UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Survey of obese men with obstructive sleep apnea : variables linked to body weight Travis, Karol Ann 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1998-0186.pdf [ 7.09MB ]
Metadata
JSON: 831-1.0099305.json
JSON-LD: 831-1.0099305-ld.json
RDF/XML (Pretty): 831-1.0099305-rdf.xml
RDF/JSON: 831-1.0099305-rdf.json
Turtle: 831-1.0099305-turtle.txt
N-Triples: 831-1.0099305-rdf-ntriples.txt
Original Record: 831-1.0099305-source.json
Full Text
831-1.0099305-fulltext.txt
Citation
831-1.0099305.ris

Full Text

S U R V E Y OF OBESE M E N WITH OBSTRUCTIVE SLEEP APNEA: VARIABLES LINKED TO BODY WEIGHT by K A R O L A N N TRAVISS B.H.E., University of British Columbia, 1979 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE F A C U L T Y OF GRADUATE STUDIES (School of Family and Nutritional Sciences) We accept this thesis as conforming to the required standards UNIVERSITY OF BRITISH COLUMBIA March, 1998 ©Karol Ann Traviss, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ^7^A^r/ 0?/lWjjJ{ ^ A ^ ^ / f ^ The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract ABSTRACT Obstructive sleep apnea (OSA), a disorder characterized by intermittent cessation of breathing during sleep, is most prevalent in obese middle-aged men and is associated with considerable morbidity and mortality. The purpose of this study was to characterize newly diagnosed obese men with OSA regarding several variables linked to body weight. Although the association between OSA and obesity is clear, very little is known about factors influencing body weight in OSA. Studies have failed to uncover a metabolic defect in obese individuals with the disorder, raising speculation that obesity may be related to lifestyle or psychological factors. Patients meeting study criteria were recruited following the overnight sleep study at which the diagnosis of OSA was confirmed. Data were collected using chart reviews, anthropometric measurements, food records, written questionnaires, and subject interviews. Forty-nine men participated, about two-thirds of whom completed and returned useable written questionnaires. A l l completed other study components. Food record analysis showed that reported energy intake was highly variable, but the mean intake reported was moderate, and should not have resulted in weight gain. However, qualitative problems with subjects' diets were evident, including high intake of meats and alternatives and fat, and low intakes of vegetables and fruits and milk products. Binge eating scores were significantly correlated with BMI, indicating increasing eating behaviour dysfunction with increasing relative weight. No subjects appeared to have the night eating syndrome (NES); however, three described eating behaviour suggestive of nocturnal binge eating (NBE). Physical activity questionnaire results indicated that many subjects were physically inactive, and few participated in recreational sports. Similar to previous studies of psychological variables in OSA, about one third of patients were found to have symptoms of depression, ii Abstract anxiety, or both. Many of the psychological and eating behaviour variables were significantly correlated. The interview component of the study revealed that the symptom profile of obese men with OSA was variable, although fatigue, unrefreshing sleep, and snoring were common. In most cases, it was a family member or friend who had prompted the subject's visit to the clinic. The majority of subjects had experienced a substantial increase in body weight since onset of OSA symptoms. Most felt their weight gain was attributable to identifiable lifestyle factors. The amount of weight gain reported was significantly correlated with binge eating and depression scores. As well, subjects who felt that fatigue or stressful life events had affected their food intake had gained significantly more weight than those who had not had these perceptions. Most subjects reported that they had not become obese until adulthood. More than two-thirds of subjects reported that weight loss attempts had been infrequent. Eating habit changes were the most predominant weight loss strategy used. Most subjects reported some initial success with weight loss efforts, but virtually all had regained weight following weight loss attempts. Overall, this study provides a profile of a patient population in which debilitating symptoms, sub-optimal diets, low physical activity, symptoms of depression and anxiety, and escalating weight are common. These patients appear to need weight gain prevention strategies, qualitative dietary changes, regular physical activity, and psychological support. A multidisciplinary and community-based approach may be necessary to assist them in making desirable lifestyle changes. Hi Table of Contents TABLE OF CONTENTS P A G E ABSTRACT ii TABLE OF CONTENTS iv LIST OF TABLES vii LIST OF FIGURES x ACKNOWLEDGEMENTS xi CHAPTER I INTRODUCTION 1. Background 1 2. Purpose of the study 1 3. Research questions 2 CHAPTER II L ITERATURE REVIEW 1. Introduction 3 2. Obstructive sleep apnea (OSA) 3 3. Diagnosis of OSA 4 4. Obesity and OSA 4 5. Treatment of OSA 6 6. Weight loss in OSA 7 7. Food intake and eating behaviour in OSA 9 8. Physical activity in OSA 10 9. Psychological variables in OSA 10 10. Obesity literature 11 CHAPTER III METHODOLOGY 1. Overview 13 2. Subjects 13 3. Sleep laboratory procedures 14 4. Recruitment procedures 15 5. Data collection a) Procedures 15 b) Health and demographic data 16 c) Anthropometric measures 16 d) Food intake 17 e) Written questionnaires i) Eating behaviour 17 iv Table of Contents P A G E ii) Physical activity 18 iii) Psychological variables 18 f) Interview guide 19 g) Data analysis i) Food intake data 19 ii) Other data 20 CHAPTER IV RESULTS 1. Recruitment : 21 2. Subject characteristics 24 3. OSA treatment status 32 4. Food intake and questionnaire data a) Food intake 34 b) Eating behaviour 41 c) Physical activity 47 d) Psychological variables 53 5. Interview data a) Symptom profile 61 b) Weight history 64 c) Perceived relationship between OSA symptoms and food intake 73 d) Nocturnal eating history 74 e) Life events and stressors 78 6. Feedback from subjects 85 CHAPTER V DISCUSSION 1. Overview 86 2. Subject characteristics 86 3. Food intake and questionnaire data a) Food intake 94 b) Eating behaviour 99 c) Physical activity 103 d) Psychological variables 104 4. Interview data 106 5. Feedback from subjects I l l 6. Study limitations H I 7. Generalizability of results .'. 113 J Table of Contents P A G E 8. Implications of findings 114 9. Directions for further research 116 10. Conclusions 117 REFERENCES 118 APPENDICES A. Rationale for methodologies used 129 B. Ethical approval, University of British Columbia 131 C. Study approval, Vancouver Hospital 132 D. Subject recruitment letter 133 E. Subject consent form 134 F. Sample subject follow-up letter 135 G. Health history and demographic profile form 136 H. Food record forms and instructions 137 I. Verbal food frequency questionnaire 142 J. Three Factor Eating Questionnaire 143 K. Binge Eating Scale .146 L. Night Eating Syndrome Scale 149 M. Paffenbarger Physical Activity Questionnaire '. 