1 HEALTH RELATED QUALITY OF LIFE, BEHAVIORAL INTENTION AND AMBIVALENCE IN OBESE ADULT PATIENTS DURING STAGES OF CHANGE IN A HEALTHY LIFESTYLE/WEIGHT LOSS PROGRAM B y JEFFERY DON GILLIAM A DISSERTATION PRESENTED TO THE GRADUA TE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY THE UNIVERSITY OF FLORIDA 2014
2 2014 Jeffery Don Gilliam
3 To the many patients I worked with through the years
4 ACKNOWLEDGMENTS I realize you never accomplish something without the help of others. I am grateful to my supervisory committee members who offered direction and encourageme nt along the way. I also thank friends and family who cheered from the sidelines. Finally, I thank my wife who has always supported my academic efforts.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGU RES ................................ ................................ ................................ .......... 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 Problem ................................ ................................ ................................ .................. 13 Purpose ................................ ................................ ................................ .................. 13 Rationale ................................ ................................ ................................ ................. 13 Research Questions ................................ ................................ ............................... 15 Hypothesis ................................ ................................ ................................ .............. 15 Delimitations ................................ ................................ ................................ ........... 15 Limitations ................................ ................................ ................................ ............... 16 Assumptions ................................ ................................ ................................ ........... 16 Summary ................................ ................................ ................................ ................ 17 2 LITERATURE REVIEW ................................ ................................ .......................... 19 Factors That Affect Health Behavior ................................ ................................ ....... 21 Biological and Environmental Factors That Affect Behavioral Change ............. 21 Metabolism ................................ ................................ ................................ ....... 22 Environmental Factors That Affect Behavior ................................ .................... 23 Self Control over Addictive Behaviors ................................ .............................. 24 Ambivalence in Health Behavior Change ................................ ......................... 25 Behavioral Intentions ................................ ................................ ........................ 27 Stages of Change and Ambivalence ................................ ................................ 27 Health Related Quality of Life, Ambivalence, and Behavioral Intention ............ 28 3 METHODS ................................ ................................ ................................ .............. 33 Research Design ................................ ................................ ................................ .... 33 Research Variables ................................ ................................ .......................... 34 Health Related Quality of Life ................................ ................................ ........... 35 Attitudinal Ambivalence ................................ ................................ .................... 35 Behavioral Intention ................................ ................................ .......................... 38 Stages of Change ................................ ................................ ............................. 38 Compliance for Exercise and Food Choices ................................ ..................... 39 Body Weight ................................ ................................ ................................ ..... 40
6 Participants ................................ ................................ ................................ ............. 40 Setting ................................ ................................ ................................ ..................... 41 Procedures ................................ ................................ ................................ ............. 41 Data Collection ................................ ................................ ................................ ....... 43 Data Analysis ................................ ................................ ................................ .......... 45 Quantitative Analysis for Parametric Data ................................ .............................. 45 Quantitative Analysis for Non Parametri c Data ................................ ....................... 46 Correlational Data ................................ ................................ ................................ ... 46 Significance of the Study ................................ ................................ ........................ 47 4 R ESULTS ................................ ................................ ................................ ............... 51 Participant Demographics ................................ ................................ ....................... 51 Body Weight ................................ ................................ ................................ ..... 52 Health Rela ted Quality of Life ................................ ................................ ........... 53 Stage of Change ................................ ................................ .............................. 53 Attitudinal Ambivalence ................................ ................................ .................... 54 Behavioral Intentions ................................ ................................ ........................ 55 Program Compliance ................................ ................................ ........................ 55 Research Question 2 ................................ ................................ .............................. 60 Research Question 3 ................................ ................................ .............................. 62 Summary ................................ ................................ ................................ ................ 64 5 DISCUSSION AND SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS .. 83 Conclusions ................................ ................................ ................................ ............ 91 Recommendations for Future Research ................................ ................................ 98 APPENDIX A IRB INFORMED CONSENT FORM ................................ ................................ ...... 108 B SF 36 HEALTH SURVEY ................................ ................................ ..................... 116 C GRIFFIN INDEX ................................ ................................ ................................ ... 120 D POSITI VE ATTITUDINAL AMBIVALENCE ASSESSMENT FORM ...................... 121 E NEGATIVE ATTITUDINAL AMBIVALENCE ASSESSMENT FORM .................... 123 F BEHAVIORAL INTENTION FOR M ................................ ................................ ....... 125 G STAGES OF CHANGE ASSESSMENT FORM ................................ .................... 126 H FOOD CHOICE & EXERCISE LOG ................................ ................................ ...... 127 I COMPLIANCE LOG FOR FOOD CHOICES AND EXERCISE ............................. 12 8
7 J FLYER ................................ ................................ ................................ .................. 129 REFERENCE LIST ................................ ................................ ................................ ...... 130 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 142
8 LIST OF TABLES Table page 3 1 Description of the instruments and variable characteristics ................................ 48 4 1 Population demographics ................................ ................................ ................... 67 4 2 Means by variable and time of measurement ................................ ..................... 69 4 3 Interactions with time ................................ ................................ .......................... 70 4 4 Pairwise Comparisons over time ................................ ................................ ........ 71 4 5 Pairwise comparisons for ambivalence r elated to (weight loss),WL,daily exercise (DE), and food diary (FD) ................................ ................................ ..... 72 4 6 Pairwise comparisons for behavioral intentions related to WL, DE, and FD ....... 73 4 7 Mean weight loss by socioeconomic ................................ ................................ .. 74 4 8 Correlation between percent change in weight and percent change in physical component score (PCS) and mental component score (MCS) ............. 75 4 9 Correlation between percent change in weight loss (WL) and behavioral intention (BI) ................................ ................................ ................................ ....... 76 4 10 Correlation between p ercent change in weight loss (WL) and ambivalence (AMB) ................................ ................................ ................................ ................. 77 4 11 Correlations between percent change in weight loss (WL) and compliance (COMP) ................................ ................................ ................................ .............. 78 4 12 Correlations between health related quality of life (HRQL) and compliance (COMP) ................................ ................................ ................................ .............. 79 4 13 Correlations between ambivalence (AMB) and behavior intentions (BI) for food diar y (FD) and daily exercise (DE) ................................ .............................. 80 4 14 Correlation between health related quality of life (HRQL) and behavioral intentions (BI) ................................ ................................ ................................ ..... 81 4 15 Correlations between health related quality of life (HRQL) and ambivalence (AMB) ................................ ................................ ................................ ................. 82 5 1 Percent change in HRQL PCS and MCS ................................ ......................... 100
9 LIST OF FIGURES Figure page 2 1 Level of attitudinal ambivalence (AA) varies across the stages of change. ........ 31 2 2 Level of ambivalence affects behavioral int entions and subsequent behavior .. 32 3 1 How factors/variables can affect outcomes: Health Related Quality of Life (HRQL) and Body weight/BMI (Weight Loss) during a Healthy Lifest yle/Weight Loss program ................................ ................................ ........... 49 3 2 Attitudinal ambivalence ................................ ................................ ....................... 50 5 1 Changes in body weight and percent change in body weight ........................... 101 5 2 Mean changes in HRQL for PCS and MCS over time ................................ ...... 102 5 3 Changes in Stage of change over time ................................ ............................. 103 5 4 Program compliance over time. ................................ ................................ ........ 104 5 5 Significance between WL, and both FD and DE for behavioral intention and attitudinal ambivalence.. ................................ ................................ ................... 105 5 6 The relationship between Program Compliance and Percent change in body weight. ................................ ................................ ................................ .............. 106 5 7 Associations between stage of change, behavioral inte ntion and ambivalence ................................ ................................ ................................ ..... 107
10 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy HEALTH RELATED QUALITY OF LIFE, BEHAVIORAL INTENTION AND AMBIVALENCE IN OBESE ADULT PATIENTS DURING STAGES OF CHANGE IN A HEALTH Y LIFESTYLE/WEIGHT LOSS PROGRAM By Jeff Gilliam May 2014 C hair: R. Morgan Pigg Jr. Major: Health and Human Performance The obesity epidemic in the United States places an enormous financial burden on our healthcare system. Despite the abundant educational information available to the public on nutrition and exercise, the obesity epidemic continues to grow. Many behavioral int ervention strategies are able to achieve short t erm changes in health behaviors H owever because of lapses in health behavior long term benefits are often not realized. Ambivalence and behavior al intentions related to weight loss and lifestyle changes ar ability to successfully change behaviors to maintain long term weight loss. My study investigate d relationships among health related quality of life, attitudinal ambivalence and behavioral intentions to ward lifestyle change that may influence the deterioration of positive health behaviors as a person progresses through the stages of a weight loss/lifestyle program. T he research methodology used was a s ingle group repeated measures design following 91 ma le and female patients who were physician referred to a n 8 week physical therapy and weight loss program Data were collected using the following five
11 assessment tools: SF 36 HRQL, Attitudinal Ambivalence assessment, Behavioral Intentions assessment, a nd S tage of Change assessment. A repeated measures ANOVA demonstrated statistically significant changes in five of the variables : body weight, HRQL, a mbivalence, stage of change and p rogram compliance (p = .0005) Correlation analyses demonstrate d statisticall y significant negative associations between a mbivalen ce related to a food diary and b ehavioral Intentions related to a food diary Also present were statistically significant negative associations between a mbivalence related to daily exercise and behaviora l i nten tions related to daily exercise. A positive association was de tected between compliance an d percent change in body weight. A negative association with a medium effect size was demonstrated between HRQL for the mental component score and a mbivalence related to a food diary A negative association was also found between the mental component score and a mbiva lence related to daily exercise. My study provides evidence of associations between ambivalence and behavioral intentions correspond ing to a reducti on in program compliance and a concurrent reduction in the rate o f weight loss during the last 2 to 4 weeks of an 8 week healthy lifestyle/weight loss program.
12 CHAPTER 1 INTRODUCTION More than 68 % of the United States (US) adult population is either overw eight or obese (Fegal et. al., 2010). Based on annual increases in prevalence, it is predicted that by 2015, 75% of the adult populat ion will be overweight or obese; 2030 predictions indicate an adult overweight/ obesity rate of 86.3% (Wang et al., 2007). T he financial burden created by the obesity epidemic will increase health care costs, potentially doubling every decade to $900 billion by 2030 (Wang et al., 2008). Chronic health conditions including hypertension, dysli pidemia, and insulin resistance resu lt from overweight/obesity and place individuals at increased risk for death from diseases of the heart, cancer and cerebrovascular diseases ( U.S. National Center for Health Statistics, 2006; Chronic Disease Prevention and Health Promotion 2007). Obesity drastically intensifies the detrimental effect of diabetes, dyslipidemia, and hypertension on medical expenditures and productivity loss in the United States (Sull ivan, Ghushchyan, & Ben Joseph 2008). Those who are overweight or obese experience reduce d quality of life resulting from an increase d number of physical problems and reduced function (Doll, Peterson, & Stewart B rown, 2000; Lopez Garcia and et al. 2003; Renzah, Wooden, and Houng, 2010). However, in obese individuals, health related quality of life (HRQL) has been shown to improve after weight loss (Kral, Sjostrom, & Sullivan, 1992; Mathus Vliegen, de Weerd, & de Wit, 2004). While biological factors can be associated with genetic predisposition to weight gain in some individuals, experts say t he rapid increase in weight over the past 4 decades in the US population stems from an environment al change, not a gen e tic one
13 (Hill et al. 2003). Despite the significant amount of nutrition and exercise information avai lable to the public, the obesity ep idemic continues with little apparent change in behavior (Herbert, 1996; Cifuentes, 2004). Even among numerous commercial and self help programs des igned to facilitate weight loss, success is limited by minimal weight loss and unsuccessful long term mainte nance of weight loss (Tsai and Wadden, 2005). Problem Many behavioral intervention strategies are able to facilitate weight loss by achieving short term changes in health behaviors H owever long term health benefits of weight loss are not often realized since health behaviors decay over time and people lapse into previously held negative health behaviors (Hunt, Barnett, and Branch, 1971, Miguez and Becona 2008). Currently the causes of behavioral lapses during weight loss program s are neither well identi fied nor understood. The persistent and recalcitrant nature of overweight/obesity in the US population cannot be addressed until our understanding of factors driving the obesity epidemic improves. Without improved research in this area of inquiry, effectiv e weight loss/maintenance interventions will remain rare Purpose The purpose of my study wa s to investigate factor s associated with the deterioration of health behaviors and intentions as a person progresses through a health y lifestyle/weight loss progra m. Once this knowledge is gained, interventions intended to attenuate health behavior decay can be designed and implemented. Rationale A ttitudinal and behavioral a mbivalence related to weight loss may be a factor associated with inability t o successfully change the behaviors necessary
14 to maintain long term weight loss. Merely knowing what behaviors promote improved health does not assure that a change in behavior will be the subsequent outcome, nor that behav ior will continue over the long t erm to realize an improvement in health status. Sparks et al (2001) examined ambivalence in the p erspective of attitudes toward food consumption Their findings indicate that ambivalence may have significance for the predictive ability of attitude inten tion behavior models, suggesting that those experiencing motivational conflicts with high levels of ambivalence demonstrate a redu ction in behavioral intentions. Armitage and Conner (2000) used a mbivalence and behavioral intentions toward eating a low fat diet to demonstrate that less ambivalent attitudes are more pred ictive of behavioral intentions. They also demonstrated that increases in ambivalent attitudes were more adaptable when confronted with convincing communication (Armitage & Conner). Additiona lly researchers have demonstrated high levels of ambivalence in the stages of change model (Prochaska 1982), namely the contemplation and preparation stages (Armitage 2003; Armitage & Arden 2007). Previous studies imply that individuals demonstrating high levels of ambivalence tend to be more adaptable (Armitage & Conner, 2000) and possibly more pliable. Also relevant to individuals exhibiting a high level of ambivalence in the contemplation and preparation stages of change is a greater likelihood to resp ond to persuasive communication ; communication that when needed could be used to refocus their behavioral intentions. Research has shown that obese individuals experience an improved health related quality of life (HRQ L) with weight loss (Kaukua, et al. 2003; Blissmer et al.
15 2006). Additionally research has demonstrated that ambivalence and behavioral intentions vary in accordance to the stages of change ( Armitage & Arden, 2007; Armi tage & Connor, 2000). Relationships among HRQL and progressive weight loss, ambivalence, and behavioral intentions through the stages of a healthy lifestyle/weight l oss program remain unanswered. Research Questions RQ1: Does HRQL affect behavioral intentions and ambivalence as one progresses through a healthy lifestyle/wei ght loss program? RQ2: Are improvements in HRQL proportional to the progression in weight loss and can associations among HRQL, ambivalence, and behavioral intent ions be identified? RQ3 : Are lapses in b ehavior the result of improved HRQL, facilitating belief that they can return to previous ly held behaviors, with a subsequent deterioration of health behaviors and health status? My study investigate d relationships among health related quality of life, ambivalence related to attitudes towar d lifestyle changes, and behavioral intentions to make lifestyle changes associated with progression through the stages of a weight loss/lifestyle program. Hypothesis There will be a n association between health related quality of life and levels of ambiv alence and behavioral intentions during progression through stages of a healthy lifestyle/weight loss program. Delimitations 1. My study was conducted at a physical therapy clinic in Gainesville, Florida. 2. Patients enrolled in the healthy lifestyle/weight lo ss program carried medical insurance for coverage. 3. Data w ere collected from May 2011 through February 2013
16 4. Variables were measured using psychometric assessment tools: The SF 36 health survey, b ehavi or Intention assessment scale, s tag e of change assessm ent, and an a mbivalence scale. 5. Psychometric assessment data were patient self report ed 6. Patient d emographic information was collected during the application process using intake medical information. Limitations 1. The physical therapy clinic selected for p articipant recruitment may not represent all physical therapy clinics in Florida or elsewhere. 2. Patients covered by medical insurance do not represent all of the population. 3. O verweight or obese patient s are candidates for a healthy lifestyle/weight loss pr ogram. 4. Data collected from May 201 1 through February 201 3 may differ from data collected during other time periods 5. The assessment tools may not fully des cribe the associated constructs. 6. Self reported responses may be based on inaccurate perceptions and may not 7. Records used to obtain patient demographic information may not be complete. Assumptions 1. The selected physical therapy clinic was considered representative of physical therapy clinics in Gainesville Flor ida. 2. Patients seen provide d an adequate diversity of orthopedic patient populations. 3. Data collected from May 201 1 through February of 20 13 was considered adequate for the intention of this study. 4. The assessment questionnaires chosen for this study were c onsidered appropriate to describe associated constructs involved with each patient. 5. Self reported responses provide an acceptable level of honesty and perception for the purpose of this study. 6. Records used to obtain demographic information about each pati ent provide adequate information about each patient.
17 Definition of Terms AMBIVALENCE Two equally intense, opposing attitudes about a construct. BEHAVIORAL INTENTION behavior. It is assumed t o be the immediate antecedent of behavior. STAGE OF CHANGE Behavior change is a process rather than an event T he change process takes place over months and years and is chara cterized by six distinct stages: pre contemplation, contemplation, preparati on, action, maintenance. SELF CONTROL The ability to exert control over oneself by withholding immediate gratification for something that may be healthier (Muraven and Baumeister, 2000) R EWARDS Gratification received after an action, designed to cause a recurrence of the process. RESTORATIVE BEHAVIOR AL COPING Specific behavioral techniques which involve overt activity that would allow for continuation of a behavioral change during periods of lapses in behavior. Examples of these are stimulus control ( throwing away the remaining food), social support (discussing the lapse with a friend), and compensation (eating fewer calories the next day). RESTORATIVE COGNITIV E COPING Specific non observable mental activities that allow for continuation of a behavio ral change during periods of lapses in behavior. and positive thoughts (self encouragement, thoughts of accomplishment). LAPSE A slight error or temporary deviation typically due to forget fulness or inattention. RELAPSE The act of or an instance of backsliding, worsening, or subsiding. HEALTHY LIFESTYLE PR OGRAM A cognitive and behavioral program designed to improve food choices and facilitate exercise in overweight/obese patients. WEIGHT LOSS A volitional effort to lose weight through dietary modifications and/or addition of regular exercise (Miller, 2010). METABOLISM The energy it takes for the body to operate its basic bodily functions (Ravussin and Bogardus, 1989). Summary Many obe sity related health issues increase the risk for associated diseases. H ealth related quality of life is detrimentally affected by obesity through reduced qu ality
18 in the areas of function and health, both physical and emotional Long term weight loss is dif ficult to maintain T hose who embark on lifestyle changes often regain much of their weight and many associated health problems. Factors affecting health behavior decay include genetic factors interacting with environmental influences. A ttitudinal ambival ence is toward a lifestyle change, which affect s behavioral intentions and behavior toward areas related to weight loss.
19 CHAPTER 2 LITERATURE REVIEW Health care costs will unquestionably increase during the next 30 years as a result of and accessing Medicare benefits. The economic and social impact of the future growth of US health care expenditures for the el derly will be significant (Rice & Fineman, 2004). Obesity will impact costs sin ce an obes e 65 year old individual entering the Medicare program require s immediate and higher health care expenditures than normal wei ght individuals (Finkelstein et al. 2008). An obese 45 year old person surviving to age 65 has significantly higher aver age lifetime Medicare costs of $163,000 compared to $117,000 for a normal weight 45 year old surviving to age 65 (Cai, Lubitz, Flegal, & Pamuk, 2010). Additionally a s obesity and resulting health effects occur in younger individuals costs to the healthc are system will be earlier and for longer than in past years when obesity developed at older ages. The reality of these costs are substantiated by recent e stimates of national healthcare expenditures for children related to obesity : $ 14.1 billion in additi onal prescription drug, emergency room and outpatient costs annually (Trasande & Chatterjee, 2009). M any obesity related chronic diseases affect the health related quality of life of obese individuals (Field et al. 2001). When obese individuals are asse ssed using the health assessment short form (SF 36), they typically exhibit low scores on the physical aspects of assessment involving functioning (Doll, Petersen, Stewart Brown, 2000). P hysical functioning is most commonly affected by orthopedic problems, that impede joint m obility and interfere with gait and other activities of daily living ( such as climbing steps, lifting groceries, and getting into and out of the bathtub ) (Gelber et al. 1999; Oliveria et al. 1999). Researchers have found a positive as sociation with the body
20 mass index (BMI) and knee osteoarthritis (OA) (Coggon et al. 2001) ; specifically knee, hip and back pain (Anderson et al. 2003). A study comparing obese /non obese individuals demonstrated a 13 fold increase in the incidence of kn ee osteoarthritis (OA) among the heaviest subjects (Felson et al. 1988). Studies have also demonstrated that hip joint replacement at younger age s is positive ly associated with body weight and hip contact stress (Recnik, Kralj Iglic, Iglic, et al 2009). In contrast, research show s that obese individuals who lo se a modest amount of weight (0 to 9.9% weight loss) have improve d SF 36 scores in the area of psychosocial functioning, physical functioning, and general health at 2 years, demonstrating a dose res ponse improvement with increasing weight loss. In my study participants who had >10% weight loss at 2 years showed an improvement in psychosocial problems, physical functioning, physical role functioning, bodily pain general health, mental health, and vit ality (Kaukau et a l. 2003). The literature also offers studies describing overweight people with radiological evidence of knee ( OA ) who upon losing 10 to 12 pounds on average reported a 30 % improvement in knee pain and experienced 24% improvement in th eir ability to perform daily activities; including the ability to walk up stairs more quickly and easily than those subjects who did not lose weight (Messier et al. 2004; Christianson et al. 2005). Long term changes in health behavior are central to last ing health benefits in ma ny disease prevention models (Lisspers et al. 2005; Wing and Phelan, 2005). Genetic components driving the interaction between behaviors and challenging environments provide obstacles that sometimes counter the efforts of healt h be havior change causing limited long term success (5+ years) of health behavior programs (Stalonas, Perri, & Kerzner, 1984; Kramer, et al. 1989). Many factors affect the
21 progression of health behavior change Long term health behavior change cannot be ac complished m erely by knowing which behaviors promote improved health outcomes (Steuart, 1967; Page and Cole, 1984 85; Becker & Joseph, 1988; Becker 1990; Silverman Pe rakyla, and Bor 1992; Ferris et al. Moreover programs that focus solely on behavior change make assumptions about the potential cause and effect of change s in health status (Lorig & Laurin, 1985). Therefore, targeting outcomes that will measur e improvements in health status is highly important. Factors T hat A ffect Health B ehavior Biological and Environmental Factors T hat Affect Behavioral C hange F actors affect ing relapse into addictive be haviors are multiple and includ e environmental factors (Brownell, Marlatt, Lichtenstein, & Wilson 1986), genetic predispo sition, and h abitual circuitry. Habitual circuitry is an internal system that reinforces cognitive a nd behavioral processes based on a rewards system involving the dopanergic pathways (Newlin and Strubler, 2007). By performing a task that stimulat es the d opanergic pathway, learned behavior occurs to reinforce further cognitive decisio ns that continue to support the repetitive behavior, thereby developing a habitual circuitry. The mesolimbic dopamine pathway, a chemical circuit in the brain is stimulated b y m any drugs and some foods (Kreek, 1996). This circuitry involves the medial forebrain bundle sometimes referred to as the pleasure center. It has been demonstrated that the following drugs intensify the action of neurotran smitters: dopamine, gamma ami no butyric acid (GABA), and serotonin (Koob, 1992). These pathways work on either a rewards principle, similar to the rewards received from heroin, alcohol, and cocaine (Rocha et such as that received fro m nicotine (Epping Jordan 1998). Studies have demonstrated that even activities such as
22 gambling and playing video games have a pronounced e ffect on stimulating the dopam ine pathway (Koepp et al 1998; Potenza, 2008). Biochemical, functional neuroimagi ning, genetic studies, an d treatment research have demonstrated a strong neurobiological link between behavioral addictions and substance use disorders (Grant, Brewer, and Protenza, 2006). Research has demonstrated b ehavioral and neurochemical similarities between binge eating and the administration of drugs with abuse potential ( Alsio Olszewsk i, Levine, & Schioth, 2012). Animal models have been used to verify similarities between bingeing on sugar and behaviors associated with abuse potential drugs (beha viors such as opiate like withdrawal signs, heightened intake following abstinence, and overlapping areas of sensitivity ) Aveena ( 2007) demonstrated the exacerbation of these neurochemical alterations when low weight animals binge d on sugar when the foo d they ingest ed was purged Given that in humans these underlying neuro physiological pathways evoke immediate and often overwhelming gratification for the individual, health behavior models theorize the need for progressive stages of self awareness in or der to help override this basic biological phenomena and to facilitate behavior change. Metabolism The amount of energy it takes for the body to ope rate its basic bodily functions ( including lung, heart, kidney and liver functions ) while at rest is defined as the basal metabolic rate The metabolic rate of an individual can be altered secondary to age, sex, lean body mass, and certain diseases. Eighty three percent of the variance in metabolic rate is attributed to fat free mass, while age and gender contri bute minimally (Ravussin and Bogardus, 1989). Activation of the sympathetic nervous system occurs predominately in response to stress and works to stimulate the catecholamines
23 ( epinephrine and norepine phrine ) from the adrenal glands which increase s the me tabolic rate. The thyroid gland produces hormones that play a major role in maintaining and altering metabolism. Environmental Factors T hat A ffect B ehavior E nvironmental changes contribute to the obesity epidemic. Environmental changes that have contribu ted to the obesity epidemic i nclude increased marketing and accessibility to food products ( predominantly fast food and other prepackaged foods that are immediately consumable and usually eaten away from home ) A transformation in the dynamics of family l ife driven by changes in social conditions ( such as dramatic increases in the proportion of women who work outside the home ) has introduced time constraints that have replaced meal preparation with convenience foods (French, Story and Jeffrey, 2001). Add itionally, other contributing factors such as increased forms of sedentary entertainment ( for example television video games, and computer use) and a reduction in walking and biking compound obesiogenic behaviors. E nvironmental factors contribute to wei ght gain and complicate efforts to maintain weight loss over time Studies of long term adherence to treatment strategies have demonstrated that most subjects regained a major portion of the w eight lost during treatment 5 years later and points to the need for new conceptual models to behavioral programs ( Jeffery, Epstein, Wilson, Drewnowski, Stunkard, & Wing, 2000; Anderson, Konz, Frederich, & Wood, 2001 ) One study reported that fewer than 2% of individuals were able to maintain 10% weight loss over 5 yea rs (French, Jeffery, Folsom, McGovern, & Williamson1996). Subjects reported numerous situational, social and emotional factors as negatively impacting their weight control efforts (S talonas, Perri, & Kerzner, 1984; Visram, Crosland, & Cording, 2009).
