Citation
Healthcare Utilization in Insomnia Patients with Comorbid Depression and/or Anxiety

Material Information

Title:
Healthcare Utilization in Insomnia Patients with Comorbid Depression and/or Anxiety
Creator:
Mosti, Caterina
Publication Date:
Language:
English

Subjects

Subjects / Keywords:
Anxiety ( jstor )
Cost estimates ( jstor )
Dysthymic disorder ( jstor )
Health care costs ( jstor )
Insomnia ( jstor )
Medications ( jstor )
Primary health care ( jstor )
Sleep ( jstor )
Sleep initiation and maintenance disorders ( jstor )
Transponders ( jstor )
Cognitive therapy
Hospital utilization
Insomnia
Genre:
Undergraduate Honors Thesis

Notes

Abstract:
Chronic insomnia is associated with significant financial costs and is often comorbid with psychiatric disorders, particularly anxiety and depression. Cognitive-behavioral therapy of insomnia (CBTi) is an efficacious treatment, but its impact on healthcare utilization (HCU) is largely unknown. This study compared HCU pre- and post-CBTi of patients with chronic insomnia with and without comorbid depression and/or anxiety diagnosis. Greater pre-treatment HCU among patients with comorbid insomnia was expected. Following successful treatment, reductions in HCU for both groups were expected, with even greater reductions for those with chronic insomnia without anxiety and/or depression. A review of records was conducted for patients treated for insomnia (N =38; age M =52.1, SD = 18.86) at a behavioral sleep medicine clinic in an academic medical center from 2005-2010. Patients with chronic insomnia were characterized into four groups: 1) treatment responders without anxiety and/or depression, 2) treatment responders with anxiety and/or depression, 3) treatment non-responders without anxiety and/or depression, and 4) treatment non-responders with anxiety and/or depression. HCU was measured six months pre/post-treatment using a Chronic Disease Score (Clark et. al, 1995). There were no significant differences in HCU pre-CBTi. Differences in post-CBTi total (p = .08) and outpatient (p = .07) HCU and primary care visits (p = .07) HCU trended towards significance among treatment responders. No significant HCU differences were found in non-responders or between groups who responded positively to treatment. Results of this study imply that CBTi may help reduce healthcare service use. ( en )
General Note:
Awarded Bachelor of Science; Graduated May 8, 2012 summa cum laude. Major: Psychology
General Note:
College of Liberal Arts and Sciences
General Note:
Advisor: Christina McCrae

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Caterina Mosti. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EZ2VUDREQ_4EUNZV INGEST_TIME 2017-06-01T19:20:37Z PACKAGE AA00057730_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

Running Head: HCU OF COMORBID INSOMNIACS 1 Health Care Utilization of Insomnia Patients with Comorbid Depression and/or Anxiety Caterina Mosti University of Florida

PAGE 2

HCU COSTS OF COMORBID INSOMNIACS 2 Abstract Chronic insomnia is associated with significant financial costs and is often comorbid with psychiatric di sorders, particularly anxiety and depression. Cognitive behavioral t herapy of insomnia (CBTi) is an efficacious treatment, but its impact on healthcare utilization (HCU) is largely unknown. This study compared HCU pre and post CBTi of patients with chroni c insomnia with and without comorbid depression and/or anxiety diagnosis. G reater pre treatment HCU among patients with comorbid insomnia was expected Following successful treatment, reductions in HCU for both groups were expected with even greater reduc tions for those with chronic insomnia without anxiety and/or depression A review of records was conducted for pat ients treated for insomnia ( N =38 ; age M =52.1, SD = 18.86 ) at a behavioral sleep medicine clinic in an academic medical center from 2005 2010 Patients with chronic insomnia were characterized into four groups : 1) treatment responders without anxiety and/or depression 2) treatment responders with anxiety and/or depression 3) treatment non responders without anxiety and/or depression and 4) t reatment non responders with anxiety and/or depression HCU was measured six months pre/post treatment using a Chronic Disease Score (Clark et. al, 1995) There were no significant differences in HCU pre CBTi D ifferences in post CBTi total ( p = .0 8 ) and o utpatient ( p = .07) HCU and primary care visits ( p = .07) HCU trend ed towards significance among treatment responders N o significant HCU differences were found in non responders or between groups who responded positively to treatment. R esults of this stud y imply that CBTi may help reduce healthcare service use

PAGE 3

HCU COSTS OF COMORBID INSOMNIACS 3 Health Care Utilization of Insomnia Patients with Comorbid Depression and/or Anxiety Economic Costs of Insomnia Currently, over 60 million Americans suffer from insomnia (Chilcott & Shapiro, 199 6) and depending on the definition of insomnia, 9% to 22 % of the general population meets criteria for a chronic insomnia disorder (Ohayon, 1997 Roth et. al, 2011 ). Accordin g to Walsh and Engelhard t (1999) direct costs of insomnia were estimated to be ap proximately $13.9 billion in 1995 ; another study that measured both direct (e.g. cost of care) and indirect c osts (e.g. workplace absenteeism lost productivity) of insomnia estimated total costs of $92.5 billion to $107.5 billion (Stoller, 1994). Recent r esearch indicate s that direct costs of insomnia are estimated to be at $ 14 billion to $ 21 billion (Daley et. al, 2009). According to the literature, the direct and indirect costs of insomnia have increased significantly within the last 15 years (Daley et. al, 2009) On average, young er and old er adults with untreated insomnia have a n increased cost bur den of $1,253 and $1,143 respectively when compared to young er and old er adults without insomnia (Ozminkowski, Wang & Walsh, 2007) Chronic insomnia has a substantial economic impact and it is important to address the impact of insomnia on individual, provider, and systemic costs, as well as the potential to reduce costs through the treatment of insomnia. Insomnia and Greater Health Care Utilization Severa l studies have aimed to identify the relationship between people with poor sleep quality and greater health care utilization (HCU). Research indicates that people with insomnia have more emergency room visits, mor e calls to their physician, use more over t he counter medications and ha ve an overall lower health related quality of life compared to people without insomnia (Hatoum et. al, 1998). P eople who suffer from insomnia also engage in higher rates of