150 N. Zung Self-rated Anxiety Scale 151 O. Zung Self-rated Depression Scale 152 P. Rosenberg Self-esteem Scale 153 Q. Subject interview guide 154 R. Diagnostic criteria for binge eating disorder (BED) 157 vi List of Tables LIST OF TABLES P A G E 1. Age, anthropometric measures and sleep apnea severity of obese men with OSA (n=49) 24 2. Correlation matrix for age, anthropometric, and polysomnographic measures in obese men with OSA (n=47-49) 25 3. Coexisting health conditions in obese men with OSA (n=49) 27 4. Medications used by obese men with OSA (n=49) 28 5. Demographic characteristics of obese men with OSA (n=49) 29 6. Occupations of obese men with OSA (n=49) 30 7. Smoking status of obese men with OSA (n=49) 31 8. Alcohol intake reported to physician by obese men with OSA (n=49) 32 9. OSA treatment recommendations and outcomes for obese men with OSA (n=49) 33 10. Mean daily energy and macronutrient intake of obese men with OSA (n=30) 35 11. Mean daily servings from each food group of Canada's Food Guide (CFG) consumed by obese men with OSA (n=30) 36 12. Correlation matrix for food intake variables for obese men with OSA (n=30).... 38 13. Correlation matrix for food intake variables, age, BMI, and measures of OSA severity 39 14. Binge Eating Scale and Three Factor Eating Questionnaire scores of obese men with OSA (n=30-31) 42 15. Comparison of mean scores on Binge Eating Scale items reflecting eating behaviour versus cognitions for obese men with OSA (n=31) 43 16. Correlation matrix for Binge Eating Scale (BES) and Three Factor Eating Questionnaire Scores (TFEQ) (n=30-31) 44 17. Correlation matrix for eating behaviour scores and age, BMI, measures of OSA severity, and food intake (n=30-31) 45 18. Night Eating Syndrome Scale responses of obese men with OSA (n=31) 46 19. Physical activity, as assessed by the Paffenbargar questionnaire, for obese men with sleep apnea (n=31) 48 20. Time spent engaging in activities of varying intensity by obese men with OSA (n=27) 50 21. Perceived changes in physical activity over the previous year of obese men with OSA(n=31) 50 vii List of Tables P A G E 22. Correlation matrix for physical activity, age, BMI, measures of OSA severity, food intake variables, and eating behaviour for obese men with OSA (n=27-30) 51 23. Zung Self-Rated Depression Scale (ZSRDS) scores for each scale item for obese men with OSA (n=31) 54 24. Zung Self-rated Anxiety Scale (ZSRAS) scores for each scale item for obese men with OSA (n=31) 55 25. Comparison of mean scores on Zung Self-rated Depression Scale (ZSRDS) and Zung Self-rated Anxiety scale (ZSRAS) items reflecting vs. not reflecting common OSA symptoms for obese men with OSA (n=31) 56 26. Correlation matrix for psychological variables for obese men with OSA (n=30-31) 57 27. Correlation matrix for psychological variables, age, BMI, measures of OSA severity, food intake, eating behaviour and physical activity for obese men with OSA(n=28-31) 58 28. Correlations between psychological and eating behaviour variables, controlling for BMI, for obese men with OSA (n=30-31) 59 29. Symptom profile of obese men with OSA (n=49) 63 30. Correlation matrix for weight gain since onset of OSA symptoms, and age, BMI, measures of OSA severity, food intake, alcohol intake eating behaviour, physical activity and psychological variables for obese men with OSA who had gained weight since symptom onset (n=22-30) 66 31. Perceived body weight at various life stages of obese men with OSA (n=49) 67 32. Perceived causes of dramatic weight gain for obese men with OSA (n=20) 68 33. Perceived causes of gradual or non-dramatic weight gain for obese men with OSA(n=29) 69 34. Frequency of weight loss attempts by obese men with OSA (n=49) 70 3 5. Methods used to attempt weight loss by men with OSA (n=39) 71 36. Most weight ever lost by obese men with OSA who had attempted weight loss (n=39) 72 37. Perceived effects of fatigue on food intake for obese men with OSA (n=20) 74 38. Characteristics of nocturnal binge eaters (n=3) 76 39. Additional characteristics of nocturnal binge eaters (n=3) 77 40. Major life events or changes at onset of symptoms for obese men with OSA (n=43) 79 41. Sources of stress at symptom onset for obese men with OSA (n=43) 80 viii List of Tables P A G E 42. Perceived effects of life events or stresses at time of symptom onset on eating habits for obese men with OSA (n=43) 82 43. Current sources of stress for obese men with OSA (n=49) 83 44. Additional sources of current stress for obese men with OSA (n=49) 84 ix List of Figures L I S T O F F I G U R E S P A G E 1. Overview of study recruitment process 23 x Acknowledgments ACKNOWLEDGEMENTS It seems like an eternity ago. It was 1993, and there I was with a one-year-old on my hip and a three-year-old at my feet, not to mention full time employment. Whatever possessed me to consider graduate studies? At that time, I didn't have the slightest hunch that I would immerse myself in the eating behaviour of men with sleep apnea, nor that such an undertaking could actually be enjoyable and enriching. Well, I did it (I'm still not exactly sure what prompted me) and I know for certain that it wouldn't have happened without the energy and support of many people around me. First, I would like to thank the colleagues, both dietitians, and sleep program physicians and staff, who helped the idea for this project to germinate. Next, I would like to thank my thesis committee members for helping me to refine this project. A special thanks to Susan Barr, my thesis advisor, who helped give me the strength to carry on with subject recruitment, when for weeks on end, it seemed that all sleep lab patients were thin or female. Without the daily support and guidance from the sleep program staff, it would have been difficult (if not impossible) to carry out this project. Although this project involved a lot of very early mornings, this time well spent even on days when there were no subjects for me. The soothing background music (selected by the technician of the day), the breathtaking mountain view, and the enlightening conversations helped to make the time worthwhile. I'm sure I learned as much about sleep apnea during these informal encounters as I did from reviewing the literature on the topic. And not to forget all the friends, colleagues, and family members who may have had to feign interest in my "tales of the sleep lab". A special thanks to my husband, Dick, who never once tried to talk me out of doing this (although it perplexed him why it should take so long). Finally, I would like to acknowledge the individuals with sleep apnea (both study subjects and members of the B C Sleep Apnea Society) who took the time to contribute to this project. If I were as exhausted as most of these people are, I'm not sure that I would take the time to participate in a study like this. I was amazed that almost all patients I approached