24 R temptations to overeat was predicted by immediate coping. However, the part icular type of coping mechanism ( cognitive versus behavioral ) made little difference. Restorative behavioral coping was elicited as a response to overeating, while restorative cognitive coping seemed elicited by the negative thoughts and feelings accompany ing behavioral lapses or temptations ( Grilo, Shiffman & Wing, 1993). Recovery from r elapse into previous be haviors demonstrated behavioral re lapses ( defined as weight regain of 5% or more ) showed that recovery from even minor weight regain was uncommon (Phelan, Hill Lang, Dibello, & Wing, 2003). Self C ontrol over Addictive B ehaviors Self control is described a s the abilit y to exert control over oneself. An example is the ability to suppress the desire for immediate gratification when faced with a behavioral decision to make the healthier choice. Self control involves overriding challenging urges, behaviors, or desires for purposes of optimizing the most advantageous long term interests of an individual The significance of self control in relation to the resolve needed to overcome urges for self gratification is unc ertain. Muraven and Baumeister (2000) said i ndividual resources for self control are limited and most likely partially consumed in the energy process required for d emonstrating self control over circumstance s M anaging stressors while simultaneously practicing self control ( including overriding dist racting thoughts, urges, and emotions) can be exacerbated while trying to focus attention and avoid distraction and fatigue from multiple extenuating factors C oping with stress ors is believed to result in decreased ability to exercise self control At t he same time, the increased pressure to exert self control results in depletion of the energy needed to
25 accomplish a particular task ( Muraven, Gagne, & Rosman, 2008). According to Muraven, Tice, and Baumeister (1998) individuals have limited physical and emotional resources to control and/or alter their behavior, and this capacity appears to be challenged by continuous and mounting demand. T hese resources can be depleted when people find themselves in circumstances requiring additional self control or when self control reserves are reduced in response to various life stressors. As a result some individuals may find fulfill ing important life goals ( particularly goals requiring self control, dis cipline and focus ) increasingly difficult. Additionally feelin g compelled to exert self control requires greater amoun ts of resolve than exertion of self control for more independent or autonomous reasons (Muraven, Gagne, Rosman, 2008). Amb ivalence in Health Behavior C hange Ambivalence is an attitude characterized by one or more of the following attributes : 1) exhibits stability over time 2) ability to influence information processing 3) with ability to resist persuasion and 4) ability to impact behavior (Krosnick & Petty, 1995) Ambivalence regarding health beh avior change is characterized by simultaneous in combination with impulses and desire to go against that understan ding in a given situation (van H arreveld, van der Pligt, & de Liver, 2009). Sparks et al. (2001) examined ambivalence in relation to food consumption attitudes. Their research cites conflicts arisi ng between the sensory rewards received from sweet and fatty foods and the and body image. As such, the relationships between ambivalence and behavior can differ as a functi on of the degrees of ambivalence. This dichotomy ( individual knowledge of what should occur versus individual desire for a particular reward ) suggests that ambivalence underlies the
26 internal struggle of choosing the immediate benefits of s hort term desires (often sensory rewards) or the long term benefits of delayed gratification ( usually health benefits ). Research findings indicate implications for ambiva lence regarding the predictive ability of attitude intention behavior models, especially when applied to health related behaviors characterized by mo tivational conflicts (Sparks et al 2001) F eelings of conflict and discomfort are most apparent when ambiv alent individuals must make a choice regarding the existing conflict. When feelings of ambivalence conflict with personal values negative affect can occur (Baron & Spranca, 1997). For example while on a weight loss/lifestyle program an individual may ex perience conflict when trying to make good food choices in the setting of important family get togethers or social gatherings centered around food. One may experience a negative affect by not participat ing when trying to make a choice. However when a cho ice between the two can be avoided, commitment to either can remain low, resulting in less discomfort, even in the presence of ambivalent feelings. Ambivalence about health issues present s in diverse forms, including cognitive ambivalence ( characterized b y mixed views), and affective ambivalence ( characterized by mixed feelings ) potential ambivalence (individuals are unaware of their existing beliefs ) or felt ambivalence ( both aware of and experiencing their feelings ) (Armitage and Arden 2007). When indi felt ambivalence in relation to a health behavior or issue emotional discomfort can result. When attempting to change health behaviors the degree and form of ambivalence may vary as one pro gresses through stages of change. While the concept of potential and felt ambivalence differs Armitage and Arden (2007) demonstrated a
27 moderate correlation between the two In a sample of the general population Armitage et al. (2003) demonstrated a quadratic relationship between the stages of chan ge and attitudinal ambivalence. Findings demonstrated significantly higher levels of ambivalence among participants in the contemplation, preparation and action stages in regarding eating a healthy diet than among people in either the precontemplation or maintenance stage s Behavioral Intentions Behavioral intentions mediat e attitudes and behavior. The Theory of Planned B ehavior asserts that attitude toward the action and perceived behavioral control best explains the variation in intention (Schifter & A jzen, 1985; Godin & Kok, 1996). However, in the presence of a mbivalent attitudes the Theory of Planned Behavior loses predictive ability in the constructs behavioral intention and behavior ( Figure 2 1 ) (Armit age & Conner, 2000; Sparks et al., 2001). Less ambivalent attitudes are more predictive of behavior, while ambivalent attitudes are less predictive of behavior, but may be more easily influenced with intervention ( Figure 2 2 ) (Armitage & Conner 2000). Stages of Change and Ambivalence The Transtheore tical model (TTM) developed by Prochaska, and DiClemente (1982) asserts movement through six stages of change ( pre contemplation, contemplation, preparation, action, maintenance and termination) which are characterized by periods of progression and relapse Prochaska and colleagues (1982; 1994) said changes in the balance between pros and cons are responsible for individual progression through contemplation, preparation and action stages of change (Prochaska, et al., 1994). The changing balance between pro s and cons also
28 characterizes the stages with the greatest ambivalence. Notably the precontemplation and maintenance stages are characterized by the least ambivalence since the pros and cons are more polarized (Armitage & Arden, 2007). The pre contem plation stage describes individuals who do not recognize a need for change. This lack of recognition does exclude developing intentions to change later. The contemplation stage includes individuals contemplating a behavior change within the next 6 months. Individuals in the preparation stage aim to take action within the next 30 days and have taken some behavioral steps in a positive direction The action stage is distinguished by evident, quantifiable changes in behavior consistently practiced for the pas t 6 months. Individuals in the action stage are at risk for re lapse Therefore successful movement through this stage requires constant attention and vigilance. When changes in behavior continue for 6 consecutive months, a person moves into the mainte nan ce stage. In this stage the behavior is now a regular action or activity. As a result, the probability for relapse decreases but remains present and requires occasional attention. Individuals enter the termination stage once the behavior change has been maintained 5 or more years and temptation to engage in the non desirable behavior has been replaced with feelings of total self efficacy when performing the new behavior (Prochaska & Velicer, 1997). Health Related Quality of Life, Ambivalence and Behavio ral I ntention N euro pathways involved in the basic mechanisms of survival are mimicked by activities that stimulate a similar pathway (the dopanergic pathway) (Pani & Gessa, 1997). Activities such as smoking ; gambling; and ingesting certain foods, alcohol and drugs are activities that present rewards through stimulation of brain areas similar to
29 those of the dopanergic pathway, despite hazard s posed to species (Koepp et al 1998; Kreek, 1996; Potenza, 2008). In contrast an individual may experience fee lings of ambivalence when significantly challenged by pain and discomfort that are linked to certain diseases or health conditions ( such as those associated with obesity ) while simultaneously facing pleasure (reward) seeking behavior through certain foods or activities that may have enabled the original condition. Indiv in relati on to their quality of life, feed their resistance to major behavioral change. To complete any type of behavioral change people must be dissatisfied with their current quality of life status. Once obese individual s become dissatisfied with their health related quality of life they become more inclined to make behavioral change s to improve their health condition and associated quality of life (Bish et al., 2007), particularly if the goal is perceived as attainable (Fujioka, 2010). As a person progresses through the lifestyle/weight loss program an improved quality of life with a reduction in problems from health issues/c onditions often o ccurs (Ross et al. 2009). As a result individuals may be more inclined to seek out previously pleasurable activities that may have contributed to the initial health issue (Miguez & Becona 2008.) This dichotomy may explain the persistent problem sustaini ng long term health behavioral changes Lapses in behavior can occur even while one is progress ing toward positive changes in health status (Phelan, Hill, Lang, Dibello, & Wing, 2003). Associated increases in ambivalence most often occur during the contemp lation, preparation, and action stages Additionally increased attitude ambivalence during these stages has been shown to be more pliable resulting in a more positive response
30 to behavioral change interventions (Armitage & Conner, 2000). Stages in which increased ambivalence occurs (contemplation, preparation, and action stages) may be times when a patient best responds to mot ivational interviewing (West et al. 2007) followed by implementation of intentions (Chapman & Armitage, 2009), in order to encoura ge continued behavior changes. The least ambivalence is experienced during the precontemplation and maintenance stages (Armitage, Povey & Arden, 2003).
31 Attitudinal Ambivalence Related to the Stages of Change Low AA High AA Hi gher AA Moderate AA Low AA Area of Increased P liability Attitudinal Ambivalence = AA Figure 2 1. L evel of attitudinal ambivalence (AA) varies across the stages of change. A lower level of ambivalence is seen both at the pre contemplation and maintenance stages, while a higher level of ambivalence is seen in the contemplation, preparation and action stages. Attitudes have increased pliability in the areas of high attitudinal ambivalence. Pre contemplation Contemplation Preparation Action Maintenance
32 Ambivalence Predicts Behavioral Intention and Subseq uent Behavior Figure 2 2. L evel of ambivalence affects behavioral intentions and subsequent behavior. Behavioral intention is less predictable during a heightened level of ambivalence compared to low levels of ambivalence. Unpredictable Behavioral Intention Behavior High Ambivalence Predicta ble Behavioral Intention Low Ambivalence Behavior more Predictable Behavior less Predictable PredicPredictable
33 CHAPT ER 3 METHODS Recently the CDC classified obesity as a chronic disease (Byas, 2013 ). T his new classification recognizes the importance of intervention programs designed to achieve long term weight loss W eight loss programs abl e to demonstrate sustainabil ity and resulting long term benefits are vital if obesity and its sequelae are to be addressed (Gilden, Tsai, Thomas, 2005). Efforts at long term weight loss and sustainability have been hampered by health behavior decay which often leads to lapses in weig ht loss efforts return to previously embraced addictive eating behaviors ( Dohm, Beattie, Aibel, Striegel Moore, 2001; Miguez and Becona 2008 ) Also complicating weight loss efforts is the time span between actual weight loss and evident health benefits. For many, the absence of immediate tangible health benefits dampens continued efforts at dietary control and increased physical activity levels. During this time frame health behavior decay is also occurring, adding to an ind waning commitment to weight loss and improved health. As a result, research identifying factors that contribute to health behavior decay is vital for the success of long term weight loss and maintenance programs. To better understand the behavioral lapses frequently associated with weight loss we examine the relationships among changes in body weight and health related quality of life ( HRQL ) ambivalence, behavioral intentions, stage of change, and program compliance Research Design Be f ore data collecti on Institutional Review Board ( IRB ) was obtained The research methodology used quantitative data collection My study used quantitative data implemented collectively on three separate occasions, with one quantitative
34 variable collected re peatedly throughout the study. My study used a prospe ctive quasi experimental single group research design, with multiple measurement s taken at three points in time. B aseline data w ere collected from all participants during the initial evaluation (T ime 1) at 4 weeks post initial evaluation (T ime 2) and again at 8 weeks post initial evaluation (T ime 3) Time lines for data collection best approximate the time of treatment for the prognosis related patient referral. A repeated measures design allows the researcher to monitor participant behavior changes over time. This design also reduces the var iance of estimates of treatment effects over time, allowing statistical inference to be made with a smaller sample Research Variables My study m easured behavio r via health related quality of life (HRQL) attitudinal ambivalence, behavioral intentions, stage of change and program compliance. These variables were itemized based on each assessment tool. The SF 36 HRQL was used to measure a physical and mental comp onent score. The Attitudinal Ambivalence Assessment Too l was used to measure ambivalence related to weight loss, keeping a food diary, and daily exercise. The Behavioral I ntentions Assessment Tool was used to assess weight loss, keeping a food diary, and d aily exercise The Stage of Change Assessment T ool measured food choices and daily exercise behaviors by individual stage of change A p rogram compliance assessment tool was used to determine program compliance Additionally weight loss was measured weekl y to assess physical changes occur ring over time. Table 3 1 shows instruments and variable characteristics.
35 Continuous variables include d health related quality of life, attitudinal ambivalence, behavioral intentions, stage of changes, program compliance and body weight. Dependent variables include d health related quality of life, attitudinal ambivalence, behavioral intentions, stage of changes, program compliance and body weight. The independent variable included intervention over time. Health Related Quality of Life S cores for the physical and mental component s were measured using th e SF 36 health assessment s cale ( Appendix B ) The SF 36 is a multi item psychometric health assessment scale that measures 8 domains of health: 1) limitations in physical a ctivities because of health problems; 2) limitations in social activities because of physical or emotional problems; 3) limitations in usual role activities because of physical health problems; 4) bodily pain; 5) general mental health; 6) limitations in us ual role activities because of emotional problems; 7) vitality (energy levels and fatigue); and 8) general health perceptions. The SF 36 is described as a valid psychometric assessment tool when used with patient population (Mchorney, Ware, & Raczek, 1993) mixed gender population s (Lim, Seubsman, & Sleigh, 2008) and obese adult patient pop ulation s (Corica et al. 2006). Attitudinal Ambivalence The Griffin index ( Appendix C ) was used to measure attitudinal ambivalence. Attitudinal ambivalence initially d escribed by Thompson (1995) and later highlighted by Armitage and Connor, (2000) i s a condition in which a person regard s an issue with an equally strong positive and negative assessment. T o determine both the positive and negative attitudes ( including in dividual feelings, thoughts and beliefs regarding three specific areas: weight loss, daily exercise, and maintaining a food diary ) patients were
36 presented with two attitudinal assessments The first assessment measured positive attitudes, feelings, and be liefs toward each above named area The second assessment measured corresponding negative attitudes. The following statement is a weight loss related example : Think about your attitude toward or evaluation of weight loss. That is an intentional reduction of your body weight through diet, exercise or both. Considering only the favorable characteristics of weight loss and ignoring the unfavorable characteristics, r ate how favorable is your evaluation of weight loss? Not at all Slightly Quite Extremel y Unfavorable Unfavorable Unfavorable Unfavorable The second statement is the same but address es feelings toward weight loss : Think about your feelings or emotions when I mention weight loss. Considering only your feelings of satisfaction toward weight loss and ignoring your feelings of dissatisfaction, how satisfied do you feel about weight loss? The third statement address es thought and beliefs toward weight loss. Think about your thoughts or beliefs when I mention weight loss. Considering only the be neficial qualities of weight loss and ignoring the harmful characteristics, how beneficial do you believe weight loss to be? These same questions were presented to evaluate attitudes toward The second assessment concerned negative attitudes, feelings, thoughts and beliefs toward weight loss, the food diary and daily exercise. The first example concern ed attitude about weight loss. Think about your attitude toward or evaluation of weight loss. That is an intentional redu ction in your body weight through diet, exercise or both. Considering only the unfavorable qualities of weight loss and ignoring the favorable characteristics, how unfavorable is your evaluation of weight loss?
37 Not at all Slightly Quite Extremely Unfa vorable Unfavorable Unfavorable Unfavorable Each question was weighted from 0 to 3 as in the Griffin formula. All three assessment areas follow a similar pattern to ascertain level of ambivalence regarding each issue ( positive ambivale nce assessment A ppendix D; negative ambivalence assessment Appendix E ) The Griffin index was designed to attain the level of similarity and the intensity of the two issue specific attitudinal ambivalence assessments, and is determined by the equation: A mbivalence = (positive + negative)/2 positive negative (A ppendix C) The Griffin I ndex has been shown to ambivalence is associated with positive and negative eval uations that are both int ense ndex the mean of the positive and negative evaluations demonstrates the intensity while the difference between the absolute value of the positive and negative evaluations indicates the similarity o f the construct evaluation Table 3 2 was developed using a weighted 0 to 3 scale t o demonstrate the possible outcomes of a construct positive ly evaluat ed and illustrate how a negative evaluation is used to determine level of ambivalence A ccording to the Griffin I ndex, construct s rated as 1 (positive) and 2 ( negative ) yield an ambivalence rating of .5, whereas ratings of 3 (positive) and 3 ( negative ) yield an ambivalence rating of 3. According to the index high ambivalence is characterized by equally intense and similar positive and negative evaluations Therefore, a higher score represents greater ambivalence than would a smaller or negative number. Thompson & Zanna, (1995), demonstrated v alidity of ambivalence as a con s truct using a posit ive self report ed measure of ambivalence among a sample of 61 female
38 psychology students. Armitage & Connor, (2000) demonstrated both acceptable reliability of .84 and discriminant validity Behavioral Intention Behavioral intention (A ppendix F ) was as sessed using two similar 7 point bipolar scale ( from 3 to +3 ) statements Each statement address es the same 3 areas used to assess attitudinal ambivalence (weight loss, daily exercise, and a food diary). Each statement is followed by a measured degree of intent or plan with response options ranging I intend to lose weight in the future. definitely do not 3, 2, 1, 0, 1, 2, 3 definitely do I plan to lose weight in the future. defin itely do not 3, 2, 1, 0, 1, 2, 3 de finitely do Past behavioral intention research by Armitage and Conner ( 2000) demonstrated test validity and reliability ranging from .83 to .88 W ilson and Rogers (2004) used the 3 item behavioral intention scale with a sample of 49 overweight/obese females and demonstrated an internal consistency of .89. Stages of Change Stage of change (Appendix G) for both exercise and food choices w as determined at T1, T2, and T3. To determine current stages of change participants were asked to choose a personal descriptor from among 5 topic relevant statements. For example Responses were numbered as indicated: 1) precontemplation stage, 2) contemplation stage, 3) preparation stage, 4) action stage, and 5) maintenance stage Similar statements were used to determine stage of change for food choices.
39 Daily exercise and food choi ce criteria were est ablished. The criteria for regular moderate exercise was exercising 5 to 7 days per week for 30 to 40 minutes. Moderate intensity was characterized as a lack of painful, increased breathing, but able to carry on a conversation while exercising, and light s weat ing Criteria for food choices were established for five areas: including eating high fiber foods (Tucker & Thomas 2009), at a frequency of 1) 5 to 9 servings of fruits and vegetables daily; 2) 2 to 3 ser vings of beans and whole grains, partaking of h igh calcium foods ; 3) 2 to 3 servings of fat free dairy, get ting enough protein (Layman, et al. 2009); avoiding high fat foods by 4) eating lean meats only and avoiding high fat fried foods and condiments; and avoiding high sugar items like 5) high sugar drinks or foods like sugary sodas and deserts (Malik, Schulz, & Hu, 2006). The es tablished criteria were used daily by participants when completing their Food Choice ( Appendix H ) T he Stage of Change measures demonstrated good v alidit y when used with both obese adults and hospital workers ( Armitage, Sheeran, Conner & Arden, 2004; Sarkin et al. 2001 ). Using the scale with a sample of 768 overweight/obese individuals demonstrated good internal consistency (.66 to .92), (Valis et al. 2 003). Complianc e for Exercise and Food C hoices xercise measured participants level of compliance with food choices and exercise. Compliance was determined by fulfillment of an established percentage of the criteria indicated for both food choices and exercise. For example, compliance with food criteri a was measured by the number of points participants received in this category. Participants received 1 point for fulfilling each food choice category and one additional point for staying within 5 points of their total daily point allocation. If they reach ed their goal in each area at a 100% they receive d 6
40 poi n ts. Fully compliant participants who reached their goals in each area (score 100%) received 6 points. For fulfill ing exercise goals of 30 to 40 minutes per day participants received 2 compliance points. Participants could receive a maximum of 8 points each for food choice and exercise compliance Participant compliance was determined using their self report Appendix I ) Body Weight Body weight and height w ere measure d using a medical scale ( Detecto medical scale Webb City, MO. ). For consistency one scale was used throughout the study for all but 2 participants whose weight exceeded the scales weight capacity. During the study period these two participants were weighed using an alternate scale (Health o M eter p rofessional scale Rye, NY ) Both scales were calibrated by BioMed Techs, Inc. ( Mt. Arlington, NJ) for reliability and validi ty. For all participants weight was recorded to the nearest .25 pounds and height was recorded to the nearest 1 inch. H eight and weight measurements were used to determine participant body mass index (BMI) values. For all participants, measurements and BM I calculations were performed at T1, T2, and T3. Participants Study participants included males and females ranging in age from 18 to 83 years. All participants received physician referrals to the weight loss/healthy lifestyle program. Before starting the study i nformed consent was obtained from each participant Participants received a copy of the informed consent for their personal records. Throughout the study HIPA A guidelines ( Health Insurance Portability and Accountability Act ) issues relating to pa tient confidentiality were strictly followed
41 Assessment data were collected c onfidentially and kept in a secured file cabinet behind locked doors. Setting My study was conducted in ReQuest Physical Therapy clinic in Gainesville, Florida. Procedures My study involve d patients referred to an orthopedic physical therapy clinic with a primary orthopedic diagnosis and a secondary diagnosis of obesity. All subjects were first assessed by a medical physician and then referred with a prescription of a primary orthopedic diagnosis and a secondary diagnosis of obesity. Inclusion criteria : all patients referred with a primary orthopedic diagnosis and a secondary diagnosis of overweight/obesity. Descriptive statistics w ere recorded for all participants and include d age, gender, race, body weight, height, body mass index and zip code Additionally diagnosis and insurance plan s w ere recorded. Measures of change in the following areas were included: health related quality of life, ambivalence, behavioral intentions, body weight and compliance with the program. My study follow ed prot ocols set forth by the institutional review board (IRB) for clinical studies. All patients receive d an informed consent release form established through confirmation process. Aft er the initial evaluation process patients were told about the study and given an opportunity to participate. Patients were referred to the physical therapy clinic by a licensed medical physician with a primary orthopedic diagnosis ( e.g. low back pain or knee pain) and a secondary diagnosis of obesity. An advertising flyer was also used to inform people about the study and include d general eligibility information, affiliation, purpose location, and contact information ( Appendix J )
42 Inclusion c riteria: Pa tients referred to physical therapy by a licensed medical physician with a primary orthopedic diagnosis and a secondary diagnosis of obesity. Exclusion c riteria: Patients who have a body mass index lower than 25 w ere not considered for the study. Patients had to be able to comprehend reading material at a 7 th grade level. This was ascertained both through the initial medical application process and the initial intake assessment for physical therapy. Additional exclusion criteria include d the following. 1. Pa tients must not have any gastrointestinal diseases that would interfere with their participation in a lifestyle/weight loss program that encourages plentiful high fiber foods such as fruits, vegetables, beans a nd high fiber cereals 2. Patients with diseases or syndromes that int erfere with the appetite ( hyperthyroidism, Prader Willi syndrome ) 3. Patients that are on medications that would signif icantly interfere with appetite ( corticosteroids, cyproheptadine, and tricyclic antidepress ants ) 4. Patient s with diseases that affect the metabolism (h ypothyroidism, hy perthyroidism ) 5. Patients with gastric bypass or similar surgeries ( such as gastric balloon or gastric band ) Each patient received an orthopedic evaluation followed by an assessme nt of body mass index and disease risk identification related to obesity performed by the physical therapist carrying out this study. Assessment forms w ere then given to patients after instruction on the healthy lifestyle/weight loss program. The physical therapist helped the patient with goal setting for weight loss and instructed on exercise and food choices. Short and medium range goals were set based on 5 to 10% initial body weight (Grundy et al. 2005) and a long range goal of 15% of their initial body weight w ere set with weight loss at the rate of approximately 1.25 pounds p er week (NHLB, 1998; Jakicic et al., 2001). All patients were involved in a cognitive and
43 behavioral lifestyle program that included a behavioral contract (Ured a, 1980); a lifestyle binder with educational information on nutrition and exercise along with daily food choice and exercise logs (A ppendix H ) (Shay et al. 2009); as well as motivational information designed to encourage, inform and challenge the patien t. Patients were weighed at least once every week and received a weight graph of their progress (Butyn, Phelan, Hill & Wing, 2007). Patients were involved with an exercise program on each of their clinical visits and were to bring their lifestyle binder to each session to enhance accountability regarding participation in the program. Patients received detailed instruction of the program on their first visit including instruction on all educational material and behavi oral components of the program. Therea fter patients were involved with therapeutic exercise each session to address their functional problems and were also instructed on progressive integration of their rehabilitation exercises into a general fitness program. Data Collection Data collection to ok place o ver a 20 month period, from May 2011 through February 2013. After orthopedic evaluation and instruction on the healthy lifestyle/weight loss program research assessment forms were then given to patients. This first set of assessments served as a baseline (Time 1) for all behavioral areas. The assessment forms were completed in the following order: SF 36, HRQL, followed by the Attitudinal Ambivalence form for positive attitude, then the Stage of change assessment form followed by the Behavioral i ntentions form and then the Negative Attitudinal Ambivalence form Patients received their second set of assessment s at 4 weeks (Time 2) completing all assessment forms in the same order as in their baseline assessment.
