 2 Correlation Coefficients Between Individual Negative Life Events and Cognitive Measures Episodic MemoryPsychomotor SpeedAttention Good friend died Occurrence.029–.062.059 Severity ratinga.016 .013 .072 Injury/illness of friend Occurrence .139** –.116* .114* Severity rating .147** –.137** .162*** New injury Occurrence –.011 –.008 –.042 Severity rating –.036 .012 –.065 Major home problem Occurrence .023 –.061 .069 Severity rating –.049 .002 .052 Less money to live on Occurrence –.008 .120* –.066 Severity rating –.037 .123* –.088 Spouse had injury Occurrence .076 –.091 .105* Severity rating .087 –.085 .122* Kids with money problems Occurrence –.061 –.035 .023 Severity rating –.044 –.002 .001 Go to hospital Occurrence .108* –.084 .045 Severity rating .115* –.068 .084 Conflict with family Occurrence .112* –.032 .092 Severity rating .115* –.071 .139** Parents injury Occurrence .054 –.143** .175*** Severity rating .043 –.133** .165*** Sibling injury Occurrence .040 –.018 .021 Severity rating –.012 –.010 .019 Crime committed on you Occurrence .085 .062 –.016 Severity rating .044 .137**–.043 Lost your pet Occurrence .040 –.040 .064 Severity rating .010 –.023 .054 (continued) at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 11

Rosnick et al. / Negative Life Events and Cognition621 Table 2 (continued) Episodic MemoryPsychomotor SpeedAttention Problem in marriage Occurrence.007–.020–.005 Severity rating.018–.016.004 Kids with new injury Occurrence.102*–.011–.020 Severity rating.104*.006.002 Friend relocated Occurrence–.036.103*–.085 Severity rating–.012.096*–.081 Someone committed suicide Occurrence.000–.028.009 Severity rating.021–.038.010 Brother/sister died Occurrence–.095*.004.012 Severity rating–.073–.025.074 Child died Occurrence–.042.051–.054 Severity rating–.042.051–.054 Parent died Occurrence–.002–.087.036 Severity rating–.007–.087.037 Spouse died Occurrence.014.062–.038 Severity rating.023.043–.028 More responsibility Occurrence–.003–.034.059 Severity rating–.002–.025.067 Stop all church activities Occurrence.011.078–.062 Severity rating.000.060–.093 Stop recreation activities Occurrence.102*–.022.015 Severity rating.112*–.045.031 a. 1 = no effect,2 = slight effect,3 = moderate effect ,4 = strong effect . *p < .05. **p < .01. ***p < .001. Furthermore,both the blocks for frequency and severity measures were statistically significant,accounting for 4.2% and 3.1% of the variance in episodic memory performance,respectively. Experiencing the injury or illness of a friend over the past year and rating the experience as having more of an effect on one’s life was associated with recalling a greater number of words. at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 12

622Journal of Aging and Health Table 3 Hierarchical Multiple Regression on the Association Between the Individual Negative Life Events and Severity Measures and Episodic MemoryModel 1 Model 2 Independent Variable Background CharacteristicsFrequencySeverity Age –.192*** Gender .292*** Education .163*** Marital status .014 Neuroticism score –.149*** Injury/illness of friend .110*.103* Hospitalization .049.023 Conflict with family .065.056 Kids with new injury .083.060 Brother/sister died –.063— Stopped recreational activities .065.077 R2.167 .209.198 R2change .167 .042.031 Significant R2change p < .001 p < .001 p < .01 Note:The dash (—) indicates that the variable was not included in the model. *p < .05. **p < .01. ***p < .001. Psychomotor Speed Table 4 provides the results of the regression for the association between the frequency and severity measures and psychomotor speed. The demographic characteristics contributed 22.2% of the variance in cognitive performance. At the individual variable level,higher age,having fewer years of education,not being married,and higher neuroticism scores were associated with increased times to finish the task. Again,both the blocks for frequency and severity measures were significant (accounting for 2.3% and 4.8% of variance,respectively). Experiencing the injury or illness of a friend over the past year and rating it as having more of an effect on one’s life was associated with faster times on the psychomotor speed tasks. However,having less money to live on over the past year and rating it as having more of an effect on one’s life was associated with slower times. Similarly,individuals who rated having a crime committed against them as having more of an effect on their lives took more time to complete the psychomotor speed tasks. at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 13

Rosnick et al. / Negative Life Events and Cognition623 Table 4 Hierarchical Multiple Regression on the Association Between the Individual Negative Life Events and Severity Measures and Psychomotor SpeedModel 1 Model 2 Independent VariableBackground CharacteristicsFrequencySeverity Age.299*** Gender–.085 Education–.202*** Marital status–.098* Neuroticism score.193*** Injury/illness of friend –.101*–.120** Less money to live on.100*.088* Parents injured–.028–.029 Crime committed on you—.148*** Friend relocated.065.054 R2.222 .245.269 R2change .222 .023.048 Significant R2change p < .001 p < .05 p < .001 Note:The dash (—) indicates that the variable was not included in the model. *p < .05. **p < .01. ***p < .001. Attention The results of the regression for the association between the frequency and severity measures and performance on the attention tasks are shown in Table 5. The demographic characteristics contributed a statistically significant amount of variance to performance (30.6%). At the individual-variable level,higher age,being male,having fewer years of education,not being married,and higher neuroticism scores were associated with fewer correct answers. Although the block of frequency measures was not statistically significant,experiencing the injury or illness of a friend was positively associated with performance on the attention tasks. The block of severity measures was statistically significant,accounting for 2.4% of the variance in performance on the attention tasks. Rating the injury or illness of a friend over the past year as having more of an effect on one’s life was associated with more correct answers as compared to those of individuals who rated it as having less of an effect. at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 14

624Journal of Aging and Health DiscussionThe goal of the present analysis was to examine the associations between negative life events,in the aggregate and individually,and cognitive performance in a population-based sample of older adults. The strengths of the current study include the use of multiple measures of cognitive performance (attention,psychomotor speed,and episodic memory),the use of a measure of negative life events that was created for an elderly population,and the ability to examine the differences between the occurrence of events and the perceived effects that the events had on participants’lives in relation to cognitive performance. Similar to past research (Grimby & Berg,1995),our research was unable to find an association between the aggregate frequency and severity measures and cognitive performance after controlling for multiple background characteristics. However,there were multiple individual negative life events and effect ratings associated with all three cognitive domains under investigation. Table 5 Hierarchical Multiple Regression on the Association Between the Individual Negative Life Events and Severity Measures and AttentionModel 1 Model 2 Independent VariableBackground CharacteristicsFrequencySeverity Age–.384*** Gender.159*** Education.248*** Marital status.106* Neuroticism score–.160*** Injury/illness of friend .087*.121** Spouse injured.052.047 Conflict with family—.064 Parents injured.035.035 R2.306 .317.330 R2change .306 .012.024 Significant R2change p < .001 p = .066 p < .01 Note:The dash (—) indicates that the variable was not included in the model. *p < .05. **p < .01. ***p < .001. at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 15

The most robust finding was that individuals who experienced the injury or illness of a friend during the past year and rated it as having more of an effect on their lives performed better on all three cognitive tasks. One possible explanation is that individuals who experienced an injury or illness of a friend during the last year may change their lifestyles and may be motivated to increase their physical activity to maintain their own physical/cognitive health. In support of a relationship between health and cognitive performance, Anstey and Christensen (2000) suggested that exercise might have an indirect effect on cognition through lowering blood pressure. Furthermore,Yaffe and colleagues (2001) observed that women who reported more physical activity and expended more calories over a 6to 8-year period experienced less cognitive decline. In contrast to the findings pertaining to experiencing the injury or illness of a friend,individuals who reported having less money to live on over the past year and individuals who rated having a crime committed against them as having more of an effect on their lives took more time to complete the psychomotor speed tasks. One possible explanation for these findings is Wegner’s theory of mental control (1994). The basic premise is that people wish to control their mental activities by suppressing unwanted thoughts. By suppressing unwanted thoughts,such persons are thereby utilizing cognitive resources that could be used for the tasks at hand. This theory appears to fit well with the current findings,given that the negative effects were restricted to the psychomotor speed tasks (see also,Klein & Boals,2001). Alternatively,the positive effect of experiencing and rating the injury or illness of a friend as having more of an effect on one’s life,the negative effect of having less money to live on over the last year,and rating having a crime committed against oneself as having more of an effect on one’s life could all be explained by the inverted-U function of stress/arousal and performance, proposed by Yerkes and Dodson (1908). The Yerkes–Dodson law posits that there is an optimal level of stress or arousal where people perform their best. If there is a lack of arousal or too much arousal,individuals perform poorly. With regard to the current results,experiencing and perceiving the injury or illness of a friend as having more of an effect on one’s life may be sufficient stress for individuals to perform optimally. However,having less money to live on and perceiving having a crime committed against oneself as having a strong effect on one’s life may be too much stress and thus explain why individuals are performing worse. Our findings support the statement by Sands (1981) that using estimates of individual stressors rather than aggregate measures increases the predictive validity of stress measurement. We found that none of the aggregate Rosnick et al. / Negative Life Events and Cognition625 at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 16

626Journal of Aging and Health life event measures were significantly associated with cognitive performance, whereas multiple individual life event measures were significantly related to cognitive functioning in older adults. This may be due to the fact that the sum scores comprise many different life events that affect people’s lives in many different ways. Also,the individual life events can have both a positive and a negative effect,which nullify each other when using the sum score of events. Several limitations to this study should be noted. First,the current sample of older adults reported experiencing approximately 4 negative life events over the past year,out of a possible 24. However,despite the limited variability in life events reported,there were significant amounts of variance explained by multiple life events. Second,self-reports of experiencing negative life events over a year may be subject to recall bias. In addition,these life events may occur multiple times throughout a year,but the LOPES responses are in a yes/no format. For example,experiencing the injury or illness of a friend may occur multiple times among multiple friends. It may be the accumulation of single events instead of the accumulation of multiple events that is driving the effect on cognitive performance. Also,we were unable to assess whether the events were chronic or acute episodes. Future research needs to address these issues. Finally,the data used in this analysis are crosssectional,so we are therefore unable to determine the direction of the associations between the variables under study. For example,it is entirely possible that individuals with lower attention scores are more likely to experience a financial loss when compared to individuals with higher attention scores. Longitudinal follow-up of participants from the Charlotte County Healthy Aging Study should prove to be valuable as participants begin to experience decrements in cognitive functioning.Note1. We reanalyzed the data,including depression and religiosity as covariates. For the latter, we used a composite score that assessed the frequency of participation in religious activities. The results were the same as those of the original models,with the exception of a nonsignificant association between the reported severity of having less money to live on during the past 12 months and psychomotor speed,when controlling for depressive symptoms (results not shown).ReferencesAlbert,M. S.,Jones,K.,Savage,C. R.,Berkman,L.,Seeman,T.,Blazer,D.,et al. (1995). Predictors of cognitive change in older persons:MacArthur studies of successful aging. Psychology and Aging,10 ,578-589. at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 17

Rosnick et al. / Negative Life Events and Cognition627 Amster,L. E.,& Krauss,H. H. (1974). The relationship between life crises and mental deterioration in old age. International Journal of Aging and Human Development , 5,51-55. Ano,G. G.,& Vasconcelles,E. B. (2005). Religious coping and psychological adjustment to stress:A meta-analysis. Journal of Clinical Psychology , 61,461-480. Anstey,K.,& Christensen,H. (2000). Education,activity,health,blood pressure,and apolipoprotein E as predictors of cognitive change in old age:A review. Journal of Gerontology: Behavioural Science Section, 46,163-177. Anstey,K. J.,Lord,S. R.,& Williams,P. (1997). Strength in the lower limbs,visual contrast sensitivity,and simple reaction time predict cognition in older woman. Psychology and Aging, 12,137-144. Bckman,L.,Small,B. J.,& Wahlin,A. (2001). Aging and memory:Cognitive and biological perspectives. In J. E. Birren & K. W.Schaie (Eds. ), Handbook of the psychology of aging (5th ed.,pp. 349-377).San Diego,CA:Academic Press . Beasley,M.,Thompson,T.,& Davidson,J. (2003). Resilience in response to life stress:The effects of coping style and cognitive hardiness. Personality and Individual Differences, 34,77-95. Benedict,R. H. B.,Schretlen,D.,Groninger,L.,& Brandt,J. (1998). Hopkins Verbal Learning Test–Revised:Normative data and analysis of inter-form and test–retest reliability. Clinical Neuropsychologist, 12,43-55. Brand,N.,Hanson,E.,& Godaert,G. (2000). Chronic stress affects blood pressure and speed of short-term memory. Perceptual and Motor Skills , 91,291-298. Brandt,J. (1991). The Hopkins Verbal Learning Test:Development of a new memory test with six equivalent forms. Clinical Neuropsychologist, 5,125-142. Bretsky,P.,Guralnik,J. M.,Launer,L.,Albert,M.,& Seeman,T. E. (2003). The role of APOE_4 in longitudinal cognitive decline:MacArthur studies of successful aging. Neurology, 60, 1077-1081. Brilman,E. I.,& Johan,O. (2001). Life events,difficulties,and onset of depressive episodes in later life. Psychological Medicine, 31,859-869. Carmack,C. L.,Boudreaux,E.,Amaral-Melendez,M.,Brantley,P. J.,& de Moor,C. (1999). Aerobic fitness and leisure physical activity as moderators of the stress–illness relation. Annals of Behavioral Medicine, 21,251-257. Costa,P. T.,& McCrae,R. R. (1989). NEO PI/FFI manual supplement . Odessa,FL: Psychological Assessment Resources. De Jong,G. M.,van Sonderen,E.,& Emmelkamp,P. M. G. (1999). A comprehensive model of stress:The roles of experienced stress and neuroticism in explaining the stress–distress relationship. Psychotherapy and Psychosomatics , 68,290-298. Dixon,R. A.,& Hultsch,D. F. (1983). Metamemory and memory for text relationships in adulthood:A cross-validation study. Journal of Gerontology , 38,689-694. Farrer,L. A.,Cupples,L. A.,Haines,J. L.,Hyman,B. T.,Kukull,W. A.,Mayeux,R.,et al. (1997). Effects of age,sex,and ethnicity between apolipoprotein E genotype and Alzheimer’s disease: A meta-analysis. Journal of the American Medical Association , 278,1349-1356. Flannery,R. B.,Jr. (1986). Major life events and daily hassles in predicting health status: Methodological inquiry. Journal of Clinical Psychology , 42,485-487. Grimby,A.,& Berg,S. (1995). Stressful life events and cognitive functioning in late life. Aging, Clinical,and Experimental Research, 7 ,35-39. Herlitz,A.,Nilsson,L.,& Backman,L. (1997). Gender differences in episodic memory. Memory and Cognition , 25,801-811. at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 18

Kim,J. J.,& Diamond,D. M. (2002). The stressed hippocampus,synaptic plasticity,and lost memories. Nature Reviews:Neuroscience, 3,453-462. Klein,K.,& Boals,A. (2001). The relationship of life event stress and working memory capacity. Applied Cognitive Psychology, 15,565-579. Krause,N. (2004). Stressors arising in highly valued roles,meaning in life,and the physical status of older adults. Journal of Gerontology:Social Sciences,59 ,S287-S297. Leserman,J.,Zhiming,L.,Yuming,H.,& Drossman,D. A. (1998). How multiple types of stressors impact on health. Psychosomatic Medicine, 60,175-181. Lutgendorf,S. K.,Reimer,T. T.,Schlechte,J.,& Rubenstein,L. M. (2001). Illness episodes and cortisol in healthy older adults during a life transition. Annals of Behavioral Medicine, 23,166-176. McEwen,B. S.,& Sapolsky,R. M. (1995). Stress and cognitive function. Current Opinion in Neurobiology, 5,205-216. Murrell,S. A.,& Norris,F. H. (1984). Resources,life events,and changes in positive affect and depression in older adults. American Journal of Community Psychology , 12,445-464. Murrell,S. A.,Norris,F. H.,& Hutchins,G. L. (1984). Distribution and desirability of life events in older adults:Population and policy implications. Journal of Community Psychology , 12, 301-311. Nacoste,D. R. B.,& Wise,E. H. (1991). The relationship among negative life events,cognitions, and depression within three generations. The Gerontologist,31,397-403. Owen,J. E.,Roth,D. L.,Stevens,A. B.,McCarty,H. J.,Clay,O. J.,Wadley,V. G.,et al. (2002). Association of life events and psychological distress in family caregivers of dementia patients. Aging and Mental Health,6 ,62-71. Reitan,R. M.,& Wolfson,D. (1985). The Halstead–Reitan Neuropsychological Test Battery. Tucson,AZ:Neuropsychology Press. Rosnick,C. B.,& Reynolds,S. L. (2003). Thinking ahead:Factors associated with executing advance directives. Journal of Aging and Health , 15,409-429. Rosnick,C. B.,Small,B. J.,Borenstein,A.,& Mortimer,J. A. (2004). The association between health and cognitive performance in a population-based study of older adults:The Charlotte County Healthy Aging Study (CCHAS). Aging,Neuropsychology,and Cognition , 11,89-99. Russell,M.,& Davey,G. C. (1993). The relationship between life event measures and anxiety and its cognitive correlates. Personality and Individual Differences , 14,317-322. Saczynski,J. S.,Rebok,G. W.,& Holtzman,R. E. (2002,November). Stressful life events and cognitive functioning in mid and late life. Paper presented at the 55th annual meeting of the Gerontological Society of America,Boston. Sands,J. D. (1981). The relationship of stressful life events to intellectual functioning in women over 65. International Journal of Aging and Human Development , 14,11-22. Sapolsky,R. M. (2000a). Glucocorticoids and hippocampal atrophy in neuropsychiatric disorders. Archives of General Psychiatry, 57,925-935. Sapolsky,R. M. (2000b). The possibility of neurotoxicity in the hippocampus in major depression:A primer on neuron death. Biological Psychiatry, 48,755-765. Small,B. J.,Graves,A. B.,McEvoy,C. L.,Crawford,F. C.,Mullan,M.,& Mortimer,J. A. (2000). Is APOE_4 a risk factor for cognitive impairment in normal aging? Neurology, 54, 2082-2088. Small, B. J.,Rosnick,C. B.,Fratiglioni,L.,& Backman,L. (2004). Apolipoprotein E and cognitive performance:A meta-analysis. Psychology and Aging , 19,592-600. Stroop,J. R. (1935). Studies of interference in serial verbal reaction. Journal of Experimental Psychology, 18,643-662. 628Journal of Aging and Health at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from

PAGE 19

Wegner,D. (1994). Ironic processes of mental control. Psychological Review, 101,34-52. Willis,L.,Thomas,P.,Garry,P. J.,& Goodwin,J. S. (1987). A prospective study of response to stressful life events in initially healthy elders. Journal of Gerontology , 42,627-630. Yaffe,K.,Barnes,D.,Nevitt,M.,Lui,L.,& Covinsky,K. (2001). A prospective study of physical activity and cognitive decline in elderly women. Archives of Internal Medicine, 161, 1703-1708. Yerkes,R. M.,& Dodson,J. D. (1908). The relation of strength of stimulus to rapidity of habitformation. Journal ofComparative Neurology and Psychology , 18,459-482. Zacks,R. T.,Hasher,L.,& Li,K. Z. H. (2000). Human memory. In F. I. M. Craik & T. A. Salthouse (Eds.), The handbook of aging and cognition (pp. 293-358). Mahwah,NJ: Lawrence Erlbaum. Zelinski,E. M.,& Burnight,K. P. (1997). Sixteen-year longitudinal and time lag changes in memory and cognition in older adults. Psychology and Aging , 12,503-513. Rosnick et al. / Negative Life Events and Cognition629 at UNIV OF FLORIDA Smathers Libraries on October 2, 2014 jah.sagepub.com Downloaded from



PAGE 1

1 ANTICHOLINERGIC MEDICATION USE AND RISK OF COGNITIVE IMPAIRMENT AND DEMENTIA IN OLDER ADULTS WITH PRECLINICAL ALZHEIMER DISEASE By SUNIL SWAMI A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PA RTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2014

PAGE 2

2 © 2014 Sunil Swami

PAGE 3

3 To my parents and sister

PAGE 4

4 ACKNOWLEDGMENTS I thank my parents and sister. I th ank Dr. Elena Andresen, ex chair, D epartment of E pidemiology and Institute on Aging , for bringing me to the University of Florida. I thank my dissertation committee chair, Dr. Todd Manini, for his mentorship, supervision and guidance. I am sure it was equa lly exciting and challenging for him to help me finish this dissertation as it was for me. I also thank my dissertation committee members, Drs. Catherine Striley, Ronald Cohen, and John Kairalla, for their encouragement, valuable recommendations, and guida nce during this enduring journey. I thank my dear friend Yi Lin for her moral and emotional support. Finally , I thank all my friends and peers who patiently supported me during this process .

PAGE 5

5 TABLE OF CONTENTS p age ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 10 ABSTRACT ................................ ................................ ................................ ................... 13 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 15 ................................ ................................ ............... 15 Back ground ................................ ................................ ................................ ...... 15 Cholinergic Hypothesis of Geriatric Memory Dysfunction ................................ . 16 ................................ ..................... 19 ................................ ........ 20 Pre ................................ ................................ ...... 21 ................................ ............................ 22 Background ................................ ................................ ................................ ...... 22 Anticholinergic Exposure Measurement Methods ................................ ............ 24 Epidemiology of Anticholinergic Use and Risk of Dementia ................................ .... 29 ................................ ....................... 31 Minimum Data Set ................................ ................................ ............................ 33 Uniform Data Set ................................ ................................ .............................. 33 Neuropsychological Database ................................ ................................ .......... 34 ............. 35 What Is New in This Study? ................................ ................................ .................... 37 New Approach ................................ ................................ ................................ .. 37 Rationale for Anticholinergic Use and Dementia ................................ .............. 38 2 ANTICHOLINERGIC DRUG USE AND RISK OF COGNIT IVE IMPAIRMENT AND DEMENTIA IN OLDER ADULTS WITH QUESTIONABLE COGNITIVE IMPAIRMENT: A CROSS SECTIONAL ANALYSIS ................................ ............... 44 Background ................................ ................................ ................................ ............. 44 Methods ................................ ................................ ................................ .................. 46 Study Participants ................................ ................................ ............................ 46 Anticholinergics Exposure ................................ ................................ ................ 47 Co gnitive Outcomes ................................ ................................ ......................... 48 Covariates ................................ ................................ ................................ ........ 49 Statistical Analysis ................................ ................................ ............................ 50

PAGE 6

6 Result s ................................ ................................ ................................ .................... 51 Discussion ................................ ................................ ................................ .............. 55 3 ANTICHOLINERGIC DRUG USE AND RISK OF COGNITIVE IMPAIRMENT AND DEMENTIA IN OLDER ADULTS WITH QUESTIONABLE CO GNITIVE IMPAIRMENT: A LONGITUDINAL STUDY ................................ ............................ 78 Background ................................ ................................ ................................ ............. 78 Methods ................................ ................................ ................................ .................. 81 Study Participants ................................ ................................ ............................ 81 Anticholinergic Exposure ................................ ................................ .................. 81 Cognitive Outcomes ................................ ................................ ......................... 82 Covariates ................................ ................................ ................................ ........ 83 Statistical Analyses ................................ ................................ .......................... 84 Results ................................ ................................ ................................ .................... 85 Di scussion ................................ ................................ ................................ .............. 88 4 LONG TERM EXPOSURE TO ANTICHOLINERGIC DRUG USE AND RISK OF COGNITIVE IMPAIRMENT AND DEMENTIA IN OLDER ADULTS WITH QUESTIONABLE COGNITIVE IMPAIRMENT: A MIXED MODEL APPROACH .. 101 Background ................................ ................................ ................................ ........... 101 Methods ................................ ................................ ................................ ................ 103 Study Participants ................................ ................................ .......................... 103 Anticholinergic Exposure ................................ ................................ ................ 104 Cognitive Outcomes ................................ ................................ ....................... 105 Covariates ................................ ................................ ................................ ...... 106 Statistical Analyses ................................ ................................ ........................ 107 Results ................................ ................................ ................................ .................. 108 Discussion ................................ ................................ ................................ ............ 111 5 CONCLUSIONS ................................ ................................ ................................ ... 142 Background ................................ ................................ ................................ ........... 142 Discussion ................................ ................................ ................................ ............ 143 Future Directions ................................ ................................ ................................ .. 146 APPENDIX A: ANTICHOLINERGIC DRUG SCALE LEVELS SAS CODE ................. 148 LIST OF REFERENCES ................................ ................................ ............................. 162 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 189

PAGE 7

7 LIST OF TABLES Table page 1 1 Staging categories of precli ................................ ............ 43 2 1 Anticholinergic Drug Scale and associated anticholinergic activity ........................ 63 2 2 Composite cognitive domains Z score ................................ ................................ ... 64 2 3 Baseline characteristics of the study population as a function of anticholinergic drug use ................................ ................................ ................................ ............. 65 2 4 Base line cognitive z score distribution of the study population as a function of anticholinergic drug use ................................ ................................ ...................... 6 7 2 5 Effects of baseline covariates and association with global cognitive scores assess ed by linear regression ................................ ................................ ............ 68 2 6 Effects of baseline covariates and association with memory scores assessed by linear regression ................................ ................................ ............................ 70 2 7 Effects of baseline covariates and association with attention scores assessed by linear regression ................................ ................................ ............................ 72 2 8 Effects of baseline covariates and association with executive function scores assessed by linear regression ................................ ................................ ............ 74 2 9 Effects of baseline covariates and association with language scores assessed by linear regression ................................ ................................ ............................ 76 3 1 Anticholinergic Drug Scale and associated anticholinergic activity ........................ 94 3 2 Method of creating specific cognitive domains from z scores of individual neuropsychological tests ................................ ................................ .................... 95 3 3 Baseline characteristics of the study population as a function of anticholinergic drug use ................................ ................................ ................................ ............. 96 3 5 Multivariable Cox prop ortional hazard regression models of time to incident dementia for global cognition ................................ ................................ .............. 99 4 1 Anticholinergic Drug Scale and associated anticholinergic activity ...................... 118 4 2 Composite cognitive domain z scores ................................ ................................ .. 119 4 3 Baseline characteristics of the study population as a function of anticholinergic drug use ................................ ................................ ................................ ........... 120

PAGE 8

8 4 4 Baseline cognitive scores of the study population as a function of anticholinergic drug use ................................ ................................ .................... 121 4 5 Multivariable linear mixed model r egression of change in global cognitive scores with time ................................ ................................ ................................ 122 4 6 Multivariable linear mixed model regression of change in memory scores with time ................................ ................................ ................................ ................... 125 4 7 Multivariable linear mixed model regression of change in attention scores with time ................................ ................................ ................................ ................... 128 4 8 Multivariable linear mixed model regression of change in executive func tion scores with time ................................ ................................ ................................ 131 4 9 Multivariable linear mixed model regression of change in language scores with time ................................ ................................ ................................ ................... 134

PAGE 9

9 LIST OF FIGURES Figure page 1 1 Conceptual framework of anticholinergic drug effects on aging and cognition. ...... 42 3 1 Kaplan Meier survival curves of dement ia free survival in older adults with questionable cognitive impairment according to anticholinergic drug use . ......... 98 4 1 Plot of adjusted estimated means of global cognitive scores over follow u p visits ................................ ................................ ................................ ................. 137 4 2 Plot of adjusted estimated means of memory scores over follow up visits ........... 138 4 3 Plot of adjusted estimated me ans of attention scores over follow up visits .......... 139 4 4 Plot of adjusted estimated means of executive function scores over follow up visits ................................ ................................ ................................ ................. 140 4 5 Plot of adjusted estimated means of language scores over follow up visits ......... 141

PAGE 10

10 LIST OF ABBREVIATIONS AC Anticholinergic ACB Anticholinergic Burden ACh Acetylcholine AChE Acetylcholinesterase AChE I Acetylcholinesterase I nhibitor AD ADC Alzheimer's Disease Centers ADS APOE4 Anticholinergic Drug Scale Apolipoprotein E4 APP Amyloid P recursor P rotein ARS Anticholinergic Risk Scale Beta A myloid P laques BNT Boston Naming Te st CAT Choline A cetyltransferase CDR Clinical Dementia Rating CDR Global Clinical Dementia Rating Global CDR SOB Clinical Dementia Rating Sum of Boxes CERAD CI Confidence Interval CNS Centra l Nervous System CSF C erebrospinal F luid CT Computed Tomography

PAGE 11

11 CTF Clinical Task Force DBI Drug Burden Index FAQ Functional Assessment Questionnaire FTD Frontal T emporal D ementia GDS Geriatric Depression Scale HR Hazard Ratio MCI M ild C ognitive I mpairment MDS Minimal Data Set MMSE Mini Mental State Exam MRI Magnetic Resonance Imaging MS Medial S eptum NACC NAPA NBM Nucleus B asalis of Meynert NFT N eurofibrillary T angle s NIA National Institute on Aging NIH National Institutes of Health NPI Q Neuropsychiatric Inventory Questionnaire NTB Neuropsychological Test Battery OR Odds Ratio PAD Preclinical AD PET Positron Emission Tomography PI Persistency Index

PAGE 12

12 QCI Quest ionable Cognitive Impairment SAA S erum A nticholinergic A ctivity SD Standard Deviation SE Standard Error SPECT Single Photon Emission Computed Tomography TM A Trail Making Test P art A TM B Trail Making Test P art B UDS Uniform Data Set VDB Vertical L imb of the Diagonal Band WAIS R Wechsler Adult Intelligence Scale Revised WMS R Wechsler Memory Scale Revised

PAGE 13

13 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requi rements for the Degree of Doctor of Philosophy ANTICHOLINERGIC MEDICATION USE AND RISK OF COGNITIVE IMPAIRMENT AND DEMENTIA IN OLDER ADULTS WITH PRECLINICAL ALZHEIMER DISEASE By Sunil Swami December 2014 Chair: Todd M. Manini Major: Epidemiology Age d ependent decline in learning and memory is associated with atrophic or degenerative changes in the brain characterized by cortical cholinergic deficits and dysfunction is also as sociated with cognitive impairment, deposition of neuritic amyloid plaques and neurofibrillary tangles in brain. Furthermore, dementia severity is associated with decrease in cholinergic neurons, cholinergic signaling, and cholinergic receptors. Older adu lts have higher prevalence of chronic disease and polypharmacy. In addition, age associated decline in central cholinergic activity makes older adults susceptible to harmful effects of anticholinergic (AC) agents. However, there is an inadequate understand ing of effects of AC drug use on progression of cognition. Recognizing individuals with very subtle clinical, cognitive , and pathological alterations years before they develop noticeable clinical signs and symptoms of P reclinical AD (PAD) and quantifying t heir anticholinergic burden might help the understand ing of the effects of anticholinergics on cognition. It is important to understand clinical and epidemiological effects of AC drug use on PAD and progression of cognitive decline.

PAGE 14

14 These cognitive impairm ents are very difficult to decipher clinically , and neuropsychological tests are especially helpful to identify individuals with questionable cognitive impairment (QCI). The c entral objective of this study was to determine the association between AC drug use and global and domain specific cognition among older adults with QCI. We hypothesize d that older adults with QCI who have high exposure to AC drugs are more likely to have lower global and domain specific cognitive scores, have a higher incidence rate of mild cognitive impairment (MCI)/AD, and have more rapid declines in global and domain specific cognition compared to older adults who have no or low exposure to AC drugs. We assessed anticholinergic load through the Anticholinergic Drug Scale (ADS) usin g temporal medication records in the National Alzheimer's Coordinating Center (NACC) database. The results showed that the o lder adults with QCI who took medications with AC properties had higher global and domain specific cognitive score as compared to those who did not take AC medication. However, the differences between the groups were not statistically significant. Furthermore, the repeated measure analysis showed that the cognitive scores were not only higher at baseline; they remained high over the follow up time. In addition, t ime to event analysis illustrates that the exposure to AC medication in older adults with QCI was associated with lower risk of progression to MCI/dementia as compared to no AC medication exposure, but the association was not statistically significant. This study provided greater understanding of the role of AC drug use and its temporal associations in cognitive impairment, which might help in reducing the public health burden of cognitive impairment and AD.

PAGE 15

15 CHAPTER 1 INTRODUC TION Background The first case of AD was described in 1907 by Alois Alzheimer in a 51 year old woman with clinical presentation of rapidly deteriorating memory and psychiatric disturbances (Alzheimer A, 1907). Since then, AD and AD associated dementia has increased at epidemic proportions , reaching the current U . S . and global burden of AD of about 5.2 million (Brookmeyer et al., 2007) and 36 million (Campbell et al., 2009), respectively. This count is projected to increase in the U . S . and globally to 15 (Brookmeyer et al., 2007 ) and 115 million (Campbell et al., 2009) by 2050, respectively. The AD associated economic burden in 2013 is estimated to be $203 billion (Uusvaara, Pitkala, Kautiainen, Tilvis, and Strandberg, 2013) . It is expected to increase up to $1.2 trillion in the U.S. alone by 2050 ( Hebert, Weuve, Scherr, & Evans, 2013 ) . Despite the best preventive measures and treatment strategies, AD is the sixth leading cause of death in the U.S ( Murphy, Xu, & Kochanek. 2013 ) . Gauss (1906) was the first to notice that anesthetic hyoscine (scopolamine) produced amnesia and that patients were unable to recall events after taking the medication (Thomson & Cotterill, 1909). Crow and Grove White (1973) examined hyoscine and found tha t it produced a significant impairment in immediate and delayed free recall. Around the same time, Davies and Maloney (1976), Bowen and colleagues (1976) , and Perry and colleagues (1977) reported decrease in choline acetyltransferase (CAT) activity in the cortex, hippocampus , patients. Previous work on postmortem brain tissue of AD patients showed that there

PAGE 16

16 was a decrease in high affinity choline uptake (Rylett et al., 1983), reduced acetylcholine (ACh) release (Nil sson et al., 1986), and weak nicotinic and muscarinic ACh receptor binding (Araujo et al., 1988; Whitehouse et al., 1988; Aubert et al., 1992; Nordberg et al., 1992; Perry et al., 1995) as compared to non pathological control brains. Perry and colleagues ( 1978) and Wilcock and colleagues (1982) further found a correlation between decrease s in CAT and the degree of cognitive impairment using premortem mental test score. Whitehouse and colleagues (1981, 1982) found selective loss in cholinergic cells count (5 0 75 percent ) and density in the Nucleus Basalis of Meynert (NBM) within the substantia innominata of cholinergic basal forebrain established an anatomical correlate with the biochemical evidence of cholinergic loss in AD. Consequently, animal model studie s also reported cholinergic cell loss in basal forebrain (Haba et al., 1988; Michalek et al., 1989; Rosene, 1993; Thal et al., 1992; Luine et al., 1986; Mesulam et al., 1987; Strong et al., 1980; Mundy et al., 1991; Voytko et al., 1995; Armstrong et al., 1 993; Fischer et al. , 1987 , 2006; Riekkinen et al., 1992). Cholinergic Hypothesis of Geriatric Memory Dysfunction Drachman and Leavitt (1974) were the first to demonstrate an association between cholinergic dysfunction and age related memory decline using S copolamine, an anticholinergic drug, among healthy young humans (Drachman and Sahakian, AD related decrease in cognitive function is associated with a decline in basal forebrain cholinergic neurotransmission. This hypothesis is supported by the fact that drugs that improve central cholinergic function also improve cognitive functions in patie nt (Bartus, 2000).

PAGE 17

17 Recent studies have supported disruptive action of scopolamine on age related memory decline and age relate decrease in cholinergic tone (Flicker et al., 1992; Molchan et al., 1992; Ray et al., 1992; Sunderland et al., 1987; Tariot et al ., 1996). Subsequent work in rodents has shown that scopolamine impairs recent memory (Andrews et al., 1994; Dodart et al., 1997; Kirkby et al., 1995; Pilcher et al., 1997; Ravel et al., 1992), reference memory (Biggan et al., 1996), conditioned memory (Se ssions et al., 1998), and conditioned learning (Flood and Cherkin, 1986). Similar work in non human primates showed that scopolamine produce s impairment in recent memory (Bartus and Johnson, 1976; Bartus, 2000), attention shifting (Davidson et al., 1999), and visual discrimination learning (Bartus and Johnson, 1976). Major support of the cholinergic hypothesis comes from studies demonstrat ing memory enhancing effects of Acetyl cholinesterase inhibitors (AChE I). Bartus (1979) and Muir and colleagues (1995) h ad shown memory enhancing effects of cholinomimetic drugs in animal models of aging and AD. Previous studies also demonstrated that the loss in CAT activity correlates with reduction in cholinergic neurons of the NBM (Henke and Lang, 1983; Bird et al., 198 3; Etienne et al., 1986). Postmortem studies have established a significant negative correlation between CAT activity and the density of senile plaques in AD brains (Perry et al., 1981; Mountjoy et al., 1984). Cholinomimetic replacement strategies have not yet attained significant benefits in alleviating cognitive deficits in AD (Gauthier, 1999, 2002; Giacobini, 2000a, b). Research by Sarter and Bruno (1998) were inconsistent in demonstrating cholinergic cell loss and its association with cognitive decline. A s imilar inconsistency was also found in the second experiment by Drachman and colleagues , in which they

PAGE 18

18 administered physostigmine to older adults (Drachman and Sahakian, 1980; Drachman, Glosser, Fleming and Longenecker, 1982). Furthermore, selective le sions of the basal forebrain cholinergic system caused impairments in nonmnemonic processes such as attention (Muir et al., 1995; Baxter et al., 1995; Voytko et al., 1994). Since CAT is not a rate limiting step in ACh synthesis, loss of CAT activity is not believed to be a causal factor of AD but only a consequence of the loss of basal forebrain cholinergic neurons (Cohen and Wurtman, 1975). These results illustrate that the understanding of changes in the central cholinergic systems and associated AD sympt oms and disease progression remains limited and controversial. In the last decade or two , researchers have emphasized understand ing the changes in various neurotransmitter systems of the AD brain and deterioration in neural pathways. Some of the important neurotransmitters of the brain implicated in AD neuropathology are acetylcholine, serotonin, noradrenaline, and dopamine. The large neurons of the basal forebrain cholinergic system extend from the medial septum (MS) through the diagonal band of Broca to t he NB M. The h ippocampus is predominantly innervated by neurons from the rostral part of MS , while the cingulate cortex is innervated by neurons in the vertical limb of the diagonal band (VDB). The n eocortex and the amygdale are predominantly innervated by t he neurons of the caudal NB M. Various studies have shown that the forebrain cholinergic neurons undergo atrophic or degenerative changes with age (Gilad et al., 1987; Fisher et al., 1987). In addition, age dependent decline in learning and memory has been associated with atrophic or degenerative changes in the AD brain (Barnes, 1990; Gallagher & Nicolle, 1993). AD cholinergic dysfunction is associated with deposition of beta amyloid plaques (A ) in the

PAGE 19

19 cerebral cortical and the hippocampal area and neurofibrillary tangles (NFT) in cytoplasm of cortical pyramidal neurons (Perry et al., 1995; Moore et al., 1995). Amyloid H D isease Several studies have shown that impairs various steps in the synthesis and release of ACh from basal forebrain cholinergic neurons (Kar et al., 1996, 1998; Hoshi 2001; for review, see Auld et al., 1998 ). It is note worthy that is also produced in the normal aging brain and therefore is not an explicit indication of neurotoxicity (Haass et al., 1992; Seubert et al., 1992). In addition, animal study have show n that higher levels of reduce hippocampal (Vaucher et al., 2001) a nd cortical synaptosomes Ach release (Wang et al., 1999) . In addition , Lee et al. (2001a) also reported induced reduction of ACh release from rat hippocampus. This suggests that endogenous might contribute to cognitive impairment in older population . Higher levels of have been shown to reduce A C h synthesis and release by reducing high affinity choline uptake in basal forebrain cholinergic neurons (Galdzicki et al., 1994; Allen et al., 1997) and by inhibiting the accumulation of internalized hippoc ampal and cortical choline (Kar et al., 1998). This might limit the neuronal access to choline, which might induce autocannabilism of cholinergic neurons to meet the demand of Ach (Nitsch et al., 1992). These indications suggest that the reduction in high affinity choline uptake is the leading mechanism of induced inhibition of ACh synthesis and release. Chronic accumulation also is believed to impair the hippocampal long term potentiation and to reduce ACh release (Cullen et al., 1997; Chen et al., 2000; Walsh et al., 2002). also has been found to reduce the ACh release from the cholinergic nerve endings (Kar et al., 1996, 1998; Wang et al., 1999;

PAGE 20

20 Lee et al., 2001a; Satoh et al., 2001; Vaucher et al., 2001). deposition impairs muscarinic receptor signaling (Kelly et al., 1996; Huang et al., 2000) and reduces nicotinic contributing to cognitive impairment (Liu et al., 2001; Pettit et al., 2001) . Tau and The tau and tangle hypothesis suggest that tangles, microtubule associated protein tau, are central to AD pathogenesis. Tangles are located in the axon of a neuron but may become highly phosphorylated and aggregate into abnormal filaments in the cell body. In AD, tau can no longer stabilize microtubules , and phosporylation of tau reduces binding to microtubule s (Lovestone & Reynolds, 1997). The tau proteins undergo structural modifications and aggregate in microtubules (Mena et al., 1996) and finally replace them by tangles (Gray E. G., 1987). Previous studies have shown an association between burden of tangles and cognitive impairment (Nagy et al., 1995). In addition, some researchers have found association between tau aggregation and amyloid deposits (Naslund et al., 2000). Neuropathological evaluations have shown increased phosphorylation and aggregation of t angles in the hippocampus and the neocortex (Braak & Braak, 1998; Braak et al., 1994). The tau hypothesis was finally validated with the discoveries of mut ations in tau protein in frontal temporal dementia (FTD) (Heutink P., 2000). Subsequently , various mi ssense mutations were found to be associated with a reduction in microtubule binding (Arrasate et al., 1999; Nacharaju et al., 1999) and tau aggregation (Rizzu et al., 2000; Perez et al., 2000). A nimal models show that despite tangles and plaques being rel ated pathogenetically, tau has an independent pathology (Lewis et al., 2001, Gotz et al., 2001). Furthermore, over expression of tau stimulates aggregation of phosphorylated tau and AD like neurofibrillary tangles (Ishihara et al., 1999, 2001) in animal st udies. In vitro studies

PAGE 21

21 have shown that nicotinic receptor activation increases tau phosphorylation (Wang et al., 2003) ; however, the mechanism remains unclear (Rubio et al., 2006). Interestingly, acetylcholine M1 muscarinic receptor activation decreases t au phosphorylation and destabilize microtubule (Fisher A, 2000). Pre D isease had primarily been used in reference to the clinical syndrome of progressive cognitive impairment, behavioral modifications and n europathological changes in dementia. However, the time lag between underlying pathophysiological processes which start years before the clinically evident impairment can be conceptualized as a continuum or trajectories of AD. To facilitate possib le early detection, diagnosis , and management of pathological process, it is important to characterize AD based upon the underlying pathophysiological disease process and the clinical stages of the illness, which may evolve in parallel or dissimilar trajectories of cognitive decline (Dubois et al., 2010). In addition, individuals with biomarker evidence of a worsening pathophysiological process are at increased risk for developing cognitive impairment and progression to AD dementia. However, the extent to which path ological biomarkers can predict the cognitive course of a cognitively normal individual remains ambiguous. Therefore, it is critical to determine factors that affect cognition, define PAD, detect biomarkers associated with PAD, and predict emergence of mil d cognitive impairment ( MCI ) and development of AD dementia. The current National Institute of Health ( NIH ) guidelines have defined PAD as having the three stages as shown in Table 1 1 .