about the study were willing to take part. Now I can get some sleep... xi Chapter I: Introduction I N T R O D U C T I O N 1. Background Obstructive Sleep Apnea (OSA), a condition characterized by the intermittent cessation of breathing during sleep, is most prevalent in obese middle aged men. OSA is associated with significant morbidity and mortality and impaired daytime functioning. For obese individuals with OSA, weight loss reduces the severity of the disorder, and is therefore a routine treatment recommendation. Considering the critical role of weight management in OSA, it is noteworthy that few dietitians appear to have established themselves as practitioners or researchers within this specialty area. A 1994 review article on weight control in OSA (Loube, Loube & Mitler) emphasizes the potential value of an interdisciplinary team approach to management of obese individuals with OSA, including the active involvement of a dietitian. Since OSA is relatively common, dietitians are likely to encounter patients with the disorder in various practice settings, including clinics for sleep disorders, hypertension, lipid disorders, diabetes, or general outpatient nutrition counselling. A descriptive study to examine the unique characteristics of obese men with OSA appears to be warranted. Findings may be of use in the development of pertinent lifestyle intervention strategies for these patients. 2. Purpose of the Study The purpose of this research project was to characterize obese men with newly diagnosed OSA, regarding several variables linked to body weight, including food intake, eating behaviour, physical activity, psychological variables, and body weight history. 1 Chapter I: Introduction 3. Research Questions The study was designed to address the following questions, in obese men with OSA: a) What is their food intake profile? i) What is their energy and macronutrient intake? ii) What is their daily energy intake distribution? iii) How does their food intake compare to the recommendations of Canada's Food Guide (CFG)? b) What are their eating behaviour characteristics? i) How prevalent are binge eating problems? ii) What is their level of intention to control food intake (dietary restraint)? iii) What is the prevalence of the night eating syndrome (NES)? iv) What is the prevalence of nocturnal binge eating (NBE)? v) How do they perceive that having the disorder has affected their food intake? c) What is their physical activity profile? i) What is their current level of physical activity? ii) How do they perceive their physical activity has changed? d) What is their psychological profile? i) How prevalent are symptoms of depression, anxiety and low self-esteem? ii) Are there traumatic or stressful life events associated with weight gain at onset of OSA? iii) How do they perceive that life events or stressors have affected their food intake? e) What is their body weight history? i) Is there a pattern of weight gain associated with onset of OSA? ii) Do they perceive themselves to have been overweight at various life stages? iii) What is their history of weight loss attempts? 2 Chapter II: Review of Literature L I T E R A T U R E R E V I E W 1. Introduction This literature review provides an overview of OSA, including its symptoms and health implications, diagnosis and treatment. The relationship of OSA to obesity is presented and studies of weight loss in OSA are described, as is literature pertaining to the variables studied. A brief summary of the general obesity literature in relation to these variables is also presented. 2. Obstructive Sleep Apnea Obstructive Sleep Apnea (OSA) is a disorder in which upper-airway obstruction during sleep results in cessation of breathing (Cetel & Guilleminault, 1994; Young, Palta, Dempseu, Skatrud, Webber, & Badr, 1993). The usual presenting complaints of individuals with OSA are loud snoring, unrefreshing sleep and daytime somnolence (Cetel & Guilleminault, 1994). OSA is a condition associated with considerable morbidity and mortality. It is a significant predictor of myocardial infarction, hypertension, stroke, and early renal dysfunction (Cetel & Guilleminault, 1994; Coy, Dimsdale, Ancoli-Israel, & Clausen, 1996; Grunstein, Stenlof, Hedner, & Sjostrom, 1995; Grunstein, Wilcox, Yang, Gould & Hedner, 1993). OSA has also been associated with high rates of sick leave, poor self-rated health, impaired work performance, and divorce (Grunstein et al., 1995). Individuals with OSA are also at increased risk for motor vehicle accidents (Stoohs, Bingham, Itoi, Guilleminault, & Dement, 1995; Wu & Yan-Go, 1996) At least 12 large studies have been conducted to date on the prevalence of sleep disordered breathing in adults (Davies & Stradling, 1995). Prevalence estimates have varied, likely due to methodological differences between studies, true differences in study populations, or a combination. An estimated 1-5% of the adult population has OSA (Davies & Stradling, 3 Chapter II: Review of Literature 1995; Kripke et al., 1997). However, OSA is known to be more prevalent in middle-aged men, particularly those who are obese. Close to ten percent of middle aged men in a study of Wisconsin state employees had symptoms or polysomnographic findings suggestive of OSA (Young et al., 1993). A more recent study in California, involving a study population of mixed ethnicity (Kripke et al., 1997), estimated OSA prevalence in middle-aged men to be even higher, at 20%. 3. Diagnosis of OSA The diagnosis of OSA is confirmed by sleep studies, in which the respiratory disturbance index (RDI) is determined. The RDI represents the average number of respiratory disturbances per hour of sleep, including apneic episodes (involving complete closure of the upper airway) and hypopneic episodes (involving a partial closure of the upper airway). An RDI of 10 or more is usually considered diagnostic of OSA (Cetel & Guilleminault, 1994). The minimum level of arterial oxygen saturation associated with respiratory events during sleep (Sa0 2), is an additional measure of OSA severity (Cetel & Guilleminault, 1994). In healthy individuals, the level of arterial oxygen saturation is about 97% (Ganong, 1977). During apneas or hypopneas, desaturations occur. Low levels of arterial oxygen saturation are indicative of severe OSA, and are associated with impaired cognitive and task performance (Cheshire, Engleman, Deary, Shapiro, & Douglas, 1992). 