PAGE 4

HCU COSTS OF COMORBID INSOMNIACS 4 workplace absenteeism, have more difficulty concentra ting at work, and report more medical problems than those who sleep well (Leger et. al, 2002). Insomnia and Psychiatric Comorbidity Chronic insomnia is often comorbid with other psychiatric disorders and a pproximately 40 50% of people with insomnia have a psychiatric di sorder (Ohayon et. al, 1997). T he most common co occurring psychiatric disorders with insomnia are anxiety and affective disorders (Roth & Roehrs, 2003). Recent research has shown that people suffering from clinical insomnia are 9.82 times and 17.35 times more likely to be clinically depressed and anxious than people without insomnia, respectively (Taylor, et. al, 2005). Additionally, people with insomnia tend to have more severe depressive symptoms and more severe anxiety than people not su ffering from insomnia (Taylor et. al, 2005). The presen ce of insomnia with a comorbid major depressive d isorder (MDD) diagnosis appears to result in higher direct and indirect costs com pared to a MDD diagnosis without insomnia. One study that compared peo ple with MDD and untreated insomnia to those with just MDD found that people with MDD had more total outpatient visits, MDD related visits, were prescribed more antidepressant medications, and had higher direct healthcare costs than people with MDD alone ( As che et al 2010). Although health care utilization costs appear to be higher in patients with comorbid insomnia and psychiatric disorders, the impact of insomnia treatment on these costs remains largely unknown. The present study aims to identify the po tential economic benefits of treating insomnia in patients with comorbid depression and/or anxiety and insomnia. Defining Insomnia Although chronic insomnia is a common disorder in the United States especially among people with depression and/or anxiety there is often disagreement amongst researchers and

PAGE 5

HCU COSTS OF COMORBID INSOMNIACS 5 clinicians regarding an operat ional definition of insomnia. Determining insomnia is based on self report measures and information gathered from clinical interviews. In previous researc h, the presence of insomnia has often been based up on measures of convenience which do not include severity, duration, or fr equency of sleep problems. The above mentioned criteria are critical in matchin g definitions with the Diagnostic and Statistical Manual of Mental Diso rders (4 th ed; text rev.; DSM IV TR ; American Psychiatric Association, 2000 ) International Classification of Sleep Disorders ( 2 nd ed. ; ICSD; American Academy of Sleep Medicine, 2005 ) and quantitative and research diagnostic criteria ( Edinger et al., 2004 ) Furthermore, researchers frequently choose to omit or partially omit the recency component in insomnia studies This lack of a standard operational definition for chronic insomnia can have a substantial impact on healthcare utilization estimates ; when t he operational definition of insomnia is vague, it can considerably complicate the process of quantifying the economic impacts of insomnia It is unlikely that a patient experiencing insomnia for the past two weeks will have a comparable HCU impact as comp ared to a patient with a lifetime history of insomnia. A person with chronic insomnia would be more likely to suffer from other medical and psychiatric conditions in conjunction with their insomnia as opposed to a person with transient insomnia that has on ly recently developed sleep difficulties It would be expected that people suffering from chronic insomnia, and in turn comorbid medical conditions, would have considerably higher HCU costs than people with transient symptoms of insomnia. Marginalizing ins omnia subtypes into an overarching general insomnia group including transient insomnia with chronic insomnia can also alter prevalence rates of comorbid disorders, such as psychiatric and affective disorders (Taylor et. al, 2005). Poor operational defin iti ons for insomnia make it difficult to accurately identify chronic insomnia and therefore calculate the

PAGE 6

HCU COSTS OF COMORBID INSOMNIACS 6 direct and indirect costs, especially for individuals with a comorbid disorder. This study extends previous research by investigating a community sample with chronic insomnia, diagnosed by a licensed clinical psychologist using criteria consistent with the DSM IV TR (APA, 2 000), ICSD (AASM 2005) criteria and established research diagnostic criteria (Edinger et al. 2004 ) It is innovative because it mea sur es HCU pre/post CBTi and HCU change among patients with and without comorbid anxiety and/or depression CBT for Insomnia Cognitive behavioral therapy for insomnia (CBTi) is a non pharm ac ological treatment that has been found to be effective in both pat ients with primary and comorbid insomnia (Stepanski & Rybarczyk, 2006). CBTi aims to implement behavioral practices to improve sleep behavior and challenge maladaptive thoughts about sleep and insomnia. Common treatment procedures that are often incorporat ed in to CBTi include sleep hygiene, stimulus control therapy, sleep restriction therapy, relaxation training, and cognitive therapy that challenges faulty beliefs about insomnia and may include paradoxical intention used to reduce anxiety surrounding sleep loss (Morin et. al, 2006). Furthermore, past research such as McCrae et al has demonstrated the effectiveness of a briefer CBTi protocol (4 sessions as opposed to the 8 10 sessions commonly used in efficacy trials) in a primary care setti ng (McCrae et. al, 2007). CBTi has been shown to be similarly effective in both patients with primary insomnia and patients with insomnia and a comorbid psychiatric disorder (Edinger et. al, 2009) with effect sizes ranging from .42 .94 for various sleep d isorder variables (Morin et. al, 1999) In addition CBTi has been shown to significantly reduce both depressive and insomnia related symptoms in patients with an insomnia diagnosis comorbid with MDD (Manber et. al, 2008). In insomnia patients with