44 The third set of assessment s were performed at 8 weeks (Time 3), with all participants completing all assessment forms in the same order. The procedure used each time a participant completed the set of assessment forms was as follows : 1. Thanking them for their willingness to participate in the study. 2. Giving participants the assessment forms and instructing them on completion of the assessments that followed a script: understand how to help people as they progress through a healthy lifestyle/weight loss progr am. I am studying changes in health status and changes in behavior as people go through this healthy lifestyle/weightloss program. As a part of my research about your current health status and different feelings yo u experience related to ambivalence and behavioral intentions, as well as where you feel you are regarding to making a healthy was followed by answering any questions they may have regarding the assessments. 3. Participants were then a sked to circle only one answer that best describes the m for each statement. 4. Reminding participants to take their time completing the assessment forms. 5. Making sure participants were not interrupted or distracted as they completed the assessment forms. 6. Ma ki ng sure I was available to answer any questions participant s may had as they completed the forms. 7. Upon completion forms were checked for any missed questions, or stray marks that confuse data results. 8. Assessment forms were then collected from participan t s and placed in a secured file cabinet. Patients then completed their therapy session. Other data collection measures involved measuring body weight. This was done weekly, using the same scale the participant used at baseline, with as close as possible t o same attire ( shoes on versus off ) Body weight was recorded to the nearest .25 pounds and placed in participants file s
45 Data w ere collected on program compliance measured each patient visit by s and exercise logs in t heir lifestyle binder. Program compliance logs were kept on all patients, for daily food choice and exercise compliance. Patient visits were recorded each visits in patient chart s. Total visits were later computed for compliance. Data Analysis Methods fo r data analysis were selected based on the nature of the data invoked by the research question. This led to methods that woul d analyze parametric versus non parametric data, and ordinal versus interval data. A T ype 1 error rate was set at .05 for all tests. Correlational analysis was performed on data to look for associations among variables. Demographic data w ere analyzed using descriptive statistics on all study participants. F requencies and percentages were reported for each variable along with the mean standard deviation, and range. Variables included age, ethnicity, BMI, socioeconomic level insurance, and diagnosis. Quantitative Analysis for Parametric Data Descriptive analysis was performed on variables for mean, range, and standard deviation Sta tistical a nalysis was performed on the effects of physical therapy and a healthy lifestyle/weight loss program intervention over time on d ependent variables W eight loss at time 1 (baseline), time 2 (4 weeks) and time 3 (8 weeks) Program compliance was an alyzed at 2 week intervals: Time 1 (1 to 2 wk), Time 2 (2 to 4 wk), Time 3 (4 to 6 wk) and Time 4 (6 to 8 wk ). Repeated m easures analysis of variance (ANOVA) was used to determine statistically significant changes in body weight including percent change i n body weight over Time 1, Time 2, and T ime 3 with a
46 bonferroni correction factor. Program compliance w as also analyzed with repeated measures ANOVA at 2 week intervals over 8 weeks. Effects size was calculated using partial eta squared for which Cohen (19 88) suggested effect sizes of .01, .06, and .14 to indicate small, medium and large effects respectively. Results of these analyses would indicate statistically significant differences in the variable between times. Quantitative Analysis for Non Paramet ric Data Descriptive analysis was also performed on variables for mean, range, and standard deviation. Statistical analysis was performed to determine the effects of physical therapy and a healthy lifestyle/weight loss program intervention over time on dep endent variables HRQL, Attitudinal Ambivalence, and Behavioral Intentions at Time 1 (baseline), Time 2 (4 weeks), and T ime 3 (8 weeks). A Friedman test was performed initially to determine significance among variables. Wilcoxon Signed Ranks was performed to determi ne pairwise comparisons with a B onferroni correction factor to p= .017 level. This statistical analysis would determine if differences were found for the variable over ti me. These statistical anal yse s were used to draw inferences on the effects of progression through a healthy lifestyle/weight loss program measured three separate times, separated by 4 week periods on the dependent variables weight loss, HRQL, attitudinal ambivalence, behavio ral intentions and program compliance Correlational Data For research questions 1 to 3 including subset questions analysis to determine associations w as performed using correlational analysis. E xamining correlations among variables weight loss, HRQL, attitudinal ambivalence, behavioral intentions, stage o f change, and progra m compliance was performed using Pearson correlation for
4 7 non parametric data analysis HRQL, Attitudinal Ambivalence, Behavioral Intentions and stage of change. Correlational studies used a two tailed test. Interpretation of strength of correlation coefficient followed Cohen (1988 ) classif ication for effect size; .0 to .09 as no effect; .10 to .23 as a small effect size ; .24 to .36 as a medium effect s ize ; and .37 to 1.00 as a large effect size Significance of the Study The obesity epidemic has amplified the urgency of improving interventions that will curtail this ongoing crisis. Once we understand the critical relationships between ambivalence to lif estyle changes and progression through a lifestyle/weight loss program, healthcare programs will insist on altering their approach to changing behaviors. By understanding causal behaviors that p erpetuate obesity in a person we can develop new ways to inte rvene, and modify current interventions. Then, progress may be realized. My study to understand relationships among HRQL, weight loss, ambivalence and behavioral intentions, so predictions in behavior may allow for optimal intervention at key times when p atients may lapse back into old behaviors. Understanding where a person is in the stages of change provides important information that allows health behaviorists a way to predict and intervene through motivational in terviewing (West et al. 2007) and suppo rtive counseling that may improve long term success in terms of health behavior change
48 Table 3 1. Description of the instruments and variable characteristics I nstrument Variable Score SF 36 HRQL Physical component score Mental component score Sum of 4 domains Sum of 4 domains Attitudinal ambivalence Weight loss Food diary Daily exercise Average of attitudes, thoughts and beliefs, and feelings from both positive and negative 0 3 point scale Behavioral intentions Weight loss Food Diary Daily Exercise Average of intent and plan from 3 to +3 bipolar scale Program compliance Food choice exercise Sum of points from 7 areas of compliance Stage of Change Stage Stage of change 1 to 5 Medical scale Body weight Body weight in pounds Continuous measu res
49 Initial Assessment Figure 3 1. H ow factors/variables can affect outcomes: Health Related Quality of Life (HRQL) and Body wei ght/BMI (Weight Loss) during a Healthy Lifestyle/Weight Loss progra m Change in HRQL Body Weight/BMI Change in HRQL Body Weight/BMI Change in HRQL Body Weight/BMI Ambivalence Stage of Change Behavioral Intention Ambivalence Stage of Change Behavioral Intention 4 week Assessment Va riables affecting Outcomes of HRQL & Body weight/BMI Ambivalence Stage of Change Behavioral Intention 8 week Assessment Cognitive & Behavioral Dietary/Exercise Program Cognitive & Behavioral Dietary/Exercise Program Intervention Age Gender Socioeconomic Status Race
50 Positive Rating Negative Rating Figure 3 2 Attitudinal a mbivalence 0 1 2 3 0 0 .5 1.0 1.5 1 .5 1 .5 0 2 1.0 .5 2 1.5 3 1.5 0 1.5 3
51 CHAPTER 4 RESULTS My study examined associations among weight loss, health related qualit y of life, ambivalence, stage of change and behavioral intentions as patients progress ed through sta ges of an 8 week healthy lifestyle/weight loss program. Data collected during my study are presented in this chapter. The data demonstrate changes in these variables that occur over an 8 week period Other factors that may influence these variables are also examined, including age, race, socioeconomic status, and program compliance. Participant Demographics Participants were referred to a physical therapy cl inic in Gainesville, Florida. Participants were referred for physical therapy and a healthy lifestyle/weight loss program by their physicians with a primary orthopedic diagnosis and a secondary diagnosis of overweight/ obesity. All participants were from t he area within and surrounding Gainesville, Florida, a community within Alachua county which is located in North Florida. healthy lifestyle/w eight loss program. All participan ts completed the usual medical intake assessment for physical therapy. P articipants involved in the study completed an IRB approved informed consent and completed the baseline assessment. There were 112 participants in the study. Nineteen patients dropped out of the program for the following reasons: 7 for medical and/or health related problems 4 for insurance /financial problems and 8 for personal or family problems Two were inco mplete and could not be used.
52 P articipants ranged from 18 to 83 years of age. The mean age was 55.6 years: 26 3 7 % of patients were 50 to 59; 26.37 % were 60 69 16.48% were 4 0 to 49; 13.18 % were 7 0 to 7 9 ; and 3.29 and 2.19% were 18 to 29 and 80 to 89 respectively There were 71 female and 20 male pa rticipants: 8.79 % were African American, 2 % were hispanic and 89% were Caucasion. Table 4 1 shows demographic data and d escriptive statistics with categories for body mass index. This data demonstrate s that at time 1, 15.38 % of patients were in the overweight category ( 25 to 29.9 ) ; 37.26 % of patients were in the obesity class I category ( 30 to 34.9 ) ; 24.17 % were in the class II obesity category ( 35 to 39.9 ) ; and 24.27 % were in the class III obesity category ( >40 ) P rimary insurance providers for patients were Blue Cross and Medicare at 47.25 and 30.76% respectively. The two prevailing diagnoses were low back and knee at 42.85 and 30.76%, respectively Table 4 7 shows socioeconomic factors (Population Studies Center ) and weight loss. Those in to 59,999 category had a 4.14 % reduction in body weight compared to 4.58 % weight loss for those in the $60,000 to Body Weight M ean body weigh t progressively decreased from Time 1 through T ime 3, (Table 4 2 ) C hanges in mean body weight using a Repeated Measures Analysis of Variance with a pairwise comparison using a Bonferroni correction supported significant differences in body weight over time (p=.0005, F=184.18, df=1.349) (Table 4 3) Pairwise comparisons for T ime s 1 to 2, p =.0005, mean difference= 6.84; T ime s 2 to 3, p=.0005, mean difference 3.21; T ime s 1 to 3, p= .0005, mean difference=10.05 pounds. M ean weight loss during the eight week period was 10.05 6.63 pounds with a range of 34.2 to 4.5 pounds. M ean weight loss from baseline to 4 weeks was 6.84 .38 pounds ;
53 and f rom 4 weeks to 8 weeks it was 3.21 3.86 pounds (Table 4 2 ) M ean percent change in weight was 4.42% from baseline to 8 weeks; 2.98% from baseline to 4 weeks ; and 1.49% from 4 weeks to 8 weeks. A Repeated Measures Analysis of Variance supported significant changes in percent body weight among times (p=.0005, F=131.112, df= 1.835 ). A pairwise comparison using a Bon ferroni correction for time s 1 to 2 p=.0005, mean difference .015 percent; Time 2 to 3, p=.0005, mea n difference .029 percent, and T ime s 1 to 3, p=.0005, mean difference .014 % Table 4 4 shows weight loss data. H ealth Related Quality of Life M ean scores for Health Related Quality of Life for the physical component score ( PCS ) and mental c omponent score (M CS ) progressively in creased from Time 1 through Time 3 (Table 4 2 ) Changes in mean scores for Health Related Quality of Life for physical component score ( PCS ) using a Friedman test demonstrated significant differences in scores, (p=.0005 c hi square = 86.89 df= 2) (Table 4 3). This was followed by Wilcoxon Sign Ranks for pairwise comparisons with a Bonferroni correction requiring p=.017 ; (T ime s 2 to 1, p=.0005, Z= 6.762; T ime s 3 to 2, p=.0005, Z= 4.692; T ime s 3 to 1, p=.0005, Z= 7. 773) ( Table 4 3 ) Changes in mean scores for Health Related Quality of Life for M CS using a Friedman test demonstrated significant differences in scores (p=.0005, F=, chi square=26.643, df=2 ) followed by Wilcoxon Sign Ranks for pairwise comparisons with a Bonfe rroni correction requiring p=.017 ; (T ime s 2 to 1, p=.0005, Z= 4.579 ; T ime s 3 to 2, p=.0 21 Z= 2.306 ; T ime s 3 to 1, p=.0005, Z= -4.706 ) ( Table 4 4) Stage of Change There was a statistically significant change in stage increasing from T ime 1 to T ime 2, and from Time 1 to Time 3 for both exercise and food choices. Stage of change
54 (Table 4 3) related to exercise using a Friedman test demonstrated significant differences in ratings among times (p=.0005, chi square=87.085, df=2) followed by Wilcoxon Sign Ra nks for pairwise comparisons with a Bonferroni correction requiring p=.017; (T ime 2 to 1 p=.0005, Z= 6.673; T ime 3 to 2, p=.537 Z= .617 ; T ime 3 to 1, p=.0005, Z= 6.257 ). The mean change in stage of change related to food choices using a Friedman test demonstrated significant differences in scores, (p=.0005, chi square= 114.902 df=2) followed by post hoc pairwise comparisons with a Bonferroni correction requiring p=. 017 ; (time 2 to 1, p=.0005, Z= 7.313 ; time 3 to 2, p= .243, Z= 1.166 ; time 3 to 1, p= .0005, Z= 7. 369) (Table 4 4). Attitudinal Ambivalence Ambivalence related to weight loss, daily exercise and food diary was calculated to assess for significant changes over time. M ean scores for ambivalence progressively decreased for weight loss, da ily exercise, and food diary from Time 1 through Time 3 (Table 4 2 ) Mean changes in ambivalence related to weight loss (WL), daily exercise ( DE ) and food diary (FD) using a Friedman test demonstrated significant differences in scores for WL, (p=.0005, chi square = 21.342 df = 2 ) followed by a Wilcoxon Sign Ranks for pairwise comparisons with a Bonferroni correction requiring p=.017; (Time 2 to 1, p=.0 59 Z= 1.891 ; T ime 3 to 2, p=.003 Z= 3.006, Time 3 to 1, p=.0005, Z= 5.956 ) ; DE ( p=.0005, chi square = 33.174 df=2 ) (Table 4.4), followed by pairwise comparisons; (time 2 to 1, p=.0005, Z= 3.801 ; T ime 3 to 2, p=.003 Z= 2.953 ; Time 3 to 1, p=.0005, Z= 5.412 ) and for FD, ( p=.0005, chi square = 41.282 df=2 ) followed by pairwise comparisons; (Time 2 to 1 p=.0005, Z= 4.152 ; T ime 3 to 2, p=.208 Z= 1.258 ; time 3 to 1, p=.0005, Z= 5.225 ) (Table 4. 4 ) Findings supported differences in ambivalence at all levels with exception of Time 3 to 2 for food diary and Time 1 to 2 for
55 weight loss When comparing di f ferences among weight loss, daily exercise, and food diar y, a difference was noted between ambivalence for weight loss and d aily exercise for T ime 1, 2, and 3, (p= .011, z= 2.550; p=002, z= 3.088; p= .0 01, z= 3.0463) and between ambivalence for weight l oss and food diary for T ime 3 (p= .012, z= 2.515) (Table 4 5) Behavioral Intentions M ean scores for b ehavioral intentions for weight daily exercise, and food d iary from Time 1 through T ime 3 are given in Table 4 2. Mean changes in behavioral intentions r elated to levels of weight loss (WL), daily exercise (DE) and food diary (FD) using a Friedman test demonstrated no statistically significa nt differences in scores for WL (p= .944, chi square= .115, df= 2 ), for DE (p= .447, chi square=1.610, df=2), and for FD (p=.113, chi square= 4.360, df= 2) ( Table 4 3) Additionally, when comparing differences among weight loss daily exercise and food diary A notable difference was found among behavioral intentions for weight loss and daily exercise for T ime s 1, 2, and 3, ( p= .0005, z= 4.591 ; p=0005 z= 3. 990 ; p= .00 4 z= 2.880 ) and among behavioral intentions for weight loss and food diary for T ime s 1, 2, and 3, ( p= .0005, z= 4.084; p= .00 0 5, z= 3.522; p= .0005, z= 3.609 ) respectively (Table 4 6) Program Compliance Program compliance (Table 4 3) was recorded in 2 week int ervals over the 8 week period represented by T imes 1 to 4. Program compliance was greatest from 2 to 4 weeks and 4 to 6 weeks. Repeated Measures Analysis of Variance with a pairwise comparison u sin g a Bonferroni correction supported significant differences in compliance (p=.008, F=4.484.18, df=2.385). Post hoc p airwise comparisons for T ime 1 to 2 (p=.012 ); and T ime 1 to 3 (p=.006 ) were statistically significant Pairwise
56 comparisons were not signifi cantly different for time s ( Times 2 to 3 (p=1.000), Times 1 to 4 ( p= .690 ), Times 2 to 4 ( p= 1.000 0 or T ime s 3 to 4 ( p= .450 ) (Table 4.4) S ocioecon omic category is shown in ( Table 4 7 ) P arti cipants in the socioeconomic category demon strated a m ean weight loss of 5.20 1.66 percent change compared to 3.91, 2.88 and 4.58, 2.54 for categories $40,000 to 59,999 and $60,000 to respectively The following research questions are listed below followed by an itemized breakdown of each question for analysis. Research Question s 1. Are the improvements in HRQL proportional to the progression in weight loss and are there associations between weight loss and HRQL, ambivalence, and behavioral intentions, which can be identifi ed? 2. Are lapses in behavior an effect of improvement in the HRQL, that facilitates a deterioration of health behaviors and health status? 3. Does HRQL affect behavioral intent ions and ambivalence as one progresses through a healthy lifestyle/weight loss program? In order to answer the research questions the following areas must be separately identified and analyzed. Are the improvements in HRQL proportional to the progression in weight loss and can assoc iations among weight loss, HRQL, ambivalence, and behavioral intentions, which can be identified? The relationship between weight loss and HRQL is first addressed by examining percent change in body weight and percent change in the physical and mental component score s of the HQRL of the SF 36. This is followed by identifying relationships among HRQL, ambivalence and behavioral intentions. Finally
57 relationship s among categories within the ambivalence and behavioral intentions (f ood diary, weight loss and daily exercise) will help identify meaningful data related to behavioral changes. I temized questions below are followed by results. Is there a correlation between weight loss and health related quality of life (SF 36)? Statistic al analysis used to determine correlation s We investigated potential associations of weight loss using percent change in weight and percent change in the physical component score (PCS) and t he mental component score (MCS) of the HRQL. Perce nt change in weight and PCS for Time 1 to 2 demonstrated a weak correlation (r= 232 p= 027 ) but n o significant correlations for Time 2 to 3 (r= .09 2 p= .383) or Time 1 to 3 (r= .107, p= .31 1 ). The correlation between weight loss and the mental compo nent score (MCS) was not statistically significant for Time 1 to 2 (r= .002, p= .988) Time 2 to 3 (r= .037, p= .725) or Time 1 to 3 (r= .173, p= .101) (Table 4 8) Are there associations between behavior plans/intentions and weight loss? We analyzed a m ean fo r planned and inte nded behavior for each category, behavior related to weight loss (WL), behavior related to daily exercise (DE) and behavior related to behavior and perce nt weight loss. There were no significant correlations between beha vior related to weight loss at T ime 1 and pe rcent change in weight between Time 1 to Time 2 (r= .129, p= .2 25). A weak correlation was found in behavior at T ime 2 and percent change in wei ght from time 2 to Time 3 (r= .219, p= .037), but no correlation was found for behavior at time 3 and percent weight loss from Time 1 to Time 3 (r=
58 .029, p= .78). Table 4 9 shows percent change in body weight and mean behavior for weight loss, food diary and daily exercise. Behavio r related to daily exercise at T ime 1 and percent change in weight from Time 1 to Time 2 showed no correlations (r= .036, p=.735) between T ime 2 to 3 for weight and T ime 2 for behavior ( r= .187, p= .076), or between Time 1 to 3 for weight and T ime 3 for behavior (r= .098, p= .355). Beh avior related to food diary at T ime 1 and pe rcent change in weight between T ime s 1 and 2 showed no correlations (r= .040, p=.707), from Time 2 to Time 3 for weight and T ime 2 for behavior ( r= .069, p= .516), or between Time 1 to 3 for weight and T ime 3 for behavior (r= .064, p= .546). Are there associations between ambivalence and change in body weight? Table 4 10 shows correlations between percent change in body weight and mean ambivalence for categories WL, FD and DE. For each category a mean was taken for subsets: thoughts/beliefs, feelings, and attitudes related to ambivalence. An average of the subsets for each category (ambi valence related to WL ambivalence related to DE and ambiva lence related to FD analyze correlations between behavior and percent weight loss. N o significant correlations were found for mean ambivalence related to weight loss at time 1 and percent change in weight fr om Time 1 to Time 2 (r= .012, p= .913), or between mean ambivalence at time 2 and percent change in weight from Time 2 to Time 3 (r= .178, p= .091), however a weak correlation for mean ambivalence at T ime 3 and percent weight loss between Time 1 and Time 3 (r= .258, p= .013). Table 4 9 shows percent change in body weight and mean ambivalence for weight loss, food diary, and daily exercise.
59 Ambivalence related to daily exercise a t T ime 1 and pe rcent change in weight from Time 1 to Time 2 showed no correlati ons (r= .015, p=.889), or between Time 2 to Time 3 for weight and T ime 2 for mean ambivalence ( r= .093, p= .378), or between Time 1 to Time 3 for weight and time 3 for mean ambivalence (r= .163, p= .124). Ambiva lence related to food diary at T ime 1 and pe rcent change in weight betwe en T ime 1 to Time 2 demonstrated no correlations (r= .057, p=.593), or between Time 2 to Time 3 for weight and T ime 2 for mean ambivalence (r= .053, p= .619), or between Time 1 to Time 3 for we ight and T ime 3 for mean ambivalen ce (r= .104, p= .326). Are there significant differences between categories within ambivalence and behavioral intentions? Table 4 5 and 4 6 demonstrates differences found between categories for ambivalence and behavior intentions. For T ime 1 a mbivalence r elated to w eight loss, d aily e xercise, and f ood d iary demonstrated a significant difference between weight loss and daily exercise ( p= .0005 z= 4.558 ) and between weight loss and food diary ( p= .001 z= 3,284 ) however no statistically significant diffe rence between food diary and daily exercise ( p= .274, 1.095 ) For T ime 2 a mbivalence r elated to w eight loss, d aily e xercise, and f ood d iary demonstrated a significant difference between weight loss and daily exercise ( p= .002, z= 3.088 ) however no signi ficant difference between weight loss and food diary ( p= 467 z= .727 ) or food diary and daily exercise ( p= 018 2.359 ) For T ime 3 a mbivalence r elated to w eight loss, d aily e xercise, and f ood d iary demonstrated a significant difference between weight loss and daily exercise ( p= .00 1 z= 3.463 ) and between weight loss and food diary ( p= 012 z= 2.515 ) however no significant difference between food diary and daily exercise ( p= 385 .870 )
60 Categories within behavioral intentions demonstrated simila r differences (Table 4 6) For T ime 1 behavioral intention related to weight loss, daily exercise, and food diary demonstrated a significant difference between weight loss and daily exercise ( p= .0005, z= 4.591) and weight loss and food diary (p= .005, z= 4.084), however no significant difference between food diary and daily ex ercise (p=.871, z= 1.62). For T ime 2 behavior intention r elated to w eight loss, d aily e xercise, and f ood d iary demonstrated a significant difference between weight loss and daily e xercise ( p= .00 05 z= 3.900) and between weight loss and food diary ( p= 0005 z= 3.522) However no significant difference between food diary and daily exercise ( p= 899 .127). For T ime 3 behavior intention r elated to w eight loss, d aily e xercise, and f ood d iary demonstrated a significant difference between weight loss and daily exercise ( p= .00 4, z= 2.880) and betwe en weight loss and food diary ( p= 0005, z= 3.609) however no significant difference between food diary and daily exercise ( p= 098 1 .656). Research Question 2 Are lapses in behavior an eff ect of improvement in the HRQL that facilitates a deterioration of health behaviors and health status? Lapses in be havior are a recognized part of regressing in terms of behavior and health status ( Elfhag & Rossner, 2005 ) Variables order to address this question. Additionally program compliance is a reflection of a lapse in behavior. E xamining potential correlations between program compliance and weight loss along with associations between Health Related Quality of Life and compliance may provide meaningful information related to this question
61 Are there associations between program compliance and change in body weight? Table 4 11 shows corre l ations between program compliance at 2 weeks, 4 weeks, 6 weeks and 8 weeks and percent change in body weight from T ime s 1 to 2, Times 2 to 3, and Times 1 to 3. No correlation was seen between Time 1 to Time 2 for percent change in body weight and compliance from Time 1 to 2 weeks (r= .140, p= .187). There was a moderate negative correlation between Time 1 to Time 2 for percent change in body weight and complianc e from weeks 2 to 4, (r= .301, p= .004). There was a weak negative correlation between percent change in body weight between Time 2 to Time 3 and compliance from 4 to 6 weeks (r= .276, p= .008) and a moderate correlation for compliance from 6 8 weeks (r= .345, p=.001). There was not a significant correlation between percent change in body weight between Time 1 to Time 3 and compliance from 1 to 2 weeks (r= .144, p= .172). However there was a moderate negative correlation between pe rcent change in body w eight at Times 1 to Times 3 and compliance from 2 to 4 weeks (r= .332, p= .001), compliance from 4 to 6 weeks (r= .304, p=.003), and compliance from 6 8 weeks (r= .318, p= .002). There was also a moderate negative correlation between average compliance over 8 weeks and percent weight loss over 8 weeks (r= .313, p= .003). Is there a correlation between program compliance and HRQL? Table 4 12 shows correlations between program compliance total s f rom baseline to 4 weeks, 4 to 8 weeks, and the 8 week avera ge; and the physical and mental component score. There was not a significant correlation between compliance from baseline to 4 weeks, 4 to 8 weeks and the 8 week average ; and the difference in mental component score from time 1 to 2, 2 to 3 and 1 to 3, ( r= .105, p= .320; r= .065, p= .542; r= .069, p= .515).
62 Also there was not a significant correlation between compliance from baseli ne to 4 weeks, 4 to 8 weeks and the 8 week average ; and the difference in the physical component sco re from T ime 1 to 2, Tim e 2 to 3 and Time 1 to 3, (r= .017, p= .872; r= .055, p= .602; r= .019, p= 855). Research Question 3 Does HRQL affect behavioral intentions and ambivalence as one progresses through a healthy lifestyle/weight loss program? To investigate this question associations between behavioral intentions and ambivalence need to be addressed. Then the relationship between mean changes in HRQL for both the MCS and PCS from Time 1 to Time 2 and Time 2 to Time 3 aff ect ambivalence or behavioral intentions needs to be examined. Finally the relationship between ambivalence and behavioral intentions was addressed. Are there associations between ambivalence and behavioral intentions? Table 4 13 shows correlations between ambivalence and behavior No correlations seen be tween ambivalence and behavior related to WL for T imes 1 and 2, (r= .172, p= .103; r= .177, p= .093) respectively, however a weak correlation seen between ambivalence and behavior related to WL for time 3, (r= .202; p= .055). A weak, strong and moderat e negative correlation was seen between ambivalence and behavior related to FD for T imes 1, 2, and 3, (r= .217, p= .039; r= .457, p= .0005; r= .371, p= .0005 ) respectively. A moderate, weak, and strong negative correlation seen between ambivalence and b ehavior related to DE for T imes 1, 2, and 3, (r= .303, p= .003 ; r= .261, p= .013 ; r= .413, p= .0005 ) respectively. Are there associations between Health related quality of life (SF 36) and behavioral intentions? Table 4 14 shows correlations among compo nents of the SF 36
63 Health Related Quality of Life (HRQL), physical component score (PCS) and mental component score (MCS) ; and behavior related to weight loss (WL), food diary (FD), and daily exercise (DE). No correlations were found between the PCS and m ean behavior related to WL for Times 1, 2, or 3 (r= .060, p= .574; r= .023, p= .828; r= .170, p= .107) respectively. There were no correlations seen between PCS and m ean behavior related to FD for Times 1, 2, or 3 (r= .010, p= .927; r= .036, p=.738; r= .0 15; p= .885) respectively. There were no correlations seen between PCS and m ean behavior related to DE for T imes 1, 2, or 3, (r= .034, p= .808; p=.109, p= .303; r=.029, p= .787) respectively. No correlations were found between the MCS and m ean behavior r elated to WL for Times 1, 2, or 3 (r= .060, p= .574; r= .023, p= .828; r= .170, p= .107) respectively. No correlations were seen between MCS and m ean behavior related to FD for times 1, 2, or 3 (r= .010, p= .927; r= .036, p=.738; r= .015; p= .885) respe ctively. No correlations were seen between MCS and m ean behavior related to DE for Times 1, 2, or 3 (r= .034, p= .808; p=.109, p= .303; r=.029, p= .787) respectively. Are there associations between health related quality of life (SF 36) and ambivalence? Table 4 15 shows correlations among components of the SF 36 Health Related Quality of Life (HRQL), physical component score (PCS) and mental component score (MCS) ; and mean ambivalence related to weight loss (WL), food diary (FD), and daily exercise (DE). No correlations were found between the PCS and mean ambivalence rel ated to WL for times 1, 2, or 3 (r = .066, p = .535; r = .112, p= .289; r= .160, p= .131) respectively. A weak negative correlation was seen between PCS and mean ambivalence related to FD for T ime 1 (r = .282, p =.007) however no
64 correlations were seen for Times 2 and 3 (r= .023, p = .830; r= .050, p=.640) respectively. No correlations were seen between PCS and mean ambivalence related to DE for Times 1, 2, or 3 (r= .089, p= .400; r = .160, p=.130; r= .137, p= .195) respectively. No correlations were found between the MCS and mean ambivalence related to WL for Times 1 and (r= .034, p= .748; r= .183, p= .083) respectively; however there was a moderate negative correlation bet ween MCS and mean ambiv alence related to WL for Time 3 (r= .337, p= .001). There were no correlations seen between MCS and mean ambivalence related to FD for Times 1 (r= 084, p= .427); however a moderate negative correlation was seen between MCS and me an ambivalence related to FD for Times 2 and 3 (r= .323, p= .002; r= .356, p= .001) respectively. No correlation was seen between MCS and mean ambivalence rela ted to DE for Time 1 (r= .152. p= .150); however a weak and moderate negative correlation wa s seen between MCS and mean ambivalence related to DE for Times 2 and 3 (r= .262, p= .012; r = .342, p= .001) respectively. Summary These findings suggest that involvement in physical therapy concurrent with a weight loss/ healthy lifestyle program can fa cilitate weight loss as demonstrated by a mean weight loss of 4.42 % over an 8 week period. My study also demonstrated statistically significant increases in the physical and mental status of a patient indicated by a 32.24 % increase in the mean physical c omponent score and a 19.42% increase in the mental component score over an 8 week period. Also demonstrated was a progressive decrease in ambivalence related to weight loss ( WL ) food diary ( FD ) and daily exercise (DE) over the 8 week pr ogram The most si gnificant decrease in
65 ambivalence was seen from baseline to 4 weeks which corresponded to a mean increase in stage of change from C ategory 3 (preparation stage) to C ategory 4 (action stage) Additionally there were statistically significant differences b etween categories of ambivalence for weight loss and daily exercise (Table 4 5) and between categories of behavioral intentions for weight loss and daily exercise as well as for weight loss and food diary (Table 4 6). The mean rating for the category weig ht loss had the lowest level of ambivalence and the highest level of behavioral intention compared to food diary or daily exercise. A ssociation s were found between compliance with food and exercise log s and weight loss demonstrating greater compliance a t increased weight l oss. My study also found negative associations between the MCS and ambiva lence related to th e food diary and daily exercise. This suggest s that as ambivalence decreases MCS increases. Additionally, negative associations were found betw een ambivalence and stage of change ; and between ambivalence and behavioral intentions for food diary and daily exercise. This suggests that as ambivalence decreases there is a resultant increase in stage of change and behavioral intentions for the specif ied areas of food diary and daily exercise. One of the strengths of my study was its experimental design using an actual clinical setting with repeated measures collected longitudinally over time using the same subjects. This effectively reduces the var ianc e of estimates of the treatment effects over time, allowing statistical inference to be made with a smaller quantity of subjects. T his t ype of design allowed for investigation of how subjects respond to stages of change during a lifestyle/weight loss p rogram in a real life setting, without being
66 encumbered by potential bias that may occur when the experimental environment has imposed restri ction s.