4. Obesity and OSA It has been postulated that obesity is the primary etiologic factor in OSA (Wittels & Thompson, 1990); however, not all individuals with OSA are obese (Cetel & Guilleminault, 1994). Non-obese individuals with OSA tend to have anatomical abnormalities of the upper airway, such as nasal obstruction, enlarged tonsils/adenoids or skeletal abnormalities (Kryger, 4 Chapter II: Review of Literature 1989; Bradley, Brown, & Brossman, 1986). The exact mechanism by which obesity contributes to OSA is not completely clear, although it is recognized that neck obesity encourages pharyngeal obstruction. In addition to neck adiposity, the weight of the chest wall and abdomen appear to be contributing factors (Bradley, 1994; Grunstein, et al., 1993; Shinohara et al., 1997). Weight reduction improves the symptoms of OSA (Aubert-Tulkens, 1989; Browman et al., 1984; Cetel & Guilleminault, 1994; Rubinstein, Colapinto, Rotstein, Brown, & Hoffstein, 1988; Wittels & Thompson, 1990). However, the amount of weight loss required for a significant improvement in symptoms varies greatly (Loube et al., 1994; Wittels & Thompson, 1990). Some obese individuals can resolve their OSA with relatively minor weight loss (Loube et al., 1994), while others who achieve major weight loss continue to have OSA symptoms (Pasqualietal., 1990). It has been speculated that individuals may have a "threshold" weight at which OSA develops (Sugerman, 1986). It has been anecdotally reported that once this weight is reached and onset of OSA occurs, rapid weight gain takes place (Sugerman, 1986; Wittels & Thompson, 1990). At present, little is known about how prevalent this pattern of weight gain is, and whether onset of OSA precipitates the weight gain, or whether weight gain due to other factors results in onset of OSA. In one retrospective study of 20 obese men with OSA, it was confirmed that a significant and progressive weight gain took place from the initial occurrence of snoring without apneas to the onset of OSA symptoms (Pasquali et al., 1990). This escalation of weight is of concern as it is likely to exacerbate OSA; however, the reasons for it are not clear. It does not seem to be due to metabolic abnormalities; both the resting metabolic rate and metabolic response to a meal by individuals with OSA appear to be normal (Richman, 1994; Ryan, Love & 5 Chapter II: Review of Literature Buckley, 1995). 5. Treatment of OSA Until the past decade, weight loss and tracheostomy were the primary treatment options for OSA (Cetel & Guilleminault, 1994). In recent years, mechanical devices (i.e., dental appliances) or surgical procedures have been used in treating individuals with OSA who have anatomical upper airway abnormalities. Nasal continuous positive airway pressure (CPAP) was proposed as a treatment for OSA in 1981, and came into common use in the mid-1980's (Aubert-Tulkens et al., 1989; Cetel & Guilleminault, 1994). This treatment consists of pressurized air which is delivered into the upper airway through a nasal mask, preventing upper airway collapse during sleep. CPAP effectively alleviates symptoms of OSA but it does not provide a cure. It has been anecdotally reported that patients using CPAP can more readily lose weight (Borak, Cieslicki, Szelenberger, Wilczak-Szadkowska, Koziej, & Zielinski, 1994; Sullivan & Grunstein, 1994), an observation that has recently been verified through systematic investigation. Loube, Loube, & Erman (1997), compared weight loss in two groups of obese OSA patients: those who reported using and not using CPAP. They found that 9 of 21 CPAP users had achieved weight loss of 4.5 kg or more in the six month period following diagnosis, whereas none of the 11 non-CP AP users had lost that amount of weight. It is not known whether this short term weight loss can be maintained, and what factors facilitate weight loss in some CPAP users. CPAP compliance can be problematic. In a study of CPAP compliance in OSA (Reeves-Hoche, Meek, and Zwillich, 1994), 9 of 47 patients (19%) abandoned CPAP within three months, and the remaining patients were found to use it for an average of 68% of total sleeping time. Considering the challenges in achieving CPAP compliance, and the fact that weight loss 6 Chapter II: Review of Literature only appears to occur in some CPAP users, lifestyle modifications still need to be advocated for obese patients with OSA. 6. Weight Loss in OSA The results of weight loss efforts in OSA appear to be dismal (Aubert-Tulkens, et al., 1989; Browman et al., 1984; Cetel & Guilleminault, 1994; Harman & Block, 1986; Loube et al., 1994; Rubinstein, et al., 1988; Wittels & Thompson, 1990; Strobel & Rosen, 1996). It has been difficult to fully explore the apparent failure of these weight loss interventions, as studies of weight loss in OSA have had numerous shortcomings, including lack of random assignment of subjects to treatment groups, confounding effects of OSA treatments, absence of untreated controls, small subject numbers, and lack of long-term follow-up (Aubert-Tulkens et al., 1989; Browman et al., 1984; Rubinstein, et al., 1988; Harman & Block, 1986; Strobel & Rosen, 1996). Details are lacking on the intake modifications used, and whether behaviour modification or exercise were incorporated. Three types of weight loss studies in OSA are reported in the literature: dietary interventions, surgical interventions, and case reports of individuals achieving weight loss. Almost all of these studies pre-date the routine use of CPAP as a therapy for OSA. Some dietary interventions in OSA have involved use of very low calorie diets (VLCD) in morbidly obese subjects. These studies show that in most cases, massive weight loss results in significant improvement in OSA symptoms (Pasquali, et al., 1990; Suratt, McTier, Findley, Pohl, & Wilhoit, 1992). However, long term weight maintenance appears to be extremely difficult for such individuals to achieve. In the^VLCD diet study with the longest followup period (Suratt et al., 1992), virtually all patients had returned to their baseline weight at two years post intervention. 7 Chapter II: Review of Literature Surgical interventions for weight loss have also focused on morbidly obese individuals. With surgical methods such as intestinal bypass and gastric banding, dramatic weight loss and improvement in OSA symptoms have been achieved (Charuzi, Lavie, Peiser, & Peled, 1992; Pasquali, et al., 1990). However, patients who have undergone such surgical procedures have had difficulties maintaining weight loss. In a seven year follow-up report of obese patients with OSA who had undergone bariatric surgery, the majority had regained a significant amount of weight, and sleep apnea symptoms had returned (Charuzi et al., 1992). There has been limited examination of moderate, gradual weight loss in OSA. Smith et al. (1985) studied the effects of relatively minor weight loss in 15 moderately obese individuals with OSA. Subjects were matched to similar individuals who had been advised to keep weight stable. Subjects lost an average of .45 kg per week, although no standardized weight loss strategy was used. Each subject was followed until body weight was 5% below baseline. Apneic episodes and daytime somnolence significantly decreased in study subjects; sleep patterns also improved. The follow-up period for the study was only 6 months, so the long-term ability of subjects to sustain weight loss and symptomatic improvement was not assessed. However, these findings do show that minor weight loss can result in significant clinical improvement in OSA in moderately obese individuals. Case reports of weight loss in OSA have invariably described improvement in OSA