PAGE 7

HCU COSTS OF COMORBID INSOMNIACS 7 comorbi d anxiety, CBTi is moderately effective for reducing anxiety levels (Belleville et. al, 2011). Specific Aims Although there has been much research on the economic impacts of insomnia and CBTi as an effective treatment for insomnia, little is known about patients. Even less is known about the relationship between CBTi and HCU in insomnia patients with a comorbid psychia tric disorder This study aimed to calculate HCU costs in chronic insomnia patients with and without a com orbid anxiety and/or depression diagnosis six months pre and post CBTi treatment. Hypothes e s 1) HCU pre CBT i will be greater in c hronic insomnia patients with anxiety and/or depression than in those without anxiety and/or depression 2) There will be p ost CBTi reductions in HCU costs among treatment responders but no such reductions for treatment non responders 3) Post CBTi there will be within group reductions in HCU for both treatment responder groups, with greater reductions in HCU for patients w ith out comorbid anxiety and/or depression Methods Participants Participants were patients treated at the Insomnia and Behavioral Sleep Medicine (IBSM) outpatient clinic at the U niversity of F lorida & Shands Sleep Disorders Center in Gaines ville, FL A ll participants were seen between 2005 and 2010 and were either referred by a physician or self referred. Eighty four patient charts were reviewed for eligibility for the current study. Forty six individuals were excluded due to not having received a suffi cient dose of Cognitive

PAGE 8

HCU COSTS OF COMORBID INSOMNIACS 8 behavioral t herapy for insomnia (CBTi), defined as attending at least three sessions of treatment. Of the remaining 38 individuals, all patients completed an initial therapy assessment in the IBSM clinic. Eight individuals did not have sufficient pre and post treatment medication data to perform all HCU analyses. All participants used in analyses ( N = 38 ) had a diagnosis of chronic insomnia consistent with DSM IV TR ( APA 2000 ; s ee below) ICSD (A ASM 2005), and established research d iagnostic criteria (Edinger et al 2009 ), and the majority ( N = 29) had a comorbid psychiatric and/or medical disorder. For the purpose of this study, the only comorbid psychi atric disorders included were depressive and anxiety disorders ( N = 13 ) Procedur e Following University of Florida IRB approval, a medical record review of patients seen in the IBSM Clinic from 2005 2010 was conducted. Patients were included in the study if they met criteria for insomnia and completed at least three session s of CBTi. Patients referred to the IBSM Clinic were initially seen for a n intake session that included a clinical interview and a battery of questionnaires (see Materials). If the patient was recommended for CBTi and chose to pursue treatment, they completed sleep diaries for 14 day s prior to their first session of CBTi Patients also completed daily sleep diaries throughout treatment, beginning the first night of treatment until their final appointment. Therapy consisted of multicomponent CBTi ( sleep hygiene, stimu lus control, sleep restriction, relaxation, and cognitive therapy). CBTi followed a treatment manual created by the director of the IBSM clinic ( Christina S. McCrae, PhD CBSM ) and was conducted by clinical psychology graduate students and pre doctoral in terns enrolled in APA approved predoctoral training program in clinical and health psychology and pre doctoral internship program, respectively. These therapists were supervised by a licensed clinical psychologist certified in b ehavioral sleep medicine ( CSM).

PAGE 9

HCU COSTS OF COMORBID INSOMNIACS 9 Materials D aily sleep d iaries Patients completed a daily sleep diary for 14 consecutive days prior to starting treatment and throughout the length of their treatment. Participants completed a sleep diary (SD)[Lichstein, Ri edel, & Means, 1999] each morning for 14 days prior to starting treatment and throughout the length of their treatment, providing subjective estimates of the following five sleep wake parameters: (1) sleep onset latency (SOL) time from initial lights out until sleep onset; (2) wake time after sleep onset (WASO) time spent awake after initial sleep onset until the last awakening; (3) time in bed (TIB) total time spent in bed was determined by taking the difference between bedtime and arise time (when patien t got out of bed for the last time); ( 4 ) total sleep time (TST) computed by subtracting total wake time (SOL + WASO + time between last awakening and arise time ) from TIB ; ( 5 ) sleep efficiency percentage (SE ) ratio of TST to TIB 100 Daily sleep diari es are a cost effective method for monitoring sleep patterns (Chesson et al., trials. Furthermore, findings that positively correlate sleep diaries to polysomnogr aphy also appear to extend to people with insomnia comorbid with depression with correlations of r= 0.33 for TST to r= 0.45 for SOL (McCall & McCall, 2011). Beck Depression Inventory II (BDI II ) The BDI II is a 21 it em questionnaire with responses scored on a scale from 0 to 3 with higher responses indicating more severe depressive symptoms The form is scored by collectively summi ng the 21 questions ; therefore making the possible score range 0 63. Scores <10 indicate minimal depression while scores >18 indicate clinically significant dep ression (Beck, Steer, & Garbin, 1988). The Beck Depression Inventory II is shown to have a high

PAGE 10

HCU COSTS OF COMORBID INSOMNIACS 10 internal consistency and a high test retest reliability of approximately 0.93 after one week (Beck et. al, 1996). State Trait Anxiety Inventory Trait scale (STAI ) The State Trait Anxiety Inventory Trait scale is comprised of 20 questions that measure tension, apprehension, and physiological signs of stress (Spielberg er, Gorsuch, Lushene, & Jacobs, 1983). Questions are scored 1 4 (higher scores indicating more s evere symptoms) and are summed for a total range of 20 0.96) and good test 0.94; Barnes, Harp, & Jung, 2002). For this study, scores Operational Definitions Insomnia Criteria used for diagnosing chronic insomnia are consistent with the DSM IV TR ( APA 2000) ICSD (AASM, 2005 ), and established research diagnostic criteria (Edinger et al., 2009 ) The predominant symptom associated with chronic i nsomnia is difficulty initiating or maintaining sleep, or experiencing non restorative sleep that causes clinically significant imp airment or dysfunction Patients had to report sleep onset or awake time >30 minutes and insomnia at least 3 nights per week for more than 6 months. All patients diagnosed with chronic insomnia were offered CBTi, regardless of the presence of a comorbid an xiety or depressive disorder Depression Patients were diagnosed with depression if 1) they had a previous diagnosis by a physician, 2) met DSM IV TR criteria ( APA 2000) for depression at intake, and/ or 3) had clinically elevated score s on the BDI II a t intake (> 18) Per DSM IV TR criteria for Major