67 Table 4 1. Population demographics Demographic Frequency = n Percent Male 20 22.0 Female 71 78.0 Age, mean SD (range) 55.6 y 14 (18 83) Age Groups (yrs) 18 29 3 3.29 30 39 11 12.09 40 49 15 16.48 50 59 24 26.37 60 69 24 26.37 70 79 12 13.18 80 89 2 2.19 Ethnicity Caucasian 81 89.01 African American 8 8.79 Hispani c 2 2.19 Annual Household Income (dollars) Mean 64,832 Median 54,607 Range 28,039 to 78,873 Annual Household Income (dollars) 6 6.97 40,000 to 59,999 28 32.55 52 60.46 Obesity classif ication BMI Overweight 25 to 29.9 14 15.38 Obesity Class I 30 to 34.9 33 36.26 Obesity Cass II 35 to 39.9 22 24.17 22 24.17 Insurance Medicare 28 30.76 Blue Cross 43 47.25
68 Table 4 1. Continued Demographic Frequ ency = n Percent Insurance Medicare 28 30.76 Blue Cross 43 47.25 Avmed 7 7.69 UHC 3 3.29 Cigna` 2 2.19 Tricare 1 1.09 Champ 1 1.09 Aetna 2 2.19 Mail Handler 1 1.09 Workers Compensation 1 1.09 Vocational Reh ab. 1 1.09 BCHO 1 1.09 Diagnosis Back pain 39 42.85 Knee pain 28 30.76 Shoulder pain 3 3.29 Neck pain 3 3.29 Ankle pain 3 3.29 Muscle weakness 3 3.29 Hip/thigh pain 3 3.29 Joint pain 2 2.19 ITB syndrome 1 1.09 Fibromyalgia 1 1.09 Achilles tendonosis 1 1.09 Abdominal strain 1 1.09 Myofascial pain 1 1.09 Pelvic arthritis 1 1.09
69 Table 4 2. Means by variable and time of measurement Physical and psychosocial variables T ime 1 M ean SD Time 2 M ean SD Time 3 mean SD Body weight 2225.2 44.9 218.4 42.9 215.1 42.9 SF 36 HRQL PCS 35.71 10.43 41.60 9.76 44.98 9. 91 MCS 47.38 12.97 52.13 10.16 53.45 10.97 Ambivalence W eight loss .676 .637 .791 .759 1.034 .547 Food diary .353 .798 .730 .777 .859 .672 Daily exercise .238 .857 .558 .676 .782 .760 Behavioral intention Weight loss 2.918 .291 2.918 .318 2.852 .535 Food diary 2.588 .728 2.676 .664 2.462 .981 Daily exercise 2.577 .702 2.659 .686 2.604 .892 Stage of Change Exercise 2.97 .936 3.80 .670 3.7 7 .747 Food choices 2.93 .800 3.86 .569 3.72 .500
70 Table 4 3. Interactions with time F df p Partial eta Squared Power Body Weight 184.18 1.349 .0005 .672 1.000 Percent Change In Body weight 131.112 1 .835 .0005 .593 1.000 Compliance 4.484 2.385 .008 .047 .815 Chi Square df Two tailed Probability PCS of HRQL 86.893 2 .0005 MCS of HRQL 26.643 2 .0005 Ambivalence Weight Loss 21.342 2 .0005 Food Diary 41.282 2 .0005 Daily Exercise 33.1 74 2 .0005 Behavioral Intentions Weight Loss .115 2 .944 Food Diary 4.360 2 .113 Daily Exercise 1.610 2 .447 Stage of Change Exercise .085 2 .0005 Food Choices 114.902 2 .0005
71 Table 4 4. Pairwise Compa risons over time Behavioral Variable T T Mean diff. p value Lower Upper Body Weight 1 2 6.840 .0005 5.179 7.962 2 3 3.205 .0005 2.219 4.192 1 3 10.046 .0005 8.351 11.741 Percent ch an ge in body w t 1 2 .015 .0005 .020 .010 2 3 .029 .0005 .025 .033 1 3 .014 .0005 .010 .018 Compliance 1 2 5.000 .012 9.248 .752 1 3 6.396 .006 11.434 1.357 1 4 3.648 .690 9.834 2.537 2 3 1.396 1.000 6.054 3.263 2 4 1.352 1.000 3.810 6.513 3 4 2.747 .450 1.368 6.862 T T Z Two tailed Prob. PCS of HRQL 2 1 6.762 .0005 3 2 4.692 .0005 3 1 7.773 .0005 MCS of HRQL 1 2 4.579 .0005 3 2 2.306 .021 1 3 4.706 .0005 Ambivalence Weight loss 2 1 1.891 .059 3 2 3.006 .003 1 3 5.956 .0005 Daily exercise 2 1 3.801 .0005 3 2 2.953 .003 1 3 5.412 .0005 Food diary 2 1 4.152 .0005 3 2 1.258 .208 1 3 5.225 .0005 Stage of change Exercise 2 1 6.673 .0005 3 2 .617 .537 1 3 6.257 .0005 Food choices 2 1 7.313 .0005 3 2 1.166 .243 1 3 7.369 .0005
72 Table 4 5 Pairwise c omparisons for a mbivalence r elated to (weight loss), WL daily exercise ( DE ) and food diary ( FD ) Time Category Z Two tailed Pr ob. Time 1 DE WL 4.558 .0005 FD DE 1.095 .274 WL FD 3.284 .001 Time 2 DE WL 3.088 .002 FD DE 2.359 .018 WL FD .727 .467 Time 3 DE WL 3.463 .001 FD DE .870 .385 WL FD .2.515 .012
73 Table 4 6. Pair wise comparisons for behavioral intentions related to WL, DE, and FD Time Category Z Two tailed Prob. Time 1 DE WL 4.591 .0005 FD DE .162 .871 WL FD 4.084 .0005 Time 2 DE WL 3.990 .0005 FD DE .127 .899 WL FD 3.522 .0005 Ti me 3 DE WL 2.880 .004 FD DE 1.656 .098 WL FD 3.609 .0005
74 Table 4 7. Mean weight loss by socioeconomic Socioeconomic Category N Mean Percent WL SD Min Max 39,999 6 5.20 1.66 3.46 7.56 40,000 59,999 28 3.91 2.88 1.59 11.14 59,999* 34 4.14 2.73 1.59 11.14 60,000 52 4.58 2.54 1.50 9.39 *Categories 1 & 2 combined
75 Table 4 8. Correlation between percent change in weight and percent change in physical comp onent score (PCS) and mental component score (MCS) Category Time Correlation coefficient Two tailed probability 1 2 .232 .027 PCS 2 3 .092 .383 1 3 .107 .311 1 2 .002 .988 MCS 2 3 .037 .725 1 3 .173 .101
76 Table 4 9. Correlation between percent change in weight loss (WL) and behavioral intention (BI) WL/BI Correlation coefficient Two tailed probability Weight loss Time 1 2/1 .129 .225 Time 1 2/2 .074 .484 Time 2 3/2 .219 .037 Time 2 3/3 .129 222 Time 1 3/1 .243 .020 Time 1 3/2 .213 .042 Time 1 3/3 .029 .788 Food diary Time 1 2/1 .040 .707 Time 1 2/2 .043 .685 Time 2 3/2 .069 .516 Time 2 3/3 .012 .913 Time 1 3/1 .037 .728 Time 1 3/2 .038 .719 Time 1 3/3 .064 .546 Daily exercise Time 1 2/1 .036 .735 Time 1 2/2 .053 .615 Time 2 3/2 .187 .076 Time 2 3/3 .090 .394 Time 1 3/1 .098 .355 Time 1 3/2 .086 .420 Time 1 3/3 .098 .355
77 Table 4 10. Correlation between percent change in weight loss (WL) and ambivalence (AMB) WL/AMB Correlation coefficient Two tailed probability Weight loss Time 1 2/1 .012 .913 Time 1 2/2 .059 .576 Time 2 3/2 .178 .091 Time 2 3/3 .285 .006 Time 1 3/1 .107 .315 Time 1 3/2 .161 .127 Time 1 3/3 .258 .013 Food diary Time 1 2/1 .057 .593 Time 1 2/2 .040 .704 Time 2 3/ 2 .053 .619 Time 2 3/3 .100 .346 Time 1 3/1 .054 .613 Time 1 3/2 .012 .911 Time 1 3/3 .104 .326 Daily exercise Time 1 2/1 .015 .889 Time 1 2/2 .083 .435 Time 2 3/2 .093 .378 Time 2 3/3 .256 .014 Time 1 3/1 .097 .359 Time 1 3/2 .081 .446 Time 1 3/3 .163 .124
78 Table 4 11. Correlations between percent change in weight loss (WL) and compliance (C OMP ) Time: WL/C OMP Correlation coefficient Two tailed probability 0 4/1 2 .140 .187 0 4/2 4 .301 .004 4 8/4 6 .276 .008 4 8/6 8 .345 .001 0 8/1 2 .144 .172 0 8/2 4 .332 .001 0 8/4 6 .304 .003 0 8/6 8 .318 .002
79 Table 4 12 Correlations between health related quality of life ( HRQL ) and compliance (COMP ) Time: HRQL/ C OMP Correlation coefficient Two tailed probability PCS 1 2 .017 .872 2 3 .055 .602 1 3 .019 .855 MCS 1 2 .105 .320 2 3 .0 65 .542 1 3 .069 .515
80 Table 4 13. Correlations between ambivalence (AMB) and behavior intentions (BI) for food diary (FD) and daily exercise (DE) AMB/BI Correlation coefficient Two tailed probability Food diary Time 1/1 .217 .039 Time 2/2 457 .0005 Time 3/3 .371 .0005 Daily exercise Time 1/1 .303 .003 Time 2/2 .261 .013 Time 3/3 .413 .0005
81 Table 4 14 Correlation between health related quality of life (HRQL) and behavioral i ntentions (BI) HRQL/BI Correlation coeffici ent Two tailed probability PCS Time 1/1 .060 .574 Weight Loss Time 2/2 .023 .828 Time 3/3 .170 .107 Food Diary Time 1/1 .010 .927 Time 2/2 .036 .738 Time 3/3 .015 .885 Daily Exercise Time 1/1 .034 .808 Time 2/2 .109 .303 Time 3/3 .02 9 .787 MCS Time 1/1 .111 .294 Weight Loss Time 2/2 .273 .009 Time 3/3 .215 .041 Food Diary Time 1/1 .041 .701 Time 2/2 .220 .036 Time 3/3 .098 .356 Daily Exercise Time 1/1 .127 .229 Time 2/2 .074 .483 Time 3/3 .168 .112
82 Table 4 15 Correlations between health related quality of life (HRQL) and ambivalence (AMB) HRQL/AMBFD Correlation coefficient Two tailed probability MCS Time 1/1 .034 .748 Weight loss Time 2/2 .183 .083 Time 3/3 .337 .001 Food d iary Time 1/1 .084 .427 Time 2/2 .323 .002 Time 3/3 .356 .001 Daily exercise Time 1/1 .152 .150 Time 2/2 .262 .012 Time 3/3 .342 .001 PCS Time 1/1 .066 .535 Weight loss Time 2/2 .112 .289 Time 3/3 .160 .131 Food d iary Time 1/1 .282 .007 Time 2/2 .023 .830 Time 3/3 .050 .640 Daily exercise Time 1/1 .089 .400 Time 2/2 .160 .130 Time 3/3 .137 .195
83 CHAPTER 5 DISCUSSION AND SUMMARY, C ONCLUSIONS, AND RECOMMENDATIONS There are many obesity related di sease. The food and exercise challenged environment genetic predisposition toward calorie dense foods ( high sugar and fat ) ( Johnson and Kenny, 2010; Berthoud, Lenard, and Shin, 2011 ; Alsio J, Olszews ki, Levine, and Schioth, 2012 ), create hardships for in dividuals trying to reach a healthy body weight My study assessed weight loss behaviors of participants enrolled in a healthy lifestyle/weight loss program. Program participants demonstrated a mea n weight loss of 4.42% ( 10 .05 pound s in 8 weeks ) a rate o f 1.25 pounds per we ek. This amount of weight loss is within suggested guidelines for adults ( Wadden, Foster and Letizia, 1994 ) (Figure 5 1) Obesity reduces quality of life in the areas of physical function and vitality and also causes bodily pain as we ll as reduced emotional health ( Ul Haq, Mack ay Fenwick, & Pell 2013). My study demonstrated that quality of life improv e ments can be obtained via participation in an 8 we ek healthy lifestyle/weight loss program when combined with physical therapy with a 28% reduction in pain, 22% improvement in physical functioning and a 33% increase in energy levels The se finding s follow ed a dose response demonstrating a negative association between mean body weight and quality of life This finding confirms previous r esearch associating the benefits of weight loss with physical therapy demonstrating a 30% reduction in pain an d a 24% improvement in physical functioning with a 5 % reduction in body weig ht ( Messi er et al. 2004; Christianson et al. 2005 ). P articipants in m y study showed a 32.4% improvement in the physical component score (PCS) and 19.32% improvement in the mental component score
84 (MCS) of the SF 36 health related quality of life (HRQL) (Table 5 1 ). Previous research ( Paans, van den Akker Scheek, Dilling, Bos van der Meer Bulstra, et al. 2013 ) examin ing the effects of weight loss and exercise on people with hip osteoarthritis demonstrated similar improvement s in physical function ( 32.6% ) In my study a s participant mean body weight decreased from T1 to T3, a corresponding incremental increase in the mean PCS and MCS occurred (Figure 5 1 and 5 2 ). H owever no correlation s were seen between percent changes in body weight an d percent changes in PCS or MCS within subjects. This suggests that other factors may influenc e participant responses in these domains including intervening factors relate d to diagnosi s For example, an individual with knee pain an d underlying osteoarthritis will respond differently to weight loss in terms of reduced pain and improved function than persons with osteoarthritis with low back or hip pain While the literature supports a reduction in pain along with improved function with weight loss for those with hip and back pain ( Lidar, Behrbalk, Regev, 2012; Paans, van den Akker Scheek, & Dilling, 2013 ) notable reduction in symptoms specific to knee osteoarthritis ( Gudbergsen, Boesen, Lohmander, 2012; Riddle & Stratford, 2013) are shown in the literature with weight loss versus weight gain This suggest s that with differ ent diagnoses responses to weight loss vary regard ing to the physical and mental component scores of the SF 36 HRQL. Additionally in terms of perceived pain improvement as result of weight loss, diagnoses such as shoulder and neck pain, fibromyalgia, a nkle/foot pain, and rheumatoid arthritis are likely to differ in their response to weight loss in terms of perceived improvement in the PCS and MCS.
85 Participant mean s tage of change score was 2.97 and 2.93 for exercise and food choices respectively at baseline These scores are indicative of S tage 3, when individuals begin making preparations to change The mean scores suggest that most of the participants were preparing to make changes and had taken small st eps in the areas of food choice and physical activity at baseline (Time 1) Since a ll patients were aware of their physician referral to physical therapy for weight loss it is plausible t hat individuals had started thinking abou t a lifestyle change in regarding weigh t loss by the time of the initial meeting. Therefore, it is likely these small changes are reflected in the baseline assessment scores for stage of change (Zimmerman, Olsen, Bosworth, 2000). During the T2 assessment the mean stage of change score for the group was 3.80 for exercise and 3.86 for food choices, indicating that most patients had entered the action stage (Figure 5 3) The mean score change was predictable, because patients were now monitoring their daily exercise and food choices with the food and exercise log to confirm active involvement ( Armitage, Sheeran, Conner & Arden, 2004). Movement from the preparation stage to the action stage represents significant increase from the baseline stage of change rating regarding both exercise and food choices. At 8 weeks progressio n through the action stage was not statistically significant (3.77 and 3.92) for exercise (3.77) and food choices (3.92). This lack of progression is to be expected since the action is characterized by evident and quantifiable behavioral changes which take within a 6 month time frame Therefore by definition, stage progression by week 8 was not expected. Work by Armitage and Arden (2007) demonstrates a linear relationship between stage of change and behavior intentions. However, they describe the relation ship
86 between stage of change and ambivalence as quadratic. As their work predicts findings from my study demonstrate a negative association between SOC and a mbivalence in relation to daily exercise at T1 (r = .352, p = .001), T2 ( r = .226, p = .031 ), a nd T3 ( r = .434, p = .0005) with corresponding medium, small and large effect size s for the associations On the other hand, the association between SOC and behavioral intentions related to exercise was positive with large and medium effect sizes at T2 ( r = .453, p = .0005 ) and T3 ( r = 3.57, p = .001) (Figure 5.3) Ambivalence ( which is two equally opposing attitudes about a construct ) incrementally decreased from baseline to 8 weeks for each cat egor y: weight loss, daily exercise and food diary This c hange was demonstrated by a statistically significant reduction in ambivalence for daily exercise from T1 to T2, T2 to T3; and from T1 to T3 Ambivalence related to weight loss also demonstrated a statistically significant reduction from T2 to T3 and from T1 to T3. A statistically significant reduction was absent from T1 to T2. Also present was a statistically significant difference in ambivalence toward maintaining a food diary from T1 to T2 and from T1 to T3. Once again, a statistically significant dec rease was absent from T2 to T3. M edium and large effect sizes were present for ambivalence related to weight loss and maintaining a food diary, weight loss and daily exercise and daily exercise and a food diary Findings relating to ambivalence and s tage of change correspond to those of Armitage & Arden (2007), in which ambivalence was lowest in pre contemplation and the maintenance stage, highest in the contemplation and preparation stage and progressing downward in the action stage. Specifically, the m ean stage of change score at ba seline was 2.97 for daily exercise and 2.93 for food choices (those scores
87 correspond to the stage ) In all categories, mean ambivalence scores were higher at T1 than at T2. Mean s tage of change at T2 for partic ipants was 3.80 and 3.86 ( which places them in the action stage ) Lower ambivalence scores are predictive of subsequent behavioral intenti o ns and behavior (Armitage and Connor, 2000). Regarding maintaining a food diary mean a mbivalence scores did not dec rease significantly from T2 to T3 This finding may i ndicate difficulties participants face when making good food choice s during later stage s of a weight loss program As such, ambivalence toward maintaining the food diary is likely reflected in the decrea sed mean compliance scores for food choice log maintenance present from T2 to T3 (Figure 5 4 ) A mbivalence was highest for daily exercise followed by food diary maintenance ; and lowest for weight loss at T 1, T 2, and T 3 Also present was a statistically si gnificant differenc e for ambivalence toward weight loss and daily exercise at T 1, T 2 and T 3 ; and for weight loss and food diary maintenance T 1 and T 3 ( F igure 5 5 ) Th ese findings suggest g reater ambivalence among weight loss program participants toward dai ly exercise and a food diary, but less ambivalence toward weight loss. This finding may individuals Behavioral intentions are a valid indicator for measuring the degree to which individuals are will ing to expend the effort required for participation in particular behavior s ( Armitage & Conner, 1991 ). For study purposes b ehavioral intention was determined by summed responses for the subcategories: plans and intentions regarding wei ght loss, daily exercise and maintaining a food diary. M ean b aseline scores were as
88 follows: weight loss 2.92, maintaining a food diary 2.59 and daily exercise 2.58. Mean behavioral intention scores for maintaining a food diary and participating in dail y exercise demonstrated an upward trend from T1 to T2 ( 2.68 and 2.60) followed by a slight reduction from T2 to T3 for plans and intentions regarding weight loss, daily exercise, and maintaining a food diary. However, for each comparison time frame mean be havioral intention scores or weight loss, maintaining a food diary, or daily exercise lacked statistical significance. The Theory of Planned Behavior asserts behavioral intentions are predictive of behavior ( Ajzen, 1991; Rosen, 2000; Armitage & Conner, 2 001 ), and studies using the T heory of Planned B ehavior to predict behavior demonstrated a linear relationship with behavioral intentions related to stage of change ( Armitage, 2006 ; Armitage & Arden, 2007). H owever my research failed to demonstrate a consi stent statistically significant difference in ratings of behavioral intentions with progression through the stage of change particularly f rom the preparation to action stage. S tatistically significant difference were found in mean scores between weight l oss and daily exercise and between weight loss and food diary for T1 T 2, and T 3 with mean behavioral intention scores greatest in the weight loss category ( F igure 5 5 ) Th ese findings suggest the presence of both participants intentions and plans to los e weight, in tandem with decreased intentions to perform the actual beha vior necessary for weight loss (that is, keeping a food diary and performing daily exercise ) This finding points to participants who are more goal oriented than process oriented. Goal oriented individuals ( Pieters, Baumgartner, & Allen, 1995; Bagozzi, Dholakia 1999) attribute importance to weight loss, and characterize weight loss programs by the amount of
89 weight they wish to lose. However inadequate attention and thought regarding th e behavioral intentions required during the process of weight loss may introduce frustration if not recognized and addressed early during a weight loss program. Ideally healthy lifestyle/weight loss program s are designed with both goal and process oriented participants in mind Specifically programs should include weight loss, daily exercise, and healthy food choice objectives for both types of participants. The appeal of com mercial weight loss programs is enhanced through extravagant weight loss c laims made t o attract goal oriented clients and depiction of food choices and exercise as processes that are seemingly effortless For new program entrants, the above mentioned claims combined with their high behavioral intentions, lessen early stage proce ss oriented ambivalence ( Gilden and Wadden, 2005). However, this entry level (goal oriented) focus cannot sustain individuals who must endure a process of weight loss which often requires months and many times more than a year. As demonstrated in this work over time ambivalence toward the process of weight loss begins to emerge, with negative effects on behavioral intention. In this study, there was a negative association between ambiva lence and behavioral intentions. These were seen at T 2 and T 3 for food diary and at T 1 and T 3 for daily exercise. As ambivalence starts to decrease from T1 (preparation stage) to T2 (action stage), behavioral intentions correspondingly increase (Figure 5 5 ) In agreement with the findings in the published literature lower le vels of ambiv alence suggest greater behavior al intentions which translate into performance of the desired behavior ( Armitage & Conner, 2000; Armitage, Povey, Arden, 2003 )
90 Process ttendance to physical therapy and p rogra m compliance Attendance to physical therapy was measured by the number of physical therapy appointments attended during the 8 week study period. The mean number of visits during the eight week study was 11.86, or approximately 1.5 per week ; which approxi mates the national average for patient visits for low back pain (Jette, Smith, Haley, & Davis, 1994). Program compliance involved maintaining a food diary and daily exercise log. Participants presented their logs during physical therapy appointments and p r ogram compliance was recorded in two week intervals over the 8 week period. Positive compliance was reflected by participants engaging in at least 30 min utes of moderate exercise daily, consuming 5 to 9 servings of fruits and vegetables daily along with re commended amounts of legumes, dairy, nuts and lean meats and whole grains, (Kushi Byers, Nestle, McTiernan, Doyle, et. al. 2006 ). During the study period a statistically significa nt increase in mean compliance from T 1 to T 2 and from T1 to T 3 was presen t (Figure 5 4) Participant c ompliance stabilized from T2 to T 3 and decreased from T 3 to T 4 however did not reach statistical significance. I ncrease d compliance from T 1 to T 2 likely reflect s the learning curve present in relation to making better food choi ces and planning to engage in some daily exercise ( Blackburn 2005 ) The stabilization from T2 to T3, followed by the slight drop in mean compliance from T 3 to T 4 may be a signal for an initial lapse in behavioral intentions and behaviors. Future research is needed to confirm this finding There wer e moderate negative associations found between the overall percent change in weight ( T 1 T 3) and program compliance at T2 T 3 and T4. There were also moderate negative associations present between percent weight loss from T 1 T 2 and
91 T2 T 3, and T3 and T4 for program compliance respectively (Figure 5 6) An association between percent change in weight and program compliance from 1 2 weeks was absent Program compliance was lowest during this 1 2 week period and may also reflect the learning curve present for participants. Following the third physical therapy visit improvements in the food diary and exercise logs were noted. This improvement is likely due to the discussion of related concerns durin nd third visits for the low level of mean program compliance during the fir st 2 weeks especially since program compliance improved over the next 2 week period. Current findings con cur with the literature through demonstrati on of reduced ambivalence associated with stage progression. Specifically d uring the last 2 4 weeks of the program significant associations were present between behavioral intentions and ambivalence in relation to maintaining a food diary and engaging in daily exercise such that ambivalence did not keep reducing at the same amount and behavioral intentions reduced their upward trend. During this same time there was a reduction in program compliance and a reduction in the percent change in body weight compared to T1 to T2 measurements. These findings confirm assertions of the Theory of Planned B ehavior demonstrating the effects of attitudinal am bivalence on behavioral intentions and ultimately behavior (Ajzen ). Concl usions The challenge of changing obesity related behaviors require s a better understanding of behavioral change theories and constructs, along with theoretically framed interventions. The physical therapy/ healthy lifestyle/weight loss program which served as the basis for this dissertation demonstrated behavioral changes which
92 promote d weight loss and a corresponding improvement in participant physical and mental status To better understand weight loss related behavioral change t his study sought to examine participant behaviors as they progressed through an 8 week physical therapy/healthy lifestyle/weight loss program. Specifically measured were attitudinal ambivalence, behavioral intentions program compliance and stage of change Associations present betw een concepts add to the current literature and inform future research efforts. High attrition rates in weight loss programs are an accepted fact Examples of attrition rate appearing in the literature include a study of 80 women randomly assigned to atten d Weight Watchers which demonstrated an attrition rate of 25% at 12 weeks (Rippe, Price, Hess et. al.1998). A 1981 study demonstrated 50% of the enrollees stopped attending Weight Watchers in the first 6 weeks and 70% stopped by 12 weeks (Volkmar, Stunka rd Woolston, and Bailey, 1981); and a large clinic based weight loss program reported attrition rates of 13 and 31percent at 8 and 16 weeks respectively The attrition rate of 16% (21) for the current study is sli ghtly lower than studies referenced above. The lower attrition rate may relate to the physical therapy/pain relief component of the program which is absent from most commercial weight loss programs. Also of note is research by Muraven & Baumeister (2000) w hich suggest individuals have limited resources for self control, resources which are partially consumed as energy during aspects of life that require self lifestyle/weight los s programs. Reasons for program non completion included death of family members or friend s taking a difficult class while completing a degree and
93 working fulltime, an illness of a child, or increased stress from their job all of which increase added effe ct on energy consum ption and decrease d the energy needed to engage in behavior change and self control efforts. Muraven and colleagues believe that individuals compelled to exert self control as that required through self dis ci pline, are required to expen d greater energy than when exerting self control for more independent or autonomous r easons (Muraven, Gagne & Rosman, 2008). Of note, the high attrition rate in weight loss programs may be minimized by attention paid to the increased energy demands faced b y program participants, especially in the early program stages. The Theory of Planned B ehavior suggests that behavior al intentions are predict ive of behavior According to the theory behavioral intentions are influence d by attitudes, subjective norms and perceived behavioral control (Ajzen). Ambivalence may influence attitudes, intentions and behavioral control as individuals progress through a healthy lifestyle/weight loss program ( Armitage and Arden, 2007 ). Previous research demonstrate s a quadratic rela tionship between ambivalence and stage of change along with a linear relationship which exists between the stage of change and behavioral intentions (Armitage, Povey & Arden, 2003) Previous work has demonstrated ambivalence as highest in the contemplation and preparation stages while lowest in the precontemplation and maintenance stages The action stage of change is characterized by ambivalence levels which are progressively declining. Weight loss and improvements in health related quality of life resul ts solely from behavioral changes ( Wing & Phelan, 2005 ). A decline in either is said to produce a decrease in behaviora l intentions and ultimately behavior As such questions arise
94 as to the cause of the decline and relationships involved, especially since changes in ambivalence can either increase or stabilize the effects of behavioral intentions and ultimately behavior itself ways to circumvent lapses in behavior through intervention strategies will be paramount B ehavior in this study was define d as program compliance to daily food and exercise logs The value of maintaining written behavioral logs was demonstrated by the relationship between compliance and percent change in bo dy weight. Findings indicate an increase in program compliance was ass ociated with a decrease in percent body weight. The importance of tracking behavioral activities in relation to weight loss/activity cannot be overlooked. In addition, participants who were not compliant with their food choice and exercise logs may be demo nstrating an early stage of ambivalence. This observation is valuable for early intervention with process oriented goals to possibly avoid program drop outs. Figure 5 6 demonstrates the association present between program compliance and percent change in b ody weight. The data demonstrate a significant reduction in ambivalence related to weight loss, food d iary and daily exercise (F igure 5 3 ) From 4 to 8 weeks a mbivalence in relation to maintaining a food diary leveled off. From baseline to 4 weeks a mbival ence in relation to weight loss changed 17% along with a 106% change for ambivalence in relation to maintaining a food diary and a 134% change for ambivalence in relation to engaging in daily exercise The percent changes from 4 weeks to 8 weeks for weigh t loss, food diary and daily exercise was 31, 18 and 40 percent. Percent at 4 to 8 week percent changes showed a marked reduction in the percent rating of ambivalence