symptoms in individuals achieving weight loss, and recurrence of symptoms upon weight gain (Aubert-Tulkens et al., 1989; Browman et al., 1984; Eveloff & Millman, 1993). Although these individual cases support the value of weight loss in OSA, they may also illustrate how rare the phenomenon of successful weight loss in OSA is. 8 Chapter II: Review of Literature 7. Food Intake and Eating Behaviour in OSA Most weight loss interventions for individuals with OSA appear to have been implemented without consideration of several relevant patient characteristics, such as usual food intake, eating patterns and eating behaviour. One systematic attempt to quantify the food intake of obese patients was undertaken as part of a study of weight loss in OSA. The total energy and macronutrient intakes of 23 subjects were examined using three-day food records augmented with diet histories (Pasquali et al., 1990). Reported intake varied from 1262 - 4280 kcal (5282 - 17,913 kJ) per day; mean intake was 2942 kcal. The high variability of these results may reflect the heterogeneity of subjects (BMI ranged from 26.6 -61 ; one subject was female), as well as the recognized difficulties in obtaining accurate food intake data, particularly from obese subjects (Briefel, McDowell, Alaimo, et al., 1995; Forbes, 1993; Lansky & Brownell, 1982). Additional intake studies appear to be necessary to gain a better understanding of food intake and intake patterns in OSA. Although rapid weight gain in absence of metabolic abnormalities might indicate the presence of binge eating, it is unknown whether binge-eating behaviour is prevalent in obese individuals with OSA. Case reports indicate that some obese individuals with OSA may have an eating disorder characterized by nocturnal binge eating (NBE) (Eveloff & Millman, 1993; Schenck, Hurwitz, Bundlie, & Mahowald, 1991; Schenck, Hurwitz, O'Connor & Mahowald, 1993); however the prevalence of this disorder in OSA is unknown. The relationship of N B E to the night eating syndrome (NES), an eating disorder characterized by uncontrolled evening snacking (Stunkard, Grace and Wolff, 1955) is also not clear. The Klein-Levin syndrome, a rare condition affecting obese adolescent males, is characterized by episodic hyperphagia and sleep apnea. Although it has been recognized for decades that compulsive overeating and sleep apnea 9 Chapter II: Review of Literature coexist in this syndrome (Cuetter, 1985), the reasons for this are unknown. It is also unknown whether other individuals with OSA exhibit similar eating behaviour abnormalities. Several dimensions of eating behaviour, previously explored in both obesity and eating disorders, have never been systematically examined in OSA. Although historical attempts at identifying a set of eating behaviours unique to obesity were not successful (Spitzer & Rodin, 1981; Stunkard, 1959), eating behaviour questionnaires developed in the 1980's (Gormally, Black, Daston, & Rardin, 1982; Stunkard & Messick, 1985) have been more successful at identifying distinctions between obese and non-obese individuals, and identifying subgroups of obese individuals more likely to be successful with weight management. Determining the eating behaviour characteristics of individuals of OSA may be useful in identifying subgroups that could potentially benefit from different approaches to intervention. Enhanced knowledge about eating behaviour in OSA may also contribute to the understanding of weight gain in OSA. 8. Physical Activity in OSA Considering the established role of physical activity in weight regulation and health (Brill, Kohl, & Blair, 1992; Garner & Wooley, 1991; Paffenbarger, Wing, & Hyde, 1978), it is noteworthy that there seems to have been no previous attempt to examine the physical activity level of individuals with OSA, or to systematically incorporate exercise into intervention programs. 9. Psychological Variables in OSA There have been several reports of psychological disturbances in OSA (Borak et al., 1994; Cassel, 1993; Millman, Fogel, McNamara, & Carlisle, 1989; Platon & Sierra, 1992), although not all studies have had this finding (Cassel, 1993). It has also been observed that 10 Chapter II: Review of Literature treatment of OSA can improve psychopathological scores (Borak et al., 1994; Millman et al., 1989; Platon & Sierra, 1992; Van Moffaert, 1994). For those studies identifying depressive symptoms in OSA, prevalence rates have varied from 24% to 56% (Borak et al., 1994; Cheshire et al., 1991; Millman, et al., 1989; Platon & Sierra, 1992). It is somewhat difficult to interpret these findings, as study populations and study methodologies have varied. Millman et al. (1989) found that about 40% of 46 middle-aged men with newly diagnosed OSA had scores on the Zung Self-rated Depression Scale that were consistent with depression. Depression has been the most widely studied psychological variable in OSA. There has been limited examination of anxiety; it does not appear that self-esteem has been assessed. In two studies examining anxiety in OSA, prevalence was found to be 17% in one (Cheshire et al., 1992) and 35% in the other (Borak et al., 1994). Further psychological characterization of obese men with OSA may be useful in concert with examination of eating behaviour and degree of obesity, Various obesity studies have shown an association between degree of obesity and both eating behaviour and psychopathology (Brody, Walsh, & Devlin, 1994; Bruce & Wilfley, 1996; Rosmond, Lapidus, Marin, et al., 1996; Yanovski et al., 1993). In OSA, such relationships have not been examined, and may be important in identifying subgroups of patients that may be resistant to lifestyle modification. 10. Obesity Literature A discussion of obesity in OSA would be incomplete without reference to the general obesity literature. Obesity is a heterogeneous and multifactorial condition (Brownell & Wadden, 1991); therefore, it is not known to what extent the current knowledge about obesity is applicable to the male OSA population. Also, much of the weight management and eating 11 Chapter II: Review of Literature behaviour literature has focussed on female populations (Jeffery, Bjornson-Benson, Rosenthal, Lindquist, Kurth & Johnson, 1984). Nevertheless, widely confirmed obesity findings have potential to be relevant to OSA. There is unequivocal evidence that most reported dietary and behavioural interventions for obesity have been ineffective for the achievement of significant and sustained weight loss (Garner & Wooley, 1991). The challenges faced by individuals with OSA to achieving and maintaining weight loss are not unique to this population. This reality underscores the importance of exploring alternatives to major weight loss which have the potential to have a positive impact on health status, such as qualitative changes to the diet, physical activity, and psychological well being. At present, baseline knowledge about diet, physical activity, and psychological well-being in OSA is limited. Considering the ineffectiveness of currently available