PAGE 11

HCU COSTS OF COMORBID INSOMNIACS 11 Depressive Disorder (MDD) ( APA 2000) patients must exhibit a depressed mood and/or loss of interest/pleasure for at least two weeks in addition to other symptoms such as significant changes in weight, sle ep disturbances, psychomotor retardation or agitation, fatigue, feelings of worthlessness, difficulty concentrating, and s uicidal thoughts. P atients are ruled out for a MDD diagnosis if they report symptoms consistent with bipolar disorder or report sympt oms that are associated with substance use or bereavement ( APA 2000). Anxiety As with depression, patients were diagnosed with anxiety if 1) they had a previous diagnosis by a physician, 2) met DSM IV TR criteria for anxiety ( APA 2000) at the intake, a nd/or 3) The predominant criteria associated with anxiety disorders is the presence of excessive anxiety or worry for at least six months that is both difficult to control and is associated with other symptoms such as restlessness, fatigue, irritability, muscle tension, and sleep disturbances. Symptoms of anxiety must not be related to another psychiatric disorder or substance use, and must cause signific ant impairment or distress (APA 2000). Treatmen t responders vs. non r esponders Treatment responders were those that experienced 1) a 50% reduction in symptoms (i.e., SOL and WASO ) (Lichstein et. al, 2003) 2) had sleep efficiency > 85% during the last two weeks of treatment (Dolan et. al, 2010 ; McCrae et al., 2007 ) or 3) had SOL/WASO less than 3 1 ( McCrae et. al, 2007 ) Non responders did not meet the above criteria.

PAGE 12

HCU COSTS OF COMORBID INSOMNIACS 12 Analyses Calculating health care u tilization Healt hc are u tilization (HCU) was calculated using direct h ealth care costs and Chronic Disease Score (CDS) an estimate of HCU based on patient medications. Direct costs were electronic medical records. Estimated HCU was meas ured during the 6 months prior to the intake appointment and 6 months post treatment. CDS was derived using Clark et al (1995) formula and medication history to predict three HCU variables estimated tota l costs, estimated outpatient costs, and number of primary care visits. Bramoweth and Taylor (2012) previously used (Clark et. al, 1995) in an insomnia sample and we also used it in the present study to total HCU, prima ry care HCU, and outpatient HCU. Each medication class, representing various chronic diseases, is weighted. The weights correspond to a cost (total and outpatient) and number of primary care visits. Medications are classified by American Hospital Formulary System (AHFS) category numbers. In addition to medications, gender and age (grouped by 10 year gaps) are weighted and are associated with the three CDS outcomes The CDS equation is: CDS = intercept + gender + age group + medication 1 + medication 2 + medication 3 + + medication n. CDS represents costs in US dollars over a six month period. S tatistical a nalyses. Preliminary exploration of the data revealed that only one patient was identified as being a treatment non responder with a comorbid p sychiatric diagnosis As is recommended for research with small patient samples ( Blair & Higgins, 1985 ), non parametric methods were used to analyze treatment outcomes. In this study w e utilized the related samples Wilcoxon signed

PAGE 13

HCU COSTS OF COMORBID INSOMNIACS 13 rank test to analyze the total difference in HCU pre and post CBTi. To analyze between group differences, we used the non parametric independent samples Mann Whitney U Test (Auble, 195 3). While there was a signi ficant amount of missing data, 38 patients had sufficient data to me asure pre and post treatment HCU and were included in the final analyses. Results Demographics Among the 38 patients included in the study, 23 (60.5%) were female and 15 (39.5%) were male. Patient ages ranged from 18 to 79 ( M =52.1, SD = 18.86). Regardi ng race 35 (92.1%) identified as White, one (2.6%) as Black/African American, one (2.6%) as Native American/Native Alaskan, and one (2.6%) as Biracial. Additionally, five patients (13.2%) indicated they were of Hispanic background. The average number of p atient medical conditions was 2.05 ( SD =2.05). Within the final patient sample, t hirteen patients had a comorbid psychiatric disorder of those 10 (26%) had a depressive disorder, four (10.5%) had an anxiety disorder, and one met criteria for both. Treatme nt Response Baseline sleep diaries indicated that there was a significant difference in SE ( p < .05) and WASO ( p < .05) among treatment responders and non responders pre CBTi. Pre treatment treatment responders reported lower SOL ( M = 43 .16, SD = 55.35), lower WASO ( M = 34.17, SD = 48.71) and higher SE ( M = 80.5, SD = 15.55) than non responders who on average reported higher SOL ( M = 50.52, SD = 35.89), higher WASO ( M = 70.75, SD = 39.11), and lower SE ( M = 64.07, SD = 14.12) values.

PAGE 14

HCU COSTS OF COMORBID INSOMNIACS 14 At the end of treatme nt, 30 ( 79 %) patients were identified as treatment responders while 8 ( 21 %) patients did not meet criteria for treatment response. Table 1 indicates treatment response for the 38 patients used in the final HCU cost analyses and their accompanying mental he alth status Based on the previously described treatment response criteria, there were significantly more individuals who were classified as treatment responders than as non responders as indicated by a chi squared test ( X 2 = 12.74, p < .001). Estimated H CU C osts P re T reatment Prior to treatment, there were no significant differences in HCU or the number of medical conditions among patients with and without anxiety and/or depression. However, there was a significant difference in age between the two groups Patients without anxiety and/or depression were significantly older ( M = 56.4 years old, SD =18.1 9 ) than p atients with anxiety and/or depression ( M = 43.7 years old, SD =16.5) The effects of age were controlled for in all analyses using the saved unstan dardized residuals from regressions with age predict ing HCU; however they did not alter the pattern of results. Estimated HCU C osts P ost T reatment Treatment responders exhibited trends towards decreased estimated HCU at post treatment total costs ( $89.33 SD = $246.86; p = .08 ), outpatient costs ( $51.33, SD = $132.36; p = .07 ), and number of primary care visits ( .23, SD = .61; p = .07 ). Additional tests revealed that there were n o significant HCU reductions for treatment non responders at post treatment Within group Wilcoxon analyses were separately run among treatment responders with no comorbid psychiatric disorder and among those with a comorbid psychiatric disorder to estimate change in HCU from pre treatment to post treatment When these groups we re analyzed separately there was no significant within group change s. Furthermore, t here were no significant