95 toward daily exercise and maintaining a food diary while demon strating a significant incr ease in percent rating for ambivalence toward weight loss. Ambiva lence and behavioral intention in relation to percent change in body weight demonstrated a small effect size for a mbivalence related to weight loss and daily exercise at T 3 and percent chang e in body weight from 4 and 8 weeks of (r = .285, p = .006; r = .256, p = .014) respectively. A small effect size was demonstrated between b ehavioral intent ions related to weight loss at T ime 1 and Time 2 and percent change in weight from 2 3 weeks and fro m 1 3 weeks respectively (r = .219, p = .037; r = .243, p = .020; r = .213, p = .042). Interestingly the relationship between ambivalence and health related quality of life (HRQL) has not been previously explored, even though changes in health sta tus may potentially affect a mbivalence through a healthy lifestyle program. S tudy findings demonstrated a negative association between HRQL MCS and a mbivalence in relation to maintaining a food diary at T 2 and T 3 respectively (r = 322, p = .002; r = 356, p = .001). Also demonstrated was a negative association between HRQL MCS and a mbivalence related to engaging in daily exercise at T 2 and T 3 respectively (r = .262, p = .012; r = .342, p = .00 1). These findings suggest that the mental s tatus of a participant as related to HRQL while participating in a healthy lifestyle/weight loss program is associated with the ir corresponding level of program activities is expected to decrease. This finding has implications for individuals engaging in long term weight loss efforts. Program directors should expect levels of
96 ambivalence as related to HRQL to decrease as the amount of weight loss begins to produce positive c hanges in health. As participant s progress th r ough healthy lifestyle/weight loss program stages research suggests the presence of a quadratic association between ambivalence and stage of change and a linear relationship between behavioral intentions and s tage of change (Armi tage & Arden, 2007). Dat a from this study demonstrate negative associations between ambivalence as related to daily exercise related stage of change at T 1, T 2, and T 3 (r = .352, p = .001; r = .226, p = .031; r = .434 p = .0005). Th ere was a positive association between behavioral i ntentions related to daily exercise and stage of change related to exercise with a large and medium effect size at T 2 and T 3 re spectively (r = .453, p = .0005 ; r = .357; p = .001). These findings point to as much as 6 10% of the variance between in these variables explained by the relationship s involved between stage of change related to exercise and both ambivalence related to daily exercise, and behavioral intentions related to daily exercise (Figure 5 7) Understanding the associatio ns between behavioral i ntentions and a mbivalence is essential for predicting behavior. In this study process oriented behaviors (maintaining a food and daily exercise log ) facilitated weight loss Also present were negative associations between a mbivalence as related to maintaining a food diary and behavioral i ntentions related to maintaining a food diary at T1, T2, and T3, (r = .217, p = .039; r = .457, p = .0005; r = .371, p = .0005). Associati ons were also demonstrated f or a mbivalence and behavioral i ntentions in relat ion to engaging in daily exercise were present at T 1, T 2, and T 3 respectively, (r = .303, p = .003; r = .261, .013; r = .413, p = .0005). The negative relationship between these two variables suggests am bivalence is
97 a factor in behavioral intentions an d behavior performed. The non significant changes seen in ambivalence toward a food diary from T2 to T3 along with the correlation seen between ambivalence and behavioral intentions toward a food diary sugge st that this may be an initial sign of indecisiveness in regards to keeping a food diary. The percent change in mean ambivalence related to maintaining a food diary and daily exercise declined from 4 weeks to 8 weeks. The change for T1 reached statistica l significance for maintaining an exercise log, but failed to do so for maintaining a food diary. While not statistically significant b ehavioral intentions decreased from T2 to T3. During the same time period there was a 100% reduction in mean percent cha nge in body weight when comparing baseline to 4 weeks to percent change from 4 weeks to 8 weeks (Figure 5 1) These associations together with reduced program compliance from 6 8 weeks would suggest the se may be initial signs of a behavior lapse or develop ing ambivalence. While these associations do not imply causation, these relationships offer reason for increased program attrition and lapses in behavior as the duration of a weight loss program extends ( Teixeira, Going, Houtkooper, Cussier, Metcalfe, Blew et al. 2004; Tur, Escudero, Romagurera, Burguera, 2013) These findings are beneficial for directing further studies to better clarify behavioral markers which can be used to predict weight loss related behaviors occurring as individuals progress through weight loss programs, either commercial or self directed. Understanding the array of variables and their relationships to each other, and to weight loss behaviors can aid clinicians when implementing and designing evidence based lifestyle/weight loss prog rams and intervention s The ability to predict lapses in weight loss/physical activity behaviors, and
98 to redirect individuals experiencing these lapses, will make positive contributions to individual efforts aimed at weight loss and physical activity. Re commendations for Future Research Paramount to building a research agenda is the ability to reproduce research methodology. Th e research design for t his study used a clinical model to investigate the study hypothesis. While use of a clinical model may limi t use of this design for some obesity related research, the rising overweight/obese ep i demic is expected to push more individuals into physician referred weight loss programs. While this treatment model has a number of advantages, a major disadvantage is p resent when clinicians have limited experience dealing with weight loss interventions. As clinics and clinicians move into the area of weight loss, the need for effective interventions will grow exponentially. Concurrent with increased need for physical th erapists to manage and treat obesity related disorders, conditions, and diseases, is the need for clinic based interventions and programs that are evidenced based and theoretically framed. Publication of these findings in peer reviewed journals can enable establishment of external validity through study replication of different settings and times and with different population. The study design was limited by the restrai nts of the clinical setting. This repeated measures longitudinal study ( 8 weeks ) was b ased on the average length of treatment for the obesity related diagnoses. Th e 8 week study provide s some evidence of the associations involv ed in behavioral changes. Future research designs will be strengthened by longer study period s such as 12 to 16 wee ks. The longer study period will provide an increased ability to identify preceding factors of behavioral relapse.
99 In my study data collection and management were labor intensive. The use of multiple assessment tools for data collection required attention to organization and timeliness in the clinical setting. Also, collection of follow up assessments added complexity to data collection and counseling. Streamlin in g the research process in this area of inquiry will be achieved through development and refine ment of a valid reliable assessment tool that will provide a broad and tailored information base of obesity related behaviors. Future research may focus on the development of an assessment tool that provided a broader base of information related to behavio ral factors studied. Professional Practice The necessity for developing effective clinical interventions for overweight and obese patients is vital for the healthcare system. Targeting specific populations such as physical therapy patients provides a ven ue within the healthcare system that allows for a ffective intervention. As more health professionals are called to work with those who are overweight/obese the need for affective tools used in interventions becomes more important The c ognitive and behavi oral lifestyle/weight loss program used in my study is informed by past and ongoing research and incorporates the concepts which are central to the philosophy of healthy education and promotion of wellness.
100 Table 5 1. Percent change in HRQL PCS and MCS Baseline to 8 weeks Baseline to 4 weeks 4 to 8 weeks PCS mean 32.24 20.98 10.43 SD 34.19 27.66 20.54 Min 17.78 21.37 40.04 Max 176.51 128.19 92.23 MCS mean 19.42 15.38 4.13 SD 37.52 27.88 21.87 Min 53.32 26.78 47.57 Max 169.84 111.07 92.45
101 Figure 5 1 Changes in body weight and percent change in body weight *p=.0005 **p=.0005
102 Figure 5 2 Mean changes in HRQL for PCS and MCS over time *p=.0005 **p=.021
103 Figure 5 3 Changes in Stage of change over time *p=.0005
104 Figure 5 4 Program compliance over time.
105 Figure 5 5 Si gnificance between WL, and both FD and DE for behavi oral intention and attitudinal ambivalence. A) Behavioral intention. B) Attitudinal ambivalence. A B **p=.0 12 *p= .0005 **p= .004
106 A B Figure 5 6 The relationship between Program Compliance and Percent change in body weight A) Progr am compliance B) Percent body weight
107 Figure 5 7 Associations between stage of change, behavioral intention and ambivalence
108 APPENDIX A IRB INFORMED CONSENT FORM I NFORMED C ONSENT F ORM to Participate in Resea rch, and A UTHORIZATION to Collect, Use, and Disclose Protected Health Information (PHI) I NTRODUCTION Name of person seeking your consent: Place of employment & position: This is a research study of Changes in Health related Quality of Life, and behavior during a healthy lifestyle change Could participating in this study offer any positive benef its to you? Yes as described on page 110 Could participating cause you any discomforts or are there any risks to you? Yes as described on page 110 Please read this form which describes the study in some detail. I or one of my co workers will also describe this study to you and answer all of your questions. Your participation is entirely voluntary. If you choose to participate you can change your mi nd at any time and withdraw from the study. You will not be penalized in any way or lose any benefits to which you would otherwise be entitled if you choose not to participate in this study or to withdraw. If you have questions about your rights as a res earch subject, please call the University of Florida Institutional Review Board (IRB) office at (352) 846 1494. If you decide to take part in this study, please sign this form on page 114 G ENERAL I NFORMATION ABOUT THIS S TUDY
109 1. Name of Participant ("Study Subject") ___________________________________________________________________ 2. What is the Title of this research study? ASSOCIATIONS BETWEEN HEALTH RELATED QUALITY OF LIFE, BEHAVIORAL INTENT ION AND AMBIVALENCE IN OBESE ADULT PATIENTS DURING STAGES OF A HEALTHY LIFESTYLE/WEIGHT LOSS PROGRAM 3. Who do you call if you have questions about this research study? Jeff G illiam; phone number: (352) 222 9545 4. Who is paying for this research study? The sponsor of this study is the University of Florida 5. Why is this research study being done? The purpose of this research study is to determine behavioral relation ships related to progression through a healthy lifestyle/weight loss program. You are being asked to be in this research study because you have been referred as a patient to participate in the Healthy Lifestyle/ Weight Loss program. W HAT C AN YOU E XPECT IF YOU P ARTICIPATE IN THIS S TUDY ? 6. What will be done as part of your normal clinical care (even if you did not participate in this research study)? You will receive an orthopedic evaluation and physical therapy treatment and will be instructed on a Healthy Lifestyle/weight loss program as part of your normal clinical care. 7. What will be done only because you are in this research study? You will be required to fill out questionnaires that provide information regarding yo ur attitude toward weight loss, exercise and a food diary. If you have any questions now or at any time during the study, please contact Jeff Gilliam in question 3 of this form. 8. How long will you be in this research study?
110 8 weeks 9. How many people are expected to take part in this research study? 80 W HAT ARE THE R ISKS AND B ENEFITS OF THIS S TUDY AND W HAT ARE Y OUR O PTIONS ? 10. What a re the possible discomforts and risks from taking part in this research study? Other p ossible risks to you may include: There are no risk additional because of involvement in this research study. Researchers will take appropriate steps to protect any information they collect about you. However, there is a slight risk that information about you could be revealed inappropriately or accidentally. Depending on the nature of the information, such a release could upset or embarrass you, or possibly affect your insurability or employability. Questions 17 21 in this form discuss what information about you will be collected, used, protected, and shared. This study may include ri sks that are unknown at this time. Participation in more than one research study or project may further increase the risks to you. If you are already enrolled in another research study, please inform Jeff Gilliam (listed in question 3 of thi s consent form) or the person reviewing this consent with you before enrolling in this or any other research study or project. Throughout the study, the researchers will notify you of new information that may become available and might affect your decisio n to remain in the study. If you wish to discuss the information above or any discomforts you may experience, please ask questions now or call the PI or contact person listed on the front page of this form. 11a. What are the potential benefits to you for taking part in this research study? Potential benefits for taking part in this research study, may be increased awareness regarding obstacles affecting long term success in the areas of weight loss and improved level of fitness.
111 11b. How could others possibly benefit from this study? Findings from this study will contribute to information necessary to help those that are overweight/obese make behavioral changes essential to achieve long termweight loss. 11c. How could the researchers benefit from this study? In general, presenting re search results helps the career of a scientist. Therefore, Jeff Gilliam may benefit if the results of this study are presented at scientific meetings or in scientific journals. Jeff Gilliam may benefit if the results of this stud y are presented at scientific meetings or in scientific journals. 12. What other choices do you have if you do not want to be in this study? You will receive the same treatment intervention if you do not want to be in this study. 13a. Can you withdraw from this study? You are free to withdraw your consent and to stop participating in this study at any time. If y ou do withdraw your consent, you will not be penalized in any way and you will not lose any benefits to which you are entitled. Yes, you are free to withdraw from this study at any time, and may continue to receive all the benefits that you n ormally would had you not participated in this research study. If you decide to withd raw your consent to participate in this study for any reason, please contact Jeff Gilliam at 352 222 9545 They will tell you how to stop your participation safely. If you have any questions regarding your rights as a resea rch subject, please call the Institutional Review Board (IRB) office at (352) 846 1494. 13b. If you withdraw, can information about you still be used and/or collected? Because the information that is gathered from this research study may benefit others in the future, could we use your information in the event that you withdraw from this research study? 13c. Can the Principal Investigator withdraw you from this study? You may be withdrawn from the study without your consent for the following reasons: If y ou have any health problems, which arise during this study, you may be required to withdraw from this study.
112 W HAT ARE THE F INANCIAL I SSUES IF Y OU P ARTICIPATE ? 14. If you choose to take part in this research study, will it cost you anything? There are no costs rela ted to participation in this research study. 15. Will you be paid for taking part in this study? No, subjects of this research study are not paid. 16. What if you are injured because of the study? Please contact the Principal Investigator listed in question 3 of this form if you experience an injury or have questions about any discomforts that you experience while participating in this s tudy. 17. How will your health information be collected, used and shared? If you agree to participate in this study, the Principal Investigator will create, collect, and use private information about you and your health. This information is called protecte d health information or PHI. In order to do this, the Principal Investigator needs your authorization The following section describes what PHI will be collected, used and shared, how it will be collected, used, and shared, who will collect, use or share i t, who will have access to it, how it will be secured, and what your rights are to revoke this authorization. Your protected health information may be collected, used, and shared with others to determine if you can participate in the study, and then as p art of your participation in the study. This information can be gathered from you or your past, current or future health records, from procedures such as physical examinations, x rays, blood or urine tests or from other procedures or tests. This informat ion will be created by receiving study treatments or participating in study procedures, or from your study visits and telephone calls. More specifically, the following information may be collected, used, and shared with others: Body weight, body mass in dex, age, food and exercise logs, questionnaires. This information will be stored in locked filing cabinets or on computer servers with secure passwords, or encrypted electronic storage de vices. Some of the information collected could be included in a "limited data set" to be used for other research purposes. If so, the limited data set will only include information
113 that does not positive ly identify you. For example, the limited data set cannot include your name, address, telephone number, social security number, photographs, or other codes that link you to the information in the limited data set. If limited data sets are created and used, agreements between the parties creating and rece iving the limited data set are required in order to protect your identity and confidentiality and privacy. 18. For what study related purposes will your protected health information be collected, used, and shared with others? Your PHI may be collected, u sed, and shared with others to make sure you can participate in the research, through your participation in the research, and to evaluate the results of the research study. More specifically, your PHI may be collected, used, and shared with others for the following study related purpose(s): Once this information is collected, it becomes part of the research r ecord for this study. 19. Who will be allowed to collect, use, and share your protected health information? Only certain people have the legal right to collect, use and share your research records, and they will protect the privacy and security of th ese records to the extent the law allows. These people include the: the study Principal Investigator, Jeff Gilliam and research staff associated with this project. other professionals at the University of Florida or Shands Hosp ital that provide study related treatment or procedures the University of Florida Institutional Review Board (IRB; an IRB is a group of people who are responsible for looking after the rights and welfare of people taking part in research). The study Prin cipal Investigator: Jeff Gilliam and research staff associated with this project. 20. Once collected or used, who may your protected health information be shared with? Your PHI may be shared with:
114 the study sponsor University of Florida United States and foreign governmental agencies who are responsible for overseeing research, such as the Food and Drug Administration, the Department of Health and Human Services, and the Office of Human Research Protections Government agencies who are responsible for overseeing public health concerns such as the Centers for Disease Control and federal, state and local health departments Malcom Randall VA Medical Center (Gainesvill e) Your insurance company for purposes of obtaining payment Otherwise, your research records will not be released without your permission unless required by law or a court order. It is possible that once this information is shared with authorized persons, it could be shared by the persons or agencies who receive it and it would no longer be protected by the federal medical privacy law. 21. If you agree to take part in this research study, how long will your protected health information be used and shared with others? Your PHI will be used and sha red with others 2 years. You are not required to sign this consent and authorization or allow researchers to collect, use and share your PHI. Your refusal to sign will not affect your treatment, payment, enrollment or eligibility for any benefits outside this research study. However, you cannot participate in this research unless you allow the collection, use and sharing of your protected health information by signing this consent and authorization. You have the ri ght to review and copy your protected health information. However, we can make this available only after the study is finished. You can revoke your authorization at any time before, during, or after your participation in this study. If you revoke it, no new information will be collected about you. However, information that was already collected may still be used and shared with others if the researchers have relied on it to complete the research. You can revoke your authorization by giving a written req uest with your signature on it to the Principal Investigator.
115 S IGNATURES participant the pu rpose, the procedures, the possible benefits, and the risks of this protected health information will be collected, used, and shared with others: Signature of Person Obtaining Consent and Authorization Date and risks; the alternatives to being in the study; and how your protected health information will be collected, used and shared with others. You have received a copy of this Form. You have been given the opportunity to ask questions before you sign, and you have been told that you can ask questions at any time. You voluntarily agree to participate in this study. You hereby a uthorize the collection, use and sharing of your protected health information as described in sections 17 21 above. By signing this form, you are not waiving any of your legal rights. Signature of Person Consenting and Authorizing Date
APPEND IX B SF 36 HEALTH SURVEY INSTRUCTIONS : This survey asks your views about your health. This information will help keep track of how you feel and how well you are able to do your usual activities. Please answer every question by marking the answer as indi cated. If you are unsure about how to answer a question please give the best answer you can.
120 APPENDIX C GRIFFIN INDEX AMBIVALENCE = (POSIT IVE + NEGATIVE)/2 POSITIVE NEGATIVE
121 APPENDIX D POSITIVE ATTITUDINAL AMBIVAL ENCE ASSESSMENT FORM Below are questions to assess your positive attitudes toward a subject. Please circle the statement which best describes your response to the question. Positive Scale for Weight Loss 1. Think about your attitude toward or evaluation of weight loss. That is an intentional reduction of your body weight through diet, exercise or both. Considering only the favorable qualities of weight loss and ignoring the unfavorable characteristics, how favorable is your evaluation of weight loss? No t at all Slightly Quite Extremely Favorable Favorable Favorable Favorable 2. Think about your feelings or emotions when I mention weight loss. Considering only your feelings of satisfaction toward weight loss and ignoring your feelings of dissatisfa ction, how satisfied do you feel about weight loss? Not at all Slightly Quite Extremely Satisfied Satisfied Satisfied Satisfied 3. Think about your thoughts or beliefs when I mention weight loss. Considering only the beneficial qualities of weight loss and ignoring the harmful characteristics, how beneficial do you believe weight loss to be? Not at all Slightly Quite Extremely Beneficial Beneficial Beneficial Beneficial Positive Scale for Daily Exercise 1. Think about your attitude toward or evaluation of daily exercise. That is exercising 30 40 minutes a day 6 7 days per week. Considering only the favorable qualities of daily exercise and ignoring the unfavorable characteristics, how favorable is your evaluation of daily exercise? Not at all Slightly Quite Extremely Favorable Favorable Favorable Favorable
122 2. Think about your feelings or emotions when I mention daily exercise. Considering only your feelings of satisfaction toward daily exercise and ignoring your feelings of dissat isfaction, how satisfied do you feel about daily exercise? Not at all Slightly Quite Extremely Satisfied Satisfied Satisfied Satisfied 3. Think about your thoughts or beliefs when I mention daily exercise. Considering only the beneficial qualities of daily exercise and ignoring the harmful characteristics, how beneficial do you believe daily exercise to be? Not at all Slightly Quite Extremely Beneficial Beneficial Beneficial Beneficial Positive Scale for Food Diary 1. Think about your attitu de toward or evaluation of a food diary. That is recording all of your daily food intakes. Considering only the favorable qualities of a food diary and ignoring the unfavorable characteristics, how favorable is your evaluation of a food diary? Not at a ll Slightly Quite Extremely Favorable Favorable Favorable Favorable 2. Think about your feelings or emotions when I mention a food diary. Considering only your feelings of satisfaction toward a food diary and ignoring your feelings of dissatisfactio n, how satisfied do you feel about a food diary? Not at all Slightly Quite Extremely Satisfied Satisfied Satisfied Satisfied 3. Think about your thoughts or beliefs when I mention food diary. Considering only the beneficial qualities of food diary and ignoring the harmful characteristics, how beneficial do you believe a food diary to be? Not at all Slightly Quite Extremely Beneficial Beneficial Benefici al Beneficial
123 APPENDIX E NEGATIVE ATTITUDINAL AMBIVALENCE ASSESSMENT FORM Below are que stions to assess your negative attitudes toward a subject. Please circle the statement which best describes your response to the question. Negative Scale for Weight Loss 1. Think about your attitude toward or evaluation of weight loss. That is an intentiona l reduction in your body weight through diet, exercise or both. Considering only the unfavorable qualities of weight loss and ignoring the favorable characteristics, how unfavorable is your evaluation of weight loss? Not at all Slightly Quite Extrem ely Unfavorable Unfavorable Unfavorable Unfavorable 2. Think about your feelings or emotions when I mention weight loss. Considering only your feelings of dissatisfaction toward weight loss and ignoring your feelings of satisfaction, how dissatisfied do you feel about weight loss? Not at all Slightly Quite Extremely Dissatisfied Dissatisfied Dissatisfied Dissatisfied 3. Think about your thoughts or beliefs when I mention weight loss. Considering only the harmful qualities of weight loss and ig noring the beneficial characteristics, how harmful do you believe weight loss to be? Not at all Slightly Quite Extremely har mful harmful harmful harmful Negative Scale for Daily Exercise 1. Think about your attitude toward or evaluation of daily exerc ise. That is exercising 30 40 minutes a day 6 7 days per week. Considering only the unfavorable qualities of daily exercise and ignoring the favorable characteristics, how unfavorable is your evaluation of daily exercise? Not at all Slightly Quite Extremely unfavorable unfavorable unfavorable unfavorable 2. Think about your feelings or emotions when I mention daily exercise.