obesity treatments, a focus on prevention is logical. In OSA, better understanding of factors leading to weight gain at onset of the condition may be useful in identifying prevention strategies. Similarly, a food intake and physical activity profile of individuals with OSA may be beneficial in identifying areas of focus for education programs that may help these individuals achieve gradual weight loss or prevent further weight gain. 12 Chapter III: Methodology METHODOLOGY 1. Overview The study was a multi-faceted survey of obese men with newly diagnosed OSA. During the data collection period (February, 1996 - January, 1997), patients undergoing respiratory sleep studies at Vancouver Hospital were screened to identify individuals meeting study inclusion criteria and to document characteristics of non-participants. Patients agreeing to participate in the study were asked to: a) provide demographic and health information; b) have anthropometric measurements taken; b) complete a three-day food record; c) complete previously validated written questionnaires pertaining to eating behaviour, physical activity and psychological variables; and e) participate in an interview with the investigator, the rationale for methodologies used is presented in Appendix A. Ethical approval for this study was received from the University of British Columbia Research Services (see Appendix B), and the Vancouver Hospital Research Committee (see Appendix C). See Appendices D and E for subject recruitment letter and consent form. 2. Subjects The proposed sample size of 50 was derived with consideration to the relative homogeneity of the study population, the number of subjects required to produce useful descriptive data, and the estimated time period required to collect the data. Study subjects were obese (BMI > 30) adult males with newly diagnosed OSA confirmed by sleep studies (RDI > 10). The following patients were excluded from participating: patients already using CPAP or other interventions for OSA; patients unable to speak, read or write English; patients following therapeutic diets; patients using psychotropic or other drugs with recognized impact on study variables; and patients with hypothyroidism, diabetes, psychiatric 13 Chapter III: Methodology disorders, or other conditions expected to influence study variables. Patients with hypertension, cardiovascular disease, and chronic respiratory diseases were not excluded, as these conditions frequently coexist with OSA; however, the presence of these conditions was noted. 3. Sleep Laboratory Procedures Patients referred to the Vancouver Hospital Sleep Disorders Program (British Columbia's provincial referral centre for sleep disorders) with suspected OSA generally have their initial appointment about a year after referral. At the clinic, a respirologist sees each patient, obtaining a medical history and performing an examination to determine whether an overnight sleep study is warranted. Sleep studies are used to assess sleep-related factors that are disturbed in OSA, including number and magnitude of respiratory events, proportion of time spent at each sleep stage, and sleep efficiency. Each night of the week, three patients are scheduled for respiratory sleep studies. Upon admission to the sleep laboratory, each patient is prepared for the sleep study by a technician. Small electrodes are attached to the body to measure the biopotentials necessary to elicit the electroencephalogram (EEG), electro-oculogram (EOG) and electromyogram (EMG) to stage sleep. The electrocardiogram (EKG), anterior tibialis E M G , and chest and abdominal respiratory effort are also measured. Flow sensors at the mouth and nose measure airflow and an oximeter attached to the index finger measures arterial oxygen saturation. Each morning at the sleep laboratory, a respirologist reviews preliminary sleep study data for each patient, and confers with the sleep laboratory technician about the likely diagnosis (final scoring of each sleep study is completed by sleep laboratory technicians within a few weeks). The respirologist then confers with each patient to discuss diagnosis and treatment recommendations. Following this, patients being sought to participate in studies can be 14 Chapter III: Methodology approached. Patients are typically fatigued following their sleep study and many have additional medical appointments and travel arrangements to attend to; therefore, time available to interact with them is limited. 4. Recruitment Procedures Each week, the list of patients attending the sleep laboratory was reviewed by the investigator; patients who clearly did not meet inclusion criteria were crossed off the list (e.g., female patients, patients on CPAP). Sleep clinic charts were screened for remaining patients to identify and eliminate those with exclusion factors. For patients remaining as potential subjects, preliminary sleep study data were reviewed following their overnight sleep study. Patients who appeared to meet all study inclusion criteria were approached at this time. For patients who met inclusion criteria but could not be approached about the study at that time due to scheduling factors or their involvement in other studies, selected patient characteristics were recorded. 5. Data Collection a) Procedures Each potential subject was briefed by the respirologist, and asked if he would be willing to speak to the investigator. The investigator then met with the patient to describe the study and seek his participation. Upon agreement to participate in the study and provision of written informed consent, subjects had their waist and hip measurements taken and were instructed on how to complete the three-day food record and written questionnaires. It was necessary to collect study data before the initiation of OSA therapies that might affect the variables studied; therefore, subjects were asked to complete food records and questionnaires promptly, and return them using the stamped addressed envelope provided. The investigator usually conducted subject interviews at the time 15 Chapter III: Methodology of recruitment. When this was not possible, interviews were conducted by telephone or in person within a few days of the sleep study. Upon receiving each subject's food records and written instruments by mail, the investigator reviewed them to ensure they were complete and correctly filled out. Subjects were contacted by telephone if queries about the food records or instruments arose. For those who did not return their written surveys by mail, a series of reminder letters was sent over several weeks (see sample, Appendix F). When it was recognized from the weekly sleep laboratory list that a subject was on CPAP or other therapy and would be undergoing a follow-up study, no further efforts were made to obtain completed surveys. Several months following each subject's sleep study, the investigator reviewed the subject's clinic chart to obtain final scores from their diagnostic sleep study, and to document treatment status. b) Health and Demographic