PAGE 15

HCU COSTS OF COMORBID INSOMNIACS 15 between group differences on any of the HCU variables when comparing changes in treatment responders with and without a comorbid psychiatric disor der at post treatment. Table 2 depicts the average estimated HCU for total costs, outpatient costs, and number of primary care visits for the six months pre and post CBTi by group. Discussion The aim of this study was to examine the impact of treatment re sponse and presence of comorbid anxiety and/or depression on HCU in patie nts with chronic insomnia who received CBTi in an outpatient setting. Although this study did not demonstrate that insomnia patients without comorbid anxiety and/or depression had lo wer HCU costs following CBTi as hypothesized, the results are still of particular interest because they suggest CBTi may contribute to reduced health care utilization and costs. Specifically, there was a trend for decreased estimated total healthcare cost s, outpatient healthcare costs, and number of primary care visits for patients successfully treated with CBTi (regardless of the presence or absence of comorbid psychiatric diagnosis) While our hypothesis that patients with chronic insomnia and comorbid depression and/or anxiety would have greater HCU prior to starting treatment was not supported, the significant difference in age among patients with chronic insomnia with and without a comorbid psychiatric disorder may help explain the results. It is like ly that among middle aged and older adults, severity of comorbid medical disorders contributed to higher HCU than among younger adults. The patients in this study with chronic insomnia and comorbid depression and/or anxiety were significantly younger, on a verage, than patients without a comorbid psychiatric disorder. While the two groups reported similar number of medical disorders, the type and/or severity of the comorbid medical disorder may have significantly impacted HCU in the non psychiatric group.

PAGE 16

HCU COSTS OF COMORBID INSOMNIACS 16 Ps ychiatric disorders may have significantly impacted the younger adults more than comorbid medical disorders. Among older individuals presenting for treatment of insomnia, medical conditions likely played a larger etiological role in their presentation for the treatment of insomnia, as has been previously documented in the literature concerning comorbid insomnia in older adults (Lichstein et. al, 2000). It may be that these individuals presented with an increased severity of medical disorders and subsequentl y more medications ; thus impacting their overall HCU. As a result, the impact of the number of medical disorders and the severity of disorders should be accounted for in future research. Despite the lack of significant difference in HCU costs among chroni c insomnia patients with and without comorbid anxiety and/or depression pre CBTi, treatment responders and non responders differed significantly in SE and WASO pre treatment. On average, treatment responders had higher SE and lower WASO than non responders in addition to lower SOL (although not statistically significant). Although this difference was not controlled for in analyses, it is nevertheless an important finding that adds to the growing interest in predicting treatment response. If we are able to identify variables on which treatment responders and non responders consistently differ, it may aid in predicting treatment response among patients in future studies. Additionally, identifying treatment responder predictor variables may help explain why pa tients often choose not to complete therapy, and ultimately may help increase retention rates. More importantly, however, significant pre treatment differences among treatment responders and non responders could have a substantial impact on HCU. For exampl e, we would expect people with greater sleep disturbances and less efficient sleep to generally use more healthcare services. If these people choose not to engage in treatment or end treatment early, however, not only can we not accurately assess HCU assoc iated with chronic insomnia, but we would also expect

PAGE 17

HCU COSTS OF COMORBID INSOMNIACS 17 greater HCU long term over those who responded positively to CBTi treatment. In essence, not receiving adequate treatment for chronic insomnia could translate to a greater usage of healthcare services. More research is needed to determine the healthcare cost impact of more severe chronic insomnia symptoms and how to more effectively retain these individuals throughout treatment. Results of this study indicate that a majority of patients who came to the c linic responded positively to treatment. These findings confirm p revious studies indicating the effectiveness of a briefer CBTi protocol used in primary care settings (McCrae et. al, 2007). Consistent with the second hypothesis that there would be post CBT i HCU cost reductions for treatment responders but not for non responders both chronic insomnia groups exhibit reductions (trends) in HCU if they responded positively to treatment. Additionally, a trend towards reduced estimated total costs, outpatient co sts, and number of primary care visits among treatment responders illustrates that CBTi may have a positive impact on HCU. These findings underline the clinical importance and potential economic benefit s of treating insomnia not only in chronic insomnia pa tients, but also in patients with comorbid psychiatric and/ or medical conditions It also highlights that insomnia patients with comorbid anxiety and/or depression respond well to CBTi. For many years, comorbid insomnia was viewed as a condition primarily driven by the co occurring condition; as a result, treatments were generally focused on the comorbid condition while neglecting the insomnia (McCrae & Lichstein, 2001). A large and growing body of research now indicates that comorbid insomnia can be succes sfully treated when targeted directly and without targeting the other condition. The current study extends that research by demonstrating that in addition to sleep improvements, cost savings are possible through the successful treatment of insomnia, even w hen the insomnia is comorbid with other conditions. Furthermore, it has also been noted that improvements in psychological symptoms are possible following the successful