124 Considering only your feelings of dissatisfaction toward daily exercise and ignoring your feelings of satisfaction, how dis satisfied do you feel about daily exercise? Not at all Slightly Quite Extremely dissatisfied dissatisfied dissatisfied dissatisfied 3. Think about your thoughts or beliefs when I mention daily exercise. Considering only the harmful qualities of dail y exercise and ignoring the beneficial characteristics, how harmful do you believe daily exercise to be? Not at all Slightly Quite Extremely harmful harmful harmful harmful Negative Scale for Food diary 1. Think about your attitude toward or evaluat ion of a food diary. That is recording all of your daily food intakes. Considering only the unfavorable qualities of a food diary and ignoring the favorable characteristics, how unfavorable is your evaluation of a food diary? Not at all Slightly Quit e Extremely unfavorable unfavorable unfavorable unfavorable 2. Think about your feelings or emotions when I mention a food diary. Considering only your feelings of dissatisfaction toward a food diary and ignoring your feelings of satisfaction, how diss atisfied do you feel about a food diary? Not at all Slightly Quite Extremely dissatisfied dissatisfied dissatisfied dissatisfied 3. Think about your thoughts or beliefs when I mention a food diary. Considering only the harmful qualities of food dia ry and ignoring the beneficial characteristics, how harmful do you believe a food diary to be? Not at all Slightly Quite Extremely harmful harmful harmful harmful
APPENDIX F BEHAVIORAL INTENTION FORM Below is a list of statements involving 3 catego ries: weight loss, daily exercise, food diary There are two statements made about each category. Your choices are from 3 to +3. Circle the number that best indicates what your intent/plan is regarding each category in the future. The future pertains to the time that you are participating in the Healthy Lifestyle/Weight loss program. 1. I intend to lose weight in the future. definitely do not 3, 2, 1, 0, 1, 2, 3 definitely do 2. I plan to lose weight in the future. definitely do not 3, 2, 1, 0, 1, 2, 3 definitely do 3. I intend to do daily exercise in the future. definitely do not 3, 2, 1, 0, 1, 2, 3 definitely do 4. I plan to do daily exercise in the future. definitely do not 3, 2, 1, 0, 1, 2, 3 de finitely do 5. I intend to do a food diary in the future. definitely do not 3, 2, 1, 0, 1, 2, 3 definitely do 6. I plan to do a food diary in the future. definitely do not 3, 2, 1, 0, 1, 2, 3 definitely do
126 APPENDIX G ST AGES OF CHANGE ASSESSMENT FORM You will be presented with two categories, (exercise and food choices). For each category you will be presented with five statements. Put an X in the box beside the statement that best describes where you are in terms of exer cise or food choices. The first category is exercise For regular moderate exercise, frequency will be set at 5 to 7 days per week for 30 to 40 minutes. Moderate intensity will be described as not being painful, increased breathing, however able to carry on a conversation while exercising, with a light sweat I currently do not exercise I currently do not exercise but I am thinking about starting. I currently exercise I currently exercise regularly but I have only begun to do so in the last six months I currently exercise The second category is food choices Criteria for food choices will be set for five areas: including eating hig h fiber foods at a frequency of 1) 5 9 servings for fruits and vegetables daily, 2) 2 3 servings of beans and whole grains; partaking of high calcium foods 3) 2 3 servings of fat free dairy; getting enough protein and avoiding high fat foods from 4) eating lean meats only and avoiding high fat and fried foods, and condiments; avoiding high sugar items like 5) high sugar drinks or foods like sugary sodas and deserts I currently do not make good food choices I currently do not make good food choices but I am thinking about starting. I currently make good food choices I currently make good food choices regularly but I have only begun to do so in the last six months I currently make good food cho ices regularly and i have done so for longer than
127 APPENDIX H FOOD CHOICE & EXERCISE LOG
128 APPENDIX I COMPLIANCE LOG FOR FOOD CHOICES AND EXERCISE Compliance for food choices and exercise will be measured by determining patients ind and exercise. For each area under the 6 criteria for food choices patients receive 1 point fo r fulfilling each area, and additional 1 point for staying within 5 points of their total daily (TP) point allocation. If they reach their goal in each area at a 100% they receive 6 points. For fulfilling exercise goals of 30 40 minutes 5 days per week, t hey receive 2 points. Patients can receive a total of 8 points for fulfilling each goal. Code : Fruits & Vegetables (F&V); B eans and who le grains (B&WG); Low fat dairy (LF dairy); Lean meats, chicken, fish, nuts & avoid fried fat foods (LMCFN & Avoid FFF); Total points (TP); Daily total (DT); Avoid sweets, sugar, deserts (Avoid SSD). Food Choices Exercise C ategory F & V B & WG LF Dairy LMCFN & Avoid FFF Avoid SSD TP 30 40 min DT Date
APPENDIX J. FLYER You May be Eligible to Participate in a Research Study Involving Healthy Lifestyle Changes and Weight Loss The Research is being performed by Jeff Gilliam who is Clinical Po sitive or of ReQuest Physical Therapy Contact Information: (352) 373 2116 Affiliation: This study is through the Health Education and Behavior Department College of Health and Human Performance University of Florida Purpose: This research is designed to study health related quality of life and behaviors related to healthy lifestyle changes and weight loss General Eligibility Criteria: Subjects who are referred by a physician with a primary orthopedic diagnosis and a secondary diagnosis of overweight Your insurance should cover your treatment. Location: The Research Study will be carried out within the ReQuest Physical Therapy Facility Time Commitment: The study is for 8 full weeks at 2 3 visits per week.
130 REFERENCE LIST Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes 50 179 211. Alsio, J., Olszewski, P. K., Levine, A. S., & Schioth, H. B. (2012). Feed forward mechanisms: addiction like behavioral and molecular adaptations in overeating. Fronti ers in Neuroendocrinology 33 (2), 127 139. Anderson, J. W., Konz, E. C., Frederich, R. C., & Wood, C. L. (2001). Long term weight loss maintenance: a meta analysis of US studies. American Journal of Clinical Nutrition, 74 579 584. Andersen, R. E., Cresp o, C. J., Bartlett, S. J., Bathon, J. M., & Fontaine, K. R. (2003) Relationship between body weight gain and significant knee, hip, and back pain in older Americans. Obes ity Res earch, 11 (10), 1159 11 62. Armitage, C. J. (2006). Evidence that implementation intentions promote transitions between the stages of change. Journal of Consulting and Clinical Psychology 74 (1), 141 151. Armitage, C. J., & Arden, M. A. (2007). Felt and potential ambivalence across the stages of change. Journal of Health Psychology, 12 (1), 149 58. Armitage, C. J., & Conner, M. (2000). Attitudinal ambivalence: a test of three key hypotheses. Personality and Social Psychology Bulletin, 26 1421 1432. Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned behavior a meta analytic review. British Journal of Social Psychology 40 471 499. Armitage, C. J., Povey, R., & Arden, M. A. (2003). Evidence for discontinuity patterns across the stages of change: A role for attitudinal ambivalence. Psychology and Health 18 (3), 373 386. Armitage, C. J., Sheeran, P., Conner, M., & Arden, M. A. (2004). Stages of change or changes of stage? Predicting transitions in transtheorectical model stages in relation to healthy food choices. Journal of Consulting and Clinical Psychology 72 (3), 491 499. Avena, N. M. (2007). Examining the addictive like properties of binge eating using an animal model of sugar dependence. Experimental Clinical Psychopharmacology 15 (5), 481 4 91. Bagozzi, R. P., & Dholakia, U. (1999). Goal setting and goal striv ing in consumer behavior. Journal of Marketing 63 19 32. Bandura, A. (1992). Self efficacy mechanism in human agency. American Psychologist 37 ,122 147.
131 Baron, J., & Spranca, M. (1997). Protected values. Organizational Behavior and Human Decision Proc esses, 70 (1), 1 16. Becker, M. H. (1990). Theoretical models of adherence and strat egies for improving adherence. Shumaker. S.A. Schron, Eleanor B., Ockene, Judith K., Parker, Christine T., Probstfield, Jeffrey L., Wolle, Joan M. (Eds.) The Handbook of Heal th Behavior Change Springer Publishing Co.New York, NY 5 43. Becker, M. H., & Joseph, J. G. (1988). AIDS and behavioral change to reduce risk: A Review. American Journal of Public Health, 78 394 410. Berthoud, H. R., Lenard, N. R., & Shin A.C., (2011). Food reward, hyperphagia, and obesity. American Journal of Physiology and Regulatory Integrative and Comparative Physiology 300 R1266 R1277. Bish, C. L., Blanck, H. M., Maynard, L. M., Serdula, M. K., Thompson, N. J., & Khan, L. K., (2007). Health re lated quality of life and weight loss practices among overweight and obese US adults, 2003 behavioral risk factor surveillance system. Medscape General Medicine. 14, 9(2), 35. Retrieved from http:/ /www.medscape.com/viewarticle/555511 Blackburn, G. L. (2005). Teaching, learning, doing: best practices in education. American Journal of Clinical Nutrition, 82 218S 221S. Blissmer, B., Riebe, D., Dye, G., Ruggieroe, L., G reene, G., & Caldwell, M. (200 6). Health related quality of life following a clinical weight loss intervention among overweight and obese adults: intervention and 24 month follow up effects. Health Quality Life Outcomes, 17 (4), 43 48. Breckler, S. J. (1994). A comparison of numerical indexes for measuring attitudinal ambivalence. Educational and Psychological Measurement, 54 350 365. Brownell, K. D, & Marlatt, G. A., Lichtenstein, E., & Wilson, G. T. (1986). Understanding and preventing relapse. American Psychology 41 765 782. But ryn, M. L., Phelan, S., Hill, J. O. & Wing, R. R. (2007). Consistent self monitoring of weight: a key component of successful weight loss maintenance. Obesity, 15 (12), 3091 3096. Byas, A. (2013). Obesity is classified as a disease. Digital Journal Retri eved from http://digitaljournal.com/article/353856 Cai, L., Lubitz, J., Flegal, K. M., & Pamuk, E. R., (2010). The predicted effects of chronic obesity in middle age on medicare costs and mortality. Medical Care 48 (6), 510 517.
132 Center for Disease Control and Prevention Chronic Disease Prevention and Health Promotion: Health consequences of obesity promotion (2007). NIH publication Bethesda,MD http://www.cdc.gov/chronicdisease/resources/publications/AAG/ob esity.htm Chapman, J., & Armitage, C. J. (2009). Evidence that boosters augment the long term impact of implementation intentions on fruit and vegetable intake Psychological Health, 27 1 17. Christensen, R., Astrup, A., & Bliddal, H. (2005). Weight loss: the treatment of choice for knee osteoarthritis? A randomized trial. Osteoarthritis Cartilage 13 (1), 20 7. Cifuentes, M., & Green, L. A., (2004). Obesity tr ial: knowledge without systems. British Medical Journal. 328 (7438), E272 E274. Coggon, D., Reading, I., Croft, P., McLaren, M., Barrett, D., & Cooper, C. (2001). Knee osteoarthritis and obesity. International Journal of Obesity and Related Metabolic Disord ers 25 (5), 622 627. Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences (2 nd ed.). Hillsdale, N. J., Lawren ce E u rlbaum Associates. 21 23. Corica, F., Corsonello, A., Apolone, G., Lucchetti, M., Melchionda, N., Marchesini, G. & Quovad is Study Group. (2006). Construct validity of the Short Form 36 Health Survey and its relationship with BMI in obese outpatients. Obesity, 14 (8), 1429 1437. Dansinger, M. L., Gleason, J. A., Griffith, J. L., Selker, H. P., & Schaefer, E.J. (2005). Compari son of the Atkins, Ornish, Weight Watchers, and Zone diets for weight loss and heart disease risk reduction: a randomized trial. JAMA, 293 43 53. Dohm, F. A., Beattie, J. A., Aibel, C., & Striegel Moore, R. H. (2001). Factors differentiating women and me n who successfully maintain weight loss from women and men who do not. Journal of Clinical Psychology, 57 (1), 105 117. Doll, H. A., Petersen, S. E., & Stewart Brown, S. L. (2000). Obesity and physical and emotional well being: associations between body m ass index, chronic illness, and the physical and mental components of the SF 36 questionnaire. Obesity Research, 8 (2), 160 170. Elfhag, K., & Rossner, S. (2005). Who succeeds in maintaining weight loss? A conceptual review of factors associated with weigh t loss maintenance and weight regain. Obesity Reviews, 6 67 85. Epping Jordan, M. P., Watkins, S. S., Koob, G. F. & Markou, A. (1998). Dramatic decreases in brain reward function during nicotine withdrawal. Nature, 393 (6680), 76 7 9.
133 Fairbank, J. C. T., & Pynsent, P.B. (2000) The Oswestry Disability Index. Spine 25 (22), 2940 295 Fegal, K. M., Carroll, M. D., Ogden, C. L., & Curtin, L. R. (2010). Prevalence and trends in obesity among US adults. JAMA, 303 235 241. Felson, D. T., Anderson, J. J., Naimar k, A., Walker, A. M., & Meenan, R. F.(1988). Obesity and knee osteoarthritis. The Framingham Stu dy. Annals of Internal Medicine 109 (1), 18 24. Ferris, F. D., von Gunten, C. F., & Emanuel, L. L. (2001). Knowledge: Insufficient for Change. Journal of Pallia tive Medicine, 4 (2),145 147. Field, A. E., Coakley, E. H., Must, A., Spadano, M. A., & Laird, N. (2001). Impact of o verweight on the r isk of d eveloping c ommon c hronic d iseases d uring a 10 y ear p eriod. Archives of Internal Medicine, 161 1581 1586. Finkelst ein, E. A., Trogdon, J. G., Brown, D. S., Allaire, B. T., Dellea, P. S., & Kamal Bahl, S. J. (2008). The lifetime medical cost burden of overweight and obesity: implications for obesity prevention. Obesity, 16 (8), 1843 1848. French, S. A., Jeffery, R. W., Folsom, A. R., McGovern, P., & Williamson, D. R. (1996). Weight loss maintenance in young adulthood; Prevalence and correlations with health behavior and disease in a population based sample of women aged 55 69 years. International Journal of Obesity, 20 303 310. French, S. A., Story, M., & Jeffery, R. W. (2001). Environmental influences on eating and physical activity. Annual Review in Public Health 22 309 335. Fujioka, K. (2010). Benefits of moderate weight loss in patients with type 2 diabetes. Di abetes Obesity Metabolism, 12 (3):186 94. Gelber, A. C., Hochberg, M. C., Mead, L.A., Wang, N.Y., Wigley, F.M., & Klag, M. J. (1999).Body mass index in young men and the risk of subsequent knee and hip osteoarthritis. American Journal of Medicine 107 (6), 5 42 8. Gilden, A. T., & Wadden, T. A. (2005). Systematic review: an evaluation of major commercial weight loss programs in the United States. Annals of Internal Medicine. 142 56 66. Godin, G., Kok, G. (1996). The theory of planned behavior: a review of i ts applications to health related behaviors. American Journal of Health Promotion, 11 (2), 87 98. Grant, J. E., Brewer, J. A., & Potenza, M. N. (2006). The neurobiology of substance and behavioral addictions. CNS Spectr., 11 (12), 924 30. Grilo, C. M., Shif fman, S., & Wing, R. R. (1993). Coping with dietary relapse crises and their aftermath. Addictive Behaviors, 18 (1):89 102.
134 Grundy, S. M., Cleeman, J. I., Daniels, S. R., Donato, K. A., & Eckel, M. D., Robert H., (2005). Diagnosis and Management of the Me tabolic Syndrome. American Heart Association, Inc., 112 285 290. Gudbergsen, H., Boesen, M., Lohmander, L. S., Christensen, R., Henriksen, M., Bartels, E. M., & Bliddal, H. (2012). Weight loss is effective for symptomatic relief in obese subjects with kne e osteoarthritis independently of joint damage severity assessed by high field MRI and radiography. Osteoarthritis Cartilage Jun, 20 (6), 495 502. Herbert, C. P. (1996). Giving information: usually necessary, but often insufficient to achieve behavior cha nge. Patient Education and Counseling, 29 229 230. Hill, J. O., Wyatt, H. R., Reed, G. W., & Peters, J. C. (2003) Obesity and the environment: where do we go from here? Science 299 (7), 853 855. S. O. (2003). Predictors of attrition in a large clinic based weight loss program. Obesity Research, 11 ,(7), 888 894. Hunt, W. A., Barnett, L. W., & Branch, L. G. (1971). Relapse rates in addiction programs. Journal of Clinical Psychology 4 455 456. Im ayama, I., Alfano, C. M., Kong, A., Foster Schubert, K. E., Bain, C. E., Xiao, L., McTiernan, A., (2011). Dietary weight loss and exercise interventions effects on quality of life in overweight/obese postmenopausal women: a randomized controlled trial. In ternational Journal of Behavioral Nutrition and Physical Activity, 8 118. Jakicic, J. M., Clark, K., Coleman, E., Donnelly, J. E., Foreyt, J., Melanson, E., Volek, J., & Volpe, S. L., (2001). Appropriate Intervention Strategies for Weight Loss and Prev ention of Weight Regain for Adults. Medicine and Science in Sports Exercise, 33 (12), 2145 2156. Jeffery, R. W., Epstein, L. H., Wilson, G. T., Drewnowski, A., Stunkard, A. J., & Wing, R. R., (2000). Long term maintenance of weight loss: current status. Hea lth Psychology 19 (1), 5 16. Jette, A. M., Smith, K., Haley, S. M., & Davis, K. D. (1994). Physical therapy episodes of care for patients with low back pain. Physical Therapy, 74 ,(2), 101 110. Johnson, P. M., & Kenny, P. J. (2010). Dopamine D2 receptors i n addiction like reward dysfunction and compulsive eating in obese rats. Nature Neuroscience 13 (5), 635 641. Kaukua, J., Pekkarinen, T., Sane, T., & Mustajoki, P., (2003). Health related quality of life in follow up study. International Journal of Obesi ty and Related Metabolic Disorders, 27 (9), 1072 1080.
135 Koepp, M. J., Gunn, R. N., Lawrence, A. D., Cunningham, V. J., Dagher, A., Jones, T., Grasby, P. M. (1998). Evidence for striatal dopamine release during a video game. Nature, 393 (6682), 266 268. Koob, G. F. (1992). Drugs of abuse: anatomy, pharmacology and function of reward pathways. Trends in Pharmacological Science 13 (5), 177 184. Kral, J. G., Sjostrom, L. V., & Sullivan, M. B. (1992). Assessment of quality of life before and after surgery for sev ere obesity. American Journal of Clinical Nutrition, 55 (2 Suppl), 611S 614S. Kramer, F. M., Jeffery, R. W., Forster, J. L. & Snell, M. K. (1989). Long term follow up of behavioral treatment for obesity: patterns of weight regain among men and women. Int ernational Journal of Obesity, 13 (2), 123 136. Kreek, M. J. (1996). Cocaine, dopamine and the endogenous opiod system. Journal of Addictive Disease 15 73 96. Kushi, L. H., Byers, T., Nestle, M., McTiernan, A., Doyle, C., Bandera, E. V., the American Can cer Society 2006 Nutrition and Physical Activity Guidelines Advisory Committee. (2006). American cancer society guidelines on nutrition and physical activity for cancer prevention: reducing the risk of cancer with healthy food choices and physical activity CA Cancer Journal for Clinicians 56 254 281. Layman, D. K., Evans, E. M., Erickson, D., Seyler, J., Weber, J., Bagshaw, D., & Kris Etherton, P. (2009). A Moderate Protein Diet Produces Sustained Weight Loss and Long Term Changes in Body Composition a nd Blood Lipids in Obese Adults. The Journal of Nutrition, 139 514 521 Lidar, Z., Behrbalk, E., Regev, G. J., Salame, K., Keynan, O., Schweiger, C., & Keidar, A. (2012). Invertebral disc height changes after weight reduction in morbidly obese patients an d its effect on quality of life and radicular and low back pain. Spine, 37 (23), 1947 1952. Lim, L. L., Seubsman, S. A., & Sleigh, A. (2008). Thai SF 36 health survey: tests of data quality, scaling assumptions, reliability and validity in healthy men and women. Health Quality Life Outcomes, 18 6 52. Lisspers, J., Sundin, O., Ohman, A., Hofman Bang, C., Ryden, L. & Nygren, A. (2005). Long term effects of lifestyle behavior change in coronary artery disease: effects on recurrent coronary events after pre cutaneous coronary intervention. Health Psychology, 24(1),41 4 8. Logue, E. E., Jarjoura, D. G., Sutton, K. S., Smucker, W. D., Baughman, K. R., & Capers, C. F., (2004). Longitudinal relationship between elapsed time in the action stages of change and weig ht loss. Obesity Research 12 (9), 1499 1508.
136 Lopez Garcia, E., Banegas, Banegas, Jr., Gutierrez Fisac, J.L., Perez Regadera, A.G., Ganan, L.D. & Rodriguez Artalejo, F. (2003). Relation between body weight and health related quality of life among the elde rly in Spain. International Journal of Obesit Related Metabolism Disorders, 27(6), 701 70 9. Lorig, K. & Laurin, J. (1985). Some notions about assumptions underlying health education. Health Education Quarterly, 12 (3), 231 243. Malik, V. S., Schulze, M. B., & Hu, F. B. (2006). Intake of sugar sweetened beverages and weight gain: a systematic review. American Journal of Clinical Nutrition 84 274 288. Mathus Vliegen, E. M., de Weerd, S., & de Wit, L. T. (2004). Health related quality of life in patients with morbid obesity after gastric banding for surgically induced weight loss. Surgery, 135 (5), 489 497. McHorney, C. A., Ware, J. E., & Raczek, A. E.(1993). The MOS 36 item short form health survy (SF 36):II. Psychometric and clinical tests of validity i n measuring physical and mental health constructs. Medical Care 31 (3), 247 263. Messier, S. P., Loeser, R. F., Miller, G. D., Morgan, T. M., Rejeski, W. J., Sevick, M. A., obese older adults with knee osteoarthritis: the Arthritis, Diet, and Activity Promotion Trial. Arthritis Rheum at ism 50 (5), 1501 1510. Miguez, M. C., Becona, E. (2008). Abstinence from smoking ten years after participation in a randomized controlled trial of s elf help program. Addictive Behaviors (10), 1369 1374. Miller, G. D., (2010). Improved nutrient intake in older obese adults undergoing a structured diet and exercise intentional weight loss program. Journal of Nutrition for Healthy Aging 14 (6), 461 466 Muraven, M., & Baumeister, R. F. (2000). Self Regulation and Depletion of Limited Resources: Does Self Control Resemble a Muscle? Psychological Bulletin, 126 (2), 247 259. Muraven, M., Gagne, M., & Rosman, H. (2008). Helpful Self Control: Autonomy Suppor t, Vitality, and Depletion. Journal of Experimental Social Psychology, 44 (3), 573 585. Muraven, M., Tice, D. W., & Baumeister, R. F.(1998). Self control as a limited resource: Regulatory depletion patterns. Journal of Personality and Social Psychology, 74 774 7 89. National Center for Health Statistics Death and Mortality. (2006). NIH publications,Bethesda, MD, http://www.cdc.gov/nchs/fastats/deaths.htm
137 National Heart, Lung and Blood Institute (1998). Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: the evidence report. Obesity Research 6 51S 209S. of substance use disorders. Substance Use Misuse, 42 (2 3), 503 526. Oliveria, S. A., Felson, D. T., Cirillo, P. A., Reed, J. I., & Walker, A. M. (1999). Body weight, body mass index, and incident symptomatic osteoarthritis of the hand, hip, and knee. Epid emiology, 10 (2), 161 166. Paans, N., van den Akker Scheek, I, Dilling R.G., Bos M., van der Meer K, Bulstra, S.K, Stevens, M. (2013). Effect of exercise and weight loss in people who have hip osteoarthritis and are overweight or obese: a prospective cohort study. Physical Therapy, Feb 93(2), 137 146. Page, R. M., Cole, G. E. (1984 framework for health education research and curriculum development. International Quarterly of Community Health Education, 5 321 329. Pelchat, M.L. (2009). Food addiction in humans. Journal of Nutrition, 139(3), 620 622. Pettman, T. L., Misan, G., Owen, K., Warren, K., Coates, A. M., Buckley, J. D., Howe, P. R. C. (2008). Self management for obesity and cardio metabolic fitne ss: description and evaluation of the lifestyle modification program of a randomized controlled trial. International Journal of Behavioral Nutrition and Physical Activ ity. 5 53. Phelan, S., Hill, J. O., Lang, W., Dibello, J. R., & Wing, R. R. (2003). Reco very from relapse among successful weight maintainers. American Journal of Clinical Nutrition, 78 (6), 1079 1084. Pieters, R., Baumgartner, H., & Allen, D. (1995). A means end chain approach to consumer goal structures. International Journal of Research in Marketing 12 227 244. Population Studies Center (2011) Institute for Social Research, University of Michigan , Ann Arbor, Michigan, http://www.psc.isr.umich.edu/dis/ census/Features/tract2zip/index.html Potenza, M. N. (2008). Review. The neurobiology of pathological gambling and drug addiction: an overview and end findings. Philosophical Trans actions of the Royal Society of London B Biological Sciences, 12 3181 318 9. Prochaska, J. O., & DiClemente, C. C. (1982). Transtheoretical therapy: Toward a more integrative model of change. Psychotherapy: Theory, Research and Practice 19 276 288.
138 Prochaska, J. O., & Velicer, W. F. (1997). The transtheoretical model of health behavior change. American Journal of Health Promotion 12 38 48. Prochaska, J. O., Velicer, W. F., Rossi, J. S., Goldstein, M. G., Rakowski, W. et. al. (1994). Stages of change and decisional balance for 12 problem behaviors. Health Psychology, 13 39 4 6. Ravussin, E., & Bogardus, C. (1989). Relationship of genetics, age, and physical fitness to daily energy expenditure and fuel utilization. American Journal of Clinical Nutrition 49 968 975. Recnik, G., Kralj Iglic, V., Iglic, A., et al. (2009). The role of obesity biomechanical constitution of the pelvis and contact joint stress in the progression of hip osteoarthritis. Osteoarthritis Cartilage 17 (7), 879 882. Renzaho, A., Wooden, M., & Houng B. (2010). Associations between body mass index and health related quality of life among Australian adults. Quality Life Research, 19(4), 515 520. Rice, D. P., & Fineman, N. (2004). Economic implications of increased longevity in the United States. Annuals of Review in Public Health. 25 457 473. Riddle, D. L., & Stratford, P. W., (2013). Body weight changes and corresponding changes in pain and function in persons with symptomatic knee osteoarthritis: a cohort study. Arthritis Care and Research 65 (1), 15 22. Rippe, J. M., Price, J. M., Hess, S. A., Klin e, G., DeMers, K. A., Damitz, S., et al. (1998). Improved psychological well being, quality of life, and health practices in moderately overweight women participating in a 12 week structured weight loss program. Obesity Research 6, 208 218. Rocha, B. A., Scearce Levie, K., Lucas, J. J., Hiroi, N., Castanon, N., Crabbe, J. C., Nestler, E. J., & Hen, R. (1998). Increased vulnerability to cocaine in mice lacking the serotonin 1B receptor. Nature, 393 (6681), 175 178. Rosen, C. S. (2000). Integrating stage a nd continuum models to explain processing of exercise messages and exercise initiation among sedentary college students. Health Psychology 19 (2), 172 180. Ross, K. M., Milsom, V. A., Rickel, K. A., Debraganza, N., Gibbons, L. M., Murawski, M. E., & Perri M. G., (2009). The contributions of weight loss and increased physical fitness to improvements in health related quality of life. Eating Behaviors, 10 (2), 84 88. Sarkin, J. A., Johnson, S. S., Prochaska, J. O., & Prochaska, J. M., (2001). Applying the transtheoretical model to regular moderate exercise in an overweight population: validation of a stages of change measure, Pre v entive Medicine, 33 462 469.