Data A health history and demographic profile form (see Appendix G) was completed for each subject, using information available in the patient sleep clinic chart. Health information collected included medical history, current medical conditions, medications, smoking status, alcohol intake, and physician's OSA treatment recommendations. Demographic information included place of residence, marital status, occupation, and employment status. c) Anthropometric Measures Height, weight, neck girth, and waist and hip measurements were used. Height and weight were measured at admission by the sleep laboratory technician using a standard balance beam scale. Patients wore light indoor clothing and no footwear. Neck girth measurements were obtained from patient charts. Respirologists measured neck girth at the level of the cricothyroid cartilage. Waist and hip measurements were taken by the investigator at 16 Chapter III: Methodology recruitment, using procedures outlined in the Operations Manual of the Canadian Standardized Tests of Fitness (1986). d) Food Intake Food intake was determined by three-day food records. Each participant was provided with food record forms and instructions, including diagrams of specific foods to illustrate portion sizes (see Appendix H). The investigator verbally reviewed the instructions with each subject. Subjects were required to record the specific times that foods or beverages were consumed. In addition to food records, an brief verbal food frequency questionnaire was administered at the interview (Appendix I). The information collected was used in cross-checking information provided on food records. If queries arose, patients would be contacted (e.g., if the patient reported using oil in cooking, but no oil was recorded on the food record, the patient would then be asked to verify preparation techniques). Questions pertaining to usual intake of alcohol and coffee were included on this questionnaire. e) Written Questionnaires Each subject was asked to complete several previously validated standardized written questionnaires on eating behaviour, physical activity, and psychological variables. From pre-testing with members of the British Columbia Sleep Apnea Society and with colleagues, it was determined that the questionnaires took about 20 minutes to complete. They are described below: i) Eating Behaviour A l l three sub-scales (restraint, disinhibition, hunger) of the Three Factor Eating Questionnaire (Stunkard and Messick, 1985) were used (see Appendix J). The restraint scale assesses cognitive control of eating behaviour. The hunger scale measures perceived hunger, 17 Chapter III: Methodology and the disinhibition scale assesses food intake disinhibition, two constructs which are believed to be related to binge eating. Three types of questions are included in this questionnaire. For the first type, respondents are required to read statements and circle whether each statement is true or false for them. For the second type of question, respondents are asked to read questions or statements and select the most applicable response from the options provided. The final type of question involves selecting the most applicable statement from a list of options. The Binge Eating Scale (Gormally and Black, 1982), designed to assess binge eating severity, consists of several sets of statements (see Appendix K). Respondents are to choose, from each set, the one they feel best describes themselves. The Night Eating Syndrome Scale (Stunkard, 1955) requires respondents to respond "yes" or "no" to five questions relevant to the syndrome (see Appendix L). ii) Physical Activity Physical activity was assessed with the Paffenbarger questionnaire (Paffenbarger, Wing, & Hyde, 1978), which includes questions on how many blocks are walked daily, frequency of vigorous activity, and whether activity levels have changed over the past year (see Appendix M). The questionnaire also has a section in which respondents estimate time spent each weekday and weekend day in various activity categories. iii) Psychological Variables Three standardized psychological scales were used (see Appendices N-P): The Zung self-rated depression scale (ZSRDS) (Zung, 1965), the Zung self-rated anxiety scale (ZSRAS) (Zung, 1971), and the Rosenberg self-esteem scale (RSES) (Rosenberg, 1979). Each consists of a single page list of statements, which respondents read and indicate the degree to which each applies to them, using a four point scale of written descriptors. Each response is scored from one to four, 18 Chapter III: Methodology higher scores being given to responses more reflective of depression, anxiety, or low self-esteem. Scores for all scale items are then totaled. ZSRDS and ZSRAS scores are adjusted so the maximum total score is 100 (raw score is divided by 80, then multiplied by 100); the RSES score is not adjusted. f) Interview Guide For study variables not addressed by the written questionnaires, an interview guide was developed. It consisted of questions pertaining to sleep apnea symptoms, weight history, history of weight loss attempts, perceived relationship between sleep apnea symptoms and food intake, nocturnal binge eating history, and life events and stressors (at onset and current). The interview guide was pre-tested with five members of the British Columbia Sleep Apnea Society, for question flow and length, and revised prior to use. The interview was designed to take about 30 minutes to complete. The interview guide is shown in Appendix Q. g) Data Analysis a) Food Intake Data The energy and macronutrient content of subjects' diets was determined using Food Processor Plus (ESHA Research, Salem, OR), a PC-based nutrient analysis program based on the Canadian Nutrient File. Food records were analyzed, as follows: i) For each food record, food intake data for two daily time categories (<5 pm and >5pm) were analyzed separately, using time categories established by Nutrition Canada (1973), to assist in determining daily intake distribution ii) For each record, total intake for each day was also determined iii) Mean intake for the three day period was determined, as well as the mean intake for each segment of the day 19 Chapter III: Methodology Food records were also manually reviewed to determine the daily number of servings from the food groups of Canada's Food Guide, estimated to the nearest half serving. These data were entered and verified as described below. Mean daily intake was computed. b) Other Data Statistical analyses were performed using SPSS for Windows, version 7.5 (SPSS Inc., Chicago, IL). Data were entered, then manually verified against source data. Any errors identified were corrected prior to conducting statistical analyses. The data entry for open interview questions was completed using a two-phase process. First, actual patient responses were coded. Next, response categories were developed and each individual response was assigned to a response category. The following types of analyses were completed: • For comparisons between two groups involving continuous data (e.g., BMI of participants vs non-participants), unpaired t-tests were used. • For within group comparisons involving continuous data (e.g., hours of vigorous exercise, weekday vs weekend), paired t-tests were used. • Bivariate and partial correlational analyses were used to identify associations between variables (e.g., binge eating scores and RDI). Because multiple comparisons were performed, increasing the likelihood of Type I errors (false positive correlations), findings were interpreted conservatively (identifying patterns of association rather than attaching significance to single correlations; correlations with lower P-values were emphasized). • For comparisons involving population proportions, chi-square was used. A l l comparisons were made at a significance level of P<0.05 20 Chapter IV: Results RESULTS 1. Recruitment An overview of the process that resulted in the recruitment of 49 subjects meeting study criteria is provided in Figure 1. From the 625 respiratory sleep studies conducted during the recruitment period, 100 patients were identified who appeared to meet all study inclusion criteria. Due to scheduling factors or patient involvement in other research studies, almost half of these patients could not be approached about the study. However, almost all patients invited to participate agreed to do so. Initially, 53 men were recruited into the study. Of these, three were subsequently eliminated due to hypothyroidism (1), diabetes (1), or use of an appetite suppressant (1), identified during data collection. One additional patient initially agreed to participate, but subsequently did not contribute any data, as he cancelled all clinic appointments and did not respond to correspondence. There were no significant differences in age, BMI, or RDI between study participants (n=49) and non-participants (those who met study criteria, but were either not invited or chose not to participate, n=48). Virtually all (48,98%) study participants had attended the sleep lab during the week, whereas a substantial proportion (20,42%) of the non-participants had attended the sleep lab on a weekend, when the investigator was generally unavailable to recruit subjects. Because of the possibility that patients seeking weekend appointments may have had different characteristics from those seeking weekday appointments, a comparison of age, BMI, RDI and occupation category (professional/technical versus other) was conducted between these patients and study participants. No significant differences in age, BMI, or RDI were found; however, a greater 21 Chapter IV: Results proportion of the patients attending the sleep lab on weekends had professional/technical occupations, 89% versus 37% (Chi-square, p=.001). Health, demographic, and interview data were obtained for all 49 subjects; written surveys and food records were obtained from 32 (65%). 22 Chapter IV: Results Figure 1. Overview of Study Recraitment Process Screened Sleep Lab Patient Lists and/or Patient Charts (n=625) Eliminated (n=460) Screened Preliminary Sleep Study Results (n=165) Eliminated patients with Inconclusive studies (n=4) No significant OSA (n=61) Female Patients: (n= 131, 21%) Male Patients: - B M K 3 0 (n=106, 17%) - Those on treatments for OSA, i.e., CPAP, oral appliance, upper airway surgery (n=T88, 30%) - Those meeting BMI criteria, but with study exclusion factors (n=35, 6%): - diabetes (n=12) - specific medications (n=3) - psychiatric illness (n=7) - hypothyroid (n=2) - various medical disorders (n=10) -no English (n=l) Patients Meeting Study Criteria (n=100) T Approached (n=55) Not approached (n=45): - Patient in another study (n=4) - Patient schedule factors (n=8) - Investigator unable to see >1 subject/day (n=6) - Investigator not available (weekends n=20; weekdays n=7) Declined to participate (n=2) Study Participants (n=49) Subsequent elimination or lack of participation (n=4): - hypothyroid (n=l) - diabetes (n=l) - use of an appetite suppressant (n=l) - no participation following recruitment (n=l) 23 Chapter IV: Results 2. Subject Characteristics Subjects' age, anthropometric measurements and OSA severity are presented in Table 1. Subjects were predominantly middle-aged. A l l were obese, with mean waist and neck circumferences suggestive of upper body obesity. Their OSA status was variable: 17 (35%) had scores suggestive of mild (RDK25), 18 (37%) moderate (RDI=25-50), and 14 (29%) severe OSA (RDI>50). Table 1. Age, anthropometric measures and sleep apnea severity of obese men with OSA (n=49) Characteristics Mean + SD Range Valid N Age (years) 50+12 21-74 49 Height (cm) 174 + 7 156-188 49 Weight (kg) 111 + 19 81 -158 49 BMI (kg/m2) 36.5 + 6.2 29.3 -52.8 49 Neck Girth (cm) 45 + 4 39-53 49 Waist Circumference (cm) 117+14 100 -157 47 Hip Circumference (cm) 111 + 10 103 -144 47 Waist Hip Ratio 1.0 + 0.1 .9-1.1 47 OSA Severity: RDI (events per hour of sleep): 40.8 + 27.2 9.5-108 49 - Apneas, obstructive 12.7+17.5 - Apneas, central .3+ .6 - Apneas, mixed .9 + 3.4 - Hypopneas 26.6+19.1 Minimum level of arterial oxygen 70.4+16.8 30.0 - 89.3 49 saturation with respiratory events (%) 24 Chapter IV: Results Bivariate analyses were conducted to identify associations between age, anthropometric measures, and OSA severity (Table 2). Age was not strongly correlated with anthropometric measures or OSA severity. Anthropometric measures (except height) were correlated with one another, and many were correlated with minimum Sa0 2 , but not RDI. The two measures of OSA severity, RDI and minimum Sa0 2 , were inversely correlated. Table 2 . Corre la t ion matrix for age, anthropometric, and polysomnographic measures in obese men wi th O S A ( n = 4 7 - 4 9 ) + Age Height Weight BMI Neck Waist Hip WHR RDI Min. sat. Age 1 Height -.19 1 Weight -.33* .33* 1 BMI -.26 -.15 .88*** 1 Neck -.15 .07 .76*** 1 Waist -.11 .03 g7*** gg*** .78*** 1 Hip -.21 .06 .85*** ,86*** gy*** .86*** 1 WHR .07 -.03 54*** .55*** 6j*** 76*** .32* 1 RDI -.24 -.14 .25 .33* .24 .20 .33* -.03 Min. O z .15 .20 -.51*** -.63*** -.56*** -.50*** -.52*** -.28 saturation T'earson correlation coefficients are shown; all significance levels were 2-tailed *p<0.05 **p<0.01 ***p<0.001 25 Chapter IV: Results The subjects' coexisting health conditions are presented in Table 3. One-third had no diagnosed coexisting conditions, while the remainder had an array of conditions, the most prevalent being hypertension, accident-related injuries, coronary artery disease, elevated cholesterol, respiratory conditions other than OSA, and arthritis or other conditions affecting joint function. About half of the subjects reported using some form of prescription medication at the time of the study (see Table 4). Anti-hypertensives were the most common medication, used by about one quarter of subjects. Considerably fewer subjects reported using other types of prescription medication, which included lipid-lowering agents, pain killers, bronchodilators/anti-inflammatory inhalers, anti-anginal drugs, gastric acid inhibitors, and antibiotics. A few subjects