PAGE 18

HCU COSTS OF COMORBID INSOMNIACS 18 treatment of comorbid insomnia (Manber et al, 2008; Taylor et al., 2007; Morawetz, 200 3). Future research should evaluate the relationship between changes in psychological symptoms and healthcare utilization, as these two domains may be interrelated among individuals with comorbid insomnia. To better estimate healthcare savings associated w ith treating insomnia, future research should involve larger sample sizes that are representative of a primary care population. Although significant within group reductions in HCU for both treatment responder groups, with greater reductions in HCU for pa tients without comorbid anxiety and/or depression were expected post CBTi no significant differences were found This lack of statistical significance may be attributed to small group sizes, and in effect a reduction of power, that occurred when treatment response was analyzed by psychiatric diagnosis. Furthermore, fairly large sta ndard deviations indicate that HCU costs varied greatly among the patient sample There were several limitations associated with this research study. It is important to note th at this study did not use data from a randomized control trial examining the efficacy of CBTi. Instead, this study used a clinical patient sample and data were extracted from a record review Because data came from patient medical records, inconsis tencies and missing data were an issue. For example, only 38 out of the original 84 patients treated at the IBSM clinic had sufficient pre and post CBTi medication data to successfully analyze changes in healthcare use. Furthermore, this sample was comprised of p rimarily middle aged and older adults, many with several comorbid medical conditions such as hypertension, cardiovascular disease, diabetes, etc The presence and frequency of comorbid medical conditions may have been an important confounding variable in a ccurately estimating HCU costs ; however in the real world these factors are unavoidable Thus, this may represent a methodological limitation on one hand, but it

PAGE 19

HCU COSTS OF COMORBID INSOMNIACS 19 may increase ecological validity on the other as the patient sample is representative of the types of insomnia patients seen in primary care settings (patients with multiple medical and/or psychiatric comorbidities) I t is possible that younger populations with better health may see greater HCU reductions post treatment. Conversely, younger perso ns in better health may already have lower HCU costs in the first place, thus justifying the importance of using CBTi and examining healthcare costs in samples such as the one used in this study, despite the methodological challenges they pose. Another lim itation was that HCU was estimated using patient medications extracted from the medical records during the six month pre and post treatment periods. Although measuring HCU based on medications is validated method for estimating HCU (Clark et. al, 1995), i t is unlikely that all medical conditions and associated costs of a patient can be accurately accounted for by looking at their medications. The HCU reported in this study is likely an underestimate of true HCU. Finally, the time interval used to estimate reduction in HCU post treatment may not have been a sufficient amount of time to see the full economic benefits of CBTi. It is possible that significant reductions in HCU post CBTi are not seen until beyond the six month time period utilized in this study. For example, those with medical and/or psychiatric comorbidities might be motivated by their success with CBTi to pursue treatment of their other medical conditions. Therefore, healthcare savings may not be observed until after the measured six months pos t CBTi. Future research should aim to estimate HCU reductions after longer time intervals to see if there are greater reductions in HCU post CBTi. In summary, w hile there were no significant pre treatment differences in HCU between chronic insomnia patient s with and without comorbid anxiety and/or depression, there w as a significant difference in age between the two groups in addition to significant pre treatment

PAGE 20

HCU COSTS OF COMORBID INSOMNIACS 20 differences in SE and WASO among treatment responders and non responders. A majority of patient s responded positively to treatment, and results demonstrated that HCU cost reductions were trending towards significance among treatment responders post CBTi. However, no significant differences were seen post CBTi among treatment non responders, nor betw een chronic insomnia patients with and without comorbid anxiety and/or depression who responded to treatment. Findings of this study indicate that CBTi may be effective in reducing healthcare costs in patients with chronic insomnia and comorbid anxiety and /or depression. Future research should aim to explore the potential economic benefits of using CBTi to treat chronic insomnia patients with other comorbid medical and psychiatric disorders, using larger and more varied patient samples. Finally, future rese arch should also aim to identify variables that could potentially predict treatment response and ultimately increase retention rates among chronic insomnia patients engaging in CBTi.

PAGE 21

HCU COSTS OF COMORBID INSOMNIACS 21 References Archer, N. Brown, R.G., Boothby, H., Foy, C., Nich olas, H., and Lovestone, S. (2006). The NEO FFI is a reliable measure of premorbid personality in patients wit h probable International Journal of Geriatric Psychology 2 1, 477 484. Asche, C.V., Joish, V.N., Camacho, F. & Drake, C.L. (2010). The direct costs of untreated comorbid insomnia in a managed care population with major depressive disorder. Current Medical Research & Opinion 26 (8), 1843 1853. Auble, D. (1953). Extended tables for the Mann Whitney statistic. Institute of Educ ational Research. Barnes, L. L. B., Harp, D., & Jung, W. S. (2002). Reliability Generalization of Scores on the Spielbe rger State Trait Anxiety Inventory. Educationa l and Psychological Measurement, 62 (4), 603 618. Beck, A. T., Steer, R. A., Garbin, M. G. (1988). Psychometric Properties of the Beck Depression Inventory: Twenty Five Years Later. Clinical Psychology Review, 8 77 100 Belleville, G., Cousineau, H., Levrier, K. & St Pierre Delorme, M. (2011). Meta analytic review of the impact of cognitive behavior therapy for insomnia on concomitant anxiety. Clinical Psychology Review 31, 638 652. Blair, R.C. & Higgins, J.J. (1985). Comparisons of the Power of the Paired Samples t Test to Rank Test Under Various Population Shap es. Psychological Bulletin, 97 (1), 119 128. Bramoweth, A. D. & Taylor, D.J. (2012). Chronic Insomnia and Health Care Utilization in Young Adults. Behavioral Sleep Medicine, 10, 1 16. Buysse, D Reynolds III, C Monk, T.H., Berman, S.R & Kupfer, D.J. (1988). The Pittsburgh