139 Schifter, D. E. & Ajzen, I. (1985). Intention, perceived control, and weight loss: an application of the theory of planned behavior. Journal of Personal and Social Psychology, 49 (3), 843 851. Shay, L. E., Seibert, D., Watts, D., Sbrocco, T., & Pagliara, C., (2009). Adherence and weight loss outcomes associated with food exercise diary preference in a military weight management program. Eating Behaviors, 10 (4), 220 227. Shiffman, S., Gnys, M., Richards, T.J., Paty, J.A., Hickox, M., & Kassel, J.D. (1996). Temptations to smoke after quitting: a comparison of lapsers and maintainers. Health Psychology. 15 (6), 455 461. Shiffman, S., Paty, J.A., Gnys, M., Kassel, J.A., & Hickox, M. (1996). First lapses to smoking: within subjects analysis of real time reports. Journal Consult Clinical Psychology, 64 (2), 366 379. Silverman, D., Perakyla, A. & Bor, R. (1992) Discussing safer sex in HIV counseling: assessing three communication formats. AIDS Care, 4 (1), 69 82. Sparks, P., Conner, M., Rhiannon, J., & Shepherd, P. R. (2001). Ambivalence about health related behaviors: An exploration in the domain of food choi ces. British Journal of Health Psychology 6 53 68. Stacy, A. & Greene, PhD., & Paul, G. (2007). Motivational Interviewing Improves Weight Loss in Women with Type 2 Diabetes, Diabetes Care, 30 (5), 1081. Stalonas, P. M., Perri, M. G. & Kerzner, A. B. (19 84). Do behavioral treatments of obesity last? A five year follow up investigation. Addictive Behaviors 9 (2), 175 1 83. Steuart, G. W. (1967). To know is not to do: the problem of health behavioral change. Journal of Occupational Medicine, 9 (12), 601 6 04. Sullivan, P. W., Ghushchyan, V., & Ben Joseph, R. H. (2008).The effect of obesity and cardiometabolic risk factors on expenditures and productivity in the United States Obesity. 16 (9), 2155 2162. Teixeira, P. J., Going, S. B., Houtkooper, L. B., Cussi er, E. C., Metcalfe, L. L., Blew, R. M., Sardina, L. B., & Lohman, T. G. (2004). Pretreatment predictors of attrition and successful weight management in women. International Journal of Obesity and Related Metabolic Disorders, 28 (9), 1124 1133. Thompson, M. M., & Zanna, M. P. (1995). The conflicted individual: personality based and domain specific antecedents of ambivalent social attitudes. Journal of Personality 63 (2), 259 288.
140 ent about (attitudinal) ambivalence. In R.E. Petty & J.A. Krosnick (Eds.) Attitude strength: Antecedents and consequences 4 361 386. Malwah, NJ: Lawrence Erlbaum. Trasande, L., & Chatterjee, S. (2009). The impact of obesity on health service utilization and costs in childhood. Obesity 17 (9), 1749 1754. Tsai, A. G. & Wadden, T. A. (2005). Systematic review: an evaluation of major commercial weight loss programs in the United States. Annals of Internal Medicine, 142 (1), 56 66. Tucker, L. A. & Thomas, K. S. (2009). Increasing Total Fiber Intake Reduces Risk of Weight and Fat Gains in Women. The Journal of Nutrition Nutritional Epidemiology, 139 576 581. Tur, J. J., Escudero, A. J., Romaguera, D., & Burguera, B. (2013). How can we predict which morbidly obese patients will adhere to weight loss prorams based on life style changes? Endocrinology and Nutrition, 60 (6), 297 302. Ul Haq, Z., Mackay, D. F, Fenwick E., & Pell J. P, (2013). Meta analysis of the associations between body mass index and health rela ted quality of life among adults, assessed by the SF 36. Obesity, 21 (3), E322 E327. Ureda, J. R. (1980). The effect of contract witnessing on motivation and weight loss in a weight control program. Health Education Quarterly, 7 (3),163 85. Vallis, M., Rugg iero, L., Greene, G., Jones, H., Zinman, B., Rossi, S., Edwards, L., Rossi, J. S., & Prochaska, J. O. (2003). Relation to demographic, eating related, health care utilization, and psychosocial factors. Diabetes Care 26 1468 1474. van Harreveld, F., van der Pligt, J., & de Liver, Y. N. (2009). The agony of ambivalence and ways to resolve it: introducing the MAID model. Perspectives in Social Psychological Review, 13 (1), 45 61. Visram, S., Crosland, A., & Cording, H. (2009). Triggers for weight gain and loss among participants in a primary care based intervention. British Journal of Community Nursing, 14 (11), 495 501. Volkmar, F. R., Stunkard, A. J., Woolston, J., & Bailey, R. A. (1981). High attrition rates in commercial weight reduction programs Arch ives of Internal Medicine 141 426 428. Wadden, T. A., Foster, G. D., & Letizia, K. A. (1994).One year behavioral treatment of obesity: comparison of moderate and severe caloric restriction and the effects of weight maintenance therapy. Journal of Consult ing Clinical Psychology 62 165 171.
141 Wang ,Y., Beydoun, M. A., Liang, L., Caballero, B., & Kumanyika, S. K. (2008). Will all Americans become overweight or obese? Estimating the progression and cost of the US obesity epidemic. Obesity 16 (10), 2323 2330. Wang, Y., & Beydoun, M. D. (2007). The obesity epidemic in the United States gender, age, socioeconomic, racial/ethnic, and geographic characteristics: a systematic review and meta regression analysis. Epidemiological Review, 29 6 28. Wang, Y. C., Cold itz, G. A. & Kuntz, K. M. (2007). Forecasting the obesity epidemic in the aging U.S. Population. Obesity, 15 (11), 2855 28 65. Ware, J. E., & Sherbourne, C.D. (1992). The MOS 36 Item Short Form Health Survey (SF 36). Medical Care, 30 (6), 473 483. West, D. S., DiLillo, V., Bursac, Z., Gore, S. A. & Greene, P. G. (2007). Motivational interviewing improves weight loss in women with type 2 diabetes. Diabetes Care, 30 (5), 1081 108 7. evisited. American Journal of Public Health 94 (6), 1027 1029. Wilson, P. M., & Rodgers, W.M. (2004). The relationship from perceived autonomy support, exercise regulations, and behavioral intentions in women. Psychology of Sport and Exercise 5 229 242. Wing, R. R., & Phelan, S. (2005). Long term weight loss maintenance. American Journal of Clinical Nutrition 82 222 225. Yeung, T. S., Wessel, J., Stratford, P. & Macdermid, J. (2009). Reliability, validity, and responsiveness of the lower extremity f unctional scale for inpatients of an orthopaedic rehabilitation ward. Journal of Orthopedic Sports Physical Therapy, 39 (6), 468 4 77. Zimmerman, G. L., Olsen, C. G., & approach to helping patients change behavio r. American Family Physician, 61 (5), 1409 1416.
1 42 BIOGRAPHICAL SKETCH Jeffery Don Gilliam received his initial undergraduate education at the University of North Carolina at Greensboro, where he earned a Bachelor of Science d egree in p hysical e ducation with a teaching certification ; and a second Bachelor of Science degree in c ommunity h ealth e ducation He later worked as a f itness t herapist at Charter Hills H ospital and then as an employee health educator at Wesley Long Hospital ( both located in Greensboro, North Carolina ) He then went back to school at East Carolina University w here he earned a Bachelor of Science degree in p hysical t herapy. After practicing physical therapy a number of years Jeff returned to graduate school at the University of Florida (G ainesville, Florida) where he earned a Master of Health Science d egree in physical therapy in the or thopedic track. During the 1990 s Jeff became interested in the association of obesity wit h many of the medical problems he was observing in the physical therapy clinic. After extensive graduate level course work in nutrition, biochemistry and exercise physiology he begin incorpo rating a healthy lifestyle program into his outpatient physical therapy practice. He later designed a cognitive and beh avioral program for adults, adolescents and teens as He is currently clinical director of ReQuest physical therapy and director Choice for We ight L oss, a behavioral program designed to promote a healthy lifestyle and facilitate weight loss. Jeff has concentrated his studies in the department o f health education and behavior at the University of Florida where he researched behavioral changes as patie nts progressed through stages of a healthy lifestyle/weight loss program.
AnEcologicalMomentaryAssessmentofRelapseCrisesinDietingRobertA.Carels,OliviaM.Douglass,HollyM.Cacciapaglia,andWilliamH.OBrienBowlingGreenStateUniversityMuchoftheresearchonrelapsecrisesindietinghasfocusedonisolatedlapseeventsandreliedheavily onretrospectiveself-reportdata.Thepresentstudysoughttoovercometheselimitationsbyusing ecologicalmomentaryassessment(EMA)techniquestoexaminesituationsofdietarytemptationand lapsewithasampleofobese,formerlysedentary,postmenopausalwomen( N 37)duringthefinalweek ofaweight-lossintervention.Moodwasassociatedwithreportsofdietarylapse.Abstinence-violation effectsweremorestronglyassociatedwithdietarylapsesthantemptations.Finally,copingresponses distinguisheddietarytemptationsfromlapses.Educationonthefactorsassociatedwithrelapsecrisesin dietingmaybeimperativeforweightlosssuccessandmaintenance.Fromapublichealthperspective,theincreasedincidenceof obesityisalarmingbecauseobesityisariskfactorforanumberof serioushealthproblems(NationalHeart,Lung,andBloodInstitute ObesityEducationInitiativeTaskForceMembers,1998).Despite thepotentialhealthbenefitsofweight-lossprograms,theseprogramsareoftenplaguedbyhighratesofrelapse(Jefferyetal., 2000).Althoughresearchhasindicatedthatanumberoffactorsare likelytocontributetosuccessfulweightloss(e.g.,self-monitoring; Brownell&Kramer,1989),thecontextual,behavioral,andemotionalfactorsthatcouldplausiblyinfluencerelapsearepoorly understood(NationalHeart,Lung,andBloodInstituteObesity EducationInitiativeTaskForceMembers,1998). Researchinvestigatingpotentialinfluencesondietingrelapse hasindicatedthatcertainfactorssuchasaffect,interpersonal stressorpressure,andspecificactivitiesmayplayakeyrolein temptationandlapseindieting(e.g.,Cummings,Gordon,&Marlatt,1980;Grilo,Shiffman,&Wing,1989;Rosenthal&Marx, 1981).However,earlystudiesreliedondatacollectedduring structuredinterviewsthatoccurred1or2monthsfollowinga weight-lossprogram(Griloetal.,1989;Rosenthal&Marx,1981). Thismethodologicalapproachmaybelimitedbyitsrelianceon retrospectiveself-reportdata,whichcanbeunreliableorbiasedin waysthatcandistortcausalinference.Specifically,mood,past behaviors,preconceivednotions,andself-imagecanbiasthese retrospectiveself-reports(e.g.,seeStone&Shiffman,1994).Also, pastresearchhastendedtofocusonasingletemptationorlapse episode,whichmayormaynotberepresentativeofotherrelapse crises(Griloetal.,1989).Finally,withoutcomparinganindividualsdietaryrelapsestosimilarconditionswhenthereisnotdietary relapse,itisnotpossibletoconcludethatfactorsassociatedwitha lapsearenotoccurringatthesamefrequencythroughouttheday. Afewpublishedstudiesofrelapsecrisesindietershaveattemptedtominimizerelianceonlengthyretrospectiveself-report dataandsimultaneouslyexaminemultiplerelapsecrises(Carelset al.,2001;Greeno,Wing,&Shiffman,2000;Johnson,Schlundt, Barclay,Carr-Nangle,&Engler,1995;Schlundt,Sbrocco,&Bell, 1989;Schlundtetal.,1990).Usingecologicalmomentaryassessment(EMA;i.e.,repeated,real-timeassessmentsinparticipants typicalenvironment;Stone&Shiffman,1994)toassessdietary relapsecrisesin30overweightcollegedieters,wecompared dietarytemptationandlapsesituationswithmomentsofminimal dietaryconsequence(i.e.,randomassessment)acrossanumberof contextual,behavioral,andemotionalfactors(Carelsetal.,2001). Carelsetal.(2001)determinedthat,comparedwithmomentsof minimaldietaryconsequence,increasednegativemoodandengagementinspecificactivities(e.g.,socializing,interpersonalconflict)tendedtopromotetemptationandlapse.Also,dietersreportedfeelinglessconfidentintheirabilitytomaintaintheirdiet afterdietarylapseswhencomparedwithmomentsofminimal dietaryconsequenceandtemptation.However,thisinvestigation didnotexamineotherimportantfactors,suchascopingresponses, thatcouldbeassociatedwithtemptationandlapseoutcomes. Additionally,thecollegesamplewasnotengagedinaformal weight-lossprogram. DiaryresearchbyGreenoetal.(2000)andSchlundtetal.(1989, 1990)hasalsoyieldedconsistentfindingsregardingtheroleof affectiveandsituationalfactorsinovereating.Inastudyofobese womenwithbingeeatingdisorder,Greenoetal.(2000)foundthat negativemood,lowalertness,feelingsofpooreatingcontrol,and cravingsforsweetsprecededbingeepisodes.Pooreatingcontrol andcravingsweetsalsopredictedbingesinwomenwithoutbinge eatingdisorder.Similarly,inresearchwithobeseindividualsparticipatinginabehavioralweight-lossprogram,Schlundtetal. (1989)foundindividualstendedtoovereatoreatunplannedmeals inresponsetothreesituations:positivesocialinteractions,negativeemotions,andphysiologicalcravings.Inaseparatestudy usingclusteranalysistoidentifygroupsofobesewomenonthe basisofeatingpatterns,emotionalovereaterswerealsoidentified asadistinctgroup(Schlundtetal.,1990).Thesediarystudies providedinsightintotheroleofmood,activities,andcontexton lapsesbutdidnotinvestigatetheroleofotherpotentiallyimportant factors,suchascopingandabstinence-violationeffects.Also, thesestudiesdidnotinvestigatedifferencesbetweensituationsin whichtheparticipantssuccessfullyavoidedalapse(i.e.,temptation)relativetoalapseoccurrence. Thepresentstudyexaminedrelapsecrisesamongasampleof obese,postmenopausalwomenduringthefinalweekofabehavRobertA.Carels,OliviaM.Douglass,HollyM.Cacciapaglia,andWilliam H.OBrien,DepartmentofPsychology,BowlingGreenStateUniversity. CorrespondenceconcerningthisarticleshouldbeaddressedtoRobert A.Carels,DepartmentofPsychology,BowlingGreenStateUniversity, BowlingGreen,OH43403.E-mail:firstname.lastname@example.orgJournalofConsultingandClinicalPsychology Copyright2004bytheAmericanPsychologicalAssociation 2004,Vol.72,No.2,341348 0022-006X/04/$12.00DOI:10.1037/0022-006X.72.2.341341
ioralweight-lossintervention.WeusedEMAtechniquestocomparetemptationsandlapseswithinstancesofminimaldietary consequenceacrossnumerousrelapsecrisisevents.EMAisa methodologicaltechniqueusedtocollectrepeatedmeasurements ofaphenomenonasitoccursinnaturalisticsettings(Stone& Shiffman,1994).Anumberofresearchinvestigationshaveused EMAwithavarietyofpatientpopulations(e.g.,smokers:Shiffman,Paty,Gnys,Kassel,&Hickcox,1996;rheumatoidarthritis: Stone,Broderick,Porter,&Kaell,1997;fibromyalgia:Afflecket al.,2001).Amajorfocusofthisinvestigationwastoidentify antecedentsandconsequencesofrelapsecrisisbyusingawithinpersondesignacrossseveraltemptationandlapseeventsinwomen completingabehavioralweight-lossprogram.Additionally,we wantedtocomparemomentsoftemptationandlapseonimportant, yetunderstudied,factors,suchascopingresponseandabstinenceviolationeffects.Method ParticipantsParticipantswere37obese,formerlysedentary,postmenopausalwomen inthefinalweekofaweight-lossintervention(Table1).Participantswere recruitedthroughlocaladvertisements(e.g.,newspaper)andfliers(e.g., distributedatwomen shealthclinics,hospitals).Womenwereincludedin theinvestigationiftheywere(a)postmenopausal(nomenstruationforat least12months),(b)obese(bodymassindex[BMI] 30kg/m2),(c) sedentary(notparticipatinginaprogramofphysicalconditioningtwoor moretimesperweek),(d)willingtoacceptrandomassignment,and(e) nonsmokers.Participantswereexcludedfromparticipationiftheyhad(a) pastorcurrentcardiovasculardisease(e.g.,myocardialinfarction),(b)surgicallyinducedmenopausewithintheprevious6months,(c)musculoskeletal problemsthatwouldpreventparticipationinmoderatelevelsofphysical activity(e.g.,osteoporosis),(d)ahistoryofinsulin-dependentdiabetes(selfreported),(e)restingbloodpressuregreaterthanorequalto160/100mmHg (assessedduringscreening),or(f)alife-limitingorcomplicatedillness.ProcedureParticipantsunderwenta24-sessionweight-lossinterventionbasedon theLEARNprogram(Brownell,2000).Duringthefinalweekofthe intervention,womencompletedEMAdiaries.Womenwereprovidedwith explanationsforthetemptation,lapse,andrandom-promptsectionsofthe diary.Womenwerealsoprovidedwithinstructionsforcompletingeach questionwithinthediaries.Participantsmadeseveralpracticediaryentries. Researchpersonnelfromtheinvestigationwereavailabletoanswerquestionsthroughouttheweekofdiaryrecording.Appropriatetimestocompleteentriesforeachsectionwerediscussed(i.e.,withinthefirst15minof atemptationorlapse;rightafterbeingrandomlypaged),andtheimportanceofcompletingentries inthemoment wasstressedthroughout training.Participantscarriedthediaries,describedabove,alongwithpagers(forthepurposeofrandompromptsfourtimesaday)for1week. FollowingtheweekofEMArecording,participantscompletedacomplianceandreactivityquestionnaire.InterventionTheLEARNprogramisacomprehensive,empiricallysupported, lifestyle-changeapproachtoweightmanagementandphysicalactivity (Brownell,2000)andhasfivecomponents:lifestyle,exercise,attitudes, relationships,andnutrition.Itisdesignedtoachievegradualweightloss, increasedphysicalactivity,andaprogressivedecreaseinenergyandfat intakethroughpermanentlifestylechanges.Theprogramemphasizes(a) self-monitoringofeatingbehavior,(b)controllingstimuliassociatedwith eating,(c)physicalactivity,(d)nutritioneducation,(e)modifyingselfdefeatingthoughtsandemotionsassociatedwithdietingandbodyimage, (f)settingrealisticgoals,(g)relationships,(h)relapsepreventionand weightmaintenance,and(i)preventing,copingwith,andgainingcontrolof temptationsandlapses.Onehalfofthewomenwererandomlyassignedto receivetheweight-lossinterventionalone,whereastheotherhalfreceived theweight-lossprogramaswellasself-controlskillstraining.TheselfcontrolskillstrainingwasbasedonBaumeister sself-controltheory (Baumeister,Heatherton,&Tice,1994).Inthisstudy,thewomenlost,on average,6.2kg( SD 4.3),theequivalentto2.4BMIunits( SD 1.7).MeasuresApaper-and-pencilEMAdiarywasusedtoexaminetemptationand lapseindieting(Carelsetal.,2001).Separatesectionswerecreatedfor temptationandlapsesituations,aswellasrandomprompts. Temptationandlapseentries. Participantswerefirstrequiredtorecord thedateandtimeofatemptationorlapseentrywhenitoccurred.Inthe caseoftemptations,participantsthenratedtheintensityoftheirtemptation usinga5-pointscale(1 notatall, 2 alittle, 3 somewhat, 4 very, and5 extremely ).Temptationsweredefinedas, Asuddenurgetobreak yourdiet(e.g.,overeat,eataforbiddenfood)inwhichyoufeltyouhad comeclosetothebrinkofbreakingyourdiet. Lapsesweredefinedas, Anincidentwhereyoufeltthatyoubrokeyourdiet(e.g.,overate,atea forbiddenfood). Participantsthencompleteditemsregardingconsumptiveactivitiesthat precededthetemptationorlapse.Itemsassessingconsumptiveactivities includedhungerratings,levelofsatietyafterprioreatingepisode,andlevel ofsatisfactionwithprioreatingepisode(eachratedonthe5-pointscale describedabove);timeoflasteatingepisodeandlengthoftimehungry (timeformat);andquantityofprioreatingepisode(checkone:snack, meal).Itemsalsoassessedparticipants location(home,work,orother), presenceofothers( yes or no ),activities(e.g.,cooking[forself/for others],shoppingforfood,conflictwithothers,eatingwithothers,exercising,eatingout,eatingameal,eatingasnack,watchingtelevision, socializing/attendingaparty,reading,orother),andmood(e.g.,frustrated, nervous,bored,content,restless,incontrol,sad,happy,stressed,tired, relaxed,andother),precedingthetemptationorlapse. Copingduringthetemptationorlapsewasassessedusing14items,rated ona4-pointscale(1 Ididnotdothis ;2 Ididthisalittle ;3 Idid thisamediumamount ;4 Ididthisalot ).Participantswereaskedto rate yourattemptstocope,ifany,duringthetemptationorlapse usingthe followingitems: removedmyselffromthesituation, distractedmyself, talkedtoagroupmemberforadviceorcomfort, talkedtoafamily memberforadviceorcomfort, talkedtoafriendforadviceorcomfort, encouragedmyself, meditated/relaxed, engagedinspiritualactivities, exercised, thoughtaboutthebenefitsassociatedwithdieting, thoughtaboutthebenefitsassociatedwithbeinghealthy, thoughtabout thenegativesassociatedwithnotdieting, thoughtaboutthenegatives associatedwithbeingunhealthy, and other. Also,copingresponses weresummedtocomputeatotalcopingscoreduringeachtemptationand lapseevent.Table1 DemographicCharacteristics Demographics n % MSD Income $30,0001540.5 Collegedegree2054.1 Caucasian3594.5 Workingfull/parttime3389.1 Age(years)54.77.9 Baselineweight(lbs)212.035.2 BaselineBMI(kg/m2) 36.45.5 Note. BMI bodymassindex.342CARELS,DOUGLASS,CACCIAPAGLIA,ANDO BRIEN