PAGE 22

22 Background Cholinergic neur ons in the brain under the action of the synthetic enzyme choline acetyltransferase (CAT) combine choline and the cofactor acetyl coenzyme A to produce ACh. ACh is packed in synaptic vesicles that fuse with the plasma membrane and release ACh into the syna ptic cleft. ACh interacts with postsynaptic cholinergic receptors and transform signal s into cellular function, either by modifying ion flux or by producing intracellular second messengers. Acetylcholinesterase ( AChE ) promptly hydrolyzes ACh. Anticholinerg ic drugs occupy receptor sites at parasympathetic nerve endings, thereby leaving fewer receptor sites free to respond to acetylcholine , leading to decrease d parasympathetic. The parasympathetic response depends upon the number of receptors blocked by antic holinergic drugs and baseline parasympathetic activity. The cholinergic inhibition by anticholinergics reduces cortical cholinergic innervations , which reduce s ACh coupling of muscarinic M1 receptors to the second messenger system. Cholinergic inhibition a lso phosphorylates tau proteins to form neurofibrillary tangles. In addition, decreased cholinergic stimulation reduces secretion of soluble amyloid precursor protein (APP), increase s amyloid production and decrease s glutamate production. P revious studies have shown that cholinergic blockade produces memory impairment, especially that of episodic memory , in recall of verbal and visuo spatial materials and retrieval deficit (Davis, 1978). Previous studies showed that higher Persistency Index (PI), which is ye ars of drug use divided by years of observation following AD diagnosis, is associated with a rapid decline in the Clinical Dementia

PAGE 23

23 Rating (CDR) scale (Rosenberg et al., 2012) and annual decline of 3.5 points on Mini Mental State Exam ( MMSE ) scores (Burns et al., 1991). AChE I bind to AChE molecules and block ACh degradation , leading to ACh accumulation in the synaptic cleft, which leads to persistent stimulation of cholinergic receptors , altering the cholinergic receptor mediated signaling pathway. These c ellular changes later translate into functional changes in tissues and organs. Many AChE I , including Tacrine, D onepezil , Galant amine , and R ivastigmine , are currently used for treatment of AD. The central therapeutic strategy is to block ACh degradation b y AChE I and to increase the persistence of synaptic ACh to prolong post synaptic cholinergic receptor activation (Taylor, 2001). This strategy to prolong the ACh post synaptic action can counter the dysfunction of neurotransmitter release or reduction in cholinergic receptors or signaling. In addition, AChE I reduce phosphorylation of tau, improve the secretion of sAPP, reduce amyloid production, normalize glutamate neurotransmission, and activate cholinergic muscarinic and nicotinic receptors. AChE I augm ent cholinergic transmission and improve the cognitive functions in mild to moderate AD patients (Blount et al., 2002; Ibach and Haen, 2004; DeLaGarza, 2003). Differences in patients genetic, neuropathological and clinical characteristics might not produc e expected benefits (Cutler and Sramek, 1997; Eagger and Harvey, 1995). O ld er patients poor ly tolera te drugs, therefore, the high incidence of adverse events might limit the use of AChE I inhibitors in certain patients with AD (Wagstaff and McTavish, 1994) . Studies have shown overall AChE activity declines in AD patients (Arendt et al., 1992; Fernandez et al., 1996; Ogane et al., 1992a; Schegg et al., 1992; Siek et al., 1990; Younkin et al., 1986). In addition, structural changes in AChE may be

PAGE 24

24 occurring (G eula and Mesulam, 1995). As AChE I attach to AChE and inhibit its, structural changes in AChE might seriously limit AChE I utilization in AD because AChE I will no longer be able to join AChE. Moreover, alteration in post synaptic muscarinic receptor respo nse and alteration in signal transduction might nullify the AChE I action in AD. Although useful, AChE I should be used carefully , research indicates . Prolonged and excessive use of AChE I can lead to cholinergic toxicity , which is characterized by stimula tion of parasympathetic end organs , leading to excessive lacrimation, salivation, sweating (Ecobichon, 2001) and meiosis (Rengstorff, 1994), and to enhanced micturition (Taylor, 2001). Anticholinergic Exposure Measurement Methods It is difficult to quantif y the anticholinergic effects of medications. The researchers tried to develop simple and uniform methods to quantify drug related AC activity that can be easily used in various clinical and research settings. Furthermore , t he utility of these methods also depends upon resources, training and the clinical experience of the researcher s to evaluate the drug related AC adverse effects. Four general methods have been proposed to assess drug related AC activity that can be used in either clinical or research set tings. The se four methods are as follows : The f irst method is assessing a nticholinergic drug lists combined with clinical judgment . This is one of the simplest methods to assess the anticholinergic activity . The clinician consults a reference list of drugs with known anticholinergic properties and applies his experience and clinical judgment to rate the anticholinergic activity of an individual. Several lists of drugs with AC properties measured both objectively and subjectively are available for refere nce (Mintzer BJ, 2000; Defilippi & Crismon, 2003; Fick et al., 2003; Mulsant et al., 2003; Flacker et al., 1998; Beers MH, 1997). In addition

PAGE 25

25 to relying upon the clinician knowledge of anticholinergic drug effects on physical and cognitive functions, thi s method relies heavily on patient report s of prescribed and over the counter medication and maintaining correct medical record s . Addition of physical or cognitive rating scales to objectively quantify drug related anticholinergic effects can reduce certai can result in better assessment of drug effect (Burnett et al., 1980; Rudd et al., 2005). Therefore, integration of subjective and objective measures helps improve decision making to counter the risk of drugs related anticholinergic activity. This method of assess ing anticholinergic burden through a drug list and clinical judgment is generating more interest than other method s for various reasons . The reasons include that it is simple to administer, i t can be used cost effective ly , it can assess both subjective and objective drug burden, the results are readily available, daily drug dose both for scheduled and as needed medications can be determined , and pre and post change in cognition can be easily evaluated. Based on this methodology, four anticholinergic scoring instruments have been developed to assess anticholinergic drug burden and to advance the risk stratification methods. The four drug scales are: Anticholinergic Drug Scale (ADS) (Carnahan, Lund, Perry, Pollock, and Culp, 2006); The Drug Burden Index (anticholinergic component) (DBI) (Hilmer et al., 2007); The Anticholinergic Risk Scale (ARS) score (Rudolph et al., 2008); The Anticholinergic Burden (ACB) scale (Fox et al., 2011). The s econd method is the m easurement of individual drug r elated anticholinergic activity. This technique measures in vitro drug anticholinergic activity using a radio receptor assay (Tune et al., 2005) standardized to anticholinergic activity of atropine

PAGE 26

26 (highly ant icholinergic drug). This method directly measures the anticholinergic activity of a drug compared to atropine equivalent (Tune et al., 1993). This is a good method to measure drug related anticholinergic activity , but it does not measure drug metabolite or The t hird method is the m easurement of total serum anticholinergic activity resulting from drugs, metabolites, and patient physiology . A drug with anticholinergic activity competes with acetylcholine for muscarinic acetylcholine (mACh) receptors. As a consequence, acetylcholine remains free and causes anticholinergic toxicities. Tune and Coyle (1980) developed a radioreceptor assay to directly measure the degree of drug binding to mACh re ceptors and to quantify the total serum anticholinergic activity (SAA) in humans. In addition to the degree of mACh competitive binding inhibition by an anticholinergic drug, the total serum anticholinergic activity also measures drug metabolites activity, drug addition or modification effect, and baseline physiology of an individual. Using this method, previous studies have shown a negative correlation between high levels of serum anticholinergic activity and cognitive functioning (Mulsant et al., 2003; Th ienhaus et al., 1990). SAA has come to be considered the gold standard in quantifying drug anticholinergic activity and burden (Carnahan, Lund, Perry, & Pollock, 2001). The f ourth method m easur es individual muscarinic receptor affinity in vitr o . This method allows researchers to ascertain the direct muscarinic receptor affinity of drugs in vitro, as a measure of drug receptor interaction, to assess the relative anticholinergic activity of drugs. In vitro measurement of drug muscarinic receptor affinity considered to be the baseline physiologic anticholinergic activity of an individual.

PAGE 27

27 Muscarinic receptor affinity of a drug is assessed by comparing the competitive binding between a drug with the m uscarinic ACh receptor and a radiolabeled muscari nic cholinergic agonist (Snyder & Yamamura, 1977; Tune & Coyle, 1980; Richelson E., 1994). N o single test comprehensively measures all the characteristics of drug related anticholinergic activity. A n ideal study, in addition to gold standard test serum an ticholinergic activity (SAA), would use a constellation of evaluation tools to gather holistic anticholinergic effects due to drugs. However, a fter careful examination of advantages and disadvantages of each methods and the limitation of the data in the NA CC database , we , in this study, select ed expert based anticholinergic drug list method to score anticholinergic burden of individuals and to differentiate between the central (Rudd et al., 2005) and peripheral (Tune & Coyle, 1980) action of drugs with anti cholinergic activity. Currently, certain researchers (Rudd et al., 2005) believe that expert based drug list method is the only clinically applicable method for assessing cognitive effects of drugs with anticholinergic activity. Among the list based antic holinergic drug scales , only ADS has been validated against the gold standard, SAA (Carnahan et al., 2006). Other scales are either not validated yet (ARS, DBI, and ACB scale) or have been developed with or for specific population. For example, ARS focused on only one healthcare system that had a primarily male population (Boustani at al., 2008). In addition, ADS includes an extensive list of prescription and over the counter medication (Carnahan et al., 2006). Therefore, ADS serves as an objective quantita tive tool to measure central and peripheral drug anticholinergic activity . Furthermore, ADS measures potential additive or synergistic

PAGE 28

28 effects of combining potentially anticholinergic drugs, which allow s clinicians to tailor AC assessment to any drug regim en. Subjective rating of anticholinergic activity using ADS based on clinicians expertise and expression of symptoms results in additional variability. Therefore, ADS is an ideal method that can increase both generalizability and reliability and can be wi dely applicable both in the research and healthcare settings with geriatric population s . Carnahan et al (2006) demonstrated the validity of ADS against the gold standard, SAA. About 95 percent of variance in SAA was explained by ADS score (R 2 =0.947). In a ddition, dose weighted ADS was significantly associated with SAA and predicted valid measure of anticholinergic burden. Total ADS scores are significantly associated with SAA, demonstrating some degree of criteria validity (Carnahan et al., 2006). Therefor e, the pragmatic method to assess and quantify anticholinergic use is to utilize ADS. The NACC recorded 100 prescriptions , non prescription drugs and vitamins supplements taken by subject s daily over two weeks. The drugs list included various medication gr oups such as antivirals, antibiotics, antidepressants, antihistaminics, antipsychotics, antianxiety, analgesics, cardiovascular drugs, muscle relaxants, H2 antagonists, corticosteroids, anticonvulsants etc. Many drugs with AC activity are not listed in the NACC medication list. I f a subject takes medications that were not in the list, the ADC staff would record that additional medication. W e can presume that if a subject took a medicine, it would be recorded in NACC dataset during the visit. Hence , the NACC dataset has strengths in collecting medication use. Overall, the NACC data provide sufficient information to assess anticholinergic drug use and to quantify drug

PAGE 29

29 exposure. In addition, the annual assessment of drug use provides comprehensive information t o evaluate cognitive impairment due to anticholinergic drug use. Epidemiology of Anticholinergic Use and Risk of Dementia C urrent U . S . data suggest that about half of the dementia patients who visit primary care providers take at least one medication with AC properties (Schubert et al., 2006). Data from a nationally representative sample, the 2006 2007 National Ambulatory Medical Care Survey and outpatient department component of the National Hospital Ambulatory Medical Care Survey , show that every year ab out 6.8 million (CI 5.27 8.44 million) ambulatory care visits are made for dementia (Bhattacharya et al., 2011). Around 43 percent (42.86%; 95% CI, 35.24 50.48) of these visits are associated with at least one AC drug use , and around 10 percent of these vi sits are associated with medication s that have high anticholinergic effects (ADS level 2 or 3) (Bhattacharya et al., 2011). Another study showed that women reported higher anticholinergic drug use than men did (9.6 percent vs. 4.4 percent ; P=0.001) (Konish i et al., 2010). Results of the previous studies suggest that the chronic use of AC drugs may accelerate the AD pathology and may have adverse effects on the clinical course of AD (Konishi et al., 2010; Lu & Tune, 2003). Carriere et al. demonstrated 3.10 p ercent and 2.01 percent incident rate of cognitive impairment and AD, respectively , among older adults taking AC medications as compared to those not taking AC medication (Carriere et al., 2009). In addition, Jessen et al. (2008) found a 12.35 percent inci dence of dementia among AC users compared to those not taking AC medication. Also , a six year longitudinal study in an Africa American (AA) population showed that the incidence of cognitive impairment among those exposed to definite AC drugs was significan tly higher by 46 percent as compared to those no t exposure to AC drugs (Campbell et al., 2010) . However, an

PAGE 30

30 eight year follow up study by Ancelin et al. (2006) found no difference between the risks of dementia among AC users and non AC users. Previous stu dies have shown that AC drug use is associated with deficits in cognitive functions like processing speed, attention, problem = solving, psychomotor performance, and language skills (Campbell et al., 2009; Corder et al., 1993; Jessen et al . , 2010; Shah et a l., 2013; Strittmatter et al., 1993). A r ecent cross sectional study by Uusvaara et al. (2013) showed that AC drug use was significantly associated with low verbal fluency (OR=1.84, CI 1.02 3.32) and low naming score s (OR=1.81, CI 1.09 3.00) but not for MM SE after controlling for covariates (Uusvaara et al., 2013). Similarly, Bottiggi et al. (2006) showed that long term AC drug use was associated with faster rate of decline in Trail Making Test part A and B; however there was no change in global cognitive s tatus (Bottiggi, 2006). In contrast , Konishi et al. (2010) showed that individuals with AC drug use had significantly lower global cognitive status ( MMSE score : 8.89 vs. 13.16, P=0.037) and specific cognitive functions like registration (1.62 vs. 2.30, P=0 .032) and recall (0.00 vs. 0.34, P=0.028) as compared to those who did not use AC drugs (Konishi et al., 2010). Another recent cross sectional analysis by Lampela at al. (2013) demonstrated that higher AC drug use was associated with impaired activities of daily living (ADL) and instrumental activities of daily living (IADL). A two year longitudinal study demonstrated a dose response relationship between anticholinergic drug burden and reduced MMSE scores (Fox et al., 2011). The study showed that two years of high AC drug use was associated with a decline of 0.34 (CI= 0.01 0.67) units of MMSE score as compared to no AC drug use. Shah et al. (2013) demonstrated in a longitudinal study that the annual rate of global cognitive

PAGE 31

31 function decline was greater in n ew AC drug users (0.096 (SE=0.007)) and chronic AC drug users (0.081 (SE=0.011)) as compared to individuals with no AC drug use (0.062 (SE=0.005)). Based on these studies , it can be concluded that one of the major factors contributing to cognitive decline among older adults is AC drug use. However, the current understanding of the effects of AC drugs on human cognition is derived largely from studies conducted on older adults with normal cognition (Corder et al., 1993; Jessen et al, 2010; Shah et al., 2013 ; Strittmatter et al., 1993). Because of this , little is known about the effects of AC drug use on cognition among an at risk population, i.e. older adults with pathophysiological characteristics of PAD. The long term effects of chronic AC drug use on glob al cognition and especially on cognitive domains warrants further study. One of the major limitations with the previous research related to the standard definition of PAD. In the absence of standardized NIH guidelines, over the years, previous studies that focused on AC drug effects on older population s used numerous classification systems. Other major gaps in the existing literature are the lack of prospective information on both cognitive domains and AC drug use. Furthermore, the relationship between the amount and the duration of AC exposure to induce clinically significant adverse cognitive outcomes has not been sufficiently examined. Our proposed study would us e the NACC dataset to identify the associations between risks of cognitive impairment and neur odegenerative biomarker change and scientifically quantified long term AC use. In 1984 the (NIH and the National Institute on Aging (NIA) established the Alzheimer's Disease Centers (ADCs) program in order to i mprove the understand ing of

PAGE 32

32 AD etiology, mechanisms, course, clinical diagnosis, neuropathologic changes, and D isease (CERAD) brought the ADCs together to formulate and to apply standardized methods to collect consistent clinical, neuropsychological, and pathological data from research participants at the ADCs (Morris et al., 1989). The NACC is a research establishment founded in 1999 of 29 collaborating ADCs that are funded by the NIA (U01 AG016976) started. To encourage and engage ADCs, NACC funded collaborat ive research. NACC provided flexibility to each ADC to develop, store , and submit its own project dataset based upon its own data elements and research needs. T hese different datasets w ere difficult to manage. With this in mind , NACC combined ADC datasets and designed a larger database, the Minimum Data Set (MDS ) that contained individual datasets with their respective data elements . T he database was made available online at www.alz.washington.edu (Beekly et al., improved the data collection process and maintained databases collected from the 29 ADCs through a newly designed data management system. The MDS was limited by retrospective collection of fewer data elements that provided inadequate information , and treatment of AD (Cronin Stubbs e t al., 2000). The Clinical Task Force (CTF) formed by the NACC expanded the MDS to create a more comprehensive database, the Uniform Data Set (UDS) , which was implemented in 2005 (Beekly et al., 2007). MDS is a database that contains datasets from 1984 to 2005, while the UDS database collects standardized information since 2005.

PAGE 33

33 Minimum Data Set The current version of data forms for MDS is Version 8, 2006. There are 74,169 participants in the MDS database , and they were enrolled from 1984 to 2005. The MDS database had 67 variables obtained retrospectively from ADC medical records. The education, marital status, residence, medical history, family history, date of initial and most recent ADC evaluation, initial and most recent MMSE score and so on . In addition, dementia, AD and non AD type dementia, clinical conditions and their cont ribution to dementia, vital status and autopsy status . It also collected information about primary and secondary neuropathological diagnosis. Furthermore, the MDS maintains information regarding the availability of banked brain tissues, cerebrospinal fluid ( CSF ) and DNA for future use. The MDS also reviewed whether ApoE genotyping, neuropsychological tests, Computed Tomography (CT), Positron Emission Tomography (PET), Single Photon Emission Computed Tomography (SPECT), and Magnetic Resonance Imaging (MRI) w ere conducted in research studies . Uniform Data Set The UDS data components were formulated to help in the diagnosis of dementia and its causes in clinical settings and to provide resources to conduct research on AD and MCI (Morris et al., 2006). The curr ent version of data forms for UDS initial, follow up , and telephone follow up visits is Version 8, February 2008. The comprehensive UDS database was implemented on September 1, 2005. As of September 1, 2012 , there were 26,452 participants , and the database had around 725 data elements. The initial and follow

PAGE 34

34 NACC MDS enrollment, living situation, level of independence, family history of dementia, severity of dementia, medical history, m edication use, physical measurements, severity of dementia using Clinical Dementia rating (CDR) (Morris, J. C., 1993) scale, Neuropsychiatric Inventory Questionnaire (NPI Q) (Cummings et al., 1994; Kaufer et al., 2000), Geriatric Depression Scale (GDS) (Ye savage, J. A., 1988), functional impairment using the Functional Assessment Questionnaire (FAQ) (Pfeffer et database has imaging and lab oratory records. A t elephone f o llow u p is done if a subject is cognitively or physically impaired for an in person visit, homebound , or institutionalized, or when a subject or informant refuse d to attend an in person visit. Due to non availability of the subject or informant, in person assessments are missing from telephone follow ups . Neuropsychological D atabase The CTF and a subcommittee designed standardized Neuropsychological Test Battery (NTB) of tests, which include 54 data elements across all ADCs. A combination of methods was us ed in selecting cognitive domains and tests to focus on cognitive markers of aging and AD, to account for continuity of previous measures , and to minimize the burden on ADCs. As the cognitive outcome focused on transformation from aging without dementia, t o MCI, and to AD, the cognitive domain selection was based on sensitivity to age related change in cognition (Petersen et al., 1992; Ivnik et al., 1990; Ivnik et al., 1996; Steinberg et al., 2005) sensitivity to the demonstrated primary cognitive impairmen ts in AD (Storandt et al., 1989; Morrison and Hof, 1997; Ferris & Kluger,1997) ability to measure change over time and to stage AD (Locascio et al., 1995) , and ability to predict progression from MCI to AD (Linn et al., 1995; Elias et

PAGE 35

35 al., 2000; Albert et al., 2001). The most common test for global severity of dementia was MMSE (Galasko et al., 1995). Digit Span Forward and Digit Span Backward were selected for attention (Parasuraman & Haxby, 1993) from the Wechsler Memory Scale Revised (WMS R) (Wechsler, 1 987) and Digit Symbol was approved for processing speed from Wechsler Adult Intelligence Scale Revised (WAIS R) (Wechsler, 1986). The Part B Trail making test (Reitan and Wolfson, 1986) was used for assessing deficit of executive functions to predict progr ession of MCI to AD. A hallmark of early AD , decline in episodic memory (Albert, 1996), was assessed by Logic Memory, Story A immediate and delayed recall from WMS R (Wechsler, 1987). Verbal Fluency and Naming were assessed by Animal and Vegetable list gen eration test s and a short version of the Boston Naming Test (Goodglass et al., 1983), respectively. Functional Assessment Questionnaire (FAQ) (Pfeffer et al., 1982) and behavioral symptoms short form NPI Q (Kaufer et al. , 2000 ; Cummings et al., 1994) and t he 15 item Geriatric Depression Scale (Sheikh & Yesavage , 1986 ) were also added to the Clinical Assessment portion of the UDS cognitive battery. Considering the length of each test and burden on subjects and their informants and on ADCs, CTF decided to kee p the NTB short enough to be administered in 30 to 40 min utes by a trained clinical professional and to administer the test annually. Current Political and Legislative Efforts Regarding The urgency to minimize AD related health and eco nomic consequences can be gauged by the fact that T a priority basis in January 2011. In addition, the federal government and the scientific co concerns are reflected in steps to immediately implement NAPA by allocating $156 million to support

PAGE 36

36 and sustain research on AD. The proposed study will specifically address the NAPA Goal 1: 5. Our study focuses on older adults who are at higher risk of cognitive decline and we will investigate the long term effects of AC medication use on cognition among this high risk population . The study results will help clinical researchers to understan d the effects of AC medication use and to formulate appropriate clinical and public health strategies to minimize AC drug use among older adults. In identifying the at risk group, we fulfill the Identify Research Pri orities and Milestones . Under Goal1, t he proposed research is highly responsive to the NAPA strategic priority Strategy 1.C: Accelerate Efforts to Identify Early and Presymptomatic Stages of by several classification systems. The primary aim of our proposed study is to identify the at risk group at an early stage based on new NIH guidelines. We would also address Strategy 1.B: Expand R esearch Aimed at Preventing and Treating Disea se . Testing Aims 1 and 2 would allow us to provide evidence about cognitive outcomes of long term anticholinergic drug use and would provide evidence to whether limiting anticholinergic use prevents and delays cognitive decline. The proposed study is also intended to address Goal 2: Enhance Care Quality and Efficiency , and Strategy 2.B: Ensure Timely and Accurate Diagnosis. It is evident that the NAPA strategic plans very clearly emphasize the significance of understanding the PAD group and its timely ident ification for appropriate intervention to minimize the dementia burden on the family, society, the health system, and the economy. In addition, the NAPA emphasizes understanding the cognitive outcomes of PAD and the long term effects of intervention s in li mit ing

PAGE 37

37 cognitive decline. The current proposed study , using one of the largest samples and data elements available, is well positioned to answer NAPA strategic priorities listed above . What Is New in This Study? New Approach Our proposed study applies a ne w approach to assess cognitive decline due to long term medication use among older adults in the context of the preclinical Disease population. While previous studies have identified the effects of AC medication use on cognition that explain co gnitive decline among older adults to some extent (Campbell et al., 2009; Mirra, 1991; Uusvaara et al., 2013 ; Nourhashémi et al., 2008 ), they have not assessed the at risk population in the preclinical stage of AD with QCI . This stage predisposes individua ls to rapid cognitive decline and the development of AD. What is unique in the proposed research is that we would evaluate not only baseline measurement s but also the long term effect of AC drug load on cognition in a very high risk population . Our sample at baseline means older adults with age greater than or equal to 60 with questionable cognitive impairment based upon Clinical Dementia Rating Sum of Boxes (CDR SOB) of 0.5 2.5. Such an approach would allow better understanding of the mechanism through whi ch AC medications influence cognitive functions, the heterogeneity of the effects, and the extent to which controlled use of AC medications would contribute in preventing AC drug associated cognitive decline. To the best of our knowledge, the proposed stud y will be the first study to assess the role of anticholinergic medication burden on cognitive decline and development of MCI/AD in older adults with QCI . The proposed study might provide groundwork for future randomized controlled studies to assess medica tion effects on wide ranges of cognitive

PAGE 38

38 outcomes among older adults. Rationale for Anticholinergic Use and Dementia Normal cognition requires the coordination of numerous complex processes, including sensory information processing, sustained and divided a ttention, short and long term memory, and executive functions. It is believed that a large portion of cognitive decline and poor cognitive performance among older adults is attributed to anticholinergic medications use (Kausler, 1994; Sunderland, Tariot, & Newhourse, 1988). These medications , which primarily act against muscarinic ACh receptors , impair various cognitive domains , including episodic and semantic memory, working memory, processing speed, language, and praxis (Molchan et al., 1992; Sunderland e t al., 1988). The m uscarinic and nicotinic AChRs are distributed throughout the central nervous system ( CNS ) (Caulfield, 1993; Dani, 2001; Felder et al., 2000; Flynn et al., 1995; Lucas Meunier et al., 2003) , and they have distinct, as well as overlapping , functions in learning and memory (Edginton & Rusted, 2003; Everitt & Robbins, 1997; Little et al., 1998; Sahakian et al., 1989; Terry et al., 1993; Wesnes & Warburton, 1984). There are five (M1 M5) muscarinic receptor subtypes (Bonner et al., 1987; Eglen, 2005; Dani, 2001), which are abundant in the cognitive areas (e.g., cortex and hippocampus) of the brain. Previous studies have demonstrated in both humans and rodents that all five receptors are important in learning and memory processes (Anagnostaras et al., 2002; Bodick et al., 1997; Messer et al., 1990; Seeger et al., 2004; Tzavara et alo, 2003), but that each muscarinic receptor has its markedly distinct regional expression pattern. M1 and M5 knockout mice models have shown a severe deficit in working (Anagnostaras et al., 2002) and recognition tasks and spatial memory (Aray et al, 2006), respectively. The mice models of receptor specific antagonistic administration have shown that M2

PAGE 39

39 blockage improves (Lachowicz et al., 2001), whereas M4 blockage dete riorates emotional learning (Jerusalinsky et al., 1998). However, M3 knockout mice seem to perform normally across various cognitive domains (Yamada et al., 2001); for example, long term administration of the M3 selective incontinence agent, darifenacin, d o es not have an adverse effect on cognition (Kay & Wesnes, 2005). These results suggest that the anticholinergic drugs may lead to varying cognitive outcome. Scopolamine, an AC prototype, administration in cognitively healthy adults has been shown to reduc e reaction time, complex attention, motor functions, verbal learning, facial recognition, praxis (Dumas et al., 2006; Flicker et al., 1992; Naranjo et al., 2000; Sunderland et al., 1987; Tariot et al., 1996; Zemishlany et al., 1991), episodic (Molchan et a l., 1992 ) , semantic and working memory (Sunderland et al., 1987; Tariot et al., 1996). In addition, AC medications have been shown to worsen cognitive impairment associated with other medical conditions like cerebrovascular disease (Nebes et al., 2005). Ac ute and chronic AC drug use is associated with rapid declines in global cognition (Lu & Tune, 2003), working memory, reaction time, attention, psychomotor speed, executive functioning, and visual memory (Bottiggi et al., 2006; Glass et al., 2003; Katz et a l., 1998). The current evidence show s that the modulation of the muscarinic cholinergic system is an important part of normal cognitive processes , and any imbalance in the neurotransmission of cholinergic system may cause domain specific cognitive impairme nt . This finding demonstrates the biological plausibility that AC medication use might be associated with cognitive impairment. Therefore, it is important to understand and to predict the effects of AC drug use on specific cognitive domains.

PAGE 40

40 Previous stud ies have shown that the cholinergic cell loss and cholinergic dysfunction in basal forebrain neurotransmission is associated with AD (Bartus, 2000). Loss in choline acetyltransferase activity correlates with reduction in cholinergic neurons (Henke & Lang, 1983). Furthermore, cholinesterase inhibitors and cholinomimetic drug replacement have been shown to have memory enhancing effects (Muir & Robbins, 1995). However, there is inconsistency in demonstrating cholinergic cell loss and its association with cogni tive decline (Sarter & Bruno, 1998). Previous cross sectional studies focusing on older adults have shown a positive association of AC drug use with cognitive decline (Cancelli et al., 2008; Cao et al., 2007; Lechevallier Michel et al., 2005; Mulsant et al ., 2003), visual and declarative memory (Lechevallier Michel et al., 2005), implicit learning (Nebes et al., 2005), and psychomotor speed (Nebes et al., 2005). Additional burden comes from age as older age is associated with numerous clinical morbidities a nd frequent drug use (Ancelin et al., 2006; Hajjar et al., 2007; Lechevallier Michel et al., 2005; Pitkala et al., 2001). Many commonly used medications have AC properties (Tune et al., 1992; Uusvaara et al., 2009) and their long term use is associated wit h cognitive decline (Ancelin et al., 2006; Carriere et al., 2009). Despite concerns regarding long term use of AC drugs, previous studies have focused primarily on acute or short term effects of prescription medication with AC properties in elderly populat ion s The possibility that over the counter AC medications might induce cognitive impairment has not been addressed sufficiently (Hilmer et al., 2007). The understanding of the association between long term u se of AC and cognitive decline among older adults with QCI remains unclear because of the lack of data with

PAGE 41

41 consistent and repeated standard assessments of both medication use and cognition overtime. The proposed hypotheses fill the scientific gap by using the NACC dataset to assess not only the cross sectional association between AC medication use and cognitive impairment but also to explore the long term trajectories of global and domain specific cognitive decline associated with AC use over time.

PAGE 42

42 Figu re 1 1. Conceptual framework of anticholinergic drug effects on aging and cognition . ; Cerebro spinal fluid.

PAGE 43

43 Table 1 1 . Staging categories Stages of PAD Description CSF) Marker of neuronal injury Evidence of subtle cognitive change Stage 1 As ymptomatic cerebral amyloidosis Positive Negative Negative Stage 2 Asymptomatic amyloidosis + Downstream neurodegeneration Positive Positive Negative Stage 3 Asymptomatic + neuronal injury + subtle cognitive/behavior decline Positive Positive Positive A bbreviations: ; PET= Positron Emission Tomography ; Cerebro spinal fluid.

PAGE 44

44 CHAPTER 2 ANTICHOLINERGIC DRUG USE AND RISK OF COGNITIVE IMPAIRMENT AND DEMENTIA IN OLDER ADULTS WITH QUESTIONABLE COGNITIVE IMPAIRMENT: A CROSS SECTIONAL ANALYSIS Background About 5.3 million people (Brookmey er et al., 2011) in the U.S. and about 35.6 ). The incidence of AD is expected to triple over the next 40 years (Alzheimer's disease facts and figures, 2012). In addition, the global aging of popu lation is driving the current epidemic of cognitive impairment and dementia (Sosa Ortiz, Acosta Castillo, & Prince , 2012 ). Notably, AD associated dementia accounts for 60 to 80 percent of total dementia cases (Mayeux & Stern, 2012; Nowrangi., Rao, & Lyket sos, 2011). The biological changes such as decrease in cholinergic neurons, signaling and receptor dysfunction are associated with a decline in learning and memory among older adults (Bird, Stranahan, Sumi, & Raskind, 1983; Etienne et al., 1986; Henke & La ng, 1983). Consequently, AD is characterized by cortical cholinergic deficits, cognitive impairment, and dementia (Bohnen et al., 2007; Tiraboschi et al., 2000). It has also been shown that AD and dementia severity is directly correlated with atrophy and d egeneration of the brain cholinergic pathways (Barnes, 1990; Gallagher & Nicolle, 1993). However, less is known about whether exogenous use of medications, particularly medications with anti cholinergic (AC) properties, is connected to age related cogniti ve impairment. There is a high rate of AC medication use among older adults . A s a result , there is an increased risk of adverse drug events (Flaherty, Perry, Lynchard, & Morley, 2000; Hanlon et al., 2006; Lyer, Naganathan, McLachlan, & Le Conteur, 2008; Marengoni, Rizzuto, Wang, Winblad, & Fratiglioni, 2009; Sergi, De Rui, Sarti, & Manzato, 2011;

PAGE 45

45 Onder et al., 2012). For example, there are known serious side effects of AC medications on cognitive (Boustani, Sachs, & Callahan, 2007; Campbell et al., 2009; Carriere et al., 2009; Fox et al., 2011) and physical abilities (Han, Agostini, and Allore, 2008; Hilmer & Gnjidic, 2009; Koyama, Steinman, Ensrud, Hillier, and Yaffe., 2013). Although, there has been considerable attention directed at monitoring and min imizing AC drug load to avoid cognitive impairment among older adults, these medications are frequently prescribed to manage specific comorbidities in older adults (Wawruch et al., 2012; Peters, 1989). While these medications can be effective for managing specific disease conditions, there is a clear biological effect on cognition through (Chew, Mulsant, & Pollock, 2005) unselective antagonism of muscarinic receptors in the brain (Peters, 1989; Nebes, Pollock, Halligan, Kirshner, and Houck, 2007). Older adu lts are likely to be more susceptible to these biological effects of AC drugs because they have lower central cholinergic activity and reduced cholinergic transmission, and because they are more susceptible to central anti muscarinic adverse effects (Feinb erg, 1993; Hammerlein et al., 1998). The pathophysiological process of AD and associated dementia begins years before the emergence of clinically detectable signs and symptoms (Price & Morris, 1999). Therefore, subtle clinical, cognitive , and pathological changes emerge years before Mild Cognitive Impairment (MCI) or AD dementia can be diagnosed. The current understanding of the effects of AC drugs on human cognition is derived largely from the studies that compare the AC medication effects on individuals with dementia to individuals with normal cognition (Corder et al., 1993; Jessen et al, 2010; Shah et al., 2013; Strittmatter et al., 1993). Consequently, there is an inadequate understanding of

PAGE 46

46 cognitive effects of AC drugs on individuals who are in the ea rly stage of cognitive impairment, i.e., questionable cognitive impairment (QCI). These individuals with QCI are at greater risk of developing dementia in the future . Therefore , it is imperative to comprehend the clinical and epidemiological effects of AC medication use on cognitive functions and the progression to dementia. Previous observational studies have reported that high AC burden is associated with greater risk of cognitive impairment in normal older adults (Ancelin ML et al., 2006; Han et al., 20 08; Lechevallier Michel et al., 2005; Rudolf, Salow, Angelini, McGlinchey, 2008). While the previous studies have identified the effects of AC medication use on cognition that explain cognitive decline among older adults to some extent (Bottiggi et al., 20 06; Campbell et al., 2009; Uusvaara et al., 2013), they have not assessed an at risk population, i.e., individuals with QCI. What is unique in this study is that we would assess the cognitive impact of varying amount s of AC burden on individuals with QCI. The objective of the proposed study is to assess the effects of AC medication use on cognitive functions among older adults with QCI. We hypothesize that older adults with QCI who have exposure to AC drugs are more likely to have poor cognitive scores as c ompared to older adults who have no exposure to AC drugs. Methods Study Participants This study is a part of larger multi center longitudinal cohort database National that includes community dwelling older adults wi th normal cognition, mild cognitive impairment (MCI), dementia , and AD. The NACC is a research establishment of 34 collaborating Alzheimer Disease Centers (ADCs) that was established in 1999 (Beekly et al., 2007). The NACC improved the

PAGE 47

47 data collection p rocess and created a more comprehensive database, Uniform Data Set (UDS) (Morris et al., 2006), to formulate and to apply the standardized research methods to understand AD etiology, course, mechanisms, clinical diagnosis, neuropathologic changes, and trea tment modalities. The NACC UDS data has been described elsewhere (Beekly et al., 2004). The sample for this study consists of individuals from the NACC UDS data who were in the study from September 2005 until March 2014. Individuals aged 60 and above, havi ng medication records, and diagnosed with QCI at the start of the enrollment in the study ( in 2005 ) were included in the study. QCI was determined by a commonly utilized dementia severity tool, the Washington University Clinical Dementia Rating Sum of Boxe s (CDR SOB) scale (Rosenberg et al., 2012). The CDR SOB is a reliable and valid scale, which is regularly used in both clinical and research settings to stage dementia severity. The CDR SOB score is a detailed quantitative rating scale , and it provides mor e comprehensive information than the CDR GS. The CDR SOB scores have a wider range of measurements (0 18) depicting the various stages of cognitive impairment , and they optimally differentiate with reasonable accuracy between individuals with QCI and those with mild cognitive impairment. For our analysis, QCI was defined as CDR SOB score between 0.5 2.5 (O'Bryant et al., 2008). The CDR SOB score is a comprehensive quantitative rating scale that helps in distinguishing preclinical cognitive impairment from dementia. Anticholinergics Exposure To assess AC drug exposure, we used the valid and reliable (Tune et al., 1993) Anticholinergic Drug Scale (ADS) (Carnahan et al., 2006), an ordinal drug scale that rates AC medications according to increasing potency of their anticholinergic activity. ADS include 413 medications which cover about 88 percent and 89 percent of regular

PAGE 48

48 and as needed drugs, respectively. ADS scores are divided into four levels according to AC drug activity (Table 2 1). We calculated each part based upon their AC medication use within two weeks prior to their visits . For the purpose of our study, we categorized the ADS drugs into three levels (Table 2 1) . Level 1 indicates drugs with no known AC properties, Level 2 indi cates drugs with low potential AC activity seen by receptor binding study, and Level 3 was created by combining ADS level 2 and 3 , which represent medications that have high AC adverse effects. follow up visits. The NACC standardized form (Form 4) was used by trained and certified researchers who recorded 100 prescription, non prescription, over the counter medications and supplements taken by the participants within two weeks prior to their vis it. The NACC recorded drugs from various medication groups such as antivirals, antibiotics, antidepressants, antihistaminic, antipsychotics, antianxiety, analgesics, cardiovascular drugs, muscle relaxants, H2 antagonists, corticosteroids, anticonvulsants. Drugs that were not listed in the NACC form were recorded using a standardized protocol by clinicians or certified ADC researchers that provided a specific drugID to the medications. So, we can confidently presume that the NACC dataset documented all the m edications taken by the participants. Therefore, the NACC dataset provided a comprehensive medication history that provides an excellent opportunity to examine the AC medication effects on cognition. Cognitive Outcomes We used standardized NACC Neuropsych ological Test Battery (NTB) that uses eight common neuropsychological tests (12 measures) to assess cognitive functions in the early stage of cognitive impairment. The NTB is sensitive to predict ing progression

PAGE 49

49 from QCI to dementia and to demonstrate age r elated changes in specific cognitive domains. Hayden et al. (2011) performed exploratory and confirmatory factor analysis of the NACC UDS NTB and grouped 12 NTB measures in four cognitive domains Memory, Attention, Executive, and Language. The NACC NTB ass essed attention by Digit Span Forward and Digit Span Backward from the Wechsler Memory Scale Revised (WMS R). Decline in episodic memory, a hallmark of early AD, was evaluated by Logic Memory, IA and delayed recall by Logic Memory, IIA. Executive functions were assessed using Trail Making Part A (TM A) and B (TM B) tests and digit symbol substitution test (DSST). Since the lower TM A and TM B scores indicate better executive functions, we reversed the sign of z scores of TM A and TM B were and added with th at of DSST to assess the executive functions in the same direction as DSST, i.e., higher scores are better than lower scores. Language (verbal fluency and naming) was assessed by Animal and Vegetable list generation test and a short version of the Boston N the four cognitive domains mentioned above, and we included one composite score for global cognition. The domain scores were calcula ted as the mean z scores of cognitive tests (Shirk et al., 2011) under that domain. The global cognition was calculated as mean of four domain scores as shown in Table 2 2. Covariates The conceptual framework and the past literature guided the selection o f independent variables. Previous studies have shown an association between dementia and important demographic factors, lifestyle behaviors, clinical conditions , and genetic biomarkers (Carriere et al., 2009; Lechevallier Michel et al., 2005; Uusvaara et a l.,

PAGE 50

50 2013). The demographic characteristics included age, sex, race/ethnicity, education, marital status, living situation, and family history. The lifestyle behaviors that affect cognition like smoking, alcohol consumption, and body mass index were include d as predictors. Clinical conditions may contribute to cognitive burden in older adults. Therefore, we adjusted models for comorbid conditions such as cardiovascular disease, cerebrovascular disease, depression, neurological diseases and diabetes , which ar e associated with cognitive impairment. Gene ApoE with its isoform E4 is the largest known genetic risk factor for impaired cognitive functions (Deary et al., 2002; Farlow et al., 2004) and late onset AD (Corder et al., 1993; Strittmatter et al, 1993). APO E4 is also associated with increased risk of cognitive impairment in the presence of anticholinergic exposure (Uusvaara et al., 2009). Therefore, we included APOE4 in the analysis to assess the effects of AC medication use on cognitive functions among olde r adults with QCI. Statistical Analysis We performed preliminary analysis to evaluate distribution patterns of means and proportions of characteristics like demographics, comorbidities and genetic predisposition at baseline by ADS levels (No, Low, and H igh AC use) of anticholinergic drug load. We checked the linear model assumptions of independence, normality, and equal variance. We checked the independence by plotting each residual against neighboring value, i.e., residual lag plot and the assumption of equal variance. We illustrated the normal probability plot to observe against normal distribution values. Furthermore, we checked for the assumption of equal variance by plotting residuals against fitted values (estimated conditional means). We planned to transform a variable

PAGE 51

51 or to conduct a non parametric test if the variables violated the assumption; however, we found the variables of interest to follow assumptions of the linear regression. We performed the analysis of variance to evaluate differences i n the means of continuous variables. We performed a chi squared test to evaluate differences in the distribution of categorical variables. Descriptive statistics allowed us to quantify and compare the distribution of different cognitive measures among AC d rug levels. In this cross sectional analysis, the potential association between lower cognitive score s on global cognition and cognitive domains among individuals taking AC drugs was examined using general linear model (GLM). GLM was fitted to estimate the effect of AC drug exposures on cognitive outcomes at baseline according to ADS levels (No, Low, and High AC use). We included demographic factors, lifestyle behaviors, clinical conditions , and genetic biomarkers as covariates in the model. Since we perfor med five separate tests, we performed the Bonferroni Correction and adjusted the statistical significance at alpha= 0.01. Results A total of 7,315 participants met the inclusion criteri records, and CDR SOB scores. The mean (M) age ± standard deviation (SD) of the entire sample was 74.84±7.91 years (range: 60 109 years), 81.58 percent individuals were white, and 51.76 percent were females. About 91 percent had high school or higher education, 63.53 percent were married, and 76.52 percent were living independently. Of the sample, 15.69 percent were taking low levels of AC medication, and 4.05 percent were taking high levels . There were a few significant differences i n population characteristics across no, low , and hi gh AC medication groups (Table 2 3). The individuals who took high or low

PAGE 52

52 AC medications were slightly older (75.55±7.80 vs. 76.59±8.23 vs. 74.46±7.81; P value= < .0001) as compared to those who did not ta ke AC medications. There were fewer males that took high or low AC medications (36.49 percent vs. 45.73 percent vs. 49.33 percent; P value= < .0001) as compared to those who took no AC medication. Older adults with QCI who took high or low AC medications h ad slightly higher BMI (27.43±5.69 vs. 27.50±5.49 vs. 27.08±5.05; P value= 0 .02 2 ) as compared to those who did not take AC medications. As compared to the older adults with QCI who did not take AC medications, the older adults with QCI who took low or high AC medications were less likely to be married and more likely to live alone. Furthermore, these older adults with QCI who took low or high AC medications were less likely to live independent or in a single residence home. These differences between the AC groups were statistically significant. The prevalence of cardiovascular disease a mong individuals who took low AC medications was about 50 percent as compared to 24 percent among those who took high AC medications , as compared to 30 percent (P value <.0001 ) among individuals who did not take medication s with AC propert ies . Similarly, the prevalence of cerebrovascular disease was 16.5 percent and 11.5 percent among individuals who took medications with low and high AC properties, respectively. Interestingly, the prevalence of cerebrovascular disease among individuals in no AC medication group and high medication AC group was similar (11.5 percent; P value=<.0001). The prevalence of metabolic disease in no AC medication group (83.88 percent) and low AC medicat ion group (84.49 percent) was similar but greater than in the high AC medication group (76.35 percent; P value=0.002). As expected, t he re was significantly higher

PAGE 53

53 prevalence of incontinence a mong individuals in low AC group (35 .47 percent ) and high AC grou p (19.35 percent) as compared to those in no AC group (15 .32 percent; P value = <.0001) . This is due to the reason that anti incontinence medications , such as Oxybutynin, are the primary mode of treatment that work s by decreasing the muscle spasms by redu cing the cholinergic drive to urinary bladder or reducing intestinal motility. The prevalence of APOE4 genotype was higher in no AC medication group ( 43.24 percent ) compared to low or high AC medication groups ( L ow AC : 41.48 and percent H igh AC : 38.78 perc ent; P value=0.355) . As expected, we found that the percent) than no and low AC groups (No AC: 3.16 percent; High AC: 2.79 percent; P value=0.233 ) but it was not significantly different across the AC groups . Anticholinergic medication use and neuropsychological tests at baseline: The mean MMSE z score in high AC medication group was higher ( 0.79±1.84) than those in no ( 1.07±2.14) or low ( 1.04±2. 05) AC medication group (Table 2 4). Similarly, the mean z scores for immediate recall, delayed recall, digit forward, digit backward, and animal and vegetable fluency tests were significantly higher in the high AC medication group as compared to the no AC medication group, which sugges ts that individuals taking high medication with AC property had better cognition than those who were not. The mean z score of DSST and BNT were higher among the high AC medications group as compared to the no AC medications group , but the difference was no t statistically significant. The mean z score differences between high and no AC medication groups were statistically significant for TM A ( 0.89±2.01 vs . 0.79±1.72) and TM B ( 1.23±1.75 vs 1.13±1.67) , indicating better cognition among individuals taking

PAGE 54

54 high AC medication as compared to those taking no AC medication. The mean z scores (Table 2 4) were statistically significantly greater in high AC medication group as compared to no AC medication group among cognitive domains of memory ( 0.63±1.18 vs 0.8 0±1.15), attention ( .033±0.89 vs 0.40±0.88), executive function (0.56±0.91 vs 0.48±0.80), and language ( 0.54±0.97 vs 0.58±0.97). Overall, the mean z scores for global cognition was significantly higher in high AC medication group ( 0.23±0.53) as compar ed to no AC medication group ( 0.32±0.53) demonstrating that the AC medication use is associated with better cognition among older adults with QCI. In the multivaria ble model (Table 2 5), after adjusting for baseline demographics, lifestyle, comorbidity, a nd genetic factors, the mean z scores for the global cognition was statistically significantly different for low AC medication group as compared to no AC medication group (P = 0.0051). The global cognitive z score among low and high AC medication groups we re higher by 0.065 (p=0.0051) and 0.069 (p=0.0903) SD, respectively, as compared to no AC medication group (reference group). We did not find a linear or quadratic trend between the ADS groups. The fully adjusted model explained about 14 percent of the var iation in the model. The mean values of standardi zed z scores for memory (Table 2 6) were 0.0838 (p=0.1046) and 0.0315 (p=0.7451) SD higher in low and high AC medication groups as compared to no AC medication group , but the difference was not statistically significant. The model explained about 14 percent of the variability in the model. The standardized mean cogniti ve z score of attention (Table 2 7) for high AC group (0.2171 SD) was significantly higher than the low (0.0323 SD) and no AC group (reference group). In addition, this mean difference had statistically significant linear trend (p=0.0038) among the AC medication groups. The

PAGE 55

55 analysis showed that the z score for exe cutive functions domain (Table 2 8) among older adults with QCI who consumed medica tions with low and high AC properties was higher by 0.057 SD as compared to individuals who did not take medications with AC properties , but this difference was not statistically significant. The model explained 11.5 percent of variability in the model. Th e cogni tive domain of language (Table 2 9) had higher mean z scores for low (0.0923 SD) and high (0.0101 SD) the AC medication groups as compared to no AC medication group, and the association was significant for the no AC group. Overall, the models sugges t that the cognitive scores among older adults with QCI taking AC medications are higher than those who were not taking any medication with AC properties. Discussion To our knowledge, this was the first population based study to assess the association of A C medication use and cognitive functions among older adults with QCI based upon their CDR SOB score. The objective of this study was to assess the effects of AC medication use on cognitive functions among older adults with questionable cognitive impairment . We hypothesized that older adults with questionable cognitive impairment who have exposure to AC drugs are more likely to have poor cognitive scores as compared to older adults who have no exposure to AC drugs. Our study results did not show that older adults with QCI who took AC medication had lower global cognitive score as compared to those who did not take AC medication ; however, the association was not statistically significant. Furthermore , older adults with QCI who took AC medication had higher sc ore for cognitive domains of memory, attention, executive functions , and language than those who did not take AC medication . Moreover, these

PAGE 56

56 associations were independent of demographic factors, clinical conditions, and APOE biomarker that might influence the cognition among older adults. In our study, we found that the cognitive scores were higher among individuals using AC medication as compared to those who did not use AC medications. In contrast with others, we found that there was no dose response re lationship between AC medication use and cognition in QCI population (Pasina et al., 2013; Uusvaara et al., 2009; Chew et al., 2005). These findings differ from the findings of previous population based studies that used the list based method to assess AC medication load among cognitively healthy older adults or demented populations (Campbell et al., 2009; Fox et al., 2011; Lechevallier Michel et al., 2013). However, our findings are compatible with previous study which used SAA to assess AC medication load that found non demented geropsychiatric patients are significantly less susceptible to the adverse cognitive effects of AC medication than those with AD (Sunderland et al., 1987). Another recent longitudinal study showed that AC medications use did not de teriorate cognition over a period of 18 months (Pasina et al., 2013). Therefore, our study results highlight a specific population that might benefit from AC medication use; however, these claims need to be further investigated. A major difference from the previous studies may be associated with the differences between the populations under study. Other studies have mostly compared the effect of AC medication use on demented individuals to those with normal cognition ( Fox et al., 2011; Jessen et al., 2010; Konishi et al., 2010; Lechevallier Michel et al., 2005; Mulsant et. al., 2003; Uusvaara et al., 2013 ). In our study, we chose a population that has below normal cognitive scores but does not manifest clinically significant

PAGE 57

57 cognitive deficiency on neuropsyc hological assessment i.e., questionable cognitive impairment. In the early stage of cognitive impairment, the preservation of cognitive functions can be attributed to the compensatory effect of non cholinergic neurons like hyperinnervation and proliferatio n of forebrain galaninergic fibers and receptors, respectively (Counts, He, Che, Ginsberg, & Mufson, 2008b). Furthermore, interaction between cholinergic system s and other neuromodulators like norepinephrine, dopamine, serotonin, GABA, opiod peptides, gala nin, substance P, and angiotensin II play a vital role in modulating the improved cognitive response in older adults (Decker, & McGaugh, 1991). Furthermore, neuropathological studies also provide evidence that the brain structures are preserved in the earl y stages of the dementia and AD. It is also possible that the pharmacodynamics interactions and effects of multiple AC medications are not additive in a linear pattern. This can be explained because there is a great variability in AC drug sensitivity rel ated to advanced age, multi morbidity, stage of cognitive impairment , and co medications ( Rudolph et al., 2008 ; Han et al., 2008). I n addition , ACh synthesizing enzyme, choline acetyltransferase, and cholinergic neurons increases in early cognitive impairm ent (Frolich 2002; Ikonomovi et al., 2003) . These compensatory mechanisms apparently preserve the cholinergic enzyme activities and cognitive functions in early stages of cognitive impairment (Ikonomovi et al., 2003). Since our study sample was at a very e arly stage of cognitive impairment, the participants might have preserved cognitive functions and have been less susceptible to AC medication side effects. Previous studies have shown that during the early phase of cognitive impairment, increased activatio n of the cholinergic system compensates for cognitive impairment by increasing the frontal brain mediated

PAGE 58

58 knowledge based detection and filtering mechanisms, i.e., top down processes (Dumas & Newhouse, 2011). Hence, it is important to note that the older a dults with intact cholinergic functions will have a greater compensation benefit from the activation of frontal networks than those with advanced dementia. Dumas and Newhouse (2011) found that scopolamine, a mACh receptor antagonists, and mecamylamine, a n ACh receptor antagonists, improved frontal and hippocampal activations as compared to a placebo, providing evidence for compensation by top down processes during the early phase of cognitive impairment. However, Oberauer (2001), Naveh Benjamin, Bray, and L evy (2007) showed that the comorbidity in older adults may alter the cholinergic functioning and may decrease the ability of the older adults to use the frontal brain mediated compensatory control of cognitive functions that impairs memory (Oberauer, 2001; Naveh Benjamin et al., 2007). Our results show that individuals that took AC medications had higher global and domains specific cognitive scores as compared to those who did not take AC medications. We theorize that our study population of older adults wi th QCI had intact cholinergic functions and benefitted from some level of compensatory response . In addition, the AC medications provided additional benefit by alleviating the comorbidity burden on cognition. Comorbidities like cardiovascular disease, cere brovascular , and diabetes each have independent debilitating effects on cognition. We speculate that the QCI population responds to the therapeutic actions of AC medications, and that the therapeutic effects of AC medications offset any potential iatrogeni c AC overload that may manifest cognitive side effects. Supporting our premise, previous studies show that clozapine, a highly potent AC drug, improved attention, reaction time, motor processing,

PAGE 59

59 executive functions, working memory, verbal fluency, and vis ual learning and memory among schizophrenic individuals (Meltzer & McGurk, 1999). Recent community based cohort studies by Qiu et al. (2006) have shown an overall decrease in incident dementia among cognitively normal older adults who were treated with ant i hypertensive drugs. Other studies have also provided evidence that anti hypertensive medication like brain penetrating angiotensin converting enzyme (ACE) inhibitors (Ohrui et al., 2004) and calcium channel blockers (CCBs) (Hanon et al., 2006) not only d ecreases the risk of dementia among individuals with hypertension but also slows the progression among individuals suffering from dementia (Hajjar et al., 2005; Ohrui et al., 2004). This evidence demonstrate s that it is likely that better cognitive scores among QCI individuals taking AC medications may partly reflect the treatment effect of AC medications , consequently reducing the cognitive burden of comorbidities. It is important to note that medications such as Captopril, Nifedipine, Diltiazem etc. that have been shown to reduce the overall burden of dementia also have low anticholinergic properties. This dual action of medications to alleviate the comorbidity effects of dementia while contributing to the anticholinergic burden on brain shows that it is v ery difficult to separate out the anticholinergic effects of medications from its potential benefits. Hence, the overall impact of AC medications can be measured by their cumulative effects on cognition in the presence of other predictors such as comorbidi ty. Controlling for these medications in the model would have reduced the power to predict the additive effects of various anticholinergic medications and cumulative effects of other cofactors on dementia . Therefore, we did not control for the treatment me dications.

PAGE 60

60 It might be possible that the neuromodulators might interact with cholinergic systems and mask AC medication effects especially in the early stages of cognitive impairment. Due to complexity of the cognitive processes and their associations be tween pathobiological and clinical manifestations, there is limited understanding the mechanism of specific vulnerability to the AC medications (Lopez et al., 2002). Therefore, besides considering independent effects of cholinergic system, it is important to include the interactions between the neuromodulators and cholinergic system. However, the lack of information about neuromodulatory systems in the NACC dataset and the cognitive profile of older adults. Our study has several based on self report s . In addition, the medication history in the NACC dataset is limited to only two weeks before the visit date. An other limitation of the NACC medication records is the absence of frequency and dosages of medication. Verification of medication history using medical record would have provided additional information about usage, frequency and doses. Even if this informa tion was available, ADS categorization does not incorporate frequency of medication use or the dosages to assess AC drug burden. Additionally, this information was not important , as w e only examined the association between current AC medication use and co gnitive scores,. Moreover, AC the drug effect is related to the accumulated effects of varying degree of AC medications and not to the dosage of each individual medication . Despite these limitations , the NACC dataset is an appropriate resource to assess A C medication burden and cognitive function . We did not adjust for the treatment medications. This

PAGE 61

61 may be a limitation as there might be some potential for the medications to differentially affec t the cognition among the AC groups ; however, controlling for the diseases for which these medications were taken could offset their differential effects across the medication groups . As the enrollments of the participants in the study were limited to ADCs, these results may not be generalizable to overall population of older adults. The strengths of this study lies in its use of the NACC data from the large population based prospective study of older adults across the nation. The NACC includes one of the largest numbers of participants in the standardized longitudina l data, which allows us to parcel out a QCI population to study . Additionally, we were able to use an accepted scale to assess AC burden. The ADS scale is a reliable and validated method to assess AC medication load with prescription and over the counter m edications. Furthermore, the NACC data allowed us to take into account various competing causes of cognitive dysfunction in older adults by adjusting for socioeconomic, demographic, health, lifestyle, and genetic factors; thus limiting any potential confou nding. These results suggest that older adults with QCI taking medications with AC properties had higher cognitive scores than those not taking AC medications. W e cannot definitely rule out the possibility of cholinergic modulators and other unknown facto rs acting to preserve cognitive reserve in QCI . However, it is probable that the therapeutic effects of AC medications are beneficial to cognition among the QCI population who took AC medications compared to those who did not. More research is needed to un derstand and assess the impact of AC medications on QCI populations. In

PAGE 62

62 addition, the future researches should also consider including cognitive modulators in the study to identify clinically meaningful effects of AC exposure groups.

PAGE 63

63 Table 2 1 . Anticholinergic Drug Scale and associated anticholinergic activity ADS Levels Anticholinergic Drug Activity Study Levels Level 0 Drugs with no known anticholinergic properties, e.g. Loperamide No AC Drugs Level 1 Drugs with potential antic holinergic activity evidenced by receptor binding studies, e.g. Ampicillin Low AC Drugs Level 2 Drugs with anticholinergic adverse effects, usually at higher doses, e.g. Ranitidine High AC Drugs Level 3 Drugs with marked anticholinergic activity, e.g. Ox ybutynin Abbreviations: ADS=Anticholinergic Drug Scale

PAGE 64

64 Table 2 2. Composite cognitive domains Z score Cognitive Domains Cognitive Tests Z score # Domain score Global Cognitive score Memory Logic Memory IA Z1 ((Z1+Z2)/2)=C1 ((C1 + C2 + C3 + C4)/4) Lo gic Memory IIA Z2 Attention Digit Forward Number Z3 ((Z3+Z4+Z5+Z6)/4)=C2 Digit Forward Length Z4 Digit Backward Number Z5 Digit Backward Length Z6 Executive Trail Making A Z7 ((Z7+Z8+Z9)/3)=C3 Trail Making B Z8 Digital Symbol Z9 Language Animal List Z10 ((Z10+Z11+Z12)/3)=C4 Vegetable List Z11 Boston Naming Z12 # Z scores are adjusted for age, sex, and education

PAGE 65

65 Table 2 3 . Baseline characteristics of the study popu lation as a function of anticholinergic drug use AC Drugs Variable No AC Drugs (n=5871, 80.26%) Low AC Drugs (n=1148, 15.69%) High AC Drugs (n=296, 4.05%) P value P value for trend Age (M±SD) 74.46±7.81 76.59±8.23 75.55±7.80 <.0001 Gender, Male, N(%) 2896(49.33) 525(45.73) 108(36.49) <.0001 <.0001 Race, Black, N(%) 776(13.29) 154(13.47) 45(15.25) 0.6233 0.4253 Hispanic N(%) 495(8.47) 73(6.38) 27(9.12) 0.0503 0.2142 Education, (M±SD) 14.94±3.60 14.72± 3 .6 0 14.64 ± 3 . 29 0. 0270 Alcohol Abuse, N(%) 339(5.79) 67(5.85) 11(3.74) 0.3292 0.3185 Smoked> 100 Cigarette, N(%) 2701(46.45) 540(47.58) 136(46.74) 0.7839 0.6045 Body Mass Index, (M±SD) 27.08±5.05 27.50±5.49 27.43±5.69 0.0218 Living Alone, N(%) 1579(26.95) 358(31.27) 98(33.11) 0.0014 0.0004 Liv e Independently, N(%) 4514(77.22) 841(73.45) 220(74.58) 0.0166 0.0120 Residence Single, N(%) 5282(90.21) 988(86.44) 253(85.76) <.0001 <.0001 Married. N(%) 3764(64.62) 684(59.74) 167(56.80) 0.0004 <.0001 1 ° Relative Dementia. N(%) 2387(41.46) 407(36.37) 129(44.64) 0.0026 0.2009 Cardio Vascular Disease, N(%) 1761(30.01) 573(49.91) 71(23.99) <.0001 <.0001 Cardiac arrest, N(%) 374(6.76) 104(10.00) 18 (6.41) 0.0009 0.0286 Chronic Heart Failure, N(%) 149(2.69) 57(5.47) 4 (1.43) <.0001 0.0289 Cardi ac Other Diseases, N(%) 379(6.46) 89(7.76) 9(3.04) 0.0123 0.5484 Cardiac Fibrillation , N(%) 281(4.82) 316(27.72) 19(6.46) <.0001 <.0001 Cardiac Angioplasty , N(%) 495(8.45) 125(10.91) 18(6.08) 0.0067 0.4594 Cardiac Implant , N(%) 153(2.61) 104(9. 08) 6(2.03) <.0001 <.0001 Cardiac Bypass , N(%) 379(6.46) 89(7.76) 9(3.04) 0.0123 0.5484 Other Cardiac Disease , N(%) 686(11.80) 173(15.22) 29(9.93) 0.0026 0.1601 Cerebro Vascular Disease , N(%) 674(11.49) 189(16.49) 34(11.49) <.0001 0.0023 Strok e, N(%) 341(5.84) 106(9.28) 17(5.76) <.0001 0.0055 Transient Ischemic Attack, N(%) 327(5.63) 98(8.60) 18(6.21) 0.0007 0.0059 Other Cerebral Disease, N(%) 82(1.41) 15(1.32) 2(0.68) 0.5691 0.3669

PAGE 66

66 Table 2 3. Continued gics ; AC=Anticholinergics; MMSE=Mini mental state exam; M=Mean; SD=Standard deviation; WAISR=Wechsler Adult Intelligence Scale Revised. AC Drugs Variable No AC Drugs (n=5871, 80.26%) Low AC Drugs (n=1148, 15.69%) High AC Drugs (n=296, 4.05%) P value P value for trend Neurological Diseas e, N(%) 828(14.11) 169(14.72) 28(9.46) 0.0603 0.2127 Seizures , N(%) 139(2.37) 31(2.70) 7(2.36) 0.8015 0.6750 Brain T rauma Brief Unconsciousness , N(%) Ext. Unconsciousness , N(%) Chronic Deficit , N(%) 441(7.60) 246(4.23) 26(0.45) 86(7.58) 48(4.21) 8 (0.70) 11(3.74) 8(2.72) 2 (0.68) 0.0478 0.4494 0.4807 0.0793 0.3636 0.2636 1 85(3.16) 32(2.79) 14(4.75) 0.2329 0.5113 Depression, N(%) 1900(32.61) 489(44.97) 103(35.03) <.0001 <.0001 Metabolic Disease, N(%) 4923(83.88) 970(84.49) 226(76.35) 0.0021 0.0362 Hypertension, N(%) 3385(57.85) 736(64.28) 153(52.04) <.0001 0.2274 H ypercholesterolemia, N(%) 3375(58.12) 623(55.08) 146(50.00) 0.0060 0.0016 Diabetes, N(%) 928(15.86) 150(13.11) 47(15.99) 0.0608 0.1401 Vitamin B12 Deficiency, N(%) 310(5.40) 84(7.49) 11 (3.77) 0.0078 0.3482 Thyroid Disease, N(%) 1083(18.64) 237 (20.70) 65(22.11) 0.1096 0.0363 Incontinence, N(%) 898(15.32) 222(19.35) 105(35.47) <.0001 <.0001 APOE4 Genotype, N(%) 1481(43.24) 297(41.48) 76(38.78) 0.3546 0.1531

PAGE 67

67 Table 2 4. Baseline cognitive z score distribution of the study population as a function of anticholinergic dr ug use mental state exam; M=Mean; SD=Standard deviation; WAISR=Wechsler Adult Intelligence Scale Revised. AC Drugs Neuropsychological Tests No AC Drugs (n=5871, 80.26%) Low AC Drugs (n=1148, 15.69%) High AC Drugs (n=296, 4.05%) P value MMSE (M±SD) 1.07±2.14 1.04±2.05 0.79±1.84 0.0619 Logic Memory Immediate (M±SD) 0.76±1.19 0.66±1.23 0. 56±1.23 0.0004 Logic Memory Delayed (M±SD) 0.83±1.19 0.70±1.21 0.70±1.22 0.0006 Digit Forward Total Trial (M±SD) 0.45±1.04 0.41±1.03 0.35±1.09 0.0690 Digit Forward Length (M±SD) 0.42±1.07 0.41±1.07 0.36±1.13 0.3962 Digit Backward Total Tr ial (M±SD) 0.38±0.99 0.30±0.97 0.28±0.97 0.0067 Digit Backward Length (M±SD) 0.36±1.01 0.31±0.99 0.31±0.95 0.1046 Fluency Animal (M±SD) 0.71±1.01 0.71±0.96 0.69±0.96 0.6215 Fluency Vegetable (M±SD) 0.03±1.20 0.12±1.18 0.34±1.21 <.0001 Trai l Making A (M±SD) 0.79±1.72 0.98±1.94 0.89±2.01 0.0054 Trail Making B (M±SD) 1.13±1.67 1.31±1.72 1.23±1.75 0.0063 WAISR ^ Digit Symbol Score (M±SD) 0.43±1.15 0.51±1.13 0.40±1.24 0.4013 Boston Naming Test (M±SD) 1.09±1.67 1.01±1.60 1.27±1. 65 0.7051 Memory (M±SD) 0.80±1.15 0.68±1.17 0.63±1.18 0.0003 Attention (M±SD) 0.40±0.88 0.36±0.85 0.33±0.89 0.0412 Executive (M±SD) 0.48±0.80 0.58±0.91 0.56±0.91 0.0004 Language (M±SD) 0.58±0.97 0.53±0.93 0.54±0.97 0.0800 Composite (M±S D) 0.32±0.53 0.24±0.52 0.23±0.53 <.0001

PAGE 68

68 Ta ble 2 5. Effects of baseline covariates and association with global cognitive scores assessed by line a r regression Variable Model 1 Model 2 Model 3 Model 4 Est. P value Est. P value Est. P value Est. P value No AC Drugs Ref Ref Ref Ref Low AC Dru gs 0.0796 <.0001 0.07 48 <.0001 0.06 16 0.000 8 0.06 36 0.005 5 High AC Drugs 0.0907 0.0053 0.05 49 0.0 886 0.0 281 0.3 988 0.06 38 0. 1146 Age 0.0006 0.4637 0.000 0 0. 9545 0.000 6 0. 4926 0.000 1 0. 9236 Sex, Male 0.2 531 <.0001 0.2 586 <.0001 0.2 621 <.0001 Race , Black 0.1 684 <.0001 0.1 504 <.0001 0.1 492 <.0001 Hispanic 0.1 838 <.0001 0. 1816 <.0001 0. 2274 <.0001 Education 0. 0264 <.0001 0. 0266 <.0001 0. 0228 0.0001 Alcohol Abuse 0.0 0 8 7 0. 7588 0.0 100 0. 7346 0.002 7 0.9 445 Smoking > 100 Cigarette 0.0 453 <.000 5 0.0 444 0.000 8 0.0 46 1 0.00 68 Body Mass Index 0.00 42 0 .001 2 0.00 35 0.0 118 0.00 39 0.0 383 Live Independently 0.2 277 <.0001 0.2 375 <.0001 0.2 248 <.0001 Married 0.0 388 0.0 10 0.03 03 0.05 06 0.0 282 0.1 591 1 ° Relative Dementia 0.0 247 0 . 061 0.0 245 0. 0690 0.0 224 0. 1990 Cardio Vascular Disease 0.05 45 0.0002 0.05 42 0.00 43 Cerebro Vascular Disease 0.0 278 0.1 781 0.0 272 0. 4019 Parkinson Disease 0.18 25 <.0001 0.15 49 0. 5867 Neurological Disease 0.01 32 0.4 226 0.0 207 0. 6656 Hypertension 0.002 4 0.8 683 0.00 99 0. 8000 Hypercholesterolemia 0.02 27 0.09 84 0.00 76 0. 7978 DIABETES 0.02 03 0. 2900 0.0 066 0. 1469 Vitamin B12 Deficiency 0.02 87 0.3 092 0.00 97 0. 7978 THYROID Disease 0.0 303 0. 0737 0.0 315 0. 1469 I ncontinence 0.070 1 0.0001 0.06 8 2 0.00 35 Depression 0.02 61 0.0 644 0.02 64 0.1 462 APOE4 Genotype 0.09 6 3 <.0001 P value: Linear Trend 0.0454 0.0053 0.02 7 0.0 886 0.01 40 0.3 988 0.03 19 0. 1146 P value: Quadratic Trend 0.0114 0.1339 0.016 0.0 38 0 0.015 9 0.04 45 0.010 6 0. 2767 R Square 0.0040 0.1 369 0.1 466 0.1 518

PAGE 69

69 1. 2. Model 1 is adjusted for age. 3. Model 2 is adjusted for age, sex, race, ethnicity (Hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia. Model 3 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of de mentia, cardio vascular disease, cerebro vascular disease, Parkinson disease, neurological disease, hypertension, hypercholesterolemia, diabetes, vitamin b12 deficiency, thyroid disease, , incontinence, depression. 4. Model 4 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia, cardio vascular disease, cerebro vascular disease, Parkinson disease, neurological disease, hyperten sion, hypercholesterolemia, diabetes, vitamin b12 deficiency, thyroid disease, incontinence, depression, APOE4.

PAGE 70

70 Table 2 6. Effects of baseline covariates and association with memory scores assessed by line a r regression Variable Model 1 Model 2 Model 3 Model 4 Est. P value Est. P value Est. P valu e Est. P value No AC Drugs Ref Ref Ref Ref Low AC Drugs 0.1382 0.0004 0.11 2 4 0.00 43 0.0 780 0.05 63 0.08 2 8 0.10 8 6 High AC Drugs 0.1786 0.0120 0.11 82 0.12 63 0.05 14 0. 5180 0.0 267 0.7 833 Age 0.0088 <.0001 0.00 71 0.000 3 0.00 92 <.0001 0.008 4 0.00 1 9 S ex, Male 0.5 188 <.0001 0.5 303 <.0001 0.5 542 <.0001 Race, Black 0.2 288 <.0001 0. 2041 < .000 1 0.2 342 0.000 4 Hispanic 0.0165 0.7737 0.0 133 0. 8213 0. 1271 0. 1257 Education 0.0277 <.0001 0. 0274 <.0001 0. 0190 0. 0019 Alcohol Abuse 0.0 160 0.8002 0.0 352 0.5903 0.0 608 0. 4885 Smoking>100 Cigarette 0. 0991 0.000 6 0. 0944 0.00 14 0.08 31 0. 0301 Body Mass Index 0. 0222 <.0001 0.0 195 <.0001 0.023 1 <.0001 Live Independently 0.4 701 <.0001 0. 4782 <.0001 0.4 315 <.0001 Married 0.1 357 <.0001 0.11 45 0.000 9 0.09 13 0.0 424 1 ° Relative Dementia 0.0 029 0. 9214 0.00 66 0. 8264 0.04 72 0.2 287 Cardiovascular disease 0.118 5 0.0003 0.153 1 0.0003 Cerebrovascular disease 0.05 30 0.2 471 0.03 51 0.5 550 Parkinson Disease 0.192 1 0.022 6 0.04 52 0. 7091 Neurological Disease 0.077 5 0.06 7 3 0. 1 0 11 0.0 678 Hypertension 0.01 65 0. 6002 0.03 46 0. 3971 Hypercholesterolemia 0.034 4 0.25 98 0.02 22 0.5 758 Diabetes 0.0 786 0.0 660 0.0 182 0.7 541 Vitamin B12 Deficiency 0.06 82 0. 2742 0.0 40 8 0.63 09 Thyroid Disease 0.01 81 0. 6308 0.00 72 0. 8821 Incontinence 0.21 56 <.0001 0.19 83 0.0002 Depression 0.0 053 0. 8661 0.00 04 0.9 926 APOE4 Genotype 0.37 7 0 <.0001 P value: Linear Trend 0.0893 0.0120 0.059 1 0.12 63 0.02 57 0. 5180 0.01 33 0.7 833 P value: Quadratic Trend 0.0163 0.3294 0.01 78 0.3 126 0.017 4 0.33 95 0.02 32 0.3 043 R Square 0.0057 0.1 220 0.12 89 0.14 42

PAGE 71

71 2. Model 1 is adjusted for age. 3. Model 2 is adjusted for age, sex, race, ethnicity (Hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia. Model 3 is adjusted for age, sex, rac e, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia, cardio vascular disease, cerebro vascular disease, Parkinson disease, neurological disease, hypertensio n, hypercholesterolemia, diabetes, vitamin b12 deficiency, thyroid disease, , incontinence, depression. 4. Model 4 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia, cardio vascular disease, cerebro vascular disease, Parkinson disease, neurological disease, hypertension, hypercholesterolemia, diabetes, vitamin b12 deficiency, thyroid disease, incontinence, depression, APOE4. 5.

PAGE 72

72 Table 2 7. Effects of baseline covariates and association with attention scores assessed by linear regression Variable Model 1 Model 2 Model 3 Model 4 Est. P value Est. P value Est. P value Est. P value No AC Drugs Ref Ref Ref Ref Low AC Drugs 0.0306 0.2938 0.04 78 0.1 167 0.04 00 0. 2072 0.03 12 0.4 289 High AC Drugs 0.0709 0.1884 0.159 9 0.00 77 0.13 25 0.032 6 0.21 36 0.00 42 Age 0.0065 <.0001 0.00 38 0.0 105 0.003 2 0.0 438 0.003 2 0. 1185 Sex, Male 0.0 448 0.0 6 57 0.0 433 0.0 958 0.0 569 0.0 912 Race, Black 0.2 871 <.0001 0.2 706 <.0001 0.2 362 <.0001 Hispanic 0. 6060 <.0001 0. 6050 <.0001 0. 6596 <.0001 Education 0 .0 2 3 1 <.0001 0.0 252 <.0001 0. 0213 <.0001 Alcohol abuse 0.0 1 3 7 0. 7791 0.0 204 0. 6880 0 .0 293 0.6 625 Smoking > 100 cigarette 0.0 756 0.000 7 0.0 757 0.000 9 0.0 876 0.00 28 Body Mass Index 0.0 104 <.0001 0.00 95 <.0001 0.0 117 0.00 03 Live independently 0.2 555 <.0001 0.2 578 <.0001 0.2 455 <.0001 Married 0.00 26 0. 9199 0.0 149 0. 5778 0. 01 20 0. 7264 1° relative dementia 0.0 777 0.00 06 0.0 706 0.00 24 0.04 90 0. 1026 Cardio vascular disease 0.058 4 0.02 18 0.034 2 0.2 953 Cerebro vascular disease 0.05 60 0. 1161 0.0 110 0.80 96 Parkinson disease 0.09 40 0.15 33 0.098 9 0.2 8 9 9 Neurologic al disease 0.018 6 0.5 724 0.03 34 0.4 306 Hypertension 0.03 16 0. 1977 0.01 81 0. 5626 Hypercholesterolemia 0.015 9 0.5 021 0.037 7 0.21 32 Diabetes 0.03 99 0. 2290 0.0 068 0. 8798 Vitamin B12 Deficiency 0.06 86 0.1 569 0.06 83 0. 2927 Thyroi d disease 0.10 62 0.000 3 0.13 15 0.000 4 Incontinence 0.0131 0.675 0 0.017 6 0.6 623 Depression 0.020 0 0.41 26 0.02 79 0.3 733 APOE4 Genotype 0.03 21 0.2 8 7 3 P value: Linear Trend 0.0355 0.1884 0.0 800 0.00 77 0.066 3 0.032 6 0.10 68 0.00 42 P val ue: Quadratic Trend 0.0016 0.8986 0.0 1 0 7 0.4 308 0.008 8 0.5 372 0.025 2 0.14 50 R Square 0.0040 0.0 657 0.0 703 0.06 93

PAGE 73

73 2. Model 1 is adjust ed for age. 3. Model 2 is adjusted for age, sex, race, ethnicity (Hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia. 4. Model 3 is adjusted for age, sex, race, ethnicity (h ispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia, cardio vascular disease, cerebro vascular disease, Parkinson disease, neurological disease, hypertension, hypercholest erolemia, diabetes, V itamin B12 deficiency, thyroid disease, incontinence, depression. 5 . Model 4 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family h istory of dementia, cardiovascular disease, cerebrovascular disease, Parkinson disease, neurological disease, hypertension, hypercholesterolemia, diabetes, V itamin B12 deficiency, thyroid disease, incontinence, depression, APOE4. 6 . Statistical signific

PAGE 74

74 Table 2 8. Effects of baseline covariates and association with executive function scores assessed by linear regression Variable Model 1 Model 2 Model 3 Model 4 Est. P value Est. P value Est. P value Est. P value No AC Dru gs Ref Ref Ref Ref Low AC Drugs 0.0909 0.0010 0. 0629 0. 0217 0. 05929 0. 0361 0. 05715 0. 1018 High AC Drugs 0.0731 0.1501 0. 0246 0. 6236 0. 02473 0. 6306 0. 05236 0. 3964 Age 0.0067 <.0001 0. 0049 0. 0003 0. 00584 <.0001 0. 00648 0. 0004 Sex, Male 0. 1760 <.00 01 0. 18428 <.0001 0. 18934 <.0001 Race, Black 0. 4106 <.0001 0. 41173 <.0001 0. 41651 <.0001 Hispanic 0. 3291 <.0001 0. 32048 <.0001 0. 35795 <.0001 Education 0. 0117 0 . 0002 0. 01062 0 .000 9 0. 01051 0 . 0111 Alcohol Abuse 0. 0076 0. 8629 0. 02192 0. 6293 0. 00641 0. 9140 Smoking>100 cigarette 0. 0262 0. 1928 0. 02595 0. 2050 0. 02232 0. 3900 Body Mass Index 0. 0049 0. 0142 0. 00676 0. 0015 0. 00682 0. 0181 Live independently 0. 2631 <.0001 0. 23387 <.0001 0. 22780 <.0001 Married 0. 0424 0. 0702 0. 02865 0. 2306 0. 03141 0. 3031 1° Relative dementia 0. 0925 <.0001 0. 08753 <.0001 0. 08050 0. 0025 Cardiovascular disease 0. 02673 0. 2398 0. 01979 0. 4944 Cerebrovascular disease 0. 12497 <.0001 0. 12804 0. 0016 Parkinson disease 0. 37515 <.0001 0. 36133 <.0001 Neurological disease 0. 03876 0. 1881 0. 10949 0. 0036 Hypertension 0. 03995 0. 0682 0. 02758 0. 3189 Hypercholesterolemia 0. 00743 0. 7255 0. 00411 0. 8782 Diabetes 0. 03776 0. 2029 0. 01646 0. 6767 Vitamin B12 Deficiency 0. 1043 0 0. 0168 0. 09665 0. 0958 Thyroid disease 0. 03772 0. 1501 0. 05304 0. 1101 Incontinence 0. 01989 0. 4754 0. 01208 0. 7346 Depression 0. 08404 0. 0001 0. 08414 0. 0024 APOE4 genotype 0. 04786 0. 0732 P value: Linear Trend 0.0365 0.1501 0. 0123 0. 45 65 0. 01237 0. 6306 0. 02618 0. 3964 P value: Quadratic Trend 0.0181 0.1285 0. 0169 0. 0488 0. 01564 0. 1998 0. 01032 0. 4857 R Square 0. 0059 0. 0973 0. 1119 0. 1101

PAGE 75

75 protein E; Ref=Referent. 2. Model 1 is adjusted for age. 3. Model 2 is adjusted for age, sex, race, ethnicity (Hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia. Model 3 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia, cardiovascular disease, cerebrovascular disease, Parkinson disease, neurologic al disease, hypertension, hypercholesterolemia, diabetes, vitamin b12 deficiency, thyroid disease, incontinence, depression. 4. Model 4 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia, cardiovascular disease, cerebrovascular disease, Parkinson disease, neurological disease, hypertension, hypercholesterolemia, diabetes, Vitamin B12 deficiency, thyroid disease, incontinence, de pression, APOE4.

PAGE 76

76 Table 2 9. Effects of baseline covariates and association with language scores assessed by linear regression Language Variable Model 1 Model 2 Model 3 Model 4 E st. P value Est. P value Est. P value Est. P value No AC Drugs Ref Ref Ref Ref Low AC Drugs 0.0617 0.0547 0. 07020 0. 0326 0. 06833 0. 0451 0. 09027 0. 0323 High AC Drugs 0.0488 0.4103 0. 07804 0. 2281 0. 06240 0. 3495 0. 09242 0. 2460 Age 0.0014 0.3364 0. 00 133 0. 4079 0. 00194 0. 2580 0. 00116 0. 5990 Sex, Male 0. 27179 <.0001 0. 27564 <.0001 0. 25165 <.0001 Race, Black 0. 50987 <.0001 0. 48133 <.0001 0. 49969 <.0001 Hispanic 0. 44588 <.0001 0. 44245 <.0001 0. 53570 <.0001 Education,100 cigarette 0. 04161 0. 0839 0. 04211 0. 0873 0. 04333 0. 1672 Body Mass Index 0. 01029 <.0001 0. 01069 <.0001 0. 01153 0.0009 Live independe ntly 0. 46741 <.0001 0. 46902 <.0001 0. 47386 <.0001 Married 0. 01865 0. 5061 0. 02381 0. 4081 0. 00701 0. 8492 1° Relative dementia 0. 11263 <.0001 0. 10539 <.0001 0. 06767 0. 0353 Cardiovascular disease 0. 06826 0. 0126 0. 04656 0. 1828 Cerebrovascular dis ease 0. 01198 0. 7545 0. 03834 0. 4321 Parkinson disease 0. 09672 0. 1729 0. 12389 0. 2147 Neurological disease 0. 04023 0. 2564 0. 06546 0. 1499 Hypertension 0. 00929 0. 7246 0. 02028 0. 5440 Hypercholesterolemia 0. 02648 0. 2989 0. 01168 0. 7 189 Diabetes 0. 01091 0. 7603 0. 01848 0. 6994 Vitamin B12 deficiency 0. 07300 0. 1628 0. 01115 0. 8730 Thyroid disease 0. 04828 0. 1253 0. 04710 0. 2391 Incontinence 0. 04865 0. 1464 0. 08092 0. 0597 Depression 0. 01117 0. 6704 0. 01319 0. 6938 APOE4 genotype 0. 03144 0. 3308 P value: Linear Trend 0.0244 0.4103 0. 03902 0. 2281 0. 03120 0. 3495 0. 04621 0. 2460 P value: Quadratic Trend 0.0124 0.3709 0. 01039 0. 4798 0. 01237 0. 4175 0. 01469 0. 4262 R Square 0.0007 0. 1057 0. 1067 0. 1094

PAGE 77

77 1. Abbreviations: Est . = Parameter Estimates; APOE4, allele producing th Ref=Referent. 2. Model 1 is adjusted for age. 3. Model 2 is adjusted for age, sex, race, ethnicity (Hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia. Model 3 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia, cardiovascular disease, cerebrovascular disease, Parkinson disease, neurological disease, hypertension, hypercholesterolemia, diabetes, Vitamin B12 deficiency, thyroid disease, , incontinence, depression. 4. Model 4 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Bo dy Mass Index, living independence, marriage status, family history of dementia, cardiovascular disease, cerebrovascular disease, Parkinson disease, neurological disease, hypertension, hypercholesterolemia, diabetes, Vitamin B12 deficiency, thyroid disea se, incontinence, depression, APOE4. 5.

PAGE 78

78 CHAPTER 3 ANTICHOLINERGIC DRUG USE AND RISK OF COGNITIVE IMPAIRMENT AND DEMENTIA IN OLDER ADULTS WITH QUESTIONABLE COGNITIVE IMPAIRMENT: A LONGITUDINAL STUDY Background Dementia is a chronic disease that is ch aracterized by progressive decline in cognitive and physical functions (American Psychiatric Association, 2000). More than 7 million older adults suffer from dementia in the U.S., and this number is expected to increase many fold in the coming decades ( Heb ert, Weuve, Scherr, & Evans, 2013 ). About half of dementia patients suffer from two additional chronic medical illnesses , and they take more than five medications (Wolff, Starfield, & Anderson, 2002). With growing number of older adults and their medicatio n intake, there is a growing concern about medication side effects among older adults (Lyer et al., 2008; Sosa Ortiz, Acosta Castillo, & Prince, 2012). Anticholinergic (AC), a class of medication, use has been shown to adversely affect cognitive functions among older adults (Han et al., 2008; Disease (AD) is the most common type of dementia, accounting for 60 to 80 percent of total dementia cases (Mayeux & Stern, 2012; Nowrangi et al., 2011). AD is characterized by cortical choliner gic deficits, cognitive impairment, and dementia (Bohnen et al., 2007; Tiraboschi et al., 2000). The biological mechanisms like decrease in cholinergic neurons, reduced cholinergic signaling , and cholinergic receptor dysfunction are associated with occurre nce of dementia among older adults (Bird et al., 1983; Etienne et al., 1986; Henke and Lang, 1983). Furthermore, the extent of atrophy and degeneration of the brain cholinergic pathways is associated with the dementia severity among older adults (Barnes, 1 990; Gallagher and Nicolle, 1993). However, there is limited evidence about the effects of exogenous

PAGE 79

79 AC drug use on cognitive functions of older adults with questionable cognitive impairment (QCI). QCI is defined as subclinical cognitive weakness that ca n be detected using neuropsychological tests but that does not manifest any clinical or functional cognitive impairment. This suggests that the biological process of dementia begins years before the emergence of clinically detectable signs and symptoms (Pr ice & Morris, 1999) and creates a lengthy susceptible period during which exposure to risk factors like AC medications can induce significant harmful effects on cognition. Individuals with QCI are at higher risk of further cognitive damage leading to mild cognitive impairment (MCI) and dementia , especially due to AC drug use. Previous studies have shown that AC medication use increases among older adults and as a result there is an increased risk of adverse drug events (Flaherty et al., 2000; Hanlon et al. , 2006; Lyer et al., 2008; Marengoni et al., 2009; Onder et al., 2012; Sergi et al., 2011). Previous studies have documented that AC medication use has serious side effects on cognitive (Boustani et al., 2007; Campbell et al., 2009; Carriere et al., 2009; Fox et al., 2011) and physical functions (Han et al., 2008; Hilmer & Gnjidic, 2009; Koyama et al., 2013). Even though, considerable effort has been directed at monitoring and minimizing AC burden in older adults to avoid cognitive impairment (Drachman, 19 77), these medications are still commonly prescribed to treat medical illnesses in older adults (Peters, 1989; Wawruch et al., 2012). Although, AC medications are effective in treating specific clinical conditions, their antagonism of muscarinic receptors in the brain leads to a strong detrimental biological effect on cognitive functions (Chew et al., 2005; Peters, 1989; Nebes et al., 2007). Furthermore, as older adults experience lower central cholinergic

PAGE 80

80 activity and reduced cholinergic transmission, they are more susceptible to central anti muscarinic adverse effects and to the detrimental biological consequences of AC drug overload (Feinberg, 1993; Hammerlein et al., 1998). The current understanding of the cholinergic hypothesis of Geriatric Memory Dysf unction and the effects of AC medications on human cognition is mainly from the comparative studies of AC medication use in individuals with dementia and normal cognition (Corder et al., 1993; Jessen et al, 2010; Shah et al., 2013; Strittmatter et al., 199 3). There is no concrete evidence of the effects of AC medication on the cognitive outcome in older adults who take AC medications to manage their multi morbidity. Furthermore, there is an insufficient understanding of effects of AC drugs on cognitive func tions among older adults with QCI . Such research is especially important because individuals with QCI are likely at greater risk of developing dementia in future . While previous studies have shown that AC drug use is associated with cognitive decline in co gnitively normal older adults to some extent (Bottiggi et al., 2006; Campbell et al., 2009; Uusvaara et al., 2013), they have not assessed its effects in older adults with QCI. Therefore, it is imperative to understand the outcomes of AC drug use on cognit ive functions and their progression to dementia in older adults with QCI. The objective of this longitudinal analysis was to assess the association between AC medication use and incident dementia among older adults with QCI. We hypothesize that older adult s with QCI who have exposure to AC drugs at baseline have a higher incident rate of MCI/Dementia as compared to older adults with QCI who have no exposure to AC drugs.

PAGE 81

81 Methods Study Participants The cohort for this study was selected form a multi center longitudinal database the University of Washington that includes community dwelling older adults with and without cognitive impairment (Beekly et al., 2004). Thirty four collaborating Alzheimer disease Centers (ADCs) contribute and share data with NACC. A comprehensive NACC database, Uniform Data Set (UDS), has participant information , including standardized neuropsychological battery, clinical diagnosis, neuropathologic changes, biomarkers, and medication use information (Morris et al., 2006). The sample for the current study consists of individuals enrolled between September 2005 and March 2014. Inclusion and valid dementia severity tool, Clinical Dementia Rating Sum of Boxes (CD R SOB) scale (Rosenberg et al., 2012). We included only those individuals in the analysis who had information from at least two clinical visits. W e used Clinical CDR SOB to identify individuals with QCI. The CDR SOB scores range from 0 to18 , where 0 means cognitively normal and 18 means having dementia. The CDR SOB describes the stages of cognitive impairment and differentiates with reasonable accuracy between individuals with QCI and MCI/Dementia. For this study, older adults with QCI were individuals with a CDR Anticholinergic Exposure AC drug exposure was assessed using a valid and reliable (Tune et al., 1993) ordinal drug scale, Anticholinergic Drug Scale (ADS) (Carnahan et al., 2006), which includ es about 88 percent prescription and 89 percent over the counter medications.

PAGE 82

82 ADS scores are categorized into four levels according to increasing potency of anticholinergic activities (Table 3 1). based upon t heir AC medication use within two weeks prior to their visits. For this study, we categorized AC drugs into three levels (Table 3 1) Level 1 includes drugs with no known AC properties (No AC Drug), Level 2 indicates drugs with low potential AC activity see n by receptor binding study (Low AC Drug), and Level 3 was created by combining ADS level 2 and 3 , which represent s medications that have high AC adverse effects (High AC Drug). The No AC Drug group served as reference category. Cognitive Outcomes The NACC database includes the standardized Neuropsychological Test Battery (NTB) with 12 cognitive measures that assess cognitive functions among older adults. The NTB is sensitive enough to comprehend the progression from QCI to MCI/dementia and to differentiate the cognitive changes in specific domains. The estimated z scores for raw scores were originally calculated adjusting for sex, age, and education based upon model created by Shirk et al. (2011). Hayden et al. performed exploratory and confirmatory factor analysis on UDS database and grouped 12 neuropsychological measures into the cognitive domains of memory, attention, executive, and language (Hayden et al., 2011). Lower Trail Making A (TM A) and Trail Making B (TM B) scores demonstrate better executive fu nctions , whereas lower DSST (Digit Symbol Substitution Test) scores reveal poorer executive functions . T herefore, we reversed the sign of TM A and TM B scores and added them with DSST scores to assess executive functions such that the higher score suggeste d factor structure approach , and we grouped neuropsychological measures in aforementioned cognitive domains and a composite score for global cognition as shown

PAGE 83

83 in Table 3 2. The cognitive outcome was defines as change in cognitive score from questionable cognitive impairment, i.e. CDR SOB score of 0.5 2.5, to MCI/Dementia which was defined as CDR SOB score Covariates Previous studies have shown that higher age, lower education, females, and blacks are associated with higher incidence of dementia. Age, sex, race, and education were recorded for each participant , and we included th ese demographic factors in the model. In addition, we included lifestyle behaviors and conditions such as alcohol use, smoking, living situation, independence, and marriage status in the model to evaluate their effects on developing dementia. The lifestyle behaviors that affect cognition like smoking, alcohol consumption, and Body Mass Index were included as predictors cardiovascular and diabetes are known to contribute to cognitive bur den in older adults (Biessels, Staekenborg, Brunner, Brayne, & Scheltens, 2006; Newman et al., 2005) . Therefore, it was necessary to examine the burden of medical conditions on cognitive functions. Thus, we adjusted the models for medical conditions that a re associated with cognitive impairment such as cardiovascular disease, cerebrovascular disease, depression, neurological diseases , and diabetes . Genetic biomarker such as ApoE gene with its isoform E4 is strongly associated with cognitive impairment and d ementia (Deary et al., 2002; Farlow et al., 2004) and AD (Corder et al., 1993; Strittmatter et al, 1993). In addition, the previous studies have shown that the presence of APOE4 further increases the risk of dementia due to AC medication exposure (Uusvaara et al., 2009). Therefore , we included the APOE4 in the model to assess its effects on cognitive functions in the presence of AC drug use in older adults with QCI.

PAGE 84

84 Statistical Analyses We performed descriptive analysis to assess distribution of means and proportions of baseline characteristics such as demographics, lifestyle behaviors, comorbidities, genetic factor, and cognitive scores by AC drug load levels, i.e., no, low, and high AC medicati o n use, measured using ADS scale. We analyzed variance to eva luate the means differences in continuous variables , and we used a chi squared test to assess differences in the distribution of categorical variables. Descriptive statistics compare and provide the significance of the distribution of baseline factors and cognitive measures among AC drug use levels. We performed Time to Event Analysis (MCI/dementia) controlling for baseline characteristics to determine the temporal trend and hazard of conversion of older adults with QCI into MCI/dementia due to AC exposure compared to no AC exposure. We eliminated from the study the participant at the time of their last evaluation or last recorded information if they died or dropped out of the study without developing MCI/dementia. We plotted the survival probability of not having a dementia event using Kaplan Meier curves. Finally, we performed the multivariable Cox proportional hazard regression model to investigate the baseline factors and model variations to better understand the cognitive change due to AC exposures and t he incidence of MCI/dementia among older adults with QCI, controlled for baseline global cognitive score, demographics, lifestyle behaviors, comorbidities , and APOE4 genetic biomarkers. We calculated hazard ratio (HR) and Wald 95 percent confidence interva l (CI) for each variable included in the regression model. The AC exposures were assessed using ADS scale and categorized into no, low, and high AC drug us e, as explained above in Table 3 1. The Cox regression method is a robust method to

PAGE 85

85 investigate time taken to develop MCI/dementia that handles discrete time. We performed log rank test to estimate the statistically significant difference between cognitive outcomes due to the AC drug exposures based upon low and high AC drug use levels as compared to no A C drug exposure. We believe that there was no non informative censoring which means that the censoring of study participant do not depend upon development of dementia during the follow up. Age, sex, race, education, lifestyle factors, health conditions, an d genetic factors were included as covariates in the model. The statistical analyses were performed using SAS 9.4 , and statistical significance was achieved at alpha= 0.05. Results A total of 5,027 participants met the inclusion criteria and were included in the study analysis. Of the total, 80.05 percent took medication without any AC property (n=4,024), 15.91 percent (n=800) took low levels and 4.04% (n=203) took medication with high AC activity. The mean age ± standard deviation (SD) of the sample was 74 .92±7.80 years (range: 60 109 years), 88.52 percent of the individuals were white, 92.30 percent were non hispanic, and 51.708 percent were females. About 91.75 percent of the individuals had high school or higher education, 65.46 percent were married, and 77.00 percent were living independently. About 46.24 percent of the population smoked cigarettes , and 5.14 percent consumed alcohol. At baseline, about 20 percent of the participants took medications with AC activity and ( similar to previous stud ies) abou t 4 percent took medications with high AC activity (Fox et al., 2011). Table 3 1 presents t he population characteristics across no, low , and high AC medication groups. As compared to the no AC drug group, individuals in high (n=800) and low (n=203) AC dru g groups were significantly older (75.42±7.39 and 76.85±8.14

PAGE 86

86 vs . 74.52±7.69; p <.0001) and had slightly higher BMI (27.47±5.50 and 27.22±5.31 vs . 27.06±4.84; p=0.1982). In comparison to the no AC group, there were significantly fewer males (35.96 percent a nd 46.63 percent vs . 50.02 percent; p <.0001) , and fewer married individuals (61.19 percent and 62.11 percent vs . 66.35 percent; p=.0108) in high and low AC groups, respectively. There were higher proportions of blacks, smokers, and individuals with family history of dementia in the high AC group , but these proportions were not significantly different from those in no AC group. The prevalence of cardiovascular disease a mong individuals in the low AC groups (50 percent) was greater than that in no and high AC groups (No AC: 29.63 percent; High AC: 22.17 percent, P value=<.0001). This is due to the reason that various medications given for the treatment of cardiovascular disease such as , Atenolol, Hydralazine, and Digoxin, have no to low anticholinergic prop erties. Similarly , newer treatment regimens for depression involve use of selective serotonin receptor blockers such as Fluoxetine. These medications have low anticholinergic properties and fewer side effects. There fore, we found a significantly higher pre valence of depression among individuals who took medications with low AC medications groups (41.87 percent) as compared to no and high AC medications groups, respectively (31.49 percent and 33.66 percent; P value= <.0001) . As anticipated , there was a signi ficantly higher prevalence of incontinence among the high AC group ( 36.45 percent) and low AC group ( 17.15 percent) as compared to no AC group (15.0 percent; P value= <.0001) . On the contrary, individuals who took medications with high AC properties had a lower prevalence of APOE4 genotype (37.58 percent) as compared to individuals in no or low AC groups (No AC: 43.12 percent and Low AC: 41.19 percent; P value=0.303) and it was not

PAGE 87

87 statistically significant. Similarly, individuals who took high AC medicati ons had greater medications but it was statistically significant (No AC: 2.85 percent and Low AC: 2.63 percent; P value=0.195). The global and domain specific cognitive z scores were different across the AC medication groups. The mean global cognition z scores among high ( 0.17±0.57) and low ( 0.29±0.59) AC medication group was significantly higher than those in no ( 0.33±0.62; p<.0001). Similarly, the mean z scores of memory, att ention, executive function, and language domains were significantly higher in high and low AC medication groups as compare to no AC medication group (Table 3 4). These results indicate that the older adults with QCI who take medications with AC properties have better cognition than those who do not. We used the Kaplan Meier survival method to estimate dementia free survival times of the participants among the AC medication groups. The maximum follow up time was about 8.15 years , and the median follow up ti me was 2.09 years. Time to event analysis showed that out of the 5,027 older adults with QCI at baseline, 2003 ( 39.84 percent) percent developed MCI/dementia. The rate of cumulative incident dementia among high, low and no AC medications groups was about 3 9.59 percent, 41.13 percent and 39.90 percent, respectively. The median survival time for no AC, low AC and high AC groups were 2.09, 2.08, and 2.45, respectively. The rate of incident MCI/Dementia per 1000 person years were 150, 154, and 130 in no, low an d high AC groups, respectively. The overall median time from enrollment into the study to the incidence of dementia was 4.27 years. Figure 3 1 Kaplan Meier survival curves,

PAGE 88

88 adjusting for confounders, compare s the incidence of dementia among AC mediation gr oups. The curves were statistically similar ( log rank test: P =0.415) among AC groups. Tables 3 5 illustrate d the age adjusted hazards of development of MCI/dementia (Table 3 5 ) among older adults with QCI . The risk of dementia among i ndividuals who took l ow (HR=1.043 f; p=0.450) or high (HR=0.925; p=0.509) levels of AC medications were similar to those who did not tak e AC medications . The final model adjusted for demographic factors, lifestyle behaviors, comorbidities , and APOE profile showed that the haza rds of incident MCI/dementia for low (HR: 0.897 ; P=0.177 ) and high (HR: 0.806 ; P=0.130 ) AC group s were lower a s comp a red to older adults with QCI who did not take AC medication . Discussion To our knowledge, this was the first longitudinal based study to ev aluate the association of AC medication use and incident dementia among older adults with QCI. The aim of this study was to assess the effects and magnitude of AC drug use on the development of dementia among older adults with QCI. We hypothesized that old er adult s with QCI who have exposure to AC drugs would have a more rapid decline in cognition score as compared to older adults who have no exposure to AC drugs. Contrary to our hypothesis, the findings indicated that exposures to AC medication use in olde r adults with QCI were not associated with increased progression to MCI/dementia in comparison to no AC medication exposure, but the association was not statistically significant. The model was adjusted for demographic factors, lifestyle behaviors, clinica l conditions, and genetic biomarkers that are known to influence the cognitive outcome in older adults.

PAGE 89

89 Our results suggest that older adults with QCI had lower cognitive scores over the eight years of follow up time but the rates of cognitive declines we re not significantly different among AC medication groups. Unlike others, our population did not show a dose response relationship between AC drug use and cognition (Chew et al., 2005; Pasina et al., 2013; Uusvaara et al., 2009). The previous stud ies found that individuals who did not have dement ia we re significantly less susceptible to the adverse cognitive effects of AC medications than those who had cognitive impairment , and that AC medications use did not deteriorate cognition over 18 months (Pasina et al., 2013; Sunderland et al., 1987). In addition, Luukkanen et al. (2011) showed that AC medication use was not independently associated with adverse outcomes like delirium and mortality. Therefore, it is likely that the other conditions like advance age, lower education , comorbidities , or genetic factors might contribute significantly more to the cognitive impairment than AC medications. Kersten et al. (2013) demonstrated that reduc ed AC medication burden did not improve cognition among nursing home indivi duals . This study emphasized that many inherent factors such as comorbidities play a very significant role in cognitive impairment. Hence, results from our study highlight a specific population that theoretically would be susceptible to the harmful effects of AC medications , but may benefit from careful monitoring and use of AC medication s for treatment of comorbidities. A major difference between our study and the previous studies is in the selection of the populations. We chose older adults who have belo w normal cognitive scores but had no clinical signs or symptoms of cognitive reduction. This might be attributed to the compensatory preservation of cognitive functions during the early stage of cognitive

PAGE 90

90 decline. Counts, Perez, and Mufson (2008a) showed t hat hyper innervation and proliferation of forebrain galaninergic fibers and receptors help in preventing cognitive functions , especially in the early stage of cognitive impairment. Therefore, it is likely that our study population had compensatory respons e s to preserve cognitive functions that would have helped in modulation of cognitive response , even in the presence of AC medications. Furthermore, there are many neuromodulators , like norepinephrine, dopamine, serotonin, GABA, opioid peptides, galanin, su bstance P, and angiotensin which might preserve cognitive functions and prevent cognitive decline (Decker and McGaugh, 1991). Additional compensatory mechanisms like increase in forebrain cholinergic neurons, upregulation of choline acetyltransferase activ ity , and an increase in acetylcholine synthesizing enzyme also play a significant role in modulating the cognitive response in early stages of cognitive impairment (DeKosky et al., 2002; Frölich, 2002; Ikonomovic et al., 2003). These compensatory mechanism s may help preserve cognitive functions and delay cognitive impairment in early stages of cognitive decline. In our study, the population of older adults with QCI was at the early stage of cognitive impairment and therefore, participants might have preser ved cognitive functions. However, there is a great variability in AC drug sensitivity related to multi morbidity which is associated with early stage of cognitive impairment. Therefore, the s susceptibility to comorbidity in cognitive impairment play s a vital balancing act in older adults. Previous studies by Oberauer (2001, 2002) and Naveh Benjamin et al. (2007) have also provided evidence that the comorbidit ies alter the cholinergic functi oning and decrease the ability

PAGE 91

91 of older adults to use the frontal brain mediated compensatory control of cognitive functions. Therefore, we speculate that the cognitive impairment in our population was primarily due to comorbidities , and the therapeutic ac tions of AC medications reduced the cognitive burden of comorbidity in the older adults , especially in the early stage of cognitive impairment. In keeping with our premise, recent community based cohort studies by Qiu et al. (2006) have shown an overall de crease in incident dementia among cognitively normal older adults who were treated with anti hypertensive drugs. Additionally, the previous studies have demonstrated that a highly potent AC drug, clozapine, improved attention, reaction time, processing spe ed, executive functions, working memory, verbal fluency, and visual learning and memory in schizophrenics (Meltzer and McGurk, 1999). Similarly, other studies have shown that anti hypertensive drugs like angiotensin converting enzyme (ACE) inhibitors and c alcium channel blockers (CCBs) slow the progression of dementia (Hajjar et al., 2005; Hanon et al., 2006; Khachaturian et al., 2006; Ohrui et al., 2004). This evidence demonstrate s that older adults with QCI taking AC medications may benefit from the thera peutic effects of the AC medications by alleviating the cognitive burden due to managing comorbidities. Previous stud ies have shown that the during the early phase of cognitive impairment, increased activation of the cholinergic system compensates for cogn itive impairment by increasing the frontal brain mediated knowledge based detection and filtering mechanisms, i.e., top down processes (Dumas & Newhouse, 2011). Hence, it is important to note that older adults with intact cholinergic functions will have gr eater compensat ory benefit from the activation of frontal networks whereas those with advanced dementia would have lost the incentive to benefit from the compensatory

PAGE 92

92 mechanisms . Dumas and Newhouse (2011) found that scopolamine , a muscarinic receptor antag onists, and mecamylamine, a nicotinic receptor antagonists, improved frontal and hippocampal activations as compared to a placebo , providing evidence for compensation by top down processes during the early phase of cognitive impairment (Dumas et al., 2010) . Therefore, older adults within the range of questionable cognitive impairment as compared to MCI/ dementia gain from compensatory mechanisms that support cognitive reserve of an individual ; h owever, additional cognitive insults may further impair the cogn ition. If the cognition levels decreases below questionable cognitive impairment threshold, it is very difficult to regain the cognition back in the range of questionable cognitive impairment. In fact, at this stage an individual has MCI and has greater li kelihood of dementia in the future. use was self report ed drug intake. Furthermore, the information about medication use was limited to only two weeks prior to the visit dat e. In addition , we lacked information about medication use frequency and dosages. Confirmation of medication use by medical record would have further contributed to the knowledge about medication effect on cognition. Despite these limitations, the AC drugs used in this analysis are typically taken chronically to alleviate common clinical conditions. Therefore, we expect this may only be a minor limitation of the study. T he NACC is a large database repository which collected population based prospective data from older adults across the nation. The neuropsychological tests and the cognitive scores in the NACC database were standardized to the national

PAGE 93

93 population , which ena bles us to accurately define the QCI population. Moreover, we used reliable and validated method to measure AC drug use and burden that covers more than 95 percent of AC prescription and over the counter medications. In addition, the annual drug records pr ovide complete information to estimate the extent of cognitive impairment due to chronic AC use. Overall, the NACC data provide sufficient information to address the hypothesis stated for anticholinergic drug use and exposure. We conclude in this longitud inal analysis that the rate of incident dementia in older adults with QCI who take drugs with AC properties was similar to that of older adults who did not take AC medications. In addition, participants who took low or high AC medications were slightly les s likely to develop dementia over eight year s but it was not statistically significant . More research needs to be done to identify the most relevant cholinergic modulators and mechanisms that modulate the cognitive profile in the older adults. In addition, it is important to simulate the results in other populations and to identify the factors that preserve cognition. We obviously cannot eliminate the possibility that cholinergic modulators that preserve cogniti on in the early stage of cognitive impairment; it is likely that the therapeutic effects of AC drug can explain the lower hazard of incident dementia among QCI population which took AC drugs in comparison to who took drugs that were not AC medication s . The future researche r should focus on mediators o f cognitive decline and modulators that preserve cognitive functions in order to provide a n independent future to our elders.

PAGE 94

94 Table 3 1. Anticholinergic Drug Scale and associated anticholinergic activity ADS Levels Anticholinergic Drug Activity Study Lev els Level 0 Drugs with no known anticholinergic properties, e.g. Loperamide No AC Drugs Level 1 Drugs with potential anticholinergic activity evidenced by receptor binding studies, e.g. Ampicillin Low AC Drugs Level 2 Drugs with anticholinergic adverse effects, usually at higher doses, e.g. Ranitidine High AC Drugs Level 3 Drugs with marked anticholinergic activity, e.g. Oxybutynin Abbreviations: ADS= Anticholinergic Drug Scale

PAGE 95

95 Table 3 2. Method of creating specific cognitive domains from z s cores of individual neuropsychological tests Cognitive Domains Cognitive Tests Z score # Domain score Global Cognitive score Memory Logic Memory IA Z1 ((Z1+Z2)/2)=C1 ((C1 + C2 + C3 + C4)/4) Logic Memory IIA Z2 Attention Digit Forward Number Z3 ((Z3+Z4 +Z5+Z6)/4)=C2 Digit Forward Length Z4 Digit Backward Number Z5 Digit Backward Length Z6 Executive Trail Making A Z7 ((Z7+Z8+Z9)/3)=C3 Trail Making B Z8 Digital Symbol Z9 Language Animal List Z10 ((Z10+Z11+Z12)/3)=C4 Vegetable Lis t Z11 Boston Naming Z12 # Z scores are adjusted for age, sex, and education

PAGE 96

96 Table 3 3. Baseline characteristics of the study population as a function of anticholinergic drug use 1. AC=Anticholinergics; SD=Standard deviation; M=Mean. AC Drugs (N=5,027) Variable No AC Drugs (n=4024, 80.05%) Low AC Drugs (n=800, 15.91%) High AC Drugs (n=203, 4.04%) P value P value for trend Age (M±SD) 74.52±7.69 76.85±8.14 75.42±7.39 <.0001 Gender, Male (N, %) 2013(50.02) 373(46.63) 73(35.96) 0.0002 <.0001 Race, Black (N, %) 459(11.45) 92(11.54) 24(11.88) 0.9804 0.8537 Hispanic (N, %): Yes 327(8.13) 41(5.13) 19(9.36) 0.0097 0.1533 Education (M±SD) 15.08±3.53 14.99±3.39 14.65±3.23 0.1088 Alcohol Abuse (N, %) 203(5.06) 48(6.01) 7(3.45) 0.2895 0.9980 Smoked>100 Cigarette (N, %) 1843(46.18) 364(46.18) 96(48.00) 0.8787 0.7329 Body Mass Index (M±SD) 27.06±4.84 27.22±5.31 27.47±5.50 0.1982 Live Independently (N, %) 3110(77.65) 587(73.65) 156(77.23) 0.0494 0.0917 Married (N, %): Yes 2650(66.35) 495(62.11) 123(61.19) 0.0305 0.0108 1 ° Relative Dementia (N, %) 1697 (42.93) 296(37.85) 96(48.98) 0.0051 0.5974 Cardiovascular Disease (N, %) 1192(29.63) 400(50.00) 45(22.17) <.0001 <.0001 Cerebrovascular Disease (N, %) 459(11.41) 129(16.13) 23(11.33) 0.0009 0.0169 Neurological Disease (N, %) 531(13.21) 118(14.75) 18(8.8 7) 0.0840 0.6204 114(2.85) 21(2.63) 10(4.95) 0.1946 0.3109 Depression (N, %) 1259(31.49) 332(41.87) 68(33.66) <.0001 <.0001 Hypertension (N, %) 2263(56.35) 495(61.95) 103(51.24) 0.0033 0.3734 Hypercholesterolemia (N, %) 2333( 58.52) 440(55.77) 103(51.50) 0.0663 0.0209 Diabetes (N, %) 601(14.97) 100(12.53) 34(16.92) 0.1339 0.5358 Vitamin B12 Deficiency (N, %) 205(5.21) 49(6.21) 6(2.99) 0.1721 0.8152 Thyroid Disease (N, %) 764(19.17) 166(20.78) 39(19.40) 0.5768 0.4602 Inconti nence (N, %) 603(15.00) 137(17.15) 74(36.45) <.0001 <.0001 APOE4 Genotype (N, %) 1203(43.12) 236(41.19) 59(37.58) 0.3026 0.1288

PAGE 97

97 Table 3 4. Baseline cognitive scores of the study population as a function of ant icholinergic drug use 1. Abbreviations: AC=Anticholinergics; SD=Standard deviation; M=Mean. 2. Statisti AC Drugs (N=5,027) Cognition No AC Drugs (n=4024, 80.05%) Low AC Drugs (n=800, 15.91%) High AC Drugs (n=203, 4.04%) P value Global (M±SD) 0.33±0.62 0.29±0.59 0.17±0.57 <.0001 Memory (M±SD) 0.76±1.33 0.70±1.31 0.46±1.29 <.0 001 Attention (M±SD) 0.43±0.91 0.39±0.88 0.33±0.90 0.0024 Executive (M±SD) 0.49±0.85 0.55±0.86 0.56±0.93 0.0015 Language (M±SD) 0.65±1.11 0.61±1.04 0.45±1.02 <.0001

PAGE 98

98 Figure 3 1. Kaplan Meier survival curves of dementia free survival of individuals with or without incident dementia in older adults taking medication with low and high AC properties relative to the older adults t aking medications without AC properties.

PAGE 99

99 Table 3 5. Multivariable Cox proportional hazard regression models of time to incident dementia for global cognition Variable Model 1 (n=4796) Model 2 (n= 4286) Model 3 (n= 4123) Model 4 (n= 2918) HR P value HR P value HR P value HR P value No AC Drugs Ref Ref Ref Ref Low AC Drugs 1.043 0.4995 0.9 80 0. 7665 0. 932 0. 3138 0. 904 0. 2082 High AC Drugs 0.925 0.5089 0. 857 0.2 180 0. 843 0. 1875 0. 815 0. 15 15 Baseline cognitive score 0.336 <.0001 0.3 39 <.0001 0. 334 <.0001 0. 332 <.0001 Age 1.023 <.0001 1.02 1 <.0001 1. 022 <.0001 1. 026 <.0001 Gender, Male 0. 806 <.0001 0. 813 0 . 0004 0. 786 0 . 0004 Race, Black 0. 749 0. 0016 0. 739 0. 0017 0. 689 0. 0013 Hispanic 0. 481 <.0001 0. 460 <.0001 0. 472 <.0001 Education 0 . 950 0. 0001 0 . 950 <.0001 0 . 946 <.0001 Alcohol Abuse 1. 191 0. 1340 1. 208 0. 1090 1. 263 0. 0813 Smoked>100 Cig 1. 046 0. 3639 1. 045 0. 3930 1.00 0 0. 9964 Body Mass Index 0. 971 <.0001 0. 969 <.0001 0. 97 0 <.0001 Live Independently 0. 441 <.0001 0. 446 <.0001 0. 432 <.0001 Married 1. 299 <.0001 1. 256 0.0003 1. 350 <.0001 1° Relative Dementia 1. 250 <.0001 1. 251 <.0001 1. 252 0.0002 Cardiovascular disease 1. 013 0. 8191 1. 052 0. 4481 Cerebrovascular di sease 1. 162 0. 0488 1. 159 0. 0960 Parkinson disease 1. 276 0. 0844 1. 213 0. 2837 Neurological disease 0. 871 0. 0749 0. 844 0. 0607 Hypertension 0. 962 0. 4702 0. 996 0. 9509 Hypercholestrolemia 0. 991 0. 8601 0. 939 0. 2995 Diabetes 1. 065 0. 43 04 1. 114 0. 2477 Vitamin B12 deficiency 1. 131 0. 2810 1. 118 0. 4108 Thyroid disease 1 . 009 0. 8946 0. 949 0. 4970 Incontinence 0. 983 0. 8107 1 . 001 0. 9950 Depression 1. 234 <.0001 1. 271 0 .0001 APOE4 1. 427 <.0001 P value for Trend 0.9669 0. 2807 0. 1117 0. 0682

PAGE 100

100 1. Abbreviations: AC=Anticholinergics; Ref=Referent. HR=Hazard Ratio apolipoprotein E. 2. Model 1 is adjusted for age. 3. Model 2 is adjusted for age, sex, race, ethnicity (hispanic), educat ion level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia . 4. Model 3 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living indepe ndence, marriage status, family history of dementia, cardiovascular disease, cerebrovascular disease, Parkinson disease, neurological disease, hypertension, hypercholesterolemia, diabetes, Vitamin B12 deficiency, thyroid disease, , incontinence, depress ion . 5. Model 4 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia, cardiovascular disease, cerebrovascular disease, Parkinson d isease, neurological disease, hypertension, hypercholesterolemia, diabetes, Vitamin B12 deficiency, thyroid disease, incontinence, depression, APOE4.

PAGE 101

101 CHAPTER 4 LONG TERM EXPOSURE TO ANTICHOLINERGIC DRUG USE AND RISK OF COGNITIVE IMPAIRMENT AND DEMENTIA IN OLDER ADULTS WITH QUESTIONABLE COGNITIVE IMPAIRMENT: A MIXED MODEL APPROACH Background Dementia is c haracterized by progressive decline in memory and language, reduced ability for planning and organizing, failure to recognize or identify objects, and impaired ability to carry out motor activities despite intact motor functions (American Psychiatric Assoc iation, 2000). About 60 to 80 percent of dementia in the older adults is D isease (AD) (Mayeux & Stern, 2012; Nowrangi et al., 2011). Currently in the U.S. , about 7 million individuals suffer from dementia; and with an increasing popul ation of older adults, this number if going to increase steadily in the next several decades (Herbert et al., 2003). Research has shown that around 50 percent of dementia patients have at least two other chronic illnesses and that these patients consume ov er five medications daily (Wolff et al., 2002). With the growing number of elderly population and the trend of increasing medication intake with age, adverse drug effects put a great burden on both health professionals and family members. As a result , ther e is growing concern about medication side effects among older adults (Iyer et al., 2008; Sosa Ortiz et al., 2012). Although various medications can cause adverse effects, medications with anticholinergic (AC) properties are suspected to produce most adve rse cognitive outcomes among older adults (Han et al., 2008; Tune, 2000). Elderly individuals are more vulnerable to anticholinergic side effects due to decrease in number of cholinergic neurons, reduced cholinergic signaling, diminished cholinergic recept or functions , and greater permeability of the blood brain barrier (Bird, Stranahan, Sumi, and Raskind,

PAGE 102

102 1983; Etienne et al., 1986; Henke and Lang, 1983). In addition , dementia severity among older adults is associated with the degree of degeneration of the brain cholinergic pathways (Barnes, 1990; Gallagher and Nicolle, 1993). However, there are is limited evidence about the effects of long term use of exogenous AC medications on cognitive functions among older adults with questionable cognitive impairment (QCI). Older individuals with QCI have very mild subclinical cognitive limitations that can be unmasked on neuropsychological tests but do not present with any clinical or functional cognitive impairment. This latent period of natural progression of dement ia obscure s the process that begins years before the occurrence of clinically detectable signs and symptoms (Price and Morris, 1999). It is important to note that this lengthy period which provides the best opportunity to stem the cognitive decline is also the most vulnerable time for the adverse cognitive effects of AC medications for the elderly . These individuals are at greater risk of added cognitive damage of AC medications leading to dementia in the future. E vidence suggest s that AC medications are co mmonly prescribed to the older adults (Peters, 1989; Wawruch et al., 2012), which diminishes cognition (Boustani et al., 2007; Campbell et al., 2009; Carriere et al., 2009; Fox et al., 2011 ) and puts the elderly at a greater risk of development of dementia (Flaherty et al., 2000; Hanlon et al., 2006; Onder G et al., 2012). Despite constant efforts to monitor and minimize AC use, these medications are still regularly prescribed and bought over the counter (Wawruch et al., 2012; Peters, 1989). Several popula tion based epidemiological studies have shown that AC drugs are associated with poor cognitive functions. Most of these studies have compar ed AC intake of individuals with dementia and normal cognition and demonstrated association

PAGE 103

103 between AC medication use and cognitive impairment (Fox et al., 2011; Jessen et al., 2010; Konishi et al., 2010; Lechevallier Michel et al., 2005; Mulsant et. al., 2003; S hah et al., 2013; Uusvaara et al., 2013). However, there is little literature on the long term effects on cogn itive outcome of AC medication in older adults with QCI, although these people are at greater risk of developing dementia in future . However, many mechanisms like AC mediated antagonism of muscarinic receptors in the brain, lower central cholinergic activi ty , and reduced cholinergic transmission have been proposed11,12(Chew et al., 2005; Feinberg, 1993; Hammerlein et al., 1998; Nebes et al., 2007; Peters, 1989 ). Nevertheless , researchers do not clearly understand the detrimental biological mechanisms by whi ch AC drugs mediate their effects on brain and cognitive functions in older adults with QCI. Hence, it is important to understand the cognitive outcomes of AC medication use in older adults with QCI. The objective of this longitudinal analysis was to asses s the association between long term AC medication use and the cognitive profile among older adults with QCI. We hypothesize d that older adults with QCI who have exposure to AC drugs at baseline have a more rapid decline in global and domain specific cognit ion as compared to older adults with QCI who have no exposure to AC drugs. Methods Study Participants Center (NACC). NACC is a research institution located at the University of Washington that collects and maintains the multi center longitudinal database of community dwelling older adults enrolled in the study through one of the 34 Alzheimer Disease Centers (ADCs) (Beekly et al., 2004, 2007). NACC includes a comprehensive datab ase, the

PAGE 104

104 Uniform Data Set (UDS) , that has participant information including a standardized neuropsychological battery, clinical diagnosis, neuropathologic changes, biomarkers, and medication use (Morris et al., 2006). This study includes individuals who we re enrolled between September 2005 and March 2014. The inclusion criteria for this study were age 60 years or older, medication records at baseline, and QCI on Clinical Dementia Rating Sum of Boxes (CDR SOB), i.e., score from 0.5 to 2.5 ( 2 008 ). In addition, the researchers required every participant to have at least two visits so as to compare the change in cognitive scores over multiple visits. Anticholinergic Exposure We used an ordinal drug scale, Anticholinergic Drug Scale (ADS), to as sess the AC medication exposure in the study population (Carnahan et al., 2006). The ADS is a valid and reliable scale that includes about 88 percent of total prescription medications and about 89 percent over the counter AC medications consumed by the old er adults (Carnahan, Lund, Perry, & Pollock, 2001) . The ADS has been validated against serum anticholinergic activity, a gold standard tool to quantify AC burden. ADS total score was significantly associated with SAA (F=20.82, P<.0001) and explained 9.5% o f variance in SAA (R 2 =.0947); therefore, showed some degree of criteria validity with SAA (Tune et al., 1992). Unlike SAA, ADS scoring does not require specialized laboratory and can be easily applied to clinical or research setting and help guide clinical interventions to reduce AC burden. ADS characterize medications into four levels based upon their increasing order of anticholinergic potency (Table 4 1). drug burden based upon the information of their AC medication use wit hin past two weeks prior to their annual follow up visits. In this study, we combined the Level 3 and 4 of the ADS scale and categorized AC drugs into a three level scoring system (Table 4 -

PAGE 105

105 1). L evel 1 indicates medications that do not have known AC propert ies (No AC drug group), the Level 2 includes medications that have low AC activity seen by the receptor binding study (Low AC drug group ). W e created Level 3 by merging ADS Levels 2 and 3 . Level 3 characterizes medications that have high AC potency (High A C drug group). The No AC drug group was chosen as the reference category. visits. The NACC recorded 100 prescription, non prescription, over the counter and supplements taken by th e participants within last two weeks of their visit. The medication list included various groups of medication like antivirals, antibiotics, antidepressants, antihistaminic, antipsychotics, antianxiety, analgesics, cardiovascular drugs, muscle relaxants, H 2 antagonists, corticosteroids, anticonvulsants etc. Many drugs with anticholinergic activity were not listed in the NACC medication list. However, if a subject took medications which were not in the NACC list then clinician or certified ADC staff recorded those additional medicat ions in standardized form (Form A4) . So, we can presume that if a subject took a medicine with AC properties, it was documented in the NACC dataset. Hence, the NACC dataset provided an excellent opportunity to examine medication his tory. Cognitive Outcomes The standardized neuropsychological test battery (NTB) in the UDS database includes 12 measures of cognition which evaluate the cognitive profile of older adults with QCI. The NTB is a sensitive tool that allows us to capture the change in the cognitive profile of an individual over the follow up time. Shirk et al. (2011) provided a model to create the standardized z scores, adjusted for sex, age, and education, for the NACC population. We used his model to calculate combined z sco re for our study

PAGE 106

106 population (Shirk et al., 2011). Hayden et al. used the exploratory and confirmatory factor analysis approach on UDS database to assess the clusters of neuropsychological measures that identify the specific cognitive domains (Hayden et al. , 2011). We grouped the 12 NTB measures into four cognitive domains : memory, attention, executive, and language as proposed by Hayden et al. (2011) (Table 4 2). A lower score on Trial Making A (TM A) and Trail Making B (TM B) illustrate better executive fu nctions , while lower scores on Digit Symbol Substitution Test (DSST) reveal poor executive functions. Therefore, we inverted the signs of TM A and TM B scores and added them with DSST tion of magnitude, with the greater value of the combined score suggesting better executive functioning . The cohort members were followed up annually , and neuropsychological assessments were performed at every visit. In addition, other relevant information like medical history and medication use were obtained and stored in the NACC database. Covariates Potentially important confounding and effect modifying covariates were included in the analysis. Our selection of predictors was guided by previous literatu re and evidence of association with the cognitive outcome. Predictors like advance age, lower education, female, and black race, family history, and lifestyle factors have been shown to be associated with lower cognitive scores and higher incidence of dem entia (Anstey et al., 2007). Therefore, we included demographic factors like age, sex, race, education , and family history in the model. In order to assess the effect of lifestyle behaviors and living environments on cognitive profile, we included alcohol use, smoking, living situation, independence, and marriage status in the model. Comorbidities like cardiovascular disease and diabetes increase cognitive burden and contribute to poor

PAGE 107

107 cognitive profile s in older adults (Biessels et al., 2006; Newman et al. , 2005) . To understand the burden of comorbidities on cognitive functions, we adjusted the models for comorbidities that are associated with cognitive impairment such as cardiovascular disease, cerebrovascular disease, diabetes , neurological diseases , and depression. Previous studies have demonstrated that the genetic biomarker ApoE4 is not only independently related to cognitive impairment and dementia (Corder et al., 1993; Strittmatter et al., 1993) in the older adults but also enhances the risk of dement ia due to AC drug exposure (Uusvaara et al., 2009). With this in mind , we adjusted the model for APOE4 genetic biomarker to evaluate the specific cognitive outcomes of AC medication use in older adults with QCI. Statistical Analyses Preliminary analysis included descriptive statistics of AC medication exposures across the baseline characteristics and potential confounders. We performed a chi squared test to estimate the difference in the distribution of categorical covariates across the AC drug use catego ries. We also ran analyses of the variance test to assess the differences in the means of continuous covariates across AC drug use groups. Descriptive statistics illustrates the significance of the distribution of the characteristics and cognitive measures across AC medication groups. We used a general linear mixed model with random effects and equal variance for random effects to evaluate the hypothesized association between baseline AC medications burden and change s in global and domain specific cognitive scores over time. To assess the association, we used a model with random intercept for individual participants. In addition, we assumed that the covariance structure of errors accounted for the dependence of subsequent observations on the same patient. Th e mixed

PAGE 108

108 modeling technique allows for difference s in the number of repeat measures across participants; therefore, we included individuals with two or more cognitive measures over the follow up duration. The outcome variables, global , and specific cognitiv e domains were assessed on regular annual intervals. We tested the normality using graphical method that might distort the usual pattern and checked for the outliers . We also tested for AC drug exposure interaction over time. We ran four separate models fo r cognitive domains and one for global cognition. We also included interaction term of AC medication use and time to assess the interaction between AC medication use times to follow up. The models were adjusted for predictors of cognitive functions , and th e potential confounders were identified a priori. The model fit was evaluated based upon lowest AIC value. All analyses were be performed using STATA 13 statistical software , and statistical significance was achieved at alpha=0.05. Results Of the 7,325 p articipants at baseline, 68.63 percent (5,027) participants met the up visit. Only those who met the criteria were included in the analysis. Of the included sample of 5,027, 4,024 (80.05 percent) took medic ation with no AC activities, 800 (15.91 percent) took medication with low AC activities and 203 (4.04 percent) took medication with high AC activity. The mean age ± standard deviation (SD) of the sample was 74.92±7.80 years (range: 60 109 years). Among the included sample population, 88.52 percent were white, 7.70% were Hispanic , and 48.29 percent were males. Around 91.75 percent of the sample population had high school or higher education , 46.24 percent of the population smoked and 5.14 percent consumed al cohol. Seventy seven percent of the population lived independently , and 65.46 were married.

PAGE 109

109 Table 4 3 describes t he baseline characteristics of the population. Individuals in the high and low AC drug groups were significantly older than the no AC drug gro up (75.42±7.39 and 76.85±8.14 vs. 74.52±7.69; p <.0001) . Body Mass Index was slightly higher in high and low AC drug groups (27.47±5.50 and 27.22±5.31 vs 27.06±4.84; p=0.198) as compared to no AC group. High and low AC groups had significantly fewer males (35.96 percent and 46.63 percent vs . 50.02 percent; p <.0001) and fewer married individuals, as compared to no AC groups (61.19 percent and 62.11 percent vs. 66.35%; p=.01 1 ), respectively . There were more black s , smokers, and individuals with a family hist ory of dementia in the high AC group as compared to no AC groups , but these differences were not statistically significantly. The prevalence of cardiovascular disease among individuals in the low AC groups (50 percent) was greater than that in no and high AC groups (No AC: 29.63 percent; High AC: 22.17 percent, P value=<.0001). This is due to the reason that various medications given for the treatment of cardiovascular disease such as, Atenolol, Hydralazine, and Digoxin, have no to low anticholinergic prop erties. Similarly , newer treatment regimens for depression involve use of selective serotonin receptor blockers such as Fluoxetine. These medications have low anticholinergic properties and fewer side effects. Therefore, we found a significantly higher pr evalence of depression among individuals who took medications with low AC medications groups (41.87 percent) as compared to no and high AC medications groups, respectively (31.49 percent and 33.66 percent; P value= <.0001). As anticipated, there was a sign ificantly higher prevalence of incontinence among the high AC group (36.45 percent) and low AC group (17.15 percent) as compared to no AC group (15.0 percent; P value= <.0001). On the contrary,

PAGE 110

110 individuals who took medications with high AC properties had a lower prevalence of APOE4 genotype (37.58 percent) as compared to individuals in no or low AC groups (No AC: 43.12 percent and Low AC: 41.19 percent; P value=0.303) and it was not statistically significant. Similarly, individuals who took high AC medicat ions had greater medications but it was statistically significant (No AC: 2.85 percent and Low AC: 2.63 percent; P value=0.195). About 87 percent participants in the high AC medi cation group stayed in the study for the second year as compared to 96 and 95 percent individuals in no and low AC medications groups, respectively. As compared to baseline, the retention rate in the third year among the high AC medication group was 57 per cent as compared to 70 and 64 percent in no and low AC medications groups, respectively. The retention rate in the fifth years among the high AC medication group was 27 percent as compared to 31 and 28 percent in no and low AC medications groups, respecti vely. The overall retention rate of the individuals taking medication with high AC properties was lower than those taking medications with no and low AC properties. Although, there was slightly lower retention rate among high AC medication group; however, the rates among AC groups were very similar. In addition, the participants remained in the study for long time and provided the history of their medication use. Therefore, it is reasonable to assume the chronic use of medications in the study population. T he mean global cognition z scores among individuals taking medications with high and low AC properties were significantly higher than those taking medications with no AC properties ( 0.17±0.57 and 0.29±0.59 vs. 0.33±0.62; p<.0001) . Similarly, the

PAGE 111

111 mean z scores in individual cognitive domains of memory, attention, executive function, and language were significantly higher in high and low AC medication groups as compare to no AC medication group (Table 4 4). The results suggest that the older adults with QC I who took medications with AC properties had better global cognition at baseline as compared to those who took medications with no AC properties (Fig. 4 1) . When considering the association between AC medication use and change in global cognitive scores (Table 4 5) over the follow up time, we found a statistically positive association between AC medication use and higher global cognitive scores. The interactions between the AC medication use groups and follow up duration were not statistically significant , suggesting that the association between higher global cognitive scores and AC medication groups was independent of time. Similarly, there was a significantly positive association between AC medication use and higher executive function scores (Table 4 8). The results for memory (Table 4 6), attention (Table 4 7) and language (Table 4 9) also showed greater cognitive scores in individuals who took medications with AC properties as compared to those who took medications without AC properties. However, these associations were significant only for the low AC medication group and not for high AC medication group. Discussion In the study cohort of older adults with questionable cognitive impairment, we observed that exposure to AC medications, as defined by the A DS scale, was independently associated with higher global cognitive scores. This association persists after adjusting for covariates including demographic factors, lifestyle behaviors, comorbidities , and genetic biomarker, suggesting that the beneficial ef fect of AC medication is independent of covariates. To our knowledge, this was the first

PAGE 112

112 population based epidemiological study to examine the longitudinal relationship between AC medication use and cognitive functions among older adults with QCI. The obje ctive of this study was to assess the long term effects of AC medication use on global and domain specific cognitive functions among older adults with QCI. Before conducting our research, w e hypothesize d that older adults with QCI who have exposure to AC d rugs at baseline have a more rapid decline in global and domain specific cognition as compared to older adults with QCI who have no exposure to AC drugs. However, contrary to our hypothesis, the study indicated that older adults with QCI who had exposure t o medication with AC properties had significantly better global cognitive score as compared to those who had exposure to medication with no AC properties. This relationship was also statistically significant for executive function. Although previous studi es have demonstrated that AC medication use is associated with cognitive impairment, we found that cognitive scores at baseline were higher among older adults taking AC medications (Campbell et al., 2009; Fox et al., 2011; Lechevallier Michel et al., 2013) . This suggests that the AC medication use have favorable long term therapeutic effect s among elderly population with QCI . A previous longitudinal study showed that AC medications use did not worsen cognitive functions over 18 months follow up (Pasina et al., 2013). These findings are of particular significance , since some studies have opposed the use of AC medications in older adults (Uusvaara et al., 2009; Chew et al., 2005). It is notable that this study focuses on specific population of older adults wh o had QCI. The practical application of the study prescribe AC medications to treat and reduce comorbidities for the benefit of older

PAGE 113

113 adults without compromising on cogniti ve functions. However, further research needs to be done to substantiate the claims. One of the reasons that our results differed from the previous studies might be related to the population that we focused on. A study by Nourhashémi et al. (2008) has show n that the comparable mean CDR SOB score for CDR Global score of 0.5 in the elderly population 3.20±.90. A CDR SOB score of 3.20 lies in the range of very mild cognitive impairment, which is outside the CDR SOB score of questionable cognitive impairment (0 .5 2.5) , which was one of our inclusion criterions for the study. These results show that for capturing at risk population early in their life course and understanding the long term effects of anticholinergic medications it is pragmatic to focus on older a dults with CDR SOB score between 0.5 and 2.5 . In addition, composite scores created by taking average of the individual test scores provide a better method to quantify the effect of AC medications on particular domains. Z scores are approximate unbiased es timate of the standard distribution and reflect reasonable estimate of the distribution across change in cognitive domains of population means . Bondi et al (2003) showed that 1 SD deviation difference in visuospatial domain was statistically significant. F urthermore, McGurk et al (2005) demonstrated that a difference of 0.5 SD on neuropsychological tests performance was significantly associated with better cognitive and physical functions. Our study measured a decline of about 0.2 0.4 SD in global cognition between the AC medication groups. Therefore , we assume that this difference of 0.2 0.4 SD appears to be clinically meaningful cognitive change over the follow up period given the fact that very old AD patients exhibit a milder degree of cognitive deficits (Bondi et al.; 2003). .

PAGE 114

114 It is important to note that although older adults in advanced age suffering from multi morbidity are highly susceptible to harmful effects of AC medications, individuals with QCI might experience cognitive compensatory response s (Kersten et al., 2013). Many compensatory biological mechanisms act to preserve and restore cognitive functions. Apparently, there is a great variability in AC drug response related to the stage of cognitive impairment and compensation mechanisms (Kerste n et al., 2013) . In the early stage of cognitive impairment , the galaninergic fibers and receptors undergo hyperinnervation and proliferation in the forebrain , which helps to maintain cognitive reserve (Counts, He, Che, Ginsberg, & Mufson, 2008b). In addit ion , various neuromodulators , like norepinephrine, dopamine, serotonin, GABA, galanin, substance P, and angiotensin II, safeguard the cognitive viability in the initial stages of cognitive compromise (Decker, & McGaugh, 1991). Furthermore, early cognitive impairment is associated with increase s in the hippocampal choline acetyltransferase, forebrain acetylcholinesterase and cholinergic neurons, which maintains the cognitive balance (Frolich 2002; Ikonomovi et al., 2003) . These compensatory mechanisms help i n preserving the cognitive functions in early stages of cognitive impairment (Ikonomovi et al., 2003). Oberauer (2001) and Naveh Benjamin, Bray and Levy (2007) demonstrated that the burden of comorbidity in older age may modify the cholinergic functioning and reduce the ability to facilitate the frontal brain mediated compensatory response (Oberauer, 2001; Naveh Benjamin et al., 2007). Since our study sample was at the very early stage of cognitive impairment, it is possible that compensatory responses migh t have maintained long term cognitive functions and AC medications might provide cognitive vitality by decreasing the cognitive burden associated with comorbidities.

PAGE 115

1 15 In our study, there was an overall increase in the cognitive scores of executive function s. This can be explained based upon the training effect the task repeatedly . The largest cognitive intervention study, ACTIVE, demonstrated that cognitive training improved cognitive abilities, especially speed and reasoning (Karle ne et al., 2002). It is possible that participants in our study became familiarized with the tests and acquired the skills to perform tests that constitute executive function, i.e., TM A, TM B, and DSST. This may explain the increase in scores for the exec utive functions that we saw . We hypothesized that individuals that take medications with AC properties are tudy do not support the hypothesis of cholinergic loss. The results of our study highlight an important point about the cholinergic hypothesis. We also discussed about the roles that compensatory mechanisms play in the incidence of cognitive impairment. Th ere seems to be a certain level of cognition threshold that needs to be breached for the cholinergic mechanisms to play a vital role in cognitive impairment. Until, the cognition or cognitive reserve is above the threshold, the compensatory mechanisms upho ld the cognitive functions. We presume that this level appears when the cognitive functions move from the stage of very mild cognitive impairment to mild cognitive impairment. Once individual are at the intersection of very mild and mild cognitive impairme nt, it is very difficult to come back to the level of questionable cognitive impairment. Therefore, the medications that help in alleviating the comorbidity effects , despite anticholinergic properties, help in retaining the cognitive threshold and thus del ay the cognitive impairment in older adults .

PAGE 116

116 Our study has several limitations. The medications with anticholinergic properties were collected as self report s . Besides, the drug history was limited only two weeks prior to each more, we do not have information about drug frequency and drug doses. The NACC database does not have information about prior long term drug history. The NACC also lacks in assessing the drug exposure in between the two patient visits. Therefore, we had li total exposure to AC drugs. A detailed medication history with frequency, doses, and long term duration would have provided useful information to evaluate AC drug exposure in more detail. However, the ADS scale uses th e name of the medication to categorize it and provide s a score accordingly. We only examined the effect of baseline AC medication use on cognitive scores, and the effect of duration and frequency of AC medication use was not used for this cross sectional a nalysis. Moreover, AC drug effect is related to accumulated effects of varying degree of AC medications and not to the dosage of each individual medication; therefore, the NACC dataset is appropriate to assess the AC medication burden. Overall, the NACC da ta provide sufficient information to address the hypothesis stated for anticholinergic drug use and exposure. It might be possible that the neuromodulators interact with the cholinergic systems and mask AC medication effects , especially in the early stage s of cognitive impairment. Due to the complexity of cognitive processes and their associations between pathobiological and clinical manifestations, there is limited understanding the mechanism of specific vulnerability to AC medications (Lopez et al., 2002 ). Therefore, besides considering independent effects of the cholinergic system, it is important to include the interactions between the neuromodulators and cholinergic system. However,

PAGE 117

117 the lack of information about neuromodulatory systems in the NACC data set , we could not the cholinergic system and the cognitive profile of older adults. The strength of this study lies in using the NACC data from the large population based prospective study of older adults across the nation. The NACC includes one of the largest numbers of participants with standardized longitudinal data, which allows us to precisely determine the QCI population. Besides, we used the ADS scale, a reliable and validated method to assess AC medication loa d that covers prescription and over the counter medications. In addition, the annual assessments of AC drug s provide comprehensive information to evaluate the magnitude of cognitive impairment related to chronic AC medication use. Furthermore, the NACC dat a allows us to take into account various competing causes of cognitive dysfunction in older adults by controlling for socioeconomic, demographic, health, lifestyle, and genetic factors -thus limiting any potential confounding. We conclude in this repeated measure mixed model analysis that older adults with QCI taking medications with AC properties had higher cognitive scores at baseline than those not taking AC medications. It is difficult to precisely account for all the factors that affect cognition in t he older adults with QCI; it is plausible that the therapeutic effects of AC medications might alleviate the comorbidity related cognitive burden on older adults. More research is required to comprehend and evaluate the influence of AC medications on QCI p opulations.

PAGE 118

118 Table 4 1. Anticholinergic Drug Scale and associated anticholinergic activity ADS Levels Anticholinergic Drug Activity Study Levels Level 0 Drugs with no known anticholinergic properties, e.g. Loperamide Level 1 (No) Level 1 Drugs with p otential anticholinergic activity evidenced by receptor binding studies, e.g. Ampicillin Level 2 (Low) Level 2 Drugs with anticholinergic adverse effects, usually at higher doses, e.g. Ranitidine Level 3 (High) Level 3 Drugs with marked anticholinergic a ctivity, e.g. Oxybutynin Abbreviations: ADS=Anticholinergic Dr ug Scale

PAGE 119

119 Table 4 2. Composite cognitive domain z scores Cognitive Domains Cognitive Tests Z score # Domain score Global Cognitive score Memory Logic Memory IA Z1 ((Z1+Z2)/2)=C1 (C1 + C2 + C3 + C4)/4 Logic Memory IIA Z2 Attention Digit Forward Number Z3 (Z3+Z4+Z5+Z6)/4=C2 Digit Forward Length Z4 Digit Backward Number Z5 Digit Backward Length Z6 Executive Trail Making A Z7 (Z7+Z8+Z9)/3=C3 Trail Making B Z8 Digital Symbo l Z9 Language Animal List Z10 (Z10+Z11+Z12)/3=C4 Vegetable List Z11 Boston Naming Z12 # Z scores are adjusted for age, sex, and education

PAGE 120

120 Table 4 3. Baseline characteristics of the study population as a function of anticholinergic drug use AC Drugs (N=5,027) Variable No AC Drugs (n=4024, 80.05%) Low AC Drugs (n=800, 15.91%) High AC Drugs (n=203, 4.04%) P value P value for trend Age (M±SD) 74.52±7.69 76.85±8.14 75.42±7.39 <.0001 Gender, Male (N, %) 2013(50.02) 373(46.63) 73(35.96) 0.0002 <.0001 Race, Black (N, %) 459(11.45) 92(11.54) 24(11.88) 0.9804 0.8537 Hispanic (N, %): Yes 327(8.13) 41(5.13) 19(9.36) 0.0097 0.1533 Education (M±SD) 15.08±3.53 14.99±3.39 14.65±3.23 0.1088 Alcohol Abuse (N, %) 203(5.06) 48(6.01) 7(3.45) 0.289 5 0.9980 Smoked> 100 Cigarette (N, %) 1843(46.18) 364(46.18) 96(48.00) 0.8787 0.7329 Body Mass Index (M±SD) 27.06±4.84 27.22±5.31 27.47±5.50 0.1982 Live Independently (N, %) 3110(77.65) 587(73.65) 156(77.23) 0.0494 0.0917 Married (N, %): Yes 2650(66.3 5) 495(62.11) 123(61.19) 0.0305 0.0108 1 ° Relative Dementia (N, %) 1697(42.93) 296(37.85) 96(48.98) 0.0051 0.5974 Cardio Vascular Disease (N, %) 1192(29.63) 400(50.00) 45(22.17) <.0001 <.0001 Cerebro Vascular Disease (N, %) 459(11.41) 129(16.13) 23(11.3 3) 0.0009 0.0169 Neurological Disease (N, %) 531(13.21) 118(14.75) 18(8.87) 0.0840 0.6204 114(2.85) 21(2.63) 10(4.95) 0.1946 0.3109 Depression (N, %) 1259(31.49) 332(41.87) 68(33.66) <.0001 <.0001 Hypertension (N, %) 2263(56. 35) 495(61.95) 103(51.24) 0.0033 0.3734 Hypercholesterolemia (N, %) 2333(58.52) 440(55.77) 103(51.50) 0.0663 0.0209 Diabetes (N, %) 601(14.97) 100(12.53) 34(16.92) 0.1339 0.5358 Vitamin B12 Deficiency (N, %) 205(5.21) 49(6.21) 6(2.99) 0.1721 0.8152 Thy roid Disease (N, %) 764(19.17) 166(20.78) 39(19.40) 0.5768 0.4602 Incontinence (N, %) 603(15.00) 137(17.15) 74(36.45) <.0001 <.0001 APOE4 Genotype (N, %) 1203(43.12) 236(41.19) 59(37.58) 0.3026 0.1288 Abbreviations: AC=Anticholinergics; SD=Standard devi ation; apolipoprotein E.

PAGE 121

121 Table 4 4. Baseline cognitive scores of the study population as a function of anticholinergic drug use Cognition AC Drugs (N=5,027) No AC Drugs (n=4024, 80.05%) Low AC Drugs (n=80 0, 15.91%) High AC Drugs (n=203, 4.04%) P value Global (M±SD) 0.33±0.62 0.29±0.59 0.17±0.57 <.0001 Memory (M±SD) 0.76±1.33 0.70±1.31 0.46±1.29 <.0001 Attention (M±SD) 0.43±0.91 0.39±0.88 0.33±0.90 0.0024 Executive (M±SD) 0.49±0.85 0. 55±0.86 0.56±0.93 0.0015 Language (M±SD) 0.65±1.11 0.61±1.04 0.45±1.02 <.0001 Abbreviations: AC=Anticholinergics; M=Mean; SD=Standard deviation; M=Mean.

PAGE 122

122 Table 4 5. Multivariable linear mixed model regression of change in global cognitive scores wit h time Covariates Model 1 Model 2 Coeff. CI P value Coeff. P value CI No AC drug Ref. Ref. Low AC drug .0658 .0417, .0899 0.000 .0618 0.000 .03 8 , 08 6 High AC drug .0852 .04123, .1291 0.000 .0553 0.0 1 4 .0 11, .099 Time (Years) .048 5 .05 87, .0382 0.000 .03 75 0.000 .04 7, .02 8 No AC drug*Time Ref. Ref. Low AC drug*Time .007 9 .017 5 , .00 1 7 0.1 0 6 .008 1 0. 109 .01 8, .00 2 High AC drug*Time .00 17 .0 193, .0159 0. 849 .00 77 0. 407 .01 1 , .026 A ge .00 10 .0003, .2630 0. 139 .000 7 0. 286 .00 2, .00 1 Sex (male) . 262 0. 000 . 285, .240 Race (black) . 1607 0.000 . 194, .127 Hispanic .1 871 0.000 . 229 , .14 6 Education . 0256 0.000 . 02 9 , .022 4 Alcohol . 0101 0. 634 .0 31, .052 Smoking .0 500 0. 000 .0 30, .070 Body Mass Inde x .0 048 0.000 .0 03 .007 Living independently . 1310 0.000 .003 , .00 7 Married . 0365 0.00 1 .1 15, .1 47 Family dementia . 0033 0. 249 .0 11, .017

PAGE 123

123 Table 4 5. Continued Covariates Model 3 Model 4 Coeff. CI P value Coeff. CI P value No AC dr ug Ref Ref Low AC drug .058 .033, .083 0.000 .064 .035, .092 0.000 High AC drug .048 .003, .094 0.038 .043 .007, .093 0.096 Time (Years) .039 .050, .031 0.000 .044 .055, .033 0.000 No AC drug*Time Ref. Ref. Low AC drug*Time .007 .017, .0 03 0.160 .009 .020, .002 0.119 High AC drug*Time .011 .008, .029 0.256 .011 .009, .031 0.276 Age .002 .003, .000 0.034 .001 .003, .000 0.165 Sex (male) .270 .293, .247 0.000 .294 .321, .267 0.000 Race (black) .152 .187, .117 0.000 .139 .181, .097 0.000 Hispanic .180 . 222, .137 0.000 .217 .270, .165 0.000 Education .026 .029, .022 0.000 ..023 .027, .019 0.000 Alcohol .006 .037, .048 0.796 .015 .035, .065 0.550 Smoking .048 .028, .068 0.000 .045 .021, .068 0.000 Body Mass Inde x .005 .003, .007 0.000 .005 .003, .008 0.000 Living independently .134 .118, .150 0.000 .117 .099, .135 0.000 Married .034 .056, .012 0.002 .029 .054, .004 0.024 Family dementia .002 .012, .016 0.744 .009 .007, .025 0.283 Cardiovascular disease .039 .020, .059 0.000 .040 .017, .061 0.001 Cerebrovascular disease .021 .006, .047 0.131 .009 .022, .039 0.573 Parkinson disease .124 .071, .177 0.000 .089 .026, .152 0.006 Neurological disease .000 .024, .025 0.975 .006 .034, .023 0.698 Hyperte nsion .001 .019, .020 0.945 .001 .023, .023 0.999 Hypercholestrolemia .004 .015, .024 0.684 .008 .014, .031 0.482 Diabetes .012 .041, .017 0.417 .014 .049, .020 0.406 Vit amin B12 deficiency .009 .046, .027 0.608 .021 .022, .063 0.340 Thyroid disease .016 .008,.0364 0.189 .005 .023, .033 0.724 Incontinence .034 .011, .056 0.003 .041 .016, .066 0.001 Depression .010 .026, .007 0.250 .002 .021, .016 0.792 APOE4 .129 .154, .104 0.000

PAGE 124

124 1. Abbreviations: AC=Anticholinergics; Ref=Refe rent. Coeff. =Coefficients; CI= 95% Confidence Interval; APOE4=allele . 2. Model 1 is adjusted for age. 3. Model 2 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, famil y history of dementia. 4. Model 3 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia, cardiovascular disease, cerebrovascular dis ease, Parkinson disease, neurological disease, hypertension, hypercholesterolemia, diabetes, Vitamin B12 deficiency, thyroid disease, incontinence, depression. 5. Model 4 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abus e, smoking, Body Mass Index, living independence, marriage status, family history of dementia, cardiovascular disease, cerebrovascular disease, Parkinson disease, neurological disease, hypertension, hypercholesterolemia, diabetes, Vitamin B12 deficiency, thyroid disease, incontinence, depression, APOE4.

PAGE 125

125 Table 4 6. Multivariable linear mixed model regression of change in memory scores with time Covariates Model 1 Model 2 Coeff. CI P value Coeff. CI P value No AC drug Ref Ref Low AC drug .117 . 065, .168 0.000 .100 .047, .153 0.000 High AC drug .149 .046, .233 0.004 .081 .015, .176 0.097 Time (Years) .060 .082 , .037 0.000 .041 .063, .020 0.000 No AC drug*Time Ref Ref Low AC drug*Time .010 .030, .012 0.326 .010 .031, .011 0.4 36 Hi gh AC drug*Time .007 .044, .030 0.725 .006 .032, .045 0.751 Age .006 .009 , .003 0.000 .007 .010 , .004 0.000 Sex (male) .553 .602 , .504 0.000 Race (black) .261 .335 , .188 0.000 Hispanic .018 .072, .109 0.691 Education .028 .035, .021 0.000 Alcohol .031 .059, .121 0.497 Smoking .096 .053, .140 0.000 Body Mass Index .022 .018, .026 0.000 Living independently .274 .240, .308 0.000 Married .125 .172 , .078 0.000 Family dementia .020 .049, .010 0.186

PAGE 126

126 T able 4 6. Continued Covariates Model 3 Model 4 Coeff. CI P value Coeff CI. P value No AC drug Ref Ref Low AC drug .077 .024,.132 0.005 .099 .038,.162 0.002 High AC drug .067 .031,.165 0.183 .053 .055,.162 0.335 Time (Years) .049 .070, .027 0.000 .054 .079, .031 0.000 No AC drug*Time Ref Ref Low AC drug*Time .003 .025,.018 0.742 .013 .037,.011 0.284 High AC drug*Time .010 .029,.049 0.633 .019 .024,.061 0.392 Age .010 .013, .006 0.000 .010 .014, .007 0.000 Sex (male) .574 .625, .523 0.000 .611 .670, .552 0.000 Race (black) .260 .336, .184 0.000 .252 .343, .161 0.000 Hispanic .019 .074,.111 0.690 .074 .190,.041 0.207 Education .027 .034,.020 0.000 .020 .029,.011 0.000 Alcohol .013 .080,.105 0.791 .019 .090, .127 0.733 Smoking .092 .048,.137 0.000 .081 .030,.132 0.002 Body Mass Index .020 .016,.025 0.000 .022 .016,.027 0.000 Living independently .280 .245,.314 0.000 .252 .213,.292 0.000 Married .115 .162, .068 0.000 .092 .147, .037 0.001 Family dement ia .017 .047,.013 0.274 .006 .028,.040 0.728 Cardio vascular disease .088 .046,.129 0.000 .097 .049,.144 0.000 Cerebro vascular disease .050 .007,.108 0.087 .028 .038,.094 0.408 Parkinson disease .146 .032,.261 0.012 .032 .105,.169 0.649 Neurologic al disease .059 .006,.113 0.030 .063 .002,.123 0.044 Hypertension .016 .028,.060 0.476 .013 .037,.063 0.618 Hypercholestrolemia .006 .049,.036 0.771 .006 .043,.055 0.814 Diabetes .031 .032,.095 0.331 .025 .050,.099 0.516 Vit amin B12 deficiency .007 .086,.071 0.852 .039 .076,.105 0.766 Thyroid disease .018 .072,.035 0.500 .040 .102,.022 0.205 Incontinence .138 .091,.185 0.000 .153 .099,.207 0.000 Depression .030 .065,.005 0.092 .021 .061,.019 0.310 APOE4 .427 .481, .372 0.000

PAGE 127

127 1. Abbreviations: AC=Anticholinergics; Ref=Referent. Coeff. =Coefficients; CI= 95% Confidence Interval; APOE4=allele 2. Model 1 is adjusted for age. 3. Model 2 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia . 4. Model 3 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia, cardiovascular disease, cerebrovascular disease, Parkinson disease, neurological disease, hypertension, hypercholesterolemia, diabetes, vitamin b12 deficiency, thyroid disease, incontinence, depression. 5. Model 4 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia, cardiovascular disease, cerebrovascular disease, Parki nson disease, neurological disease, hypertension, hypercholesterolemia, diabetes, Vitamin B12 deficiency, thyroid disease, incontinence, depression, APOE4. 5. Statistical significance achieved

PAGE 128

128 Table 4 7. Multivariable linear mixed model regression of change in attention scores with time Covariates Model 1 Model 2 Coeff. CI P value Coeff. CI P value No AC drug Ref Ref Low AC drug .023 .018,.064 0.264 .020 .022,.062 0.35 8 High AC drug .044 .032,.119 0.256 .038 .038,.115 0.326 Time (Years) .074 .089, .058 0.000 .058 .074, .044 0.000 No AC drug*Time Ref Ref Low AC drug*Time .002 .015,.020 0.777 .006 .012,.024 0.515 High AC drug*Time .011 .043,.020 0.483 . 003 .030,.037 0.852 Age .007 .005,.009 0.000 .003 .001,.051 0.012 Sex (male) .076 .112, .040 0.000 Race (black) .286 .341, .232 0.000 Hispanic .631 .698, .564 0.000 Education .024 .029,.018 0.000 Alcohol .010 .059,.078 0.784 Smoking .089 .057,.121 0.000 Body Mass Index .008 .011, .005 0.000 Living Independently .171 .144,.200 0.000 Married .011 .025,.046 0.560 Family Dementia .021 .004,.046 0.097

PAGE 129

129 Table 4 7. Continued Covariates Model 3 Model 4 Coe ff. CI P value Coeff. CI P value No AC Drug Ref Ref Low AC Drug .028 .015,.071 0.21 .026 .023,.075 0.300 High AC Drug .031 .048,.110 0.440 .050 .037,.137 0.260 Time (Years) .055 .070, .040 0.000 .063 .079, .047 0.000 No AC Drug*Time Ref Ref Low AC Drug*Time .003 .015,.022 0.724 .007 .014,.027 0.516 High AC Drug*Time .007 .027,.042 0.673 .003 .038,.040 0.861 Age .002 .000,.005 0.042 .002 .001,.005 0.129 Sex (Male) .072 .110, .035 0.000 .100 .141, .055 0.000 Race (Black) .26 3 .319, .208 0.000 .233 .299, .167 0.000 Hispanic .615 .683, .548 0.000 .662 .746, .579 0.000 Education .026 .031,.021 0.000 .022 .028,.016 0.000 Alcohol Use .010 .059,.080 0.770 .018 .063,.099 0.664 Smoking .092 .060,.125 0.000 .100 .063,. 138 0.000 Body Mass Index .006 .010, .003 0.000 .009 .013, .005 0.000 Living Independently .172 .143,.200 0.000 .162 .130,.194 0.000 Married .001 .035,.037 0.940 .007 .034,.048 0.742 Family Dementia .022 .003,.048 0.082 .016 .012,.045 0.265 C ardiovascular Disease .049 .017,.082 0.003 .038 .001,.074 0.045 Cerebrovascular Disease .069 .113, .024 0.002 .036 .087,.014 0.156 Parkinson Disease .030 .058,.117 0.509 .045 .061,.150 0.406 Neurological Disease .036 .077,.006 0.094 .029 .076, .019 0.233 Hypertension .050 .083, .016 0.003 .039 .076, .001 0.042 Hypercholestrolemia .002 .030,.035 0.884 .002 .035,.039 0.904 Diabetes .064 .111, .017 0.008 .048 .103,.007 0.088 Vitamin B12 deficiency .046 .106,.014 0.133 .033 .037,.10 2 0.355 Thyroid Disease .090 .050,.130 0.000 .096 .050,.141 0.000 Incontinence .018 .055,.019 0.341 .019 .061,.023 0.370 Depression .023 .051,.005 0.109 .017 .049,.015 0.296 APOE4 .060 .099, .020 0.003

PAGE 130

130 1. Abbreviations: AC=Anticholinergic s; Ref=Referent. Coeff. =Coefficients; CI= 95% Confidence Interval; APOE4=allele 2. Model 1 is adjusted for age. 3. Model 2 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smo king, Body Mass Index, living independence, marriage status, family history of dementia . 4. Model 3 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, fami ly history of dementia, cardiovascular disease, cerebrovascular disease, Parkinson disease, neurological disease, hypertension, hypercholesterolemia, diabetes, Vitamin B12 deficiency, thyroid disease, incontinence, depression. 5. Model 4 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia, cardio vascular disease, cerebro vascular disease, Parkinson disease, neurological diseas e, hypertension, hypercholesterolemia, diabetes, vitamin b12 deficiency, thyroid disease, incontinence, depression, APOE4.

PAGE 131

131 Table 4 8. Multivariable linear mixed model regression of change in executive function scores with time Covariates Model 1 Model 2 Coeff. Cl P value Coeff. CI P value No AC drug Ref Ref Low AC drug .073 .033,.113 0.000 .065 .024,.105 0.002 High AC drug .124 .049,.199 0.001 .102 .026,.177 0.008 Time (Years) .079 .068,.090 0.000 .055 .045,.066 0.000 No AC drug*Time Ref Ref Low AC drug*Time .007 .025,.0115 0.467 .010 .028,.008 0.284 High A C drug*Time .011 .046,.024 0.551 .016 .051,.020 0.380 Age .006 .004,.008 0.000 .005 .003,.007 0.000 Sex (male) .168 .199, .136 0.000 Race (black) .426 .378,.474 0.000 Hispanic .236 .176,.296 0.000 Education .217 .159,.275 0.000 Alco hol .034 .095,.027 0.278 Smoking .004 .033,.025 0.781 Body Mass Index .008 .011, .005 0.000 Living independently .253 .280, .226 0.000 Married .063 .095, .030 0.000 Family dementia .017 .043,.009 0.198

PAGE 132

132 Table 4 8. C ontinued Covariates Model 3 Model 4 Coeff. CI P value Coeff. CI P value No AC drug Ref Ref Low AC drug .054 .012,.097 0.011 .057 .008,.105 0.022 High AC drug .104 .027,.182 0.009 .101 .014,.187 0.024 Time (Years) .048 .037,.059 0.000 .050 .038,.6 20 0.000 No AC drug*Time Ref Ref Low AC drug*Time .008 .027,.011 0.395 .010 .031,.011 0.348 High AC drug*Time .012 .049,.024 0.499 .002 .040,.037 0.937 Age .005 .003,.008 0.000 .006 .004,.009 0.000 Sex (male) .174 .208, .140 0.000 .186 .225, .147 0.000 Race (black) .419 .370,.468 0.000 .392 .333,.452 0.000 Hispanic .225 .163,.286 0.000 .230 .154,.306 0.000 Education .222 .163,.282 0.000 .270 .199,.340 0.000 Alcohol .037 .099,.025 0.246 .003 .077,.071 0.932 Smoking .004 .034,. 025 0.782 .006 .040,.028 0.723 Body Mass Index .009 .012, .005 0.000 .006 .009, .003 0.001 Living independently .244 .272, .217 0.000 .244 .276, .213 0.000 Married .055 .089, .023 0.001 .064 .102, .025 0.001 Family dementia .017 .043,.0 09 0.200 .016 .046,.014 0.297 Cardio. disease .014 .045,.016 0.348 .009 .043,.026 0.624 Cerebro. disease .124 .083,.164 0.000 .091 .049,.138 0.000 Parkinson disease .327 .245,.408 0.000 .283 .184,.381 0.000 Neurological disease .028 .068,.011 0 .165 .064 .109, .020 0.005 Hypertension .024 .006,.055 0.119 .0111 .024,.046 0.531 Hypercholestrolemia .020 .049,.010 0.186 .003 .037,.031 0.857 Diabetes .052 .009,.095 0.017 .007 .049,.058 0.771 Vit amin B12 deficiency .048 .007,.103 0.089 .0 44 .020,.108 0.179 Thyroid disease .038 .074, .001 0.042 .060 .102, .017 0.006 Incontinence .003 .031,.038 0.850 .006 .045,.034 0.783 Depression .058 .031,.086 0.000 .058 .027,.089 0.000 APOE4 .046 .011,.082 0.018

PAGE 133

133 1. Abbreviations: AC=Antic holinergics; Ref=Referent. Coeff. =Coefficients; CI= 95% Confidence Interval; APOE4=allele 2. Model 1 is adjusted for age. 3. Model 2 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia . 4. Model 3 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage st atus, family history of dementia, cardiovascular disease, cerebrovascular disease, Parkinson disease, neurological disease, hypertension, hypercholesterolemia, diabetes, Vitamin B12 deficiency, thyroid disease, incontinence, depression. 5. Model 4 is ad justed for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia, cardio vascular disease, cerebro vascular disease, Parkinson disease, neurologica l disease, hypertension, hypercholesterolemia, diabetes, Vitamin B12 deficiency, thyroid disease, incontinence, depression, APOE4.

PAGE 134

134 Table 4 9. Multivariable linear mixed model regression of change in language scores with time Covariates Model 1 Model 2 Coeff. CI P value Coeff. CI P value No AC drug Ref Ref Low AC drug .045 .005, .086 0.028 0.028 .016,.099 0.006 High AC drug .029 .043,.102 0.429 0.429 .066,.084 0.815 Time (Years) .137 .005,.086 0.028 0.028 .016,.099 0.006 No AC drug*Time Ref. Ref. Low AC drug*Time .012 .027,.004 0.141 .015 .030,.001 0.074 High AC dr ug*Time .010 .018,.038 0.469 .018 .011,.048 0.215 Age .000 .027,.004 0.141 .309 .350 .268 0.000 Sex (male) .474 .537, .411 0.000 Race (black) .304 .383, .225 0.000 Hispanic .119 .043,.195 0.002 Education .028 .047,.103 0.460 Alc ohol .046 .001,.083 0.013 Smoking .012 .009,.016 0.000 Body Mass Index .207 .181,.234 0.000 Living independently .019 .019,.058 0.326 Married .006 .016,.028 0.606 Family dementia .015 .030,.001 0.074

PAGE 135

135 Table 4 9. Continued Cov ariates Model 3 Model 4 Coeff. CI P value Coeff. CI P value No AC drug Ref Ref Low AC drug .067 .024,.110 0.002 .062 .040,.111 0.012 High AC drug .004 .080,.073 0.996 .012 .096,.073 0.786 Time (Years) .118 .137, .099 0.000 .126 .147, .104 0.000 No AC drug*Time Ref Ref Low AC drug*Time .017 .033,.001 0.037 .015 .033,.004 0.116 High AC drug*Time .020 .010,.049 0.190 .009 .022,.041 0.567 Age .002 .005,.001 0.166 .000 .003,.003 1.000 Sex (male) .314 .357, .271 0.000 .337 .387, .287 0.000 Race (black) .465 .530, .400 0.000 .445 .523, .367 0.000 Hispanic .230 .379, .220 0.000 .341 .442, .241 0.000 Education .113 .035,.189 0.004 .063 .029,.156 0.177 Alcohol .017 .060,.094 0.658 .007 .084,.098 0.881 Smoking .04 2 .005,.079 0.028 .039 .005,.082 0.080 B ody M ass I ndex .013 .009,.017 0.000 .013 .009,.017 0.000 Living independently .210 .183,.236 0.000 .200 .170,.230 0.000 Married .019 .020,.058 0.338 .024 .021,.069 0.300 Family dementia .001 .021,.024 0.904 .003 .028,.023 0.836 Cardio. disease .018 .016,.051 0.305 .014 .024,.055 0.464 Cerebro. disease .014 .061,.033 0.566 .031 .085,.022 0.254 Parkinson disease .045 .049,.138 0.346 .058 .053,.167 0.306 Neurological disease .021 .022,.064 0.337 .0 22 .027,.071 0.378 Hypertension .012 .022,.050 0.438 .010 .031,.051 0.640 Hypercholestrolemia .036 .001,.071 0.044 .019 .021,.059 0.352 Diabetes .018 .071,.035 0.514 .010 .052,.072 0.748 Vit amin B12 deficiency .036 .027,.100 0.263 .052 .022,.1 27 0.170 Thyroid disease .028 .016,.073 0.215 .026 .028,.077 0.328 Incontinence .011 .028,.049 0.586 .031 .012,.075 0.157 Depression .024 .052,.003 0.084 .009 .040,.022 0.567 APOE4 .070 .116, .023 0.003

PAGE 136

136 1. Abbreviations: AC=Anticholinergi cs; Ref=Referent. Coeff. =Coefficients; CI= 95% Confidence Interval; APOE4=allele 2. Model 1 is adjusted for age. 3. Model 2 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, sm oking, Body Mass Index, living independence, marriage status, family history of dementia . 4. Model 3 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, fam ily history of dementia, cardiovascular disease, cerebrovascular disease, Parkinson disease, neurological disease, hypertension, hypercholesterolemia, diabetes, Vitamin B12 deficiency, thyroid disease, , incontinence, depression. 5. Model 4 is adjusted for age, sex, race, ethnicity (hispanic), education level, alcohol abuse, smoking, Body Mass Index, living independence, marriage status, family history of dementia, cardiovascular disease, cerebrovascular disease, Parkinson disease, neurological diseas e, hypertension, hypercholesterolemia, diabetes, Vitamin B12 deficiency, thyroid disease, incontinence, depression, APOE4.

PAGE 137

137 Figure 4 1. Plot of adjusted estimated means of global cognitive scores over follow up visits No AC 4 024 3872 2835 1992 1262 843 Low AC 777 738 499 330 220 136 High AC 196 170 111 65 53 19

PAGE 138

138 Figure 4 2. Plot of adjusted estimated means of memory scores over follow up visits No AC 4017 3784 2808 1971 1247 832 Low AC 759 729 491 326 219 135 High AC 193 167 110 65 52 19

PAGE 139

139 Fig ure 4 3. Plot of adjusted estimated means of attention scores over follow up visits No AC 4 024 3822 2819 1979 1254 828 Low AC 765 734 495 328 220 136 High AC 192 170 111 65 52 19

PAGE 140

140 Figure 4 4. Plot of adjusted estimated means of executive function scores over follow up visits No AC 4 016 3834 2763 1945 1224 812 Low AC 767 7 20 485 3 12 2 14 1 31 High AC 195 1 66 1 05 65 5 1 19

PAGE 141

141 Figure 4 5. Plot of adjusted estimated means of languag e scores over follow up visits No AC 4 024 3862 2821 1982 1 259 839 Low AC 775 7 37 497 3 29 2 20 1 36 High AC 195 1 69 11 1 65 53 19

PAGE 142

142 CHAPTER 5 CONCLUSIONS Background Since the first case of AD was described, researchers have tried to understand the mechanism of the disease in detail. They know a lot about the pathology and the biochemical mechanisms t hat play a major role in the development and the progression of the disease. However, the fundamental question still remains unanswered what is the underlying etiology, various theorie s have been proposed to describe the loss of memory and cognition, the hallmark of AD. One of the oldest and the most promising theories that attempts to explain the mechanism of memory and cognitive changes in the AD patients AD related cognitive changes are associated with a decline in basal forebrain cholinergic neurotransmission. According to this hypothesis, acetylcholine plays a critical role in cerebral cortical de velopment, cognitive performance, learning , and memory (Lu & Tune, 2003). What is more interesting is that the drugs which improve central cholinergic functions tend to improve cognitive functions in patients suffering from AD (Schubert et al., 2006). This evidence supports the view that acetylcholine has an important role in the development of cognitive impairment and dementia. AD is the most common cause of dementia and the sixth leading cause of death in the U . S. As the U.S. population ages, the preval ence of AD and associated comorbidities and mortality is bound to continue to increase. Because of the urgency to minimize AD A) were created on priority basis in

PAGE 143

143 to accomplish that goal is the underlying pathology of AD starts years before the appearance of the clinical sign and symptoms. While aiming to achieve NAPA goals, this study focused on older adults who are at greater risk of cognitive impairment. Previous studies have shown that anticholinergics act against the cholinergic system in the brain and are related to development of dementia (Bird, Stranahan, Sumi, & Raskind, 1983; Etienne et al., 1986; Henke & Lan g, 1983). We designed our study to see the effects of dementia among a very selective population of older adults with QCI. These individuals are at the borderline , and a significant event in life may push these individuals toward dementia. Therefore, it is crucial that we understand the effects of AC medications in this population. Discussion We hypothesized that the older adults with QCI who take AC medications will have lower cognitive scores on neuropsychological tests as compared to those who do not ta ke AC medications. We also hypothesized that individuals who take AC medications are more likely to develop dementia in the future and are more likely to have reduced cognitive score over time. In Chapter 2, we investigated the relationship between AC medi cation use and cognition using a cross sectional analysis of baseline data. The results highlighted that older adults with QCI who were taking AC medications, contrary to our hypothesis, had better cognitive scores at baseline compared to those not taking AC medications. Th e s e results prompt us to think of the

PAGE 144

144 reasons and mechanisms that might be involved in actions of anticholinergic drugs that might improve cognition. To further investigate, we performed time to event analysis in Chapter 3. The results of th is analysis showed that there was no significant difference in the rate of incident dementia among the individuals who took AC medications as compared to those who did not. This clarified that the long term risk of develop ing dementia was similar among individuals who took AC medication as compared to those who did not. This suggests that AC medications do not have differentially harmful long term effects on cognition compared to those who do not take the medication. To further explore the trajectories o f change in cognitive scores, we conducted a more sophisticated model using a mixed model repeated measure analysis in Chapter 4. We found that over the follow up duration , the cognitive scores were generally higher among AC medication groups compared to t he no AC medication group. These results suggest that AC medication use in this specific population is associated with an attenuated decline in cognition levels as compared to those not taking AC medications. In addition, over follow up time the cognitive scores in AC medication group remain higher than the no AC medication group. One of the possible explanations is that the AC medications are given to treat comorbidities. These medical conditions have their own independent deleterious effects on cognition. It is possible that the AC medications while decreasing the burden of comorbidity , helps elevate cognition. As older adults suffer from chronic diseases and regularly take medications, the beneficial effects of AC medications can be seen in our longitudin al analysis. We used standardized z scores , estimated against cognitively normal population in the NACC dataset, to estimate the distribution of cognitive scores and assess change in cognitive outcome. Standardized

PAGE 145

145 z score is a better measure to evaluate c ognitive outcomes over a large population. A standard deviation change in global and domain specific cognitive score s among older adults reflects a larger change in cognitive functions as compared to the normal population . Therefore, we presume that there was a significant cognitive decline in our population of older adults with questionable cognitive impairment over the follow up period. We hypothesized that older adults with questionable cognitive impairment that take medications with anticholinergic properties are more likely to have poor cognitive functions. Our hypothesis was based upon the theory which substantiated that anticholinergic compounds are associated with memory loss and might be implicated in Alz Leavitt, 1974). However, the results of the study do not completely support the hypothesis of cholinergic loss. The results of these studies focus on an important issue about the theory of cholinergic hypothesis of memory los s that at what level of cognition anticholinergic effects start to impede cognitive functions and aid in the process of dementia and AD. Keeping in mind the roles that compensatory mechanisms play in the preventing cognitive functions, the anticholinergic medications could have a debilitating effect on cognitive functions only when a certain level of cognition threshold has been attained for the anticholinergic medications to induce cognitive impairment. As long as the cognitive functions are above that thr eshold, the cumulative effect of anticholinergic medications might not be potent enough to counter the compensatory mechanisms. Therefore, while we did not specifically test this hypothesis, we presume that medications helping to alleviatie disease effects might preserve cognitive function,

PAGE 146

146 which could delay cognitive decline in older adults with questionable cognitive impairment. Future Directions Many gaps exist in the literature about the development of dementia and the mechanism through which AC medic ations affect the cognitive profiles in older adults. Our future research will focus on fill ing some of those gaps. The NACC population was selected from the hospital, patient and physician referrals which might not be the true representation of the commun ity dwelling older adults in the United States. We plan to replicate our results in other community dwelling populations. In addition, we would like to evaluate data in randomized controlled trials so that we can allocate participants to AC medications and analyze the cognitive change over time. Another important issue within the domain of cognitive impairment is role of neuromodulators and other compensatory mechanism s that modulate the cognitive outcomes in older adults. It is important to include these factors in the analysis and assess their independent effects on cognition among older adults. It will be advantageous to perform moderator analysis to assess the effects of individual compensatory mechanisms and their accumulative effects on cognitive func tions among at risk population. In addition to compensatory mechanisms, comorbidities mediate the process and course of dementia. Previous studies have shown that the diseases have their independent effects on cognitive impairment and play a vital role in dementia outcome. We plan to conduct mediator analysis to separate out the effects on comorbidities on dementia in the presence of anticholinergic medications. This will help

PAGE 147

147 us to understand whether anticholinergic medications delay the process of dementi a by alleviating the effect of comorbidity and to quantify that effect, if any. Biomarkers are important tool to objectively assess the impact of anticholinergic medication use on cognition among older adults. Amyloid plaques and neurofibrillary tangles de posits in the human brain are the hallmark of AD. We plan to retrospectively explore the association between anticholinergic drug use and neuropathological deposits of neurofibrillary plaques and tangles in postmortem brains of older adults. Based upon our previous knowledge, we would expect that higher anticholinergic exposure is associated with higher deposits of neurofibrillary plaques and tangles in postmortem brains. However, c onsidering our results , it will be interesting to learn whether AC medicatio n exposure protects against neurofibrillary deposits in the brain. This dissertation makes several useful contributions to the field of Alzheimer D isease , especially in the preclinical phase of AD . Our study results are in line with NAPA goals to identif y vulnerable population s and to prevent cognitive decline in an at risk population. We have provide d new epidemiological evidence about AC medication use and cognitive impairment. The work also opens up a new debate about the utility of AC medications in o lder adults , especially those with QCI. In the quest for scientific evidence, we hope that future work will identify risk factors that contribute to the etiology of AD that will aid in development of effective interventions to prevent cognitive impairment.

PAGE 148

148 APPENDIX A ANTICHOLINERGIC DRUG SCALE LEVELS SAS CODE data X YZ ; SET ABC ; ADS = . ; if acarbose = 1 then ADS = 0 ; if acetaminophen = 1 then ADS = 0 ; if acetazolamide = 1 then ADS = 0 ; if aceticacidtopical = 1 then ADS = 0 ; if acyclovir = 1 then ADS = 0 ; if adenosine = 1 then ADS = 0 ; if albuterol = 1 then ADS = 0 ; if alendronate = 1 then ADS = 0 ; if allopurinol = 1 then ADS = 0 ; if alprazolam = 1 then ADS = 2 ; if aluminumcarbonate = 1 then ADS = 0 ; if aluminumhydroxide = 1 then ADS = 0 ; if amantadine = 1 then ADS = 2 ; if amiloride = 1 then ADS = 0 ; if amiodarone = 1 then ADS = 0 ; if amitriptyline = 1 then ADS = 3 ; if amlodipine = 1 then ADS = 0 ; if ammoniumlactatetopical= 1 then ADS = 0 ; if amoxicillin = 1 then ADS = 0 ; if amoxicillinclavulanate = 1 then ADS = 0 ; if ampicillin = 1 then ADS = 2 ; if anagrelide = 1 then ADS = 0 ; if anastrozole = 1 then ADS = 0 ; if anileridine = 1 then ADS = 0 ; if apapdichloralphenazoneisomethep= 1 th en ADS = 0 ; if apraclonidineophthalmic= 1 then ADS = 0 ; if ascorbicacid = 1 then ADS = 0 ; if aspirin = 1 then ADS = 0 ; if atenolol = 1 then ADS = 0 ; if atorvastatin = 1 then ADS = 0 ; if a tropine = 1 then ADS = 3 ; if azathioprine = 1 then ADS = 2 ; if azithromycin = 1 then ADS = 0 ; if bacitracinophthalmic= 1 then ADS = 0 ;

PAGE 149

149 if bacitracintopical = 1 then ADS = 0 ; if baclofen = 1 then ADS = 0 ; if balsamperutopical = 1 then ADS = 0 ; if beclomethasone = 1 then ADS = 0 ; if beclomethasonenasal = 1 then ADS = 0 ; if benazepril = 1 then ADS = 0 ; if benzocainetopical = 1 then ADS = 0 ; if benzonatate = 1 then ADS = 0 ; if benztropine = 1 then ADS = 3 ; if betacarotene = 1 then ADS = 0 ; if betamethasoneclotrimazoletopica= 1 then ADS = 0 ; if betamethasonetopical 1 th en ADS = 0 ; if betaxololophthalmic = 1 then ADS = 0 ; if bethanechol = 1 then ADS = 0 ; if bicalutamide = 1 then ADS = 0 ; if bisacodyl = 1 then ADS = 0 ; if bismuthsubsalicylate= 1 then ADS = 0 ; if bisoprolol = 1 then ADS = 0 ; if brimonidineophthalmic= 1 then ADS = 0 ; if brinzolamideophthalmic= 1 then ADS = 0 ; if bromocriptine = 1 then ADS = 2 ; if brompheniramine = 1 then ADS = 3 ; if bu desonide = 1 then ADS = 0 ; if budesonidenasal = 1 then ADS = 0 ; if bumetanide = 1 then ADS = 0 ; if bupropion = 1 then ADS = 0 ; if buspirone = 1 then ADS = 0 ; if butabarbital = 1 th en ADS = 0 ; if butalbital = 1 then ADS = 0 ; if caffeine = 1 then ADS = 0 ; if calaminetopical = 1 then ADS = 0 ; if calcipotrienetopical= 1 then ADS = 0 ; if calcitonin = 1 then ADS = 0 ; if calcitriol = 1 then ADS = 0 ; if calciumacetate = 1 then ADS = 0 ; if calciumandvitamind = 1 then ADS = 0 ; if calciumcarbonate = 1 then ADS = 0 ; if camphormentholtopical= 1 then ADS = 0 ; if cand esartan = 1 then ADS = 0 ; if captopril = 1 then ADS = 2 ; if carbacholophthalmic = 1 then ADS = 0 ;

PAGE 150

150 if carbamazepine = 1 then ADS = 3 ; if carbamideperoxideotic= 1 then ADS = 0 ; if carbidopa = 1 then ADS = 0 ; if carbinoxamine = 1 then ADS = 3 ; if carisoprodol = 1 then ADS = 0 ; if carvedilol = 1 then ADS = 0 ; if casanthranol = 1 then ADS = 0 ; if casanthranoldocusate= 1 then ADS = 0 ; if cascarasagrada = 1 then ADS = 0 ; if castoroil = 1 then ADS = 0 ; if cefaclor = 1 then ADS = 0 ; if cefamandole = 1 then ADS = 2 ; if cefazolin = 1 then ADS = 0 ; if cefixime = 1 then ADS = 0 ; if cefoxitin = 1 then ADS = 2 ; if ceftibuten = 1 then ADS = 0 ; if ceftriaxone = 1 then ADS = 0 ; if cefuroxime = 1 then ADS = 0 ; if celecoxib = 1 then ADS = 0 ; if cephalexin = 1 then ADS = 0 ; if cephalothin = 1 then ADS = 1 ; if cerivastatin = 1 then ADS = 0 ; if cetirizine = 1 then ADS = 0 ; if cetylpyridiniumtopical= 1 then ADS = 0 ; if chloralhydra te = 1 then ADS = 0 ; if chlorambucil = 1 then ADS = 0 ; if chlordiazepoxide = 1 then ADS = 1 ; if chlorhexidinetopical= 1 then ADS = 0 ; if chlorothiazide = 1 then ADS = 0 ; if chlorpheniramine = 1 then ADS = 3 ; if chlorpromazine = 1 then ADS = 3 ; if chlorpropamide = 1 then ADS = 0 ; if chlorthalidone = 1 then ADS = 1 ; if chlorzoxazone = 1 then ADS = 0 ; if cholestyramine = 1 th en ADS = 0 ; if chondroitin = 1 then ADS = 0 ; if ciclopiroxtopical = 1 then ADS = 0 ; if cilastatin = 1 then ADS = 0 ; if cilostazol = 1 then ADS = 0 ; if cimetidine = 1 then ADS = 3 ; if ciprofloxacin = 1 then ADS = 0 ;

PAGE 151

151 if cisapride = 1 then ADS = 0 ; if citalopram = 1 then ADS = 0 ; if clarithromycin = 1 then ADS = 0 ; if clavulanate = 1 then ADS = 0 ; if clemastine = 1 then ADS = 3 ; if clindamycin = 1 then ADS = 1 ; if clindamycintopical = 1 then ADS = 0 ; if clobazam = 1 then ADS = 0 ; if clodronate = 1 then ADS = 0 ; if clomipramine = 1 t hen ADS = 3 ; if clonazepam = 1 then ADS = 1 ; if clonidine = 1 then ADS = 0 ; if clopidogrel = 1 then ADS = 0 ; if clorazepate = 1 then ADS = 1 ; if clotrimazole = 1 then ADS = 0 ; if cloxacillin = 1 then ADS = 0 ; if clozapine = 1 then ADS = 3 ; if codeine = 1 then ADS = 1 ; if colchicine = 1 then ADS = 0 ; if colestipol = 1 then ADS = 0 ; if collagenasetopical = 1 then ADS = 0 ; if conjugatedestrogens = 1 then ADS = 0 ; if cortisone = 1 then ADS = 1 ; if cranberry = 1 then ADS = 0 ; if cromolyn = 1 then ADS = 0 ; if cyanoco balamin = 1 then ADS = 0 ; if cyclobenzaprine = 1 then ADS = 3 ; if cyclophosphamide = 1 then ADS = 0 ; if cycloserine = 1 then ADS = 1 ; if cyclosporine = 1 then ADS = 1 ; if cyproheptadine = 1 then ADS = 3 ; if danazol = 1 then ADS = 0 ; if dantrolene = 1 then ADS = 0 ; if darifenacin = 1 then ADS = 3 ; if demeclocycline = 1 then ADS = 0 ; if desipramine = 1 then ADS = 3 ; if desmopressin = 1 then ADS = 0 ; if desonidetopical = 1 then ADS = 0 ; if desoximetasonetopical= 1 then ADS = 0 ; if dexamethasone = 1 then ADS = 1 ; if dexamethasonenasal = 1 then ADS = 0 ;

PAGE 152

152 if dexamethasoneophthalmic= 1 then ADS = 0 ; if dexamethasonetopical= 1 then ADS = 0 ; if dextromethorphan = 1 then ADS = 0 ; if diazepam = 1 then ADS = 1 ; if diclofenac = 1 then ADS = 0 ; if dicyclo mine = 1 then ADS = 3 ; if dienestroltopical = 1 then ADS = 0 ; if diflunisal = 1 then ADS = 0 ; if digitoxin = 1 then ADS = 1 ; if digoxin = 1 then ADS = 1 ; if dihydroxyaluminumsodiumcarbona t= 1 then ADS = 0 ; if diltiazem = 1 then ADS = 1 ; if dimenhydrinate = 1 then ADS = 3 ; if diphenhydramine = 1 then ADS = 3 ; if diphenoxylate = 1 then ADS = 0 ; if dipivefrinophthalmic= 1 then ADS = 0 ; if dipyridamole = 1 then ADS = 1 ; if dirithromycin = 1 then ADS = 0 ; if disopyramide = 1 then ADS = 3 ; if divalproexsodium = 1 then ADS = 1 ; if dobutamine = 1 then ADS = 0 ; if docusate = 1 then ADS = 0 ; if donepezil = 1 then ADS = 0 ; if dopamine = 1 then ADS = 0 ; if dorzolamideophthalmic= 1 then ADS = 0 ; if doxazosin = 1 then ADS = 0 ; if doxepin = 1 the n ADS = 3 ; if doxycycline = 1 then ADS = 0 ; if duloxetine = 1 then ADS = 0 ; if econazoletopical = 1 then ADS = 0 ; if edrophonium = 1 then ADS = 0 ; if enalapril = 1 then ADS = 0 ; if enoxaparin = 1 then ADS = 0 ; if entacapone = 1 then ADS = 0 ; if epoetinalfa = 1 then ADS = 0 ; if ergocalciferol = 1 then ADS = 0 ; if ergoloidmesylates = 1 then ADS = 0 ; if erythromycin = 1 then ADS = 0 ; if escitalopram = 1 then ADS = 0 ; if esomeprazole = 1 then ADS = 0 ; if estazolam = 1 then ADS = 1 ;

PAGE 153

153 if esterifiedestrogens = 1 then ADS = 0 ; if estradiol = 1 then ADS = 0 ; if estradioltopical = 1 then ADS = 0 ; if estropipate = 1 then ADS = 0 ; if ethambutol = 1 then ADS = 0 ; if ethinylestradiol = 1 then ADS = 0 ; if etidronate = 1 then ADS = 0 ; if et odolac = 1 then ADS = 0 ; if famotidine = 1 then ADS = 1 ; if felbamate = 1 then ADS = 0 ; if felodipine = 1 then ADS = 0 ; if fenofibrate = 1 then ADS = 0 ; if fentanyl = 1 then ADS = 1 ; if ferrousgluconate = 1 then ADS = 0 ; if ferroussulfate = 1 then ADS = 0 ; if fexofenadine = 1 then ADS = 0 ; if filgrastim = 1 then ADS = 0 ; if finasteride = 1 then ADS = 0 ; if f lavoxate = 1 then ADS = 3 ; if flecainide = 1 then ADS = 0 ; if fluconazole = 1 then ADS = 0 ; if fludrocortisone = 1 then ADS = 0 ; if flumazenil = 1 then ADS = 0 ; if flunisolide = 1 then ADS = 0 ; if fluocinonidetopical = 1 then ADS = 0 ; if fluoridetopical = 1 then ADS = 0 ; if fluorometholoneophthalmic= 1 then ADS = 0 ; if fluoxetine = 1 then ADS = 1 ; if fluphenazine = 1 then ADS = 1 ; if flurazepam = 1 then ADS = 1 ; if flutamide = 1 then ADS = 0 ; if fluticasone = 1 then ADS = 0 ; if fluticasonesalmeterol= 1 then ADS = 0 ; if fluvastatin = 1 then ADS = 0 ; if fluvoxamine = 1 then ADS = 1 ; if folicacid = 1 then ADS = 0 ; if fosinopril = 1 then ADS = 0 ; if furosemide = 1 then ADS = 1 ; if gabapentin = 1 then ADS = 0 ; if galantamine = 1 th en ADS = 0 ; if gemfibrozil = 1 then ADS = 0 ;

PAGE 154

154 if gentamicin = 1 then ADS = 1 ; if gentamicinophthalmic= 1 then ADS = 0 ; if gentamicintopical = 1 then ADS = 0 ; if ginkgo = 1 then ADS = 0 ; if glimepiride = 1 then ADS = 0 ; if glipizide = 1 then ADS = 0 ; if glucagon = 1 then ADS = 0 ; if glucosamine = 1 then ADS = 0 ; if glyburide = 1 then ADS = 0 ; if glycerintopical = 1 then ADS = 0 ; if guaifenesin = 1 then ADS = 0 ; if guanfacine = 1 then ADS = 0 ; if halcinonidetopical = 1 then ADS = 0 ; if haloperidol = 1 then ADS = 0 ; if heparin = 1 then ADS = 0 ; if hydralazine = 1 then ADS = 1 ; if hydrochlorothiazide = 1 then ADS = 0 ; if hydrocodone = 1 then ADS = 0 ; if hydrocortisone = 1 then ADS = 0 ; if hydrocortisoneophthalmic= 1 then ADS = 0 ; if hydrocortisoneotic = 1 then ADS = 0 ; if hydrocortisonetopical= 1 then ADS = 0 ; if hydromorphone = 1 then ADS = 0 ; if hydroxychloroquine = 1 then ADS = 0 ; if hydroxypropylmethylcelluloseoph= 1 then ADS = 0 ; if hydroxyurea = 1 then ADS = 0 ; if hydroxyzine = 1 then ADS = 3 ; if hyoscyamine = 1 then ADS = 3 ; if ibuprofen = 1 then ADS = 0 ; if imipenem = 1 then ADS = 0 ; if imi pramine = 1 then ADS = 3 ; if indapamide = 1 then ADS = 0 ; if indomethacin = 1 then ADS = 0 ; if insulin = 1 then ADS = 0 ; if ipratropium = 1 then ADS = 0 ; if irbesartan = 1 then A DS = 0 ; if ironpolysaccharide = 1 then ADS = 0 ; if isoniazid = 1 then ADS = 0 ; if isosorbide = 1 then ADS = 1 ; if isosorbidedinitrate = 1 then ADS = 1 ; if isosorbidemononitrate= 1 then ADS = 1 ;

PAGE 155

155 if isradipine = 1 then ADS = 0 ; if ketoconazoletopical = 1 then ADS = 0 ; if ketoprofen = 1 then ADS = 0 ; if ketotifenophthalmic = 1 then ADS = 1 ; if labetalol = 1 then ADS = 0 ; if lactase = 1 then ADS = 0 ; if lactulose = 1 then ADS = 0 ; if lamotrigine = 1 then ADS = 0 ; if lanolinmineraloiltopical= 1 then ADS = 0 ; if lansoprazole = 1 then ADS = 0 ; if latanoprostophthalmic= 1 then ADS = 0 ; if leuprolide = 1 then ADS = 0 ; if levobunololophthalmic= 1 then ADS = 0 ; if levodopa = 1 then ADS = 0 ; if levofloxacin = 1 then ADS = 0 ; if levothyroxine = 1 then ADS = 0 ; if lidocaine = 1 then ADS = 0 ; if lindanetopical = 1 then ADS = 0 ; if liothyronine = 1 then ADS = 0 ; if lisinopril = 1 then ADS = 0 ; if lithium = 1 then ADS = 0 ; if loperamide = 1 then ADS = 1 ; if loratadine = 1 then ADS = 0 ; if lorazepam = 1 then ADS = 1 ; if losartan = 1 then ADS = 0 ; if loteprednolophthalmic= 1 then ADS = 0 ; if lovastatin = 1 then ADS = 0 ; if loxapine = 1 then ADS = 3 ; if lvpsolution = 1 then ADS = 0 ; if lysine = 1 then ADS = 0 ; if magnesiumpreparations= 1 then ADS = 0 ; if mannitol = 1 then ADS = 0 ; if meclizine = 1 then ADS = 3 ; if medroxyprogesterone = 1 then ADS = 0 ; if megestrol = 1 then ADS = 0 ; if meperidine = 1 then ADS = 3 ; if meprobamate = 1 then ADS = 0 ; if mesalamine = 1 then ADS = 0 ; i f metaxalone = 1 then ADS = 0 ; if metformin = 1 then ADS = 0 ; if methazolamide = 1 then ADS = 0 ;

PAGE 156

156 if methenamine = 1 then ADS = 0 ; if methotrexate = 1 then ADS = 0 ; if methotrimeprazine = 1 then ADS = 3 ; if methyclothiazide = 1 then ADS = 0 ; if methylcellulose = 1 then ADS = 0 ; if methyldopa = 1 then ADS = 0 ; if methyleneblue = 1 then ADS = 0 ; if methylphenidate = 1 then ADS = 0 ; if methylprednisolone = 1 then ADS = 1 ; if methylprednisolonetopical= 1 then ADS = 0 ; if methyltestosterone = 1 then ADS = 0 ; if metoclopramide = 1 then ADS = 0 ; if metolazone = 1 then ADS = 0 ; if metoprolol = 1 then ADS = 0 ; if metronidazole = 1 then ADS = 0 ; if mexiletine = 1 then ADS = 0 ; if miconazoletopical = 1 then ADS = 0 ; if midazolam = 1 then ADS = 1 ; if midodrine = 1 then ADS = 0 ; if mineraloil = 1 then ADS = 0 ; if minocycline = 1 then ADS = 0 ; if mirtazapine = 1 then ADS = 0 ; if misoprostol = 1 then ADS = 0 ; if moexipril = 1 th en ADS = 0 ; if molindone = 1 then ADS = 2 ; if mometasonenasal = 1 then ADS = 0 ; if montelukast = 1 then ADS = 0 ; if morphine = 1 then ADS = 1 ; if moxifloxacin = 1 then ADS = 0 ; if mu ltivitamin = 1 then ADS = 0 ; if mupirocintopical = 1 then ADS = 0 ; if nabumetone = 1 then ADS = 0 ; if nadolol = 1 then ADS = 0 ; if naloxone = 1 then ADS = 0 ; if naproxen = 1 then ADS = 0 ; if nateglinide = 1 then ADS = 0 ; if nefazodone = 1 then ADS = 0 ; if neomycinophthalmic = 1 then ADS = 0 ; if niacin = 1 then ADS = 0 ; if nifedipine = 1 then ADS = 1 ; if nisold ipine = 1 then ADS = 0 ;

PAGE 157

157 if nitrofurantoin = 1 then ADS = 0 ; if nitroglycerin = 1 then ADS = 0 ; if nizatidine = 1 then ADS = 1 ; if norepinephrine = 1 then ADS = 0 ; if norfloxacin = 1 then ADS = 0 ; if nortriptyline = 1 then ADS = 3 ; if nystatin = 1 then ADS = 0 ; if octreotide = 1 then ADS = 0 ; if ofloxacin = 1 then ADS = 0 ; if olanzapine = 1 then ADS = 1 ; if olo patadineophthalmic= 1 then ADS = 0 ; if omeprazole = 1 then ADS = 0 ; if orphenadrine = 1 then ADS = 3 ; if oxazepam = 1 then ADS = 1 ; if oxcarbazepine = 1 then ADS = 2 ; if oxybutynin = 1 then ADS = 3 ; if oxycodone = 1 then ADS = 1 ; if oxymetazolinenasal = 1 then ADS = 0 ; if pamidronate = 1 then ADS = 0 ; if pancrelipase = 1 then ADS = 0 ; if pancuronium = 1 then ADS = 1 ; if pantoprazole = 1 then ADS = 0 ; if papaverine = 1 then ADS = 0 ; if paroxetine = 1 then ADS = 1 ; if penicillin = 1 then ADS = 0 ; if pentoxifylline = 1 then ADS = 0 ; if pergolide = 1 then ADS = 0 ; if perindopril = 1 then ADS = 0 ; if permethrintopical = 1 then ADS = 0 ; if perphenazine = 1 then ADS = 1 ; if petrolatumtopical = 1 then ADS = 0 ; if phenazopyridine = 1 t hen ADS = 0 ; if phenelzine = 1 then ADS = 1 ; if phenobarbital = 1 then ADS = 0 ; if phenylephrine = 1 then ADS = 0 ; if phenylpropanolamine = 1 then ADS = 0 ; if phenylsalicylate = 1 then ADS = 0 ; if phenytoin = 1 then ADS = 0 ; if phytonadione = 1 then ADS = 0 ; if pilocarpineophthalmic= 1 then ADS = 0 ; if pimozide = 1 then ADS = 2 ;

PAGE 158

158 if pindolol = 1 then ADS = 0 ; if pioglitazone = 1 then ADS = 0 ; if piperacillin = 1 then ADS = 1 ; if pirbuterol = 1 then ADS = 0 ; if piroxicam = 1 then ADS = 0 ; if pivampicillin = 1 then ADS = 0 ; if polycarbophil = 1 then ADS = 0 ; if polyethyleneglycolelectrolyteso= 1 then ADS = 0 ; if polymyxinbophthalmic= 1 then ADS = 0 ; if potassiumbicarbonate= 1 then ADS = 0 ; if potassiumchloride = 1 then ADS = 0 ; if potassiumcitrate = 1 then ADS = 0 ; if pramipexole = 1 then ADS = 0 ; if pramoxinetopical = 1 then ADS = 0 ; if pravastatin = 1 then ADS = 0 ; if prazosin = 1 then ADS = 0 ; if prednisolone = 1 then ADS = 1 ; if prednisoloneophthalmic= 1 then ADS = 0 ; if prednisone = 1 then ADS = 1 ; if primidone = 1 then ADS = 0 ; if probenecid = 1 then ADS = 0 ; if procainamide = 1 then ADS = 1 ; if prochlorperaz ine = 1 then ADS = 1 ; if procyclidine = 1 then ADS = 3 ; if progesterone = 1 then ADS = 0 ; if promethazine = 1 then ADS = 3 ; if propafenone = 1 then ADS = 0 ; if propantheline = 1 then ADS = 3 ; if propoxyphene = 1 then ADS = 0 ; if propranolol = 1 then ADS = 0 ; if propylthiouracil = 1 then ADS = 0 ; if protriptyline = 1 then ADS = 3 ; if pseudoephedrine = 1 then ADS = 0 ; if psyllium = 1 then ADS = 0 ; if pyrazinamide = 1 then ADS = 0 ; if pyridostigmine = 1 then ADS = 0 ; if pyrilamine = 1 then ADS = 3 ; if quetiapine = 1 then ADS = 0 ; if quinapril = 1 then ADS = 0 ; if quinidine = 1 then ADS = 0 ; if quinine = 1 then ADS = 0 ;

PAGE 159

159 if rabeprazole = 1 then ADS = 0 ; if raloxifene = 1 then ADS = 0 ; if ramipril = 1 then ADS = 0 ; if ran itidine = 1 then ADS = 2 ; if repaglinide = 1 then ADS = 0 ; if reserpine = 1 then ADS = 0 ; if rifampin = 1 then ADS = 0 ; if rimantadine = 1 then ADS = 0 ; if rimexoloneophthalmic= 1 t hen ADS = 0 ; if risedronate = 1 then ADS = 0 ; if risperidone = 1 then ADS = 0 ; if rofecoxib = 1 then ADS = 0 ; if ropinirole = 1 then ADS = 0 ; if rosiglitazone = 1 then ADS = 0 ; if s alicylicacidtopical= 1 then ADS = 0 ; if salmeterol = 1 then ADS = 0 ; if salsalate = 1 then ADS = 0 ; if scopolamine = 1 then ADS = 3 ; if selegiline = 1 then ADS = 0 ; if seleniumsulfidetopical = 1 then ADS = 0 ; if senna = 1 then ADS = 0 ; if sertraline = 1 then ADS = 1 ; if silversulfadiazinetopical= 1 then ADS = 0 ; if simethicone = 1 then ADS = 0 ; if simvastatin = 1 then AD S = 0 ; if sodiumbicarbonate = 1 then ADS = 0 ; if sodiumchloride = 1 then ADS = 0 ; if sodiumphosphate = 1 then ADS = 0 ; if sodiumsulfacetamideophthalmic= 1 then ADS = 0 ; if sotalol = 1 then ADS = 0 ; if spironolactone = 1 then ADS = 0 ; if succinylcholine = 1 then ADS = 0 ; if sucralfate = 1 then ADS = 0 ; if sulfamethizole = 1 then ADS = 0 ; if sulfamethoxazole = 1 then ADS = 0 ; if sulindac = 1 then ADS = 0 ; if tacrine = 1 then ADS = 0 ; if tamoxifen = 1 then ADS = 0 ; if tamsulosin = 1 then ADS = 0 ; if temazepam = 1 then ADS = 1 ; if terazosin = 1 then ADS = 0 ;

PAGE 160

160 if terbinafinetopical = 1 then ADS = 0 ; if terbutaline = 1 then ADS = 0 ; if terconazoletopical = 1 then ADS = 0 ; if tetracycline = 1 then ADS = 0 ; if theophylline = 1 then ADS = 1 ; if thiamine = 1 then ADS = 0 ; if thioridazine = 1 then ADS = 3 ; if thiothixene = 1 then ADS = 1 ; if thyroiddesiccated = 1 then ADS = 0 ; if ticlopidine = 1 then ADS = 0 ; if timolol = 1 then ADS = 0 ; if tobramycinophthalmic= 1 then ADS = 0 ; if tolbutamide = 1 then ADS = 0 ; if tolcapone = 1 then ADS = 0 ; if tolterodine = 1 then ADS = 3 ; if topiramate = 1 then ADS = 0 ; if torsemide = 1 then ADS = 0 ; if tramadol = 1 then ADS = 1 ; if trandolapril = 1 then ADS = 0 ; if trazodone = 1 then ADS = 0 ; if triamcinolone = 1 then ADS = 1 ; if triamcinolonenasal = 1 then ADS = 0 ; if triamcinolonetopical= 1 then ADS = 0 ; if triamterene = 1 then ADS = 1 ; if triazolam = 1 then ADS = 1 ; if trichlormethiazide = 1 then ADS = 0 ; if triethanolaminepolypeptideo leat= 1 then ADS = 0 ; if trifluoperazine = 1 then ADS = 1 ; if trihexyphenidyl = 1 then ADS = 3 ; if trimethoprim = 1 then ADS = 0 ; if trimipramine = 1 then ADS = 3 ; if troglitazone = 1 then ADS = 0 ; if trypsin = 1 then ADS = 0 ; if tuberculinpurifiedproteinderiva= 1 then ADS = 0 ; if ursodiol = 1 then ADS = 0 ; if valproicacid = 1 then ADS = 1 ; if valsartan = 1 then ADS = 0 ; if vancomycin = 1 then ADS = 1 ; if vecuronium = 1 then ADS = 0 ; if venlafaxine = 1 then ADS = 0 ; if verapamil = 1 then ADS = 0 ;

PAGE 161

161 if vitamine = 1 then ADS = 0 ; if warfarin = 1 then ADS = 1 ; if zafirlukast = 1 then ADS = 0 ; if zaleplon = 1 then ADS = 0 ; if zincgluconate = 1 then ADS = 0 ; if zincsulfate = 1 then ADS = 0 ; if zolpidem = 1 then ADS = 0 ; if zopiclon e = 1 then ADS = 0 ; RUN ; QUIT;

PAGE 162

162 LIST OF REFERENCES Albert, M. S. (1996). Cognitive and neurobiologic markers of early Alzheimer disease. Proceedings of the National Academy of Sciences , 93 (24), 13547 13551. Albert, M. S., Moss, M. B., Tanzi, R., & Jones, K. (2001). Preclinical prediction of AD using neuropsychological tests. Journal of the International Neuropsychological Society , 7 (5), 631 639. Allen, D.D., Galdzicki, Z., Brining, S.K., Fukuyama, R., Rapoport, S.I., & Smith, Q.R. (1997). Be ta amyloid induced increase in choline flux across PC12 cell membranes. Neurosci. Lett., 234 (1), 71 73. Alzheimer's disease facts and figures (201 4 ). Alzheimer's Association 10 (2). Retrieved from http://www.alz.org/downloads/Facts_Figures_2014.pdf Ameri can Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders DSM IV TR Fourth Edition (Text Revision). American Psychiatric Pub. Anagnostaras, S. G., Murphy, G. G., Hamilton, S. E., Mitchell, S. L., Rahnama, N. P., Nathanson, N. M., & Silva, A. J. (2002). Selective cognitive dysfunction in acetylcholine M1 muscarinic receptor mutant mice. Nature neuroscience , 6 (1), 51 58. Ancelin, M. L., Artero, S., Portet, F., Dupuy, A. M., Touchon, J., & Ritchie, K. (2006). Non degenerative mild cognitive impairment in elderly people and use of anticholinergic drugs: longitudinal cohort study. Bmj , 332 (7539), 455 459. Andrews, J.S., Jansen, J.H., Linders, S., & Princen, A. (1994). Effects of disrupting the cholinergic system on short term s patial memory in rats. Psychopharmacology, 115 , 485 494. Anstey, K. J., von Sanden, C., Salim, A., & O'Kearney, R. (2007). Smoking as a risk factor for dementia and cognitive decline: a meta analysis of prospective studies. American journal of epidemiolo gy , 166 (4), 367 378. Araujo, D.M., Lapchak, P.A., Robitaille, Y., Gauthier, S., & Quirion, R. (1988). Differential alteration of various cholinergic markers in cortical and subcortical regions of human J. Neurochem. 50 , 1914 1923. Araya, R., Noguchi, T., Yuhki, M., Kitamura, N., Higuchi, M., Saido, T. C., & Yamada, M. (2006). Loss of M5 muscarinic acetylcholine receptors leads to cerebrovascular and neuronal abnormalities and cognitive deficits in mice. Neurobiology of diseas e , 24 (2), 334 344.

PAGE 163

163 Arendt, T., Brückner, M. K., Lange, M., & Bigl, V. (1992). Changes in acetylcholinesterase and butyrylcholinesterase in Alzheimer's disease resemble embryonic development a study of molecular forms. Neurochemistry international, 21 (3), 381 396. Armstrong, D.M., Sheffield, R., Buzsaki, G., Chen, K.S., Hersh, L.B., Nearing, B., & Gage, F.H. (1993). Morphologic alterations of choline acetyltransferase positive neurons in the basal forebrain of aged behaviorally characterized Fisher 344 rat s. Neurobiol. Aging, 14 , 457 470. Arrasate, M., Pérez, M., Armas Portela, R., & Ávila, J. (1999). Polymerization of tau peptides into fibrillar structures. The effect of FTDP 17 mutations. FEBS letters , 446 (1), 199 202. Association of postoperative deli rium with raised serum levels of anticholinergic drugs. The Lancet , 318 (8248), 651 653. Atiya, M., Hyman, B. T., Albert, M. S., & Killiany, R. (2003). Structural magnetic resonance imaging in established and prodromal Alzheimer disease: a review. Alzheime r Disease & Associated Disorders, 17 (3), 177 195. Aubert, I., Araujo, D.M., Cecyre, D., Robitaille, Y., Gauthier, S., & Quirion, R. (1992). J. Neuro chem, 58 , 529 541. Auld, D.S., Kar, S., & Quirion, R. (1998). Beta amyloid peptides as direct cholinergic neuromodulators: a missing link. Trends Neurosci., 21 , 43 49. gress. Journal of Neurology, Neurosurgery & Psychiatry , 67 (4), 558 558. Barnes, C. A. (1990). Animal models of age related cognitive decline. In F. Boller and J. Grafman (Eds.), Handbook of Neuropsychology, (pp. 169 196). Amsterdam: Elsevier Science Publ ishers B.V. Bartus, R.T. (1979). Physostigmine and memory: effects in young and aged nonhuman primates. Science, 270 , 1087 1089. Bartus, R.T. (2000). On neurodegenerative diseases, models, and treatment strategies: lessons learned and lessons forgotten a generation following the cholinergic hypothesis. Exp. Neurol., 163 , 495 529. Bartus, R.T., & Johnson, H.R. (1976). Short term memory in the rhesus monkey: disruption from the anti cholinergic scopolamine. Pharmacol. Biochem. Behav., 5 , 39 46. Bartus, R.T., Dean III, R.L., Beer, B., & Lippa, A.S. (1982). The cholinergic hypothesis of geriatric memory dysfunction. Science, 217 , 408 414.

PAGE 164

164 Baxter, M.G., Bucci, D.J., Gorman, L.K., Wiley, R.G., & Gallagher, M. (1995). Selective immunotoxic lesions of basal f orebrain cholinergic cells: effects on learning and memory in rats. Behavioral Neuroscience ., 109 , 714 722. Beekly, D. L., Ramos, E. M., Lee, W. W., Deitrich, W. D., Jacka, M. E., Wu, J., & Kukull, W. (2007). The National Alzheimer's Coordinating Center (NACC) database: the uniform data set. Alzheimer Disease & Associated Disorders , 21 (3), 249 258. Beekly, D. L., Ramos, E. M., van Belle, G., Deitrich, W., Clark, A. D., Jacka, M. E., & Kukull, W. A. (2004). The National Alzheimer's Coordinating Center (NA CC) Database: an Alzheimer disease database. Alzheimer Disease & Associated Disorders , 18 (4), 270 277. Beers, M. H. (1997). Explicit criteria for determining potentially inappropriate medication use by the elderly: an update. Archives of Internal Medicine , 157 (14), 1531 1536. Bertram, L., Lill, C. M., & Tanzi, R. E. (2010). The genetics of Alzheimer disease: back to the future. Neuron , 68 (2), 270 281 . Bhattacharya, R., Chatterjee, S., Carnahan, R. M., & Aparasu, R. R. (2011). Prevalence and predictors of anticholinergic agents in elderly outpatients with dementia. The American Journal of Geriatric Pharmacotherapy , 9 (6), 434 441. Biessels, G. J., Staekenborg, S., Brunner, E., Brayne, C., & Scheltens, P. (2006). Risk of dementia in diabetes mellitus: a sys tematic review. The Lancet Neurology , 5 (1), 64 74. Biggan, S.L., Ingles, J.L., & Beninger, R.J. (1996). Scopolamine differentially affects memory of 8 and 16 month old rats in the double Y maze. Neurobiol. Aging, 17 , 25 30. Bird, T.D., Stranahan, S., S umi, S.M., & acetyltransferase activity in brain tissue from clinical and pathological subgroups. Annals of neurology , 14 (3), 284 293. Blount, P.J., Nguyen, C.D., & McDeavitt, J.T. (2002). Clinical use of cholinomimetic agents: a review. J. Head Trauma Rehabil., 17 (4), 314 321. Bodick, N. C., Offen, W. W., Levey, A. I., Cutler, N. R., Gauthier, S. G., Satlin, A., & Paul, S. M. (1997). Effects of xanomeline, a selective muscarinic receptor agonist, on cogn itive function and behavioral symptoms in Alzheimer disease. Archives of Neurology , 54 (4), 465 473. Bohnen, N. I., Kaufer, D. I., Hendrickson, R., Constantine, G. M., Mathis, C. A., & Moore, R. Y. (2007). Cortical cholinergic denervation is associated wit h depressive symptoms in Journal of Neurology, Neurosurgery & Psychiatry , 78 (6), 641 643.

PAGE 165

165 Bondi, M. W., Houston, W. S., Salmon, D. P., Corey Bloom, J., Katzman, R., Thal, L. J., & Delis, D. C. (2003). Neuropsy chological deficits associated with Alzheimer's disease in the very old: Discrepancies in raw vs. standardized scores. Journal of the International Neuropsychological Society , 9 (05), 783 795. Bonner, T. I., Buckley, N. J., Young, A. C., & Brann, M. R. (19 87). Identification of a family of muscarinic acetylcholine receptor genes. Science , 237 (4814), 527 532. Bottiggi, K. A., Salazar, J. C., Yu, L., Caban Holt, A. M., Ryan, M., Mendiondo, M. S., & Schmitt, F. A. (2006). Long term cognitive impact of anticho linergic medications in older adults. American Journal of Geriatric Psych , 14 (11), 980 984. Boustani, M., Campbell, N., Munger, S., Maidment, I., & Fox, C. (2008). Impact of anticholinergics on the aging brain: a review and practical application. Aging He alth , 4 (3), 311 320. Boustani, M., Sachs, G., & Callahan, C. M. (2007). Can primary care meet the biopsychosocial needs of older adults with dementia?. Journal of general internal medicine , 22 (11), 1625 1627. Bowen, D.M., Smith, C.B., White, P., & Daviso n, A.N. (1976). Neurotransmitter related enzymes and indices of hypoxia in senile dementia and other abiotrophies. Brain, 99, 459 496. Braak, F., Braak, H., & Mandelkow, E. M. (1994). A sequence of cytoskeleton changes related to the formation of neurof ibrillary tangles and neuropil threads. Acta neuropathologica , 87 (6), 554 567. Braak, H., & Braak, E. (1998). Evolution of neuronal changes in the course of Alzheimer's disease. Journal of neural transmission. Supplementum , 53 , 127 140. Brookmeyer, R., J ohnson, E., Ziegler Graham, K., & Arrighi, H. M. (2007). Forecasting the Alzheimer's & dementia , 3 (3), 186 191. Brookmeyer, R., Evans, D. A., Hebert, L., Langa, K. M., Heeringa, S. G., Plassman, B. L., & Kukull, W. A. United States. Alzheimer's & Dementia , 7 (1), 61 73. Burnett, G. B., Prange Jr, A. J., Wilson, I. C., Jolliff, L. A., Creese, I. C., & Snyder, S. H. (1980). Adverse effects of anti cholinergic antiparkinsonian drugs in tardive dyskinesia. Neuropsychobiology , 6 (2), 109 120. Burns, A., Jacoby, R., & Levy, R. (1991). Progression of cognitive impairment in Alzheimer's disease. Journal of the American Geriatrics Society , 39 , 39 45.

PAGE 166

166 Camp bell, N., Boustani, M., Limbil, T., Ott, C., Fox, C., Maidment, I., ... & Gulati, R. (2009). The cognitive impact of anticholinergics: a clinical review. Clinical interventions in aging , 4 , 225 233. Campbell, N. L., Boustani, M. A., Lane, K. A., Gao, S., Hendrie, H., Khan, B. A., ... & Hall, K. (2010). Use of anticholinergics and the risk of cognitive impairment in an African American population. Neurology , 75 (2), 152 159. Cancelli, I., Gigli, G. L., Piani, A., Zanchettin, B., Janes, F., Rinaldi, A., & Va lente, M. (2008). Drugs with anticholinergic properties as a risk factor for cognitive impairment in elderly people: a population based study. Journal of clinical psychopharmacology , 28 (6), 654 659. Cao, Y. J., Mager, D. E., Simonsick, E. M., Hilmer, S. N ., Ling, S. M., Windham, B. G., & Abernethy, D. R. (2007). Physical and cognitive performance and burden of anticholinergics, sedatives, and ACE inhibitors in older women. Clinical Pharmacology & Therapeutics , 83 (3), 422 429. Carnahan, R. M., Lund, B. C., Perry, P. J., & Pollock, B. G. (2001). A critical appraisal of the utility of the serum anticholinergic activity assay in research and clinical practice. Psychopharmacology bulletin , 36 (2), 24 39. Carnahan, R. M., Lund, B. C., Perry, P. J., Pollock, B. G ., & Culp, K. R. (2006). The Anticholinergic Drug Scale as a measure of drug related anticholinergic burden: associations with serum anticholinergic activity. The Journal of Clinical Pharmacology , 46 (12), 1481 1486. Carrière, I., Fourrier Reglat, A., Dart igues, J. F., Rouaud, O., Pasquier, F., Ritchie, K., & Ancelin, M. L. (2009). Drugs with anticholinergic properties, cognitive decline, and dementia in an elderly general population: the 3 city study. Archives of internal medicine , 169 (14), 1317 1324. Cau lfield, M. P. (1993). Muscarinic receptors characterization, coupling and function. Pharmacology & therapeutics , 58 (3), 319 379. Chen, Q.S., Kagan, B.L., Hirakura, Y., & Xie, C.W. (2000). Impairment of hippocampal long term potentiation by Alzheimer amylo id beta peptides. J. Neurosci. Res., 60 , 65 72. Chetelat, G., Villemagne, V. L., Bourgeat, P., Pike, K. E., Jones, G., Ames, D., & Rowe, C. C. amyloid deposition in Alzheimer disease. Annals of neurology , 67 (3), 317 324. . Chew, M. L., Mulsant, B. H., & Pollock, B. G. (2005). Serum anticholinergic activity and cognition in patients with moderate to sever e dementia. The American journal of geriatric psychiatry , 13 (6), 535 538.

PAGE 167

167 Chew, M. L., Mulsant, B. H., Pollock, B. G., Lehman, M. E., Greenspan, A., Mahmoud, R. A., & Gharabawi, G. (2008). Anticholinergic activity of 107 medications commonly used by older adults. Journal of the American Geriatrics Society , 56 (7), 1333 1341. Cohen, E.L., & Wurtman, R.J. (1975). Brain acetylcholine: increase after systemic choline administration. Life Sci., 16 , 1095 1102. Corder, E. H., Saunders, A. M., Strittmatter, W. J., Schmechel, D. E., Gaskell, P. C., Small, G., & Pericak Vance, M. A. (1993). Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science , 261 (5123), 921 923. Counts, S. E., Perez, S. E., & Mufson, E. J. Neuroinhibitory or neuroprotective?. Cellular and molecular life sciences: CMLS , 65 (12), 1842 1853. Counts, S. E., He, B., Che, S., Ginsberg, S. D., & Mufson, E. J. (2008b). Galanin hyperinnervation upregulates cho line acetyltransferase expression in cholinergic basal Neurodegenerative Diseases , 5 (3 4), 228 231. Cronin Stubbs, D., DeKosky, S. T., Morris, J. C., & Evans, D. A. (2000). Promoting interactions with basic scient ists and clinicians: the NIA Alzheimer's Disease Data Coordinating Center. Statistics in medicine , 19 (11 12), 1453 1461. Crow, T.J., & Grove White, I.G. (1973). An analysis of the learning deficit following hyoscine administration to man. British Journal of Pharmacology , 49 (2), 322 327. Cullen, W.K., Suh, Y.H., Anwyl, R., & Rowan, M.J. (1997). Block of LTP in rat hippocampus in vivo by beta amyloid precursor protein fragments. NeuroReport, 8 , 3213 3217. Cummings, J. L., Mega, M., Gray, K., Rosenberg Thom pson, S., Carusi, D. A., & Gornbein, J. (1994). The Neuropsychiatric Inventory comprehensive assessment of psychopathology in dementia. Neurology , 44 (12), 2308 2308. Cutler, N.R., & sease. In J.D.Brioni , M.W. Decker (Eds.), Molecular and Neurobiological Foundations (pp. 367 387). New York, NY: Wiley Liss. Dani, J. A. (2001). Overview of nicotinic receptors and their roles in the cent ral nervous system. Biological psychiatry , 49 (3), 166 174. Davidson, M.C., Cutrell, E.B., & Marrocco, R.T. (1999). Scopolamine slows the orienting of attention in primates to cued visual targets. Psychopharmacology, 142 , 1 8. Davies, P., & Maloney, A.J. disease. Lancet, 2 , 1403 1403.

PAGE 168

168 Davis, K. L. (1978). Physostigmine: improvement of long term memory processes in normal humans. Science. 201 , 272 274. Deary, I. J., Whiteman, M. C., P attie, A., Starr, J. M., Hayward, C., Wright, A. F., & Whalley, L. J. (2002). Ageing: Cognitive change and the APOE ϵ 4 allele. Nature , 418 (6901), 932 932. Decker, M. W., & McGaugh, J. L. (1991). The role of interactions between the cholinergic syste m and other neuromodulatory systems in learing and memory. Synapse , 7 (2), 151 168. Defilippi, J. L., & Crismon, M. L. (2003). Drug interactions with cholinesterase inhibitors. Drugs & aging , 20 (6), 437 444. DeKosky, S. T., Ikonomovic, M. D., Styren, S. D. , Beckett, L., Wisniewski, S., Bennett, D. A., & Mufson, E. J. (2002). Upregulation of choline acetyltransferase activity in hippocampus and frontal cortex of elderly subjects with mild cognitive impairment. Annals of neurology , 51 (2), 145 155. DeLaGarza, American family physician , 68 (7), 1365 1372. Dodart, J.C., Mathis, C., & Ungerer, A. (1997). Scopolamine induced deficits in a two trial object recognition task in mice. NeuroReport, 8 , 1173 1178. Drachman, D. A. (1977). Memory and cognitive function in man . Does the cholinergic system have a specific role? Neurology , 27 (8), 783 790. Drachman, D. A., Glosser, G., Fleming, P., & Longenecker, G. (1982). Memory decline in the aged: T reatment with lecithin and physostigmine. Neurology, 32, 944 950. Drachman, D.A., & Leavitt, J. (1974). Human memory and the cholinergic system. A relationship to aging? Arch. Neurol,. 30 , 113 121. Drachman, D.A., & Sahakian, B.J. (1980). Memory and cog nitive function in the elderly. A preliminary trial of physostigmine. Archives of neurology , 37 (10), 674 675. Dubois, B., Feldman, H. H., Jacova, C., Cummings, J. L., DeKosky, S. T., Barberger Gateau, P., & Scheltens, P. (2010). Revising the definition of Alzheimer's disease: a new lexicon. The Lancet Neurology , 9 (11), 1118 1127. Dumas, J. A., & Newhouse, P. A. (2011). The cholinergic hypothesis of cognitive aging revisited again: cholinergic functional compensation. Pharmacology Biochemistry and Behavior , 99 (2), 254 261. Dumas, J. A., McDonald, B. C., Saykin, A. J., McAllister, T. W., Hynes, M. L., West, J. D., & Newhouse, P. A. (2010). Cholinergic modulation of hippocampal activity during episodic

PAGE 169

169 memory encoding in postmenopausal women: a pilot study. Menopause (New York, NY) , 17 (4), 852 859. Dumas, J., Hancur Bucci, C., Naylor, M., Sites, C., & Newhouse, P. (2006). Estrogen treatment effects on anticholinergic induced cognitive dysfunction in normal postmenopausal women. Neuropsychopharmacology , 31 (9) , 2065 2078. Eagger, S. A., & Harvey, R. J. (1995). Clinical heterogeneity: responders to cholinergic therapy. Alzheimer Disease & Associated Disorders , 9 , 37 42. Ecobichon, D.J. ( 1996 ). Toxic effects of pesticides. asic Science of Poisons (Klaassen CD, Doull J, eds). 5th ed. New York: MacMillan , 643 689. Edginton, T., & Rusted, J. M. (2003). Separate and combined effects of scopolamine and nicotine on retrieval induced forgetting. Psychopharmacology , 170 (4), 351 357 . Eglen, R. M. (2005). Muscarinic receptor subtype pharmacology and physiology. Progress in medicinal chemistry , 43 , 105 136. Elias, M. F., Beiser, A., Wolf, P. A., Au, R., White, R. F., & D'Agostino, R. B. (2000). The preclinical phase of Alzheimer dise ase: a 22 year prospective study of the Framingham Cohort. Archives of Neurology , 57 (6), 808 813. Etienne, P., Robitaille, Y., Wood, P., Gauthier, S., Nair, N.P., & Quirion, R. (1986). Nucleus basalis neuronal loss, neuritic plaques and choline acetyltran sferase activity in Neuroscience, 19 , 1279 1291. Everitt, B. J., & Robbins, T. W. (1997). Central cholinergic systems and cognition. Annual review of psychology , 48 (1), 649 684. Farlow, M. R., He, Y., Tekin, S., Xu, J., Lane , R., & Charles, H. C. (2004). Impact of APOE in mild cognitive impairment. Neurology , 63 (10), 1898 1901. Feinberg, M. (1993). The problems of anticholinergic adverse effects in older patients. Drugs & aging , 3 (4), 335 348. Felder, C. C., Bymaster, F. P. , Ward, J., & DeLapp, N. (2000). Therapeutic opportunities for muscarinic receptors in the central nervous system. Journal of medicinal chemistry , 43 (23), 4333 4353. Fernandez, H. L., Moreno, R. D., & Inestrosa, N. C. (1996). Tetrameric (G4) acetylcholine sterase: structure, localization, and physiological regulation. Journal of neurochemistry , 66 (4), 1335 1346 . Ferris, S. H., & Kluger, A. (1997). Assessing cognition in Alzheimer disease research. Alzheimer disease and associated disorders , 11 , 45 49.

PAGE 170

170 Fic k, D. M., Cooper, J. W., Wade, W. E., Waller, J. L., Maclean, J. R., & Beers, M. H. (2003). Updating the Beers criteria for potentially inappropriate medication use in older adults: results of a US consensus panel of experts. Archives of internal medicine , 163 (22), 2716 2724. Fischer, W., Gage, F. H., & Björklund, A. (2006). Degenerative changes in forebrain cholinergic nuclei correlate with cognitive impairments in aged rats. European Journal of Neuroscience , 1 (1), 34 45. Fischer, W., Wictorin, K., Björk lund, A., Williams, L. R., Varon, S., & Gage, F. H. (1987). Amelioration of cholinergic neuron atrophy and spatial memory impairment in aged rats by nerve growth factor. Nature , 329 (6134), 65 68. Fisher, A. (2000). Therapeutic Strategies in Alzheimer's Di sease. M1 Muscarinic Agonists. The Japanese Journal of Pharmacology , 84 (2), 101 112. Flacker, J. M., Cummings, V., Mach Jr, J. R., Bettin, K., Kiely, D. K., & Wei, J. (1998). The association of serum anticholinergic activity with delirium in elderly medic al patients. The American Journal of Geriatric Psychiatry , 6 (1), 31 41. Flaherty, J. H., Perry, H. M., Lynchard, G. S., & Morley, J. E. (2000). Polypharmacy and hospitalization among older home care patients. The Journals of Gerontology Series A: Biologic al Sciences and Medical Sciences , 55 (10), M554 M559. Flaherty, J. H., Tariq, S. H., Raghavan, S., Bakshi, S., Moinuddin, A., & Morley, J. E. (2003). A model for managing delirious older inpatients. Journal of the American Geriatrics Society , 51 (7), 1031 1 035. Flicker, C., Ferris, S. H., & Serby, M. (1992). Hypersensitivity to scopolamine in the elderly. Psychopharmacology , 107 (2 3), 437 441. Flood, J.F., & Cherkin, A. (1986). Scopolamine effects on memory retention in mice: a model of dementia? Behaviora l and neural biology , 45 (2), 169 184. Flynn, D. D., Ferrari DiLeo, G., Mash, D. C., & Levey, A. I. (1995). Differential regulation of molecular subtypes of muscarinic receptors in Alzheimer's disease. Journal of neurochemistry , 64 (4), 1888 1891. Fox, C. , Richardson, K., Maidment, I. D., Savva, G. M., Matthews, F. E., Smithard, D., & Brayne, C. (2011). Anticholinergic medication use and cognitive impairment in the older population: the medical research council cognitive function and ageing study. Journal of the American Geriatrics Society , 59 (8), 1477 1483.

PAGE 171

171 Frölich, L. (2002). The cholinergic pathology in Alzheimer's disease discrepancies between clinical experience and pathophysiological findings. Journal of neural transmission , 109 (7 8), 1003 1013. Gal asko, D., Edland, S. D., Morris, J. C., Clark, C., Mohs, R., & Koss, E. (1995). The Consortium to Establish a Registry for Alzheimer's Disease (CERAD). Part XI. Clinical milestones in patients with Alzheimer's disease followed over 3 years. Neurology , 45 (8 ), 1451 1455. Galdzicki, Z., Fukuyama, R., Wadhwani, K.C., Rapoport, S.I., & Ehrenstein, G. (1994). Beta amyloid increases choline conductance of PC12 cells: possible mechanism of toxicity in Brain Res earch ., 646 , 332 336. Gallagher, M., & Nicolle, M. M. (1993). Animal models of normal aging: relationship between cognitive decline and markers in hippocampal circuitry. Behavioural brain research , 57 (2), 155 162. Gauss, C. J. (1906). Geburten in kunstlichem Dâmmerschlaf. Arch Gynaekol, 78, 579 631. Expert opinion on investigational drugs , 8 (10), 1511 1520. Canadian M edical Association Journal , 166 (5), 616 623. Geula, C., & Mesulam, M. M. (1995). Cholinesterases and the pathology of Alzheimer disease. Alzheimer Disease & Associated Disorders , 9 , 23 28. Giacobini, E. (2000a). Cholinesterase inhibitors stabilize Alzhe Annals of the New York Academy of Sciences , 920 (1), 321 327. disease. Alzheimer Disease & Associated Disorders , 14 (1), S3 S10. Gilad, G. M., Rabey , J., Tizabi, Y., & Gilad, V. H. (1987). Age dependent loss and compensatory changes of septohippocampal cholinergic neurons in two rat strains differing in longevity and response to stress. Brain research , 436 (2), 311 322. Glass, J. R., Sproule, B. A., H errmann, N., Streiner, D., & Busto, U. E. (2003). Acute pharmacological effects of temazepam, diphenhydramine, and valerian in healthy elderly subjects. Journal of clinical psychopharmacology , 23 (3), 260 268. Goodglass, H., Kaplan, E., and Barresi, B. (19 83). Boston diagnostic aphasia examination (BDAE). Philadelphia, PA: Lea and Febiger .

PAGE 172

172 Götz, J., Chen, F., Van Dorpe, J ., & Nitsch, R. M. (2001). Formation of neurofibrillary tangles in Science , 293 (5534), 1491 1495. Gray, E. G., Paula Barbosa , M., & Roher, A. (1987). Alzheimer's disease: paired helical filaments and cytomembranes. Neuropathology and applied neurobiology , 13 (2), 91 110. Haass, C., Koo, E.H., Mellon, A., Hung, A.Y., & Selkoe, D.J. (1992). Targeting of cell surface beta amyloid precursor protein to lysosomes: alternative processing into amyloid bearing fragments. Nature, 357 , 500 503. Haba, K., Ogawa, N., Kawata, M., & Mori, A. (1988). A method for parallel determination of choline acetyltr ansferase and muscarinic cholinergic receptors: application in aged rat brain. Neurochem. Res., 13 , 951 955. Hajjar, E. R., Cafiero, A. C., & Hanlon, J. T. (2007). Polypharmacy in elderly patients. The American journal of geriatric pharmacotherapy , 5 (4) , 345 351. Hajjar, I., Catoe, H., Sixta, S., Boland, R., Johnson, D., Hirth, V., & Eleazer, P. (2005). Cross sectional and longitudinal association between antihypertensive medications and cognitive impairment in an elderly population. The Journals of Ger ontology Series A: Biological Sciences and Medical Sciences, 60 (1), 67 73. Hammerlein, A., Derendorf, H., & Lowenthal, D. T. (1998). Pharmacokinetic and pharmacodynamic changes in the elderly: clinical implications. Clinical pharmacokinetics , 35 (1), 49 64 . Hampel, H., Frank, R., Broich, K., Teipel, S. J., Katz, R. G., Hardy, J., & Blennow, K. (2010). Biomarkers for Alzheimer's disease: academic, industry and regulatory perspectives. Nature Reviews Drug Discovery , 9 (7), 560 574 . Han, L., Agostini, J. V., & Allore, H. G. (2008). Cumulative anticholinergic exposure is associated with poor memory and executive function in older men. Journal of the American Geriatrics Society , 56 (12), 2203 2210. Hanlon, J. T., Pieper, C. F., Hajjar, E. R., Sloane, R. J., Lind blad, C. I., Ruby, C. M., & Schmader, K. E. (2006). Incidence and predictors of all and preventable adverse drug reactions in frail elderly persons after hospital stay. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences , 61 (5), 511 515. Hanon, O., Pequignot, R., Seux, M. L., Lenoir, H., Bune, A., Rigaud, A. S., & Girerd, X. (2006). Relationship between antihypertensive drug therapy and cognitive function in elderly hypertensive patients with memory complaints. Journal of hyperte nsion , 24 (10), 2101 2107.

PAGE 173

173 Hayden, K. M., Jones, R. N., Zimmer, C., Plassman, B. L., Browndyke, J. N., Pieper, C., & Welsh Centers uniform dataset neuropsychological battery: a n evaluation of invariance between and within groups over time. Alzheimer disease and associated disorders , 25 (2), 128 137. Hebert, L. E., Weuve, J., Scherr, P. A., & Evans, D. A. (2013). Alzheimer disease in the United States (2010 2050) estimated using the 2010 census. Neurology , 80 (19), 1778 1783. Henke, H., & Lang, W. (1983). Cholinergic enzymes in neocortex, hippocampus and basal forebrain of non neurological and senile dementia of Alzheimer type patients. Brain research , 267 (2), 281 291. Heutink, P . (2000). Untangling tau related dementia. Human molecular genetics , 9 (6), 979 986. Hilmer, S. N., & Gnjidic, D. (2009). The effects of polypharmacy in older adults. Clinical pharmacology and therapeutics , 85 (1), 86 88. Hilmer, S. N., Mager, D. E., Simon sick, E. M., Cao, Y., Ling, S. M., Windham, B. G., & Abernethy, D. R. (2007). A drug burden index to define the functional burden of medications in older people. Archives of internal medicine , 167 (8), 781 787. Hilmer, S. N., McLachlan, A. J., & Le Couteur , D. G. (2007). Clinical pharmacology in the geriatric patient. Fundamental & clinical pharmacology , 21 (3), 217 230. Hoshi, M., Takashima, A., Murayama, M., Yasutake, K., Yoshida, N., Ishiguro, K., Hoshino, T., & Imahori, K. (1997). Non toxic amyloid beta peptide 1 42 suppresses acetylcholine J. Biol. Chem., 272 , 2038 2041. Huang, H.M., Ou, H.C., & Hsieh, S.J. (2000). Amyloid beta peptide impaired carbachol but not glutamate media ted phosphoinositide pathways in cultured rat cortical neurons. Neurochemical research , 25 (2), 303 312. Hyman, B. T., & Trojanowski, J. Q. (1997). Editorial on consensus recommendations for the postmortem diagnosis of Alzheimer disease from the National I nstitute on Aging and the Reagan Institute Working Group on diagnostic criteria for the neuropathological assessment of Alzheimer disease. Journal of Neuropathology & Experimental Neurology , 56 (10), 1095 1097 . Hyman, B. T., Phelps, C. H., Beach, T. G., Bi gio, E. H., Cairns, N. J., Carrillo, M. C., & Montine, T. J. (2012). National Institute on Aging Alzheimer's Association guidelines for the neuropathologic assessment of Alzheimer's disease. Alzheimer's & Dementia , 8 (1), 1 13.