PAGE 22

HCU COSTS OF COMORBID INSOMNIACS 22 Sleep Quality Index: A New Instrument for Psy chiatric Practice and Research. Psychiatry Research 28 193 213. Chesson, A., Hartse, K., McDowell, A., Davila, D., Johnson, S., Littner, M., Wise., M. & Rafecas, J. (2000). Practice P arameters for the Evaluation of Chronic Insomnia. An American Academy of Sleep Medicine Report. Standards of Practice Committee of the American Academy of Sleep Medicine. SLEEP, 23 (2), 237 241. Chilcott L A Shapiro C M. (1996). The socioeconomic impact of insomnia. An overview. Pharmacoeconomics 10 (1), 1 14 Clark, D.O., Von Korff, M., Saunders, K., Baluch, W.M. & Simon, G.E. (1995). A Chronic Disease Score with Empirically Derived Weights Medical Care, 33 (8), 783 795. Costa, P. T., & McCrae R. R. (1 992). Revised NEO personality inventory (NEO PI R) and NEP five factor inventory (NEO FFI): Professional Manual Psyc hological Assessment Resources. Daley, M., Morin, C. M., LeBlanc, M., Gregoire, J. P., & Savard, J. (2009). The economic burden of insomn ia: Direct and indirect costs for individuals with insomnia syndrome, insomnia symptoms, and good sleepers. S LEEP 32 (1), 55 64. Diagnostic and statistical manual of mental disorders: DSM IV TR. (4th ed.). (2000). Washington, DC: American Psychiatric Asso ciation. Dolan, D.C., Taylor, D.J., Bramoweth, A.D. & Rosenthal, L.D. (2010). Cognitive behavioral therapy for insomnia: a clinical case series study of patients with co morbid disorders and using hypnotic medications. Behavior and Research Therapy, 48 3 21 327. Edinger, J.D., Bonnet, M.H., Bootzin, R.R., Doghramji, K., Dorsey, C.M., Espie, C.A., Jamieson,

PAGE 23

HCU COSTS OF COMORBID INSOMNIACS 23 A.O., McCall, W.V., Morin, C.M. & Stepanksi, E.J. (2004). Derivation of Research Diagnostic Criteria for Insomnia: Report of an American Academy of Sl eep Medicine Work Group. SL EEP 27 (8), 1567 1596. Edinger, J.D., Olsen, M.K., Stechuchak, K. M., Means, M.L., Lineberger, M.D., Kirby, A. & Camey, C.E. (2009). Cognitive Behavioral Therapy for Patients with Primary Insomnia or Insomnia Associated Predomina ntly with Mixed Psychiatric Disorde rs a Randomized Clinical Trial. SLEEP 32 (4), 499 510. Espie, C Inglis, S.J., Harvey, L & Tessier, S. (2000). properties of the Dysfunctional Beliefs and Attitudes about Sleep Sc ale and the S leep Disturbance Questionnaire. Journal of Psychosomatic Research 48 141 148. Hatoum, H.T., Dong, S.X., Darnia, C.M., Wong, J.M. & Mendelson, J.R. (1998). Insomnia, health related quality of life, and healthcare resource consumption; a stu dy of managed care enrollees. Pharmacoeconomics 14 629 647. Hauri P., Chernik, D., Hawkins, D. & Mendels, J. (1974). Sleep of depressed patients in remission. Arch Gen Psychiatry, 31 (3), 386 391. The International classification of sleep disorders: diag nostic and coding manual. (2nd ed.). (2005). Westchester, Illinois: American Academy of Sleep Medicine. Lacks, P. & Morin, C.M. (1992). Recent advances in the assessment and treatment of insomnia. Journal of Consulting and Clinical Psychology 60 (4), 586 594. Leger, D., Guilleminaut, C., Bader, G., Levey, E. & Palliard, M. (2002). Medical and socio professional impact of insomnia. S LEEP 25, 625 629. Lichstein, K. L., Riedel, B. W., & Means, M. K. (1999). Psychological treatment of late life

PAGE 24

HCU COSTS OF COMORBID INSOMNIACS 24 insomnia. I n R. Schulz, G. Maddox, & M. P. Lawton (Eds.), Annual review of gerontology and geriatrics: vol. 18. Focus on interventions research with older adults (pp. 74 110). New York: Springer. Lichstein, K. L., Durrence, H. H., Taylor, D. J., Bush, A. J., & Riedel B. W. (2003). Quantitative criteria for insomnia. Behaviour Research and Therapy 41, 427 445. Lichstein, K.L., Wilson, N.M. & Johnson C.T. (2000). Psychological treatment of secondary insomnia. Psychology and Aging, 15 232 240. Manber, R., Edinger J.D., Gress, J.L., San Pedro Salcedo, M.G., Kuo, T.F. & Kalista, T. (2008). Cognitive Behavioral Therapy for Insomnia Enhances Depression Outcome in Patients with Comorbid Major Depressive Disorder and Insomnia. SLEEP 31 (4), 489 495. McCall, C. & M cCall W. V (2011). Comparison of actigraphy with polysomnography and slee p logs in depressed insomniacs. Journal of Sleep Research 10 1111. McCrae, C.S. McGovern, R., Lukefahr, R. & Stripling, A.M. (2007). Research evaluating brief behavioral sleep treatments for rural elderly (RESTORE): A preliminary examination of effectiveness. The American Journal of Geriatric Psychiatry 15 (11), 979 982. McCrae C.S. & Lichstein, K.L. (2001). Secondary insomnia: diagnostic challenges and intervention opportuni ties. Sleep Med Rev ., 5 (1), 47 6. Means, M.K., Lichstein, K.L., Epperson, M.T., & Johnson, C.T. (2000). Relaxation therapy for inso mnia: nighttime and day time. Behaviour Research and Therapy 38 665 678. Morawetz, D. (2003). Insomnia and depression: Whi ch comes first? Sleep Research Online (Vol. 5, pp. 77 81): WebSciences. Morin, C. M., Kowatch, R. A., Barry, T. Walton, E (1993). Cognitive behavior therapy for late