Abstinence-violationeffectswereassessedbyrequestingparticipantsto indicatetheirlevelofagreementwiththefollowingeightstatements,rated ona5-pointscale(i.e.,1 stronglydisagree, 2 disagree, 3 neutral, 4 agree, 5 stronglyagree ).Participantswereaskedto characterize yourreactiontothetemptationorlapse withthefollowingitems:(a) Iam unlikelytobetemptedorlapseagain, (b) Iknowmydietwillbe successful, (c) Iamworriedaboutmaintainingmydiet, (d) Ifeellike Ifailedmydiet, and(e) Ifeelguiltybecauseofmytemptationorlapse. Inaddition,participantswereaskedto ratethefollowingstatementregardingyourtemptation/lapse withthefollowingthreeitems:(a) Iam responsibleforthetemptation/lapse, (b) IcancontrolwhatIeatinthe future, and(c) Ihavewillpower. Randomprompts. Aswithalldiaryentries,participantsrecordedthe dateandtimeoftheirentry.Inaddition,participantswereaskedifthiswas alsoatemptationand/orlapse( yes or no ).Thewords temptation / lapse werereplacedbytheword prompt asrequired. Complianceandreactivity. FollowingtheweekofEMArecording, participantscompletedacomplianceandreactivityquestionnaire.Compliancewiththediarywasratedona6-pointscale(1 none, 2 once duringtheweek, 3 severaltimesduringtheweek, 4 aboutonceaday, 5 25timesaday, 6 greaterthan5timesaday ).Participantswere asked Howmanydiaryentriesdidyoumissorskip? and Howmany diaryentriesdidyoucompletelongerthan15minafterthelapse,temptation,orprompt? Reactivity(i.e.,modificationoftypicalbehaviorin responsetothedata-collectionmodality)wasassessedwithfourquestions. Threequestionswereratedona5-pointscale(1 stronglydisagree, 2 disagree, 3 neutral, 4 agree, 5 stronglyagree ).Participants respondedtothestatement comparedtoyournormaldietingroutine, (a) Iwasmorelikelytobetemptedwhilekeepingthediary, (b) Iwasmore likelytolapsewhilekeepingthediary, and(c) Iwasmoreawareofmy behaviorwhilekeepingadiary. Onequestion(1 notatall, 2 alittle, 3 somewhat, 4 very, 5 extremely )askedparticipants Didkeeping adiaryinfluenceyoureatingbehaviors? DataAnalysesPriorresearchsuggeststhatthemooditemsmayloadonpositiveand negativemoodfactors(Matthews,Owens,Allen,&Stoney,1992).Therefore,thepositiveandnegativemooditemsweresubmittedtoafactor analysisusingavarimaxrotationprocedure.Onefactoremergedwithhigh positiveloadingsforrelaxed,happy,content,andincontrol(i.e.,positive mood; .84).Asecondfactoremergedwithhighpositiveloadingsfor sad,nervous,frustrated,andrestless(i.e.,negativemood; .80). Stressedandtireddidnotpositivelyloadoneitherfactor.Therefore, positiveandnegativemooditemswerecombinedtoformglobalpositive andnegativemoodvariables. Theadditionofself-control-skillstraininghadnosignificanteffectson anyinterventionoutcomes,includingthenumberoflapsesandtemptations.Therefore,datafrombothtreatmentgroupswerecombined.Generalizedestimatingequations(GEE;SASInstitute,1997)wereusedto comparesituationsofminimaldietaryconsequence(i.e.,randomprompts) withdietarytemptationsandlapsesonrecentconsumptiveactivities, presenceofothers,location,activities,mood,abstinence-violationeffects, andcoping(lapsesvs.temptationsonly).GEEswerealsousedtocompare dietarytemptationsanddietarylapses.GEEallowsforvaryingnumbersof observationsperparticipant,whilecontrollingforautocorrelation(Zeger, Liang,&Albert,1988).GEEsuseamultiple-step,maximum-likelihood approachtoestimationandtesting.Anessentialfeatureofthistechniqueis thatitispossibletorecognizethatEMAdatahavetworandomcomponents:oneduetothesamplingofpersonsandtheotherduetothesampling ofrepeatedmeasurementswithinpersons. Temptations,lapses,andrandompromptswereincludedasindependent dichotomousvariables.Recentconsumptiveactivities,presenceofothers, location,activities,mood,abstinence-violationeffects,andcopingwere includedasdependentvariables.Analphaof.05wasadjustedwiththe Benjamini Hochbergprocedureforcontrollingthefalse-positiveratein multiplecomparisons(Benjamini&Hochberg,1995;Thissen,Steinberg, &Kuang,2002;Williams,Jones,&Tukey,1999).UsingtheBenjamini Hochbergprocedure,theadjustedalphais.006.Results FrequencyofTemptations,Lapses,andRandomPromptsOnaverage,participantsreported2.7lapses( SD 1.9;range 0 10),3.0temptations( SD 2.8;range 0 13),and19.1 randomprompts( SD 4.7;range 10 28)overthe7-day recordingperiod.Six(16.2%)participantsexperienced0lapses,19 (51.4%)participantsexperienced1 2lapses,and12(32.4%)participantsexperiencedgreaterthantwolapsesduringtherecording period.1Ten(27.1%)participantsexperienced0temptations,14 (37.8%)participantsexperienced1 2temptations,and13(35.1%) participantsexperiencedgreaterthan2temptationsduringthe recordingperiod.Table2liststhemeansandstandarddeviations forallvariablesacrosslapses,temptations,andsituationsofminimaldietaryconsequence.ComplianceandReactivitySeventypercentoftheparticipantsreportedskippinglessthan twodiaryentriesduringtheweek,and53%reportedcompleting lessthantwodiaryentriesgreaterthan15minafteralapse, temptation,orprompt.Eighty-eightpercentoftheparticipants disagreedorstronglydisagreedwiththestatement Iwasmore likelytobetemptedwhilekeepingadiary, whereas91%disagreedorstronglydisagreedwiththestatement Iwasmorelikely tolapsewhilekeepingadiary. Only21%ofparticipantsendorsed Very or Extremely tothequestion Didkeepingadiary influenceyoureatingbehaviors? However,77%oftheparticipantsagreedorstronglyagreedwiththestatement Iwasmore awareofmybehaviorwhilekeepingadiary. ComparisonofTemptations,Lapses,andRandom PromptsRecentconsumptiveactivity,location,presenceofothers,mood, andactivity. Therewerenosignificantdifferencesinrecent consumptiveactivitiesorlocationbetweenmomentsoftemptation, lapse,andminimaldietaryconsequence(seeTables3,4,and5). Comparedwithmomentsofminimaldietaryconsequence,lapses weresignificantlyassociatedwithreportinggreaterpositivemood ( p .006)andsignificantlyassociatedwithreportinggreater negativemood( p .006;Table4).Comparedwithmomentsof minimaldietaryconsequence,lapsesandtemptationswereless likelytooccurwhileexercising( p .006;Tables3and4).There werenoothersignificantdifferencesinreportedactivitiesbetweenmomentsoftemptationandlapseorminimaldietary consequences. 1Participantswererandomlypagedfourtimeseachday.However,many participantsreportedthattheydidnotreceivefourpageseachday.Some pagersdidnotappeartobefunctioningreliably.Additionally,someparticipantsmayhave,attimes,beenoutsideofthepager sgeographical range.Althoughthemajorityofparticipantsreportedrespondingtoallof thepagesthattheyreceived,participantsonlyrespondedto19.1( SD 4.7)of28potentialprompts(68.2%).Giventhesedifficulties,weare limitedinourabilitytoobjectivelyassesscompliancewithrandompages.343RELAPSECRISESINDIETING
Coping. Comparedwithlapses,temptationswereassociated withgreateruseofthecopingstrategies:removingoneselffrom thesituation( p .006),distractingoneself( p .006),and encouragingoneself( p .006).Comparedwithlapses,temptationswerealsoassociatedwithmoreuseofthecopingstrategiesof thinkingofthebenefitsofdieting( p .006),benefitsofbeing healthy( p .006),negativesofnotdieting( p .006),and negativesofbeingunhealthy( p .006).Inaddition,temptations wereassociatedwithahighertotalcopingscore(indicatinggreater copingeffort)thanlapses( p .006;Table5). Abstinence-violationeffects. Comparedwithmomentsofminimaldietaryconsequence,temptationswereassociatedwithfeelingmorelikelytobetemptedormorelikelytolapseagain( p .006;Table3).Comparedwithmomentsofminimaldietaryconsequence,lapseswereassociatedwithfeeling(a)morelikelytobe temptedortolapseagain,(b)lessconfidentthedietwillbea success,(c)lesswillpower,(d)agreatersenseoffailureregarding thediet,and(e)moreguilt( p .006;Table4).Comparedwith momentsoftemptation,lapseswereassociatedwithfeelingasense offailureregardingthediet,moreguiltbecauseofthetemptation Table2 Mean(andStandardDeviation)ofVariablesDuringLapse,Temptation,andRandomPrompts VariableLapseTemptationRandomprompt Consumptiveactivity Howlongagodidyoulasteat(inminutes)?146.1(122.04)180.2(112.67)126.6(57.11) Howsatisfiedafterlastmeal?3.4(0.75)3.4(0.70)3.6(0.54) Howfullafterlastmeal?3.4(0.75)3.4(0.76)3.4(0.51) Howlongwereyouhungry(inminutes)?96.3(87.65)81.5(112.08)77.7(70.90) Howhungry?2.4(1.03)2.9(0.64)2.6(0.92) Typeoffood:Meal(%oftotalentries)83.1(32.5)85.9(40.3)77.8(17.3) Location(%ofentries) Work21.8(31.77)30.0(38.12)23.6(20.18) Home46.4(40.72)32.5(35.16)46.2(23.16) Other31.9(37.45)37.5(39.90)30.3(17.50) Mood Positive10.9(3.24)11.3(4.10)9.3(1.87) Negative7.7(3.08)7.0(3.10)6.1(1.67) Activity(%oftotalentries) Cooking0.8(4.49)3.7(19.25)3.5(5.80) Shopping8.9(22.87)9.3(27.86)2.8(4.13) Inconflict1.6(8.98)6.2(20.75)0.8(2.71) Eatingwithothers3.0(10.69)4.2(19.34)2.3(3.78) Exercise0.0(0.00)0.0(0.00)2.3(3.58) Eatingout0.0(0.00)0.9(4.81)2.1(3.52) Eatingameal6.6(15.45)9.3(27.86)5.9(7.87) Eatingasnack5.4(14.52)1.2(6.42)1.9(4.00) WatchingTV18.5(29.47)9.6(16.55)8.9(9.93) Socializing/attendingaparty17.8(31.18)16.3(30.23)6.1(6.63) Reading5.9(19.43)7.8(26.65)6.7(7.21) Otherspresent61.2(43.67)56.4(40.42)50.5(26.13) Coping Removedmyselffromsituation1.3(0.49)2.4(1.13) Distractedmyself1.4(0.50)2.7(1.00) Talkedtoagroupmember1.1(0.20)1.0(0.20) Talkedtoafamilymember1.1(0.29)1.1(0.30) Talkedtoafriend1.1(0.20)1.0(0.02) Encouragedmyself1.5(0.49)2.9(0.83) Meditated/relaxed1.1(0.30)1.4(0.55) Engagedinspiritualactivities1.0(0.00)1.1(0.33) Exercised1.1(0.27)1.2(0.50) Thoughtaboutthefollowing Benefitsofdieting1.5(0.66)3.0(0.90) Benefitsofbeinghealthy1.5(0.67)2.8(1.01) Negativesofnotdieting1.5(0.55)2.1(0.97) Negativesofbeingunhealthy1.3(0.48)2.1(1.23) Coping(total)14.6(2.94)20.6(5.88) Abstinence-violationeffect Unlikelytobetempted/tolapse2.0(0.74)2.2(1.03)2.7(0.92) Dietwillbeasuccess3.3(0.54)3.5(0.60)3.6(0.54) Worriedaboutmaintainingdiet3.2(0.87)3.0(0.90)2.8(0.95) FeellikeIfailedmydiet2.8(1.01)1.8(0.84)2.0(0.90) Feelguiltyaboutlapse/temptation3.2(0.89)1.9(0.83)2.0(0.91) Ihavewillpower3.2(0.81)3.6(0.68)3.5(0.64) I mresponsibleforlapse/temptation4.3(0.51)3.7(0.94) Icancontrolfutureeating3.3(0.65)3.8(0.62)344CARELS,DOUGLASS,CACCIAPAGLIA,ANDO BRIEN
orlapse,moreresponsibilityforthetemptationorlapse,lessable tocontroleatinginthefuture,andlesswillpower( p .006;Table5).DiscussionUsingEMAwithwomencompletinga6-monthbehavioral weight-lossprogram,weexaminedinthepresentstudycognitive, behavioral,emotional,andcontextualfactorsthatcouldplausibly precedeandfollowtemptationsandlapses.Althoughwomencompletingabehavioralweight-lossprogramarelikelytohaveacquiredanumberofskillsforcopingwithdietarytemptationand lapse,thewomenneverthelessevidencedconsiderablevariability intheirresponsestothesedietaryevents.Forexample,moodand abstinence-violationeffectsweresignificantlyassociatedwithreportsoflapses.Copingwasmorestronglyassociatedwithdietary temptationsthanlapses. Thisinvestigationconfirmsandextendspriorresearchexaminingrelapsecrisesamongdieters;however,someincongruitieswith priorresearchwerenoted.Consistentwithpreviousresearchexaminingrelapsecrisesindieters(Greenoetal.,2000;Johnsonet al.,1995;Schlundtetal.,1989;Schlundtetal.,1990),greaterpositive andnegativemoodstateswereassociatedwithagreaterlikelihoodof experiencinglapses.Positiveandnegativemoods,however,didnot distinguishtemptationsfromlapsesorrandomprompts. Nospecificlocationincreasedthelikelihoodofexperiencinga temptationorlapse.Inpriorresearchwithdieterstryingtolose weightwithouttheaidofaformalprogram(Carelsetal.,2001) andwithindividualsdiagnosedwithbingeeatingdisorder(Johnsonetal.,1995),lapsesorovereatingwasmorelikelytooccurat home.Becausethisassessmentwasperformednearthecompletion ofaweight-lossprogram,womenmayhavesuccessfullylimited theavailabilityoftemptingsnacksandfoodathome. Inthisinvestigation,activitydidnotincreasethelikelihoodof experiencingatemptationorlapse.Inpriorresearch,Griloetal. (1989)foundthatobese,TypeIIdiabeticsenrolledinauniversitybasedweight-lossprogramweresignificantlymorelikelytoexperiencearelapseduringsocialinteractions,whileeating,orwhile watchingtelevision.Again,womencompletingaweight-lossprogrammayhavesuccessfullymonitoredfoodintakewhileeating, watchingtelevision,orsocializing. Temptationsandlapsesareoftenassociatedwithhunger(Carels etal.,2001;Griloetal.,1989).Inthisinvestigation,therewasno associationbetweenhungerandlapses.Again,thesewomenmay havebeenactivelymonitoringdailyfoodintakeinordertominTable3 EffectsofRecentConsumptiveActivity,Location,Mood,Activity,andAbstinence-Violation Effects,ComparingTimesofMinimalDietaryConsequenceWithTimesofTemptation VariableEstimate SEzp Recentconsumptiveactivity Howlongagodidyoulasteat(inseconds)? 475.211,911.05 0.25.804 Howsatisfiedafterlastmeal?0.100.081.22.221 Howfullafterlastmeal?0.010.090.09.925 Howlongwereyouhungry(inseconds)? 2,333.802,489.09 0.94.348 Howhungry? 0.230.12 1.91.056 Snackvs.meal 0.030.05 0.69.488 Location(%oftotalentries) Work 0.090.06 1.58.114 Home0.100.061.72.085 Other 0.010.07 0.12.903 Otherspresent 0.010.06 0.10.917 Mood Positive1.330.552.43.015 Negative 1.010.39 2.57.010 Activity(%oftotalentries) Cooking0.010.011.16.247 Shopping 0.010.02 0.38.702 Inconflict 0.050.04 1.32.185 Eatingwithothers 0.000.02 0.08.938 Exercise0.020.013.89.000* Eatingout0.010.010.60.551 Eatingameal0.010.030.48.633 Eatingasnack0.010.020.58.565 WatchingTV 0.040.03 1.35.177 Socializing/attendingaparty 0.090.03 2.62.009 Reading0.010.030.45.655 Abstinence-violationeffect Unlikelytobetemptedortolapse0.280.102.74.006* Dietwillbeasuccess0.040.070.61.539 Worriedaboutmaintainingdiet 0.120.06 1.98.048 FeellikeIfailedmydiet 0.030.10 0.30.761 Feelguiltyaboutlapse/temptation 0.090.14 0.64.522 Ihavewillpower 0.110.07 1.52.128 p .006.345RELAPSECRISESINDIETING
imizetemptationsandlapsesand/ormayhavegrownaccustomed tomoderatelevelsofhunger. Toourknowledge,nopreviousstudieshaveusedEMAto examinetheenactmentofcopingduringdietarytemptationsand lapses.Inallcircumstances,dietarylapseswereassociatedwith diminishedcoping.Forexample,comparedwithdietarylapses, temptationswereassociatedwithgreatercognitivecoping(e.g., encouragingoneselforremindingoneselfaboutthebenefitsassociatedwithdieting)andgreaterbehavioralcoping(e.g.,leavingthe situationordistractingoneself).Infact,comparedwithtemptations,dietarylapseswereassociatedwithengaginginfewercoping strategiesoverall.Theonlycopingcategoriesthatdidnotdistinguishdietarytemptationsfromlapsesweresocialsupport,meditationorrelaxation,andengagementinspiritualactivities.Our findingsarequiteconsistentwithresearchexaminingrelapsecrisesindieters(Griloetal.,1989;Grilo,Shiffman,&Wing,1993) andinsmokers(Shiffmanetal.,1996).Forexample,Griloetal. foundthatperformanceofcopingwasastrongcorrelateofoutcomeindietaryrelapsecrisis,andShiffmanetal.foundthat participantswere12timeslesslikelytoreportcopingduring smokinglapsesthanduringtemptations. Inthisinvestigation,comparedwithdietarytemptationsand momentsofminimaldietaryconsequence,abstinence-violation effectsweresignificantlygreaterfollowingdietarylapses.Adietarylapsemayhavecontributedtotherealizationthatdietary lapsesareformidable,likelytooccurinthefuture,andachallenge tosuccessfullongertermmaintenance.Examiningmomentsof diminishedself-efficacyandlong-termdietaryadherencemaybe animportantareaofresearchforfutureinvestigations. Inthisinvestigation,dietarylapseswerelimitedinnumber ( M 2.7, SD 1.9)andreflectedsubjectiveexperiencesonly. Futureresearchmightbenefitfromalongerdiary-recordingperiod,giventhatthesmallnumberoflapsesreportedbyparticipants maylimitthegeneralizabilityofthefindings.Inthisinvestigation, alapsewasdefinedas anincidentwhereyoufeltthatyoubroke yourdiet. Dietersneededtoonlybelievethattheyhadbroken theirdiet.Despitesomesubtledifferences,thereappearstobe considerablesimilarityinthesubjectiveandobjectivedeterminantsoflapsewithdieters(Wing,Shiffman,Drapkin,&McDermott,1995).Whetherobjectiveorsubjective,lapsesappeartobe importanteventsfromthedieter sperspective. AswithotherresearchusingEMAtechniques,thisinvestigation issusceptibletononcomplianceandexperimentalreactivity.For example,inEMAresearchusingadebriefinginterviewwith treatedalcoholicstoassesscompliance,delayedoromittedeventcontingentandsignal-contingentrecordingwascommon(Litt, Table4 EffectsofRecentConsumptiveActivity,Location,Mood,GeneralActivity,andAbstinenceViolationEffects,ComparingTimesofMinimalDietaryConsequenceWithTimesofLapse VariableEstimate SEzp Recentconsumptiveactivity Howlongagodidyoulasteat(inseconds)? 1,194.091,160.97 1.03.304 Howsatisfiedafterlastmeal?0.090.091.05.294 Howfullafterlastmeal? 0.010.09 0.14.892 Howlongwereyouhungry(inseconds)? 1,164.48673.20 1.73.084 Howhungry? 0.10 0.13 0.79.428 Snackormeal0.060.051.16.247 Location(%oftotalentries) Work0.020.050.43.670 Home 0.050.07 0.70.482 Other0.020.080.24.812 Otherspresent(yes)0.150.062.34.020 Mood Positive0.820.213.92.000* Negative 1.400.33 4.29.000* Activity(%oftotalentries) Cooking 0.020.01 1.45.147 Shopping0.030.031.23.218 Inconflict0.000.010.31.758 Eatingwithothers0.010.020.70.458 Exercise0.020.013.89.000* Eatingout 0.020.01 0.31.756 Eatingameal0.020.030.82.411 Eatingasnack0.030.021.46.144 WatchingTV0.120.042.67.008 Socializing/attendingaparty0.110.042.58.009 Reading0.010.040.26.797 Abstinence-violationeffect Unlikelytobetemptedortolapse 0.410.10 4.17.000* Dietwillbeasuccess 0.220.06 3.54.000* Worriedaboutmaintainingdiet0.180.082.17.030 FeellikeIfailedmydiet0.950.156.27.000* Feelguiltyaboutlapse/temptation1.270.148.97.000* Ihavewillpower 0.280.09 3.21.001* p .006.346CARELS,DOUGLASS,CACCIAPAGLIA,ANDO BRIEN
Cooney,&Morse,1998).Similarly,paper-and-pencilEMAmethodsappeartobemoresusceptibletodelayedreportingthanmore sophisticatedhandheldcomputers(Stone,Shiffman,Schwartz, Broderick,&Hufford,2002).Inourstudy,aboutonehalfofthe participantsreportedcompletingmorethanonediaryentryin excessof15minaftertheevent.Althoughthisnumberisrelatively smallcomparedwiththeoverallnumberofdiaryentries,some elementofretrospectiverecallbiasisintroduced.Likewise,althoughonly21%oftheparticipantsstatedthatkeepingadiary influencedtheireatingbehaviors,77%oftheparticipantsstated thattheyweremoreawareoftheirbehavior.Thisincreasedawarenessofbehaviorandfactorsthatinfluencebehaviorisawellknownphenomenoninself-monitoringresearch(Haynes& O Brien,1999).Thereactivityofself-monitoringmayresultin participantsreportingfewerlapseepisodesbecausethediary wouldtendtocuethemtouseadaptivecoping.Self-monitoring Table5 EffectsofRecentConsumptiveActivity,Location,Mood,GeneralActivity,andAbstinenceViolationEffects,ComparingTimesofLapseWithTimesofTemptation VariableEstimate SEzp Recentconsumptiveactivity Howlongagodidyoulasteat(inseconds)?613.592,249.310.27.785 Howsatisfiedafterlastmeal? 0.010.13 0.05.964 Howfullafterlastmeal?0.000.120.01.990 Howlongwereyouhungry(inseconds)? 144.792,635.82 0.05.956 Howhungry? 0.290.15 1.89.059 Snackormeal 0.010.06 0.20.838 Location(%oftotalentries) Work 0.110.08 1.42.156 Home0.110.081.39.164 Other0.000.090.05.959 Otherspresent 0.180.08 2.31.021 Mood Positive0.150.580.26.791 Negative0.580.481.22.224 Activity(%oftotalentries) Cooking 0.000.02 0.02.987 Shopping0.020.040.64.523 Inconflict 0.060.05 1.17.243 Eatingwithothers0.010.030.45.650 Exercise 0.130.11 1.17.241 Eatingout 0.010.01 1.01.310 Eatingameal0.040.031.36.173 Eatingasnack0.040.021.69.092 WatchingTV0.070.051.51.132 Socializing/attendingaparty0.020.060.35.727 Reading0.010.040.15.884 Coping Removedmyselffromsituation 0.950.21 4.51.000* Distractedmyself 1.270.19 6.57.000* Talkedtoagroupmember0.010.020.57.567 Talkedtoafamilymember 0.020.05 0.35.727 Talkedtoafriend0.040.031.16.244 Encouragedmyself 1.200.17 6.92.000* Meditated/relaxed 0.260.10 2.48.013 Engagedinspiritualactivities 0.090.06 1.63.104 Exercised 0.130.11 1.17.241 Thoughtaboutthefollowing Benefitsofdieting 1.200.16 7.54.000* Benefitsofbeinghealthy 0.950.21 4.59.000* Negativesofnotdieting 0.630.18 3.49.001* Negativesofbeingunhealthy 0.630.21 2.95.003* Coping(total) 5.030.81 6.18.000* Abstinence-violationeffects Unlikelytobetemptedortolapse 0.180.15 1.18.236 Dietwillbeasuccess 0.230.08 2.71.007 Worriedaboutmaintainingdiet0.070.070.90.370 FeellikeIfailedmydiet1.080.176.53.000* Feelguiltyaboutlapse/temptation1.330.1210.95.000* Iamresponsiblefortemptation/lapse0.560.202.85.004* IcancontrolwhatIeatinfuture 0.350.11 3.20.001* Ihavewillpower 0.410.08 5.20.000* p .006.347RELAPSECRISESINDIETING
mayalsotendtoincreaseparticipantawarenessofthefactorsthat precede,co-occur,orfollowdietaryevents.Ofcourse,participants mayhavelimitedawarenessoftheirreactivity,particularlyfor subtlechangesintheireatingpatterns. Priorresearchontemptationandlapseindietinghasbeen limitedbyitsrelianceonretrospectiveself-reportandsingularlapsesituations.Therefore,thestrengthofthecurrentinvestigation isitsinclusionofreal-time,ecologicallyvalidassessmentsof temptationandlapseindieting(includingmomentsofminimal dietaryconsequence).Theresultsfromthisinvestigationprovide evidencethatmomentaryfactorshaveasubstantialinfluenceon theexperienceoftemptationandlapseindieters.However,wedo notknowwhetherthenumberofcontextualfactorsassociatedwith temptationorlapseswillinfluencelong-termweightlossoutcomes.Furtherinvestigationofthesefactorswillbeimportant. Despitenearingthecompletionofaformalweight-lossprogram,mood,copingresponse,andabstinence-violationeffects weresignificantlyassociatedwithdietary-relapsecrises.Recent consumptiveactivities,location,andotheractivitieswerenot typicallyassociatedwithrelapsecrises.Thesefindingssuggestthat weight-lossprogramparticipantsmaybenefitfromadditional skillstrainingthatencouragestheenactmentofcoping,aswellas teachesparticipantsskillsthatreduceabstinence-violationeffects andnegativemoodsduringrelapsecrises.Giventhattheassociationsbetweenmood,copingresponse,abstinence-violationeffects, andrelapsecriseswereevidentneartheendofaformalweightlossprogram,thesefactorsmayrequiregreaterattentioninfuture weight-losstreatment.Suchfindingsarelikelytohavecritical clinicalandpublichealthimplications.Awarenessofthefactors associatedwithdietarytemptationandlapse,understandingofthe commoncognitiveandaffectiveresponsestorelapsecrises,and developmentofstrategiestoreduceabstinence-violationeffects maybeessentialtoachievingpersonalandsocietalgoalsofweight lossandmaintenance.ReferencesAffleck,G.,Tennen,H.,Zautra,A.,Urrows,S.,Abeles,M.,&Karoly,P. (2001).Women spursuitofpersonalgoalsindailylifewithfibromyalgia:Avalue expectancyanalysis. JournalofConsultingandClinical Psychology,69, 587 596. Baumeister,R.F.,Heatherton,T.F.,&Tice,D.M.(1994). Losingcontrol: Howandwhypeoplefailatself-regulation. SanDiego,CA:Academic Press. Benjamini,Y.,&Hochberg,Y.(1995).Controllingforfalsediscovery rate:Apracticalandpowerfulapproachtomultipletesting. Journalof theRoyalStatisticalSociety,SeriesB,57, 289 300. Brownell,K.D.(2000). TheLEARNprogramforweightmanagement 2000. Dallas,TX:AmericanHealth. Brownell,K.D.,&Kramer,F.M.(1989).Behavioralmanagementof obesity. MedicalClinicsofNorthAmerica,73, 185 201. Carels,R.A.,Hoffman,J.,Collins,A.,Raber,A.,Cacciapaglia,H.,& O Brien,W.H.(2001).Ecologicalmomentaryassessmentoftemptation andlapseindieting. EatingBehaviors,2, 307 321. Cummings,C.,Gordon,J.R.,&Marlatt,G.A.(1980).Relapse:Prevention andprediction.InW.Miller(Ed.), Theaddictivebehaviors:Treatment ofalcoholism (pp.291 322).Elmsford,NY:Pergamon. Greeno,G.G.,Wing,R.R.,&Shiffman,S.(2000).Bingeantecedentsin obesewomenwithandwithoutbingeeatingdisorder. JournalofConsultingandClinicalPsychology,68, 95 102. Grilo,C.M.,Shiffman,S.,&Wing,R.R.(1989).Relapsecrisesand copingamongdieters. JournalofConsultingandClinicalPsychology, 57, 488 495. Grilo,C.M.,Shiffman,S.,&Wing,R.R.(1993).Copingwithdietary relapsecrisesandtheiraftermath. AddictiveBehaviors,18, 89 102. Haynes,S.N.,&O Brien,W.H.(1999). Principlesandpracticeof behavioralassessment. NewYork:KluwerAcademic/Plenum. Jeffery,R.W.,Drewnowski,A.,Epstein,L.H.,Stunkard,A.J.,Wilson, G.T.,&Wing,R.R.(2000).Long-termmaintenanceofweightloss: Currentstatus. HealthPsychology,19, 5 16. Johnson,W.G.,Schlundt,D.G.,Barclay,D.R.,Carr-Nangle,R.E.,& Engler,L.B.(1995).Anaturalisticfunctionalanalysisofbingeeating. BehaviorTherapy,26, 101 118. Litt,M.D.,Cooney,N.L.,&Morse,P.(1998).Ecologicalmomentary assessmentwithtreatedalcoholics:Methodologicalproblemsandpotentialsolutions. HealthPsychology,17, 48 52. Matthews,K.A.,Owens,J.F.,Allen,M.T.,Stoney,C.M.(1992).Do cardiovascularresponsestolaboratorystressrelatetoambulatoryblood pressurelevels?Yes,insomepeople,someofthetime. Psychosomatic Medicine,54, 686 697. NationalHeart,Lung,andBloodInstituteObesityEducationInitiative TaskForceMembers.(1998). Clinicalguidelinesontheidentification, evaluation,andtreatmentofoverweightandobesityinadults:The evidencereport (NIHpublicationNo.98 4083).Bethesda,MD:NationalInstitutesofHealth. Rosenthal,B.,&Marx,R.D.(1981).Determinantsofinitialrelapse episodesamongdieters. ClinicalandBariatricMedicine,10, 94 97. SASInstitute.(1997). SAS/STATsoftware:Changesandenhancements throughRelease6.12. Cary,NC:Author. Schlundt,D.G.,Sbrocco,T.,&Bell,C.(1989).Identificationofhigh-risk situationsinabehavioralweightlossprogram:Applicationofthe relapsepreventionmodel. InternationalJournalofObesity,13, 223 234. Schlundt,D.G.,Taylor,D.,Hill,J.O.,Sbrocco,T.,Pope-Cordle,J., Kasser,T.,&Arnold,D.(1990).Abehavioraltaxonomyofobese femaleparticipantsinaweightlossprogram. AmericanJournalof ClinicalNutrition,53, 1151 1158. Shiffman,S.,Paty,J.A.,Gnys,M.,Kassel,J.D.,&Hickcox,M.(1996). Firstlapsestosmoking:Within-subjectsanalysisofreal-timereports. JournalofConsultingandClinicalPsychology,64, 366 379. Stone,A.A.,Broderick,J.E.,Porter,L.S.,&Kaell,A.T.(1997).The experienceofrheumatoidarthritispainandfatigue:Examiningmomentaryreportsandcorrelatesoveroneweek. ArthritisCareandResearch, 10 (3),185 193. Stone,A.A.,&Shiffman,S.(1994).Ecologicalmomentaryassessmentin behavioralmedicine. AnnalsofBehavioralMedicine,16, 199 202. Stone,A.A.,Shiffman,S.,Schwartz,J.E.,Broderick,J.E.,&Hufford, M.R.(2002).Patientnoncompliancewithpaperdiaries. BritishMedical Journal,324, 1193 1194. Thissen,D.,Steinberg,L.,&Kuang,D.(2002).QuickandeasyimplementationoftheBenjamin Hochbergprocedureforcontrollingthefalse positiverateinmultiplecomparisons. JournalofEducationalandBehavioralStatistics,27, 77 83. Williams,V.S.L.,Jones,L.V.,&Tukey,J.W.(1999).Controllingerror inmultiplecomparisons,withexamplesfromstate-of-artdifferencesin educationalachievement. JournalofEducationalandBehavioralStatistics,24, 42 69. Wing,R.R.,Shiffman,S.,Drapkin,R.G.,&McDermott,M.(1995). Moderateversusrestrictivediets:Implicationsforrelapse. Behavior Therapy,26, 5 24. Zeger,S.L.,Liang,K.,&Albert,P.S.(1988).Modelsforlongitudinal data:Ageneralizedestimatingequationapproach. Biometrics,44, 1049 1060.ReceivedAugust22,2002 RevisionreceivedFebruary28,2003 AcceptedApril2,2003 348CARELS,DOUGLASS,CACCIAPAGLIA,ANDO BRIEN