PAGE 174

174 Ibach, B., & Haen, E. (2004) Curr. Pharm. Des., 10 (3), 231 251. Ikonomovic, M. D., Mufson, E. J., Wuu, J., Cochran, E. J., Bennett, D. A., & DeKosky, S. T. (2003). Cholinergic plasticity in hippocampus of individuals with mil d cognitive impairment: correlation with Alzheimer's neuropathology. Journal of Alzheimer's Disease , 5 (1), 39 48. Ishihara, T., Hong, M., Zhang, B., Nakagawa, Y., Lee, M. K., Trojanowski, J. Q., & Lee, V. M. Y. (1999). Age dependent emergence and progress ion of a tauopathy in transgenic mice overexpressing the shortest human tau isoform. Neuron , 24 (3), 751 762. Ishihara, T., Zhang, B., Higuchi, M., Yoshiyama, Y., Trojanowski, J. Q., & Lee, V. M. Y. (2001). Age dependent induction of congophilic neurofibri llary tau inclusions in tau transgenic mice. The American journal of pathology , 158 (2), 555 562. Ivnik, R. J., Malec, J. F., Smith, G. E., Tangalos, E. G., & Petersen, R. C. (1996). Neuropsychological tests' norms above age 55: COWAT, BNT, MAE token, WRAT R reading, AMNART, STROOP, TMT, and JLO. The Clinical Neuropsychologist , 10 (3), 262 278. Ivnik, R. J., Malec, J. F., Tangalos, E. G., Petersen, R. C., Kokmen, E., & Kurland, L. T. (1990). The Auditory Verbal Learning Test (AVLT): norms for ages 55 years and older. Psychological Assessment: A Journal of Consulting and Clinical Psychology , 2 (3), 304 312 . Iyer, S., Naganathan, V., McLachlan, A. J., & Le Conteur, D. G. (2008). Medication withdrawal trials in people aged 65 years and older. Drugs & aging , 25 ( 12), 1021 1031. Jerusalinsky, D., Kornisiuk, E., Alfaro, P., Quillfeldt, J., Alonso, M., Verde, E. R., & Harvey, A. (1998). Muscarinic toxin selective for m4 receptors impairs memory in the rat. Neuroreport , 9 (7), 1407 1411. Jessen, F., Kaduszkiewicz, H. , Daerr, M., Bickel, H., Pentzek, M., Riedel Heller, S., & Maier, W. (2010). Anticholinergic drug use and risk for dementia: target for dementia prevention. European archives of psychiatry and clinical neuroscience , 260 (2), 111 115. Kantarci, K., & Jack J r, C. R. (2003). Neuroimaging in Alzheimer disease: an evidence based review. Neuroimaging Clinics of North America, 13 (2), 197 209. Kar, S., Issa, A.M., Seto, D., Auld, D.S., Collier, B., & Quirion, R. (1998). Amyloid beta peptide inhibits high affinity choline uptake and acetylcholine release in rat hippocampal slices. J. Neurochem., 70 , 2179 2187.

PAGE 175

175 Kar, S., Seto, D., Gaudreau, P., & Quirion, R. (1996). Beta amyloid related peptides inhibit potassium evoked acetylcholine release from rat hippocampal slic es. J. Neurosci., 16 , 1034 1040. Katz, I. R., Sands, L. P., Bilker, W., DiFilippo, S., Boyce, A., & D'Angelo, K. (1998). Identification of medications that cause cognitive impairment in older people: the case of oxybutynin chloride. Journal of the America n Geriatrics Society , 46 (1), 8 13. Kaufer, D. I., Cummings, J. L., Ketchel, P., Smith, V., MacMillan, A., Shelley, T., & DeKosky, S. T. (2000). Validation of the NPI Q, a brief clinical form of the Neuropsychiatric Inventory. The Journal of neuropsychiatr y and clinical neurosciences , 12 (2), 233 239. Kausler, D. H. (1994). Learning and memory in normal aging (pp. 276 305). San Diego, CA: Academic Press. Kay, G. G., & Wesnes, K. A. (2005). Pharmacodynamic effects of darifenacin, a muscarinic M3 selective r eceptor antagonist for the treatment of overactive bladder, in healthy volunteers. BJU international , 96 (7), 1055 1062. Kelly, J.F., Furukawa, K., Barger, S.W., Rengen, M.R., Mark, R.J., Blanc, E.M., Roth, G.S., & Mattson, M.P. (1996). Amyloid beta peptid e disrupts carbachol induced muscarinic cholinergic signal transduction in cortical neurons. Proceedings of the National Academy of Sciences , 93 (13), 6753 6758. Kersten, H., Molden, E., Tolo, I. K., Skovlund, E., Engedal, K., & Wyller, T. B. (2013). Cogni tive effects of reducing anticholinergic drug burden in a frail elderly population: a randomized controlled trial. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences , 68 (3), 271 278. Khachaturian, A. S., Zandi, P. P., Lyketsos, C. G., Hayden, K. M., Skoog, I., Norton, M. C., & Breitner, J. C. (2006). Antihypertensive medication use and incident Alzheimer disease: the Cache County Study. Archives of neurology , 63 (5), 686 692. Kirkby, D.L., Jones, D.N., & Higgins, G.A. (1995). In fluence of prefeeding and scopolamine upon performance in a delayed matching to position task. Behavioural brain research , 67 (2), 221 227. Konishi, K., Hori, K., Uchida, H., Watanabe, K., Tominaga, I., Kimura, M., & Hachisu, M. (2010). Adverse effects of anticholinergic activity on cognitive functions in Alzheimer's disease. Psychogeriatrics , 10 (1), 34 38. Koyama, A., Steinman, M., Ensrud, K., Hillier, T. A., & Yaffe, K. (2013). Ten Year Trajectory of Potentially Inappropriate Medications in Very Old Wom en: Importance of Cognitive Status. Journal of the American Geriatrics Society , 61 (2), 258 263.

PAGE 176

176 & Klaschka, J. (2001). Amyloid beta peptide 1 40 and the function of rat hippocampal hemicholinium 3 sensitive choline carrier s: effects of a proteolytic degradation in vitro. Neurochemical research , 26 (3), 203 212. Kruskal, W. H., & Wallis, W. A. (1952). Use of ranks in one criterion variance analysis. Journal of the American statistical Association , 47 (260), 583 621. Lachowicz , J. E., Duffy, R. A., Ruperto, V., Kozlowski, J., Zhou, G., Clader, J., & Coffin, V. (2001). Facilitation o f acetylcholine release and improvement in cognition by a selective M2 muscarinic antagonist, SCH 72788. Life sciences , 68 (22), 2585 2592. Lampela , P., Lavikainen, P., Garcia Horsman, J. A., Bell, J. S., Huupponen, R., & Hartikainen, S. (2013). Anticholinergic Drug Use, Serum Anticholinergic Activity, and Adverse Drug Events Among Older People: A Population Based Study. Drugs & aging , 1 10. Lecheva llier Michel, N., Molimard, M., Dartigues, J. F., Fabrigoule, C., & Fourrier Réglat, A. (2005). Drugs with anticholinergic properties and cognitive performance in the elderly: results from the PAQUID Study. British journal of clinical pharmacology , 59 (2), 143 151. Lee, T., Shiao, Y.J., Chen, C.F., & Wang, L.C. (2001). Effect of ginseng saponins on beta amyloid suppressed acetylcholine release from rat hippocampal slices. Planta Med., 67 , 634 637. Levene, H. (1960). In Contributions to Probability and Stat istics: Essays in Honor of Harold Hotelling . In I. Olkin et al. (eds.), (pp. 278 292).Stanford, CA: Stanford University Press. Lewis, J., Dickson, D. W., Lin, W. L., Chisholm, L., Corral, A., Jones, G., & McGowan, E. (2001). Enhanced neurofibrillary dege neration in transgenic mice expressing mutant tau and APP. Science , 293 (5534), 1487 1491. Linn, R. T., Wolf, P. A., Bachman, D. L., Knoefel, J. E., Cobb, J. L., Belanger, A. J., & D'Agostino, R. B. (1995). The 'preclinical phase' of probable Alzheimer's d isease: a 13 year prospective study of the Framingham cohort. Archives of Neurology , 52 (5), 485 490. Little, J. T., Johnson, D. N., Minichiello, M., Weingartner, H., & Sunderland, T. (1998). Combined nicotinic and muscarinic blockade in elderly normal vol unteers: cognitive, behavioral, and physiologic responses. Neuropsychopharmacology , 19 (1), 60 69. Liu, Q., Kawai, H., & Berg, D.K. (2001). Beta amyloid peptide blocks the response of alpha7 containing nicotinic receptors on hippocampal neurons. Proceeding s of the National Academy of Sciences of the United States of America , 98 , 4734 4739. Locascio, J. J., Growdon, J. H., & Corkin, S. (1995). Cognitive test performance in detecting, staging, and tracking Alzheimer's disease. Archives of Neurology , 52 (11), 1087 1099.

PAGE 177

177 Lopez, O. L., Becker, J. T., Wisniewski, S., Saxton, J., Kaufer, D. I., & DeKosky, S. T. (2002). Cholinesterase inhibitor treatment alters the natural history of Alzheimer's disease. Journal of Neurology, Neurosurgery & Psychiatry , 72 (3), 310 3 14. Lovestone, S., & Reynolds, C. H. (1997). The phosphorylation of tau: a critical stage in neurodevelopment and neurodegenerative processes. Neuroscience , 78 (2), 309 324. Lu, C. J., & Tune, L. E. (2003). Chronic exposure to anticholinergic medications adversely affects the course of Alzheimer disease. The American journal of geriatric psychiatry , 11 (4), 458 461. Lucas Meunier, E., Fossier, P., Baux, G., & Amar, M. (2003). Cholinergic modulation of the cortical neuronal network. Pflügers Archiv , 446 (1) , 17 29. Luine, V.N., Renner, K.J., Heady, S., & Jones, K.J. (1986). Age and sexdependent decreases in ChAT in basal forebrain nuclei. Neurobiology of aging , 7 (3), 193 198. Luukkanen, M. J., Uusvaara, J., Laurila, J. V., Strandberg, T. E., Raivio, M. M., Tilvis, R. S., & Pitkälä, K. H. (2011). Anticholinergic drugs and their effects on delirium and mortality in the elderly. Dementia and geriatric cognitive disorders extra , 1 (1), 43 50. Lyketsos, C. G., Colenda, C. C., Beck, C., Blank, K., Doraiswamy, M. P., Kalunian, D. A., & Yaffe, K. (2006). Position statement of the American Association for Geriatric Psychiatry regarding principles of care for patients with dementia resulting from Alzheimer disease. The American journal of geriatric psychiatry , 14 (7), 561 573. Marengoni, A., Rizzuto, D., Wang, H. X., Winblad, B., & Fratiglioni, L. (2009). Patterns of chronic multimorbidity in the elderly population. Journal of the American Geriatrics Society , 57 (2), 225 230. Mayeux, R., & Stern, Y. (2012). Epidemiolog y of Alzheimer disease. Cold Spring Harbor perspectives in medicine. Cold spring Harbor Perspectives in medicine . Retrieved from http://perspe ctivesinmedicine.cshlp.org/content/early/2012/04/10/cshperspect.a006239.f ull.pdf+html McGurk, S. R., Mueser, K. T., & Pascaris, A. (2005). Cognitive training and supported employment for persons with severe mental illness: one year results from a random ized controlled trial. Schizophrenia Bulletin , 31 (4), 898 909. Meltzer, H. Y., & McGurk, S. R. (1999). The effects of clozapine, risperidone, and olanzapine on cognitive function in schizophrenia. Schizophrenia Bulletin , 25 (2), 233 255. Mena, R., Edwards , P. C., Harrington, C. R., Mukaetova Ladinska, E. B., & Wischik, C. M. (1996). Staging the pathological assembly of truncated tau protein into paired helical Acta neuropathologica , 91 (6), 633 641.

PAGE 178

178 Messer Jr, W. S., Bohne tt, M., & Stibbe, J. (1990). Evidence for a preferential involvement o f M l muscarinic receptors in representational memory. Neuroscience letters , 116 (1), 184 189. Mesulam, M.M., Mufson, E.J., & Rogers, J. (1987). Age related shrinkage of cortically proje cting cholinergic neurons: a selective effect. Annals of neurology , 22 (1), 31 36. Michalek, H., Fortuna, S., & Pintor, A. (1989). Age related differences in brain choline acetyltransferase, cholinesterases and muscarinic receptor sites in two strains of r ats. Neurobiology of aging , 10 (2), 143 148. Mintzer, J., & Burns, A. (2000). Anticholinergic side effects of drugs in elderly people. JRSM , 93 (9), 457 462. Mirra, S. S., Heyman, A., McKeel, D., Sumi, S. M., Crain, B. J., Brownlee, L. M. & Berg, L. (1991) . The Consortium to Establish a Registry for Alzheimer's Disease (CERAD) Part II. Standardization of the neuropathologic assessment of Alzheimer's disease. Neurology , 41 (4), 479 479. Molchan, S. E., Martinez, R. A., Hill, J. L., Weingartner, H. J., Thomps on, K., Vitiello, B., & Sunderland, T. (1992). Increased cognitive sensitivity to scopolamine with age and a perspective on the scopolamine model. Brain research reviews , 17 (3), 215 226. Moore, A. R., & induced cognitive impai rment in the elderly. Drugs & aging , 15 (1), 15 28. Moore, H., Sarter, M., & Bruno, J. P. (1995). Bidirectional modulation of cortical acetylcholine efflux by infusion of benzodiazepine receptor ligands into the basal forebrain. Neuroscience letters , 189 (1 ), 31 34. Morris, J. C. (1993). The Clinical Dementia Rating (CDR): current version and scoring rules. Neurology , 2412 2414 Morris, J. C., Heyman, A., Mohs, R. C., Hughes, J. P., van Belle, G., Fillenbaum, G., & Clark, C. (1989). The consortium to establ ish a registry for alzheimer's disease (CERAD). part I. clinical and neuropsychological assessment of alzheimer's disease. Neurology, 39 (9), 1159 1165. Morris, J. C., Weintraub, S., Chui, H. C., Cummings, J., DeCarli, C., Ferris, S., & Kukull, W. A. (200 6). The Uniform Data Set (UDS): clinical and cognitive variables and descriptive data from Alzheimer Disease Centers. Alzheimer Disease & Associated Disorders , 20 (4), 210 216. Morrison, J. H., & Hof, P. R. (1997). Life and death of neurons in the aging br ain. Science , 278 (5337), 412 419.

PAGE 179

179 Mountjoy, C.Q., Rossor, M.N., Iversen, L.L., & Roth, M. (1984). Correlation of cortical cholinergic and GABA deficits with quantitative neuropathological findings in senile dementia. Brain, 107 , 507 518. Muir, J.L., Ever itt, B.J., & Robbins, T.W. (1995). Reversal of visual attentional dysfunction following lesions of the cholinergic basal forebrain by physostigmine and nicotine but not by the 5 HT3 receptor antagonist, ondansetron. Psychopharmacology, 118 , 82 92. Mulsant , B. H., Pollock, B. G., Kirshner, M., Shen, C., Dodge, H., & Ganguli, M. (2003). Serum anticholinergic activity in a community based sample of older adults: relationship with cognitive performance. Archives of general psychiatry , 60 (2), 198 203. Mundy, W ., Tandon, P., Ali, S., & Tilson, H. (1991). Age related changes in receptor mediated phosphoinositide hydrolysis in various regions of rat brain. Life sciences , 49 (14), PL97 PL102. Murphy, S. L., Xu, J. Q., & Kochanek, K. D. (2013). Deaths: final data f or 2010. National vital statistics reports , 61 (4), 1 118. Nacharaju, P., Lewis, J., Easson, C., Yen, S., Hackett, J., Hutton, M., & Yen, S. H. (1999). Accelerated filament formation from tau protein with specific FTDP 17 missense mutations. FEBS letters , 447 (2), 195 199. Nagy, Z. S., Esiri, M. M., Jobst, K. A., Morris, J. H., King, E. F., McDonald, B., & Smith, A. D. (1995). Relative roles of plaques and tangles in the dementia of Alzheimer's disease: correlations using three sets of neuropathological cri teria. Dementia and Geriatric Cognitive Disorders , 6 (1), 21 31. Naranjo, C. A., Fourie, J., Herrmann, N., Lanctôt, K. L., Birt, C., & Yau, K. K. (2000). Probing peripheral and central cholinergic system responses. Journal of Psychiatry and Neuroscience , 2 5 (4), 325 336. Näslund, J., Haroutunian, V., Mohs, R., Davis, K. L., Davies, P., Greengard, P., & Buxbaum, J. D. (2000). Correlation peptide in the brain and cognitive decline. JAMA: the journal of the American Medical Association , 283 (12), 1571 1577. Naveh Benjamin, M., Brav, T. K., & Levy, O. (2007). The associative memory deficit of older adults : the role of strategy utilization. Psychology and aging , 22 (1), 202 208. Nebes, R. D., Pollock, B. G., Halligan, E. M., Kirshner, M. A., & Houck, P. R. (2007). Serum anticholinergic activity and motor performance in elderly persons. The Journals of Geron tology Series A: Biological Sciences and Medical Sciences , 62 (1), 83 85.

PAGE 180

180 Nebes, R. D., Pollock, B. G., Meltzer, C. C., Saxton, J. A., Houck, P. R., Halligan, E. M., & DeKosky, S. T. (2005). Serum anticholinergic activity, white matter hyperintensities, an d cognitive performance. Neurology , 65 (9), 1487 1489. Newman, A. B., Fitzpatrick, A. L., Lopez, O., Jackson, S., Lyketsos, C., Jagust, W., & Kuller, L. H. (2005). Dementia and Alzheimer's disease incidence in relationship to cardiovascular disease in the Cardiovascular Health Study cohort. Journal of the American Geriatrics Society , 53 (7), 1101 1107. Nilsson, L., Nordberg, A., Hardy, J., Wester, P., & Winblad, B. (1986). Physostigmine restores 3H acetylcholine efflux from Alzheimer brain slices to normal level. J. Neural Transm, 67 , 275 285. Nitsch, R.M., Blusztajn, J.K., Pittas, A.G., Slack, B.E., Growdon, J.H., & Wurtman, R.J. (1992). Evidence for a membrane defect in Alzheimer disease brain. Proceedings of the National Academy of Sciences of the United States of America, 89 , 1671 1675. Nordberg, A., Alafuzoff, I., & Winblad, B. (1992). Nicotinic and muscarinic subtypes in the human brain: changes with aging and dementia. Journal of neuroscience research , 31 (1), 103 111. Nourhashémi, F., Ousset, P. J., Gillette Guyonnet, S., Cantet, C., Andrieu, S., & Vellas, B. (2008). A 2 year follow up of 233 very mild (CDR 0.5) Alzheimer's disease patients (REAL. FR cohort). International journal of geriatric psychiatry , 23 (5), 460 465. Nowrangi, M. A., Rao, V., & L yketsos, C. G. (2011). Epidemiology, assessment, and treatment of dementia. Psychiatric Clinics of North America , 34 (2), 275 294. & Doody, R. (2008). Staging dementia us ing Clinical Dementia Rating Scale Sum of Boxes scores: a Texas Alzheimer's research consortium study. Archives of neurology , 65 (8), 1091 1095. Oberauer, K. (2001). Removing irrelevant information from working memory: a cognitive aging study with the modif ied Sternberg task. Journal of Experimental Psychology: Learning, Memory, and Cognition , 27 (4), 948 957. Oberauer, K. (2002). Access to information in working memory: exploring the focus of attention. Journal of Experimental Psychology: Learning, Memory, and Cognition , 28 (3), 411 421. Ogane, N., Giacobini, E., & Struble, R. (1992). Differential inhibition of acetylcholinesterase Brain research , 589 (2), 307 312.

PAGE 181

181 Ohrui, T., Tomita, N., Sato Nakagawa , T., Matsui, T., Maruyama, M., Niwa, K., & Sasaki, H. (2004). Effects of brain penetrating ACE inhibitors on Alzheimer disease progression. Neurology , 63 (7), 1324 1325. Onder, G., Liperoti, R., Fialova, D., Topinkova, E., Tosato, M., Danese, P., & Landi, F. (2012). Polypharmacy in nursing home in Europe: results from the SHELTER study. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences , 67 (6), 698 704. Parasuraman, R., & Haxby, J. V. (1993). Attention and brain function in Alz heimer's disease: A review. Neuropsychology , 7 (3), 242 272. Pasina, L., Djade, C. D., Lucca, U., Nobili, A., Tettamanti, M., Franchi, C., & Mannucci, P. M. (2013). Association of anticholinergic burden with cognitive and functional status in a cohort of h ospitalized elderly: comparison of the anticholinergic cognitive burden scale and anticholinergic risk scale. Drugs & aging , 30 (2), 103 112. Perez, M., Lim, F., Arrasate, M., & Avila, J. (2002). The FTDP 17 Linked Mutation R406W Abolishes the Interaction of Phosphorylated Tau with Microtubules. Journal of neurochemistry , 74 (6), 2583 2589. The British Journal of Psychiatry , 152 (6), 737 740. Perry , E. K., Tomlinson, B. E., Blessed, G., Bergmann, K., Gibson, P. H., & Perry, R. H. (1978). Correlation of cholinergic abnormalities with senile plaques and mental test scores in senile dementia. British Medical Journal , 2 (6150), 1457 1459. Perry, E.K., B lessed, G., Tomlinson, B.E., Perry, R.H., Crow, T.J., Cross, A.J., Dockray, G.J., Dimaline, R., & Arregui, A. (1981). Neurochemical activities in human temporal lobe Neurobiology of aging , 2 (4), 251 256. Per ry, E.K., Morris, C.M., Court, J.A., Cheng, A., Fairbairn, A.F., McKeith, I.G., Irving, D., Brown, A., & ne uropathology. Neuroscience, 64 (2) 385 395. Perry, E.K., Perry, R.H., Blessed, G., & Tomlinson, B.E. (1977). Necropsy evidence of central cholinergic deficits in senile dementia. Lancet, 1 , 189 189. Peters, N. L. (1989). Snipping the thread of life: anti muscarinic side effects of medications in the elderly. Archives of internal medicine , 149 (11), 2414 2420. Petersen, R. C., Smith, G., Kokmen, E., Ivnik, R. J., & Tangalos, E. G. (1992). Memory function in normal aging. Neurology , 42 (2), 396 396.

PAGE 182

182 Pettit, D.L., Shao, Z., & Yakel, J.L. (2001). Beta amyloid(1 42) peptide directly modulates nicotinic receptors in the rat hippocampal slice. The Journal of neuroscience: the official journal of the Society for Neuroscience , 21 (1), RC120 RC120. Pfeffer, R. I., Ku rosaki, T. T., Harrah, C. H., Chance, J. M., & Filos, S. (1982). Measurement of functional activities in older adults in the community. Journal of Gerontology , 37 (3), 323 329. Pilcher, J.J., Sessions, G.R., & McBride, S.A. (1997). Scopolamine impairs spat ial working memory in the radial maze: an analysis by error type and arm choice. Pharmacology Biochemistry and Behavior , 58 (2), 449 459. Pitkala, K. H., Strandberg, T. E., & Tilvis, R. S. (2001). Is it possible to reduce polypharmacy in the elderly?. Dr ugs & aging , 18 (2), 143 149. Price, J. L., & Morris, J. C. (1999). Tangles and plaques in nondemented aging and "preclinical" Alzheimer's disease. Annals of neurology , 45 (3), 358 368. Qiu, C., Winblad, B., Marengoni, A., Klarin, I., Fastbom, J., & Fratig lioni, L. (2006). Heart failure and risk of dementia and Alzheimer disease: a population based cohort study. Archives of Internal Medicine , 166 (9), 1003 1008. Ravel, N., Vigouroux, M., Elaagouby, A., & Gervais, R. (1992). Scopolamine impairs delayed match ing in an olfactory task in rats. Psychopharmacology, 109 , 439 443. Ray, P.G., Meador, K.J., Loring, D.W., Zamrini, E.W., Yang, X.H., & Buccafusco, J.J. (1992). Central anticholinergic hypersensitivity in aging. Journal of geriatric psychiatry and neurolo gy , 5 (2), 72 77. Reitan, R. M., & Wolfson, D. (1986). The Halstead Reitan neuropsychological test battery. The Neuropsychology Handbook: Behavioral and Clinical Perspectives. New York, NY: Springer . Rengstorff, R.H. (1994). Vision and ocular changes foll owing accidental exposure to organophosphates. Journal of Applied Toxicology , 14 (2), 115 118. Richelson, E. (1994, November). Pharmacology of antidepressants -characteristics of the ideal drug. In Mayo Clinic proceedings. Mayo Clinic (Vol. 69, No. 11, pp . 1069 1081). Elsevier. Riekkinen Jr., P., Riekkinen, M., Sirvio, J., Miettinen, R., & Riekkinen, P. (1992). Loss of cholinergic neurons in the nucleus basalis induces neocortical electroencephalographic and passive avoidance deficits. Neuroscience, 47 , 8 23 831. Rizzu, P., Joosse, M., Ravid, R., Hoogeveen, A., Kamphorst, W., Van Swieten, J. C., & Heutink, P. (2000). Mutation dependent aggregation of tau protein and its selective

PAGE 183

183 depletion from the soluble fraction in brain of P301L FTDP 17 patients. Human molecular genetics , 9 (20), 3075 3082. Rosenberg, P. B., Mielke, M. M., Han, D., Leoutsakos, J. S., Lyketsos, C. G., Rabins, P. V., & Tschanz, J. T. (2012). The association of psychotropic medication use with the cognitive, functional, and neuropsychiatric trajectory of Alzheimer's disease. International journal of geriatric psychiatry , 27 (12), 1248 1257. Rosene, D.L., (1993). Comparing age related changes in the basal forebrain and hippocampus of th e rhesus monkey. Neurobiol. Aging, 14 , 669 670. Rubio, A., Perez, M., & Avila, J. (2006). Acetylcholine receptors and tau phosphorylation. Current molecular medicine , 6 (4), 423 428. Rudd, K. M., Raehl, C. L., Bond, C. A., Abbruscato, T. J., & Stenhouse , A. C. (2005). Methods for Assessing Drug Related Anticholinergic Activity. Pharmacotherapy: The Journal of Human Pharmacology and Drug Therapy , 25 (11), 1592 1601. Rudolph, J. L., Salow, M. J., Angelini, M. C., & McGlinchey, R. E. (2008). The anticholine rgic risk scale and anticholinergic adverse effects in older persons. Archives of Internal Medicine , 168 (5), 508 513. Rylett, R.J., Ball, M.J., & Colhoun, E.H. (1983). Evidence for high affinity choline transport in synaptosomes prepared from hippocampus disease. Brain research , 289 (1), 169 175. Sadock, B. J., & Sadock, V. (2008). Kaplan and Sadock's Concise Textbook of Clinical Psychiatry . Philadelphia, PA: Lippincott Williams & Wilkins. Sahakian, B., Jones, G. , Levy, R., Gray, J., & Warburton, D. A. V. I. D. (1989). The effects of nicotine on attention, information processing, and short term memory in patients with dementia of the Alzheimer type. The British Journal of Psychiatry , 154 (6), 797 800. Sarter, M., & Bruno, J.P. (1998). Age related changes in rodent cortical acetylcholine and cognition: main effects of age versus age as an intervening variable. Brain research reviews , 27 (2), 143 156. Satoh, Y., Hirakura, Y., Shibayama, S., Hirashima, N., Suzuki, T. , & Kirino, Y. (2001). Beta amyloid peptides inhibit acetylcholine release from cholinergic pre synaptic nerve endings isolated from an electric ray. Neurosci ence Lett ers , 302 , 97 100. Schegg, K.M., Harrington, L.S., Neilsen, S., Zweig, R.M., & Peacock, J .H., ( 1992 ) . Soluble and membrane Neurobiology of aging , 13 (6), 697 704.

PAGE 184

184 Schubert, C. C., Boustani, M., Callahan, C. M., Perkins, A. J., Carney, C. P., Fox, C., & Hendrie, H. C. (2006). Co morbidity profile of dementia patients in primary care: are they sicker? Journal of the American Geriatrics Society , 54 (1), 104 109. Seeger, T., Fedorova, I., Zheng, F., Miyakawa, T., Koustova, E., Gomeza, J., & Wess, J. (2004). M2 muscarinic acetylcholin e receptor knock out mice show deficits in behavioral flexibility, working memory, and hippocampal plasticity. The Journal of neuroscience , 24 (45), 10117 10127. Sergi, G., De Rui, M., Sarti, S., & Manzato, E. (2011). Polypharmacy in the Elderly. Drugs & a ging , 28 (7), 509 518. Sessions, G.R., Pilcher, J.J., & Elsmore, T.F. (1998). Scopolamine induced impairment in concurrent fixed interval responding in a radial maze task. Pharmacology Biochemistry and Behavior , 59 (3), 641 647. Seubert, P., Vigo Pelfrey, C., Esch, F., Lee, M., Dovey, H., Davis, D., Sinha, S., Schlossmacher, M., Whaley, J., & Swindlehurst, C. (1992). Isolation and quantification peptide from biological fluids. Nature, 359 , 325 327. Shah, R. C., Janos, A. L., Kl ine, J. E., Yu, L., Leurgans, S. E., Wilson, R. S., & Tsao, J. W. (2013). Cognitive Decline in Older Persons Initiating Anticholinergic Medications. PloS one , 8 (5), e64111. Shapiro, S. S., & Wilk, M. B. (1965). An analysis of variance test for normality ( complete samples). Biometrika , 52 (3/4), 591 611. Shaw, L. M., Vanderstichele, H., Knapik Czajka, M., Clark, C. M., Aisen, P. S., Petersen, R. C., & Trojanowski, J. Q. (2009). Cerebrospinal fluid biomarker signature in Alzheimer's disease neuroimaging initiative subjects. Annals of neurology , 65 (4), 403 413 . Shirk, S. D., Mitchell, M. B., Sha ughnessy, L. W., Sherman, J. C., Locascio, J. J., Weintraub, S., & Atri, A. (2011). A web based normative calculator for the uniform data set (UDS) neuropsychological test battery. , 3 (6), 32 40. Siek, G. C., Katz, L. S., Fis hman, E. B., Korosi, T. S., & Marquis, J. K. (1990). Molecular forms of acetylcholinesterase in subcortical areas of normal and Alzheimer disease brain. Biological psychiatry , 27 (6), 573 580. Snyder, S. H., & Yamamura, H. I. (1977). Antidepressants and th e muscarinic acetylcholine receptor. Archives of General Psychiatry , 34 (2), 236 239. Sosa Ortiz, A. L., Acosta Castillo, I., & Prince, M. J. (2012). Epidemiology of dementias and 608.

PAGE 185

185 Steinber g, B. A., Bieliauskas, L. A., Smith, G. E., & Ivnik, R. J. (2005). Mayo's Older Americans Normative Studies: Age and IQ adjusted norms for the trail making test, the stroop test, and MAE controlled oral word association test. The Clinical Neuropsychologist , 19 (3 4), 329 377. Storandt, M., & Hill, R. D. (1989). Very mild senile dementia of the Alzheimer type: II. Psychometric test performance. Archives of Neurology , 46 (4), 383 386. Strittmatter, W. J., Saunders, A. M., Schmechel, D., Pericak Vance, M., Eng hild, J., Salvesen, G. S., & Roses, A. D. (1993). Apolipoprotein E: high avidity binding to beta amyloid and increased frequency of type 4 allele in late onset familial Alzheimer disease. Proceedings of the National Academy of Sciences , 90 (5), 1977 1981. Strong, R., Hicks, P., Hsu, L., Bartus, R.T., & Enna, S.J. (1980). Age related alterations in the rodent brain cholinergic system and behavior. Neurobiology of aging , 1 (1), 59 63. Sunderland, T., Tariot, P. N., & Newhouse, P. A. (1988). Differential respo nsivity of mood, behavior and cognition to cholinergic agents in elderly neuropsychiatric populations. Brain Research Reviews , 13 (4), 371 389. Sunderland, T., Tariot, P. N., Cohen, R. M., Weingartner, H., Mueller III, E. A., & Murphy, D. L. (1987). Antich olinergic sensitivity in patients with dementia of the Alzheimer type and age matched controls: a dose response study. Archives of general psychiatry , 44 (5), 418 426. Tariot, P. N., Patel, S. V., Cox, C., & Henderson, R. E. (1996). Age related decline in central cholinergic function demonstrated with scopolamine. Psychopharmacology , 125 (1), 50 56. Taylor, P. (2001). Anticholinesterase agents. In J.G. Hardman, L.E. Limbird, & A.G. Gilman (Eds.), The Pharmacological Basis of Therapeutics , 10th ed., (pp. 175 191)., New York, NY: McGraw Hill. Terry Jr, A. V., Buccafusco, J. J., & Jackson, W. J. (1993). Scopolamine reversal of nicotine enhanced delayed matching to sample performance in monkeys. Pharmacology Biochemistry and Behavior , 45 (4), 925 929. Thal, L.J ., Gilbertson, E., Armstrong, D.M., & Gage, F.H. (1992). Development of the basal forebrain cholinergic system: phenotype expression prior to target innervation. Neurobiology of aging , 13 (1), 67 72. Thienhaus, O. J., Allen, A., Bennett, J. A., Chopra, Y. M., & Zemlan, F. P. (1990). Anticholinergic serum levels and cognitive performance. European archives of psychiatry and clinical neuroscience , 240 (1), 28 33.

PAGE 186

186 Thomson, H.T., & Cotterill, D. (1909). Note on the use of scopol amine morphine combination as an anaesthetic adjunct. Edinburgh Medical Journal (New Series), 3 , 548 554. Tiraboschi, P., Hansen, L. A., Alford, M., Sabbagh, M. N., Schoos, B., Masliah, E., & Corey Bloom, J. (2000). Cholinergic dysfunction in diseases with Lewy bodies. Neurology , 54 (2) , 407 411. Tune, L. E. (2000). Anticholinergic effects of medication in elderly patients. The Journal of clinical psychiatry , 62 , 11 14. Tune, L. E. (2000, April). Serum anticholinergic activity levels and delirium in the elderly. In Seminars in clinica l neuropsychiatry (Vol. 5, No. 2, p. 149). Tune, L., & Coyle, J. T. (1980). Serum levels of anticholinergic drugs in treatment of acute extrapyramidal side effects. Archives of general psychiatry , 37 (3), 293 297. Tune, L., Carr, S., Cooper, T., & Klug, B . (1993). Association of anticholinergic activity of prescribed medications with postoperative delirium. The Journal of Neuropsychiatry and Clinical Neurosciences, 5 (2), 208 210 . Tune, L., Carr, S., Hoag, E., & Cooper, T. (1992). Anticholinergic effects o f drugs commonly prescribed for the elderly: potential means for assessing risk of delirium. American Journal of Psychiatry , 149 (10), 1393 1394. Tzavara, E. T., Bymaster, F. P., Felder, C. C., Wade, M., Gomeza, J., Wess, J., & Nomikos, G. G. (2003). Dysre gulated hippocampal acetylcholine neurotransmission and impaired cognition in M2, M4 and M2/M4 muscarinic receptor knockout mice. Molecular psychiatry , 8 (7), 673 679. Uusvaara, J., Pitkala, K. H., Kautiainen, H., Tilvis, R. S., & Strandberg, T. E. (2013). Detailed Cognitive Function and Use of Drugs with Anticholinergic Properties in Older People. Drugs & aging , 1 6. Uusvaara, J., Pitkala, K. H., Tienari, P. J., Kautiainen, H., Tilvis, R. S., & Strandberg, T. E. (2009). Association Between Anticholinergic Poorer Cognitive Function in Older Cardiovascular Patients: A Cross Sectional Study. Journal of the American Geriatrics Society , 57 (3), 427 431. Vaucher, E., Aumont, N., Pearson, D., Rowe, W., Poirier, J., & Kar, S. (2001). Amyloid beta peptide levels and its effects on hippocampal acetylcholine release in aged, cognitively impaired and unimpaired rats. Journal of chemical neuroanatomy , 21 (4), 323 329. Voytko, M.L., Olton, D.S., Richardson, R.T., Gorman, L.K., To bin, J.R., & Price, D.L. (1994). Basal forebrain lesions in monkeys disrupt attention but not learning and memory. The Journal of neuroscience , 14 (1), 167 186.

PAGE 187

187 Voytko, M.L., Sukhov, R.R., Walker, L.C., Breckler, S.J., Price, D.L., & Koliatsos, V.E. (1995) . Neuronal number and size are preserved in the nucleus basalis of aged rhesus monkeys. Dementia, 6 , 131 141. Wagstaff, A.J., & McTavish, D., ( 1994 ) . Tacrine. A review of its pharmacodynamics and pharmacokinetic properties, and therapeutic efficacy in Al Drugs Aging, 4 , 510 540. Walsh, D.M., Klyubin, I., Fadeeva, J.V., Cullen, W.K., Anwyl, R., Wolfe, M.S., Rowan, M.J., & Selkoe, D.J. (2002). Naturally secreted oligomers of amyloid beta protein potently inhibit hippocampal long term pote ntiation in vivo. Nature, 416 , 535 539. Wang, H. Y., Li, W., Benedetti, N. J., & amyloid peptide induced tau protein phosphorylation. Journal of Biological Chemistry , 278 (34), 31547 31553. Wang, H.Y., Wild, K.D., Shank, R.P., & Lee, D.H. (1999). Galanin inhibits acetylcholine release from rat cerebral cortex via a pertussis toxin sensitive G(i)protein. Neuropeptides, 33 , 197 205. Wawruch, M., Macugova, A., Kostkova, L., Luha, J., Dukat, A ., Murin, J., & Kuzelova, M. (2012). The use of medications with anticholinergic properties and risk factors for their use in hospitalised elderly patients. Pharmacoepidemiology and drug safety , 21 (2), 170 176. Wechsler, D. (1986). WAIS R: Wechsler Adult Intelligence Scale, Revised . Psychological Corporation. Wechsler, D. (1987). WMS R: Wechsler Memory Scale -Revised: manual . San Antonio: Psychological Corporation. Wesnes, K., & Warburton, D. M. (1984). Effects of scopolamine and nicotine on human rapid information processing performance. Psychopharmacology , 82 (3), 147 150. Whitehouse, P.J., Martino, A.M., Wagster, M.V., Price, D.L., Mayeux, R., Atack, J.R., & Kellar, nd Neurology, 38 , 720 723. Whitehouse, P.J., Price, D.L., Clark, A.W., Coyle, J.T., & DeLong, M.R. (1981). Alzheimer disease: evidence for selective loss of cholinergic neurons in the nucleus basalis. A nnal s of neurology, 10(2), 122 126. Whitehouse, P.J., Price, D.L., Struble, R.G., Clark, A.W., Coyle, J.T., & Delon, M.R., (1982). Science, 215 , 1237 1239.

PAGE 188

188 WHO (facts sheets). Dementia. WHO . Retrieved from http://www.who.int/mediacentre/factsheets/fs362/en/ Wilcock, G.K., Esiri, M.M., Bowen, D.M., & Correlation of corti cal choline acetyltransferase activity with the severity of dementia and histological abnormalities. Journal of the neurological sciences , 57 (2), 407 417. Wolff, J. L., Starfield, B., & Anderson, G. (2002). Prevalence, expenditures, and complications of multiple chronic conditions in the elderly. Archives of internal medicine , 162 (20), 2269 2276. Yamada, M., Miyakawa, T., Duttaroy, A., Yamanaka, A., Moriguchi, T., Makita, R., & Wess, J. (2001). Mice lacking the M3 muscarinic acetylcholine receptor are hy pophagic and lean. Nature , 410 (6825), 207 212. Yesavage, J. A. (1988). Geriatric depression scale. Psychopharmacology Bulletin, 24 (4), 709 711. Yesavage, J. A., & Sheikh, J. I. (1986). 9/Geriatric Depression Scale (GDS) Recent Evidence and Development o f a Shorter Violence. Clinical gerontologist , 5 (1 2), 165 173. Younkin, S.G., Goodridge, B., Katz, J., Lockett, G., Nafziger, D., Usiak, M.F., & Younkin, L.H., ( 1986 ) In Federation proceed ings (Vol. 45, No. 13, pp. 2982 2988). Zemishlany, Z., & Thorne, A. E. (1991). Anticholinergic challenge and cognitive functions: A comparison between young and elderly normal subjects. Israel journal of psychiatry and related sciences , 28 (3), 32 41.

PAGE 189

189 B IOGRAPHICAL SKETCH Sunil Swami is pursuing his doctorate degree in the field of e pidemiology and a graduate certificate in gerontology from the College of Public Health and Health Professions at the University of Florida. He received his degree in medicine , MBBS, from Pt. BD Sharma University of Health Science, India; and his M aster in Public Health from East Tennessee State University. He is currently working as a graduate research assistant in the Department of Aging and Geriatric Research at the Institut e on Aging, University of Florida. Prior to joining the doctoral program, he has worked as a clinician in India and as a clinical research coordinator at East Tennessee State University. His , delirium , dementia, and aging.