PAGE 25

HCU COSTS OF COMORBID INSOMNIACS 25 life insomnia. Journal of Consulting and Clinical Psychology 61 (1), 137 146 Morin, C ., Bootzin, R.R., Buysse, D.J., Edinger, J.D., Espie, C.A. & Lichstein, K.L. (2006). Psychological and behavioral treatment of insomnia: update of the recent evidence (1998 2004). SLEEP, 29 (11), 1398 1414 Morin, C. M., Hauri, P. J., Espie, C. A., Spielman, A. J., Buysse, D. J., & Bootzin, R. R. (1999). Nonpharmacologic treatment of chronic insomnia. An American Academy of Sleep Medicine review. S LEEP 22 (8), 1134 1156. Morin, C. M., & Espie, C. A. (2003). Insomnia: A clinical guide to assessment and treatme nt. New York: Kluwer Academic/Plenum Publishers. Ohayon M M. (1997). Prevalence of DS M IV diagnostic criteria of in somnia: distinguishing insomnia related to mental disorders from sleep disorders. J Psychiatr y Res ., 31 333 3 46. Ozminkowski R.J., Wang, S & Walsh, J.K. (2007). Direct and indirect costs o f untreated insomnia in adults. SLEEP, 30 (3), 263 273. Roth, T., & Roehrs, T. (2003). Insomnia: Epidemiology, Characteristics, and Consequences. Clinical Cornerstone 5 (3), 5 15. Roth., T., Coulouvrat., C., Hajak, G., Lakoma, M.D., Sampson, N.A., Shahly, V., Shillington, A.C., Stephenson, J.J., Walsh, J.K. & Kessler, R.C. (2011). Prevalence and Perceived Health Associated with Insomnia Based on DSM IV TR; International Statistical Classification of Diseas es and Related Health Problems, Tenth Revision; and Research Diagnostic Criteria/International Classification of Sleep Disorders, Second Edition Criteria: Results from the America Insomnia Survey. Biological Psychiatry, 69 592 600. Sp ielbe rger, C. D., G orsuch, R. L., Lushene, R., & Jacobs, G. A. (1983). Manual for the Stait

PAGE 26

HCU COSTS OF COMORBID INSOMNIACS 26 Trait Anxiety Inventory: STAI (form Y). Palo Alto, CA. Consulting Psychologists Press, Inc. Stepanski, E.J. & Rybarczyk B. (2006). Emerging research on the treatment and etiology of secondary or comorbid insomnia. Sleep Medicine Reviews, 10 7 18. Stoller M K. (1994). Economic effects of insomnia. Clinical Therapy, 16, 873 8 97 Taylor D Lichstein K Durrence H Rie del B & Bush A J (2005). Epidemiology of Insomnia, Depression and Anxiety. S LEEP 28 (11), 1457 1464. Taylor, D.J., Lichstein, K.L., Weinstock, J., Sanford, S., & Temple, J.R. (2007). A pilot study of cognitive behavioral therapy of insomnia in people with mild depression. Behav Ther, 38 (1), 49 57. Walsh J K & Eng elhardt C L. (1999). The direct economic costs of insomnia in the United States for 1995. SLEEP, 22 (2), 386 393. Weissman M M Greenwald S Nino Murcia G & Dement W C. (1997). The morbidity of insomnia uncomplicated by psychiatric disorders. Gen Hosp Psychiatry 19, 245 2 50 World Health Organization. (2008). ICD 10: International statistical classification of diseases and related health problems (10th Rev. ed.). New York, NY.

PAGE 27

Running Head: HCU OF COMORBID INSOMNIACS 27 Table 1 CBTi Treatment Response Among Chronic Insomnia Outpatients wi th and without Comorbid Anxiety and/or Depression Treatment Responders Treatment Non Responders Depression and/or Anxiety Diagnosis 12 1 No Depression and/or Anxiety Diagnosis 18 7 Note: Treatment responders experienced 1) a 50% reduction in sympt oms (i.e., SOL and WASO) (Lichstein et. al, 2003), 2) had sleep efficiency > 85% during the last two weeks of treatment (Dolan et. al, 2010), or 3) had SOL/WASO less than 31 per week (McCrae et. al, 2007). Non responders did not meet any of the above criteria. Insomnia, anxiety, and depression diagnoses were based on the Diagnostic and Statistical Manual of Mental Disorders (4 th ed.; text rev.; DSM IV TR ) criteri a (APA, 2000). Cognitive b ehavioral therapy for insomnia (CBTi) included sleep hygiene, stimulus control, sleep restriction, relaxation, and cognitive therapy

PAGE 28

Running Head: HCU OF COMORBID INSOMNIACS 28 Table 2 Average Estimated Outpatient HCU Costs Pre and Post CBTi Treatment Note: Total HCU and Outpatient HCU are measured in US dollars while Primary Care HCU is measured in the number of patient visits. Note: Treatment responders experienced 1) a 50% reduction in symptoms (i.e., SOL and WASO) (Li chstein et. al, 2003), 2) had sleep efficiency > 85% during the last two weeks of treatment (Dolan et. al, 2010), or 3) had SOL/WASO less than 31 minutes, nights per week (McCrae et. al, 2007). Non responders did not meet any of the above criteria. In somnia, anxiety, and dep ression diagnoses were based on the Diagnostic and Statistical Manual of Mental Disorders (4 th ed.; text rev.; DSM IV TR) criteria (APA, Non R esponders Without Depression/Anxiety Non Responders With Depression/Anxiety Responders Without Depression/Anxiety Responders With Depression/Anxiety Total HCU Pre $ 618.28 ($512.83) $0 $ 769.27 ($430.88) $ 290.74 ($201.67) Post $ 662.34 ($560.60) $0 $ 614.62 ($539.84) $290.24 ($201.56) Outpatient HCU Pre $ 314.88 ($164.37) $0 $ 382.25 ($123.57) $ 218.51 ($146.17) Post $ 311.82 ($183.59) $0 $ 299.80 ($206.25) $ 218.26 ($146.18) Primary Care HCU Pre 1.60 (1.27) 0 1.84 (0.98) 1.31 (1.09) Post 1.68 (1 .40) 0 1.48 (1.34) 1.06 (.72)

PAGE 29

HCU COSTS OF COMORBID INSOMNIACS 29 2000). Cognitive b ehavioral therapy for insomnia (CBTi ) included sleep hygiene, stimulus contro l, sleep restriction, relaxation, and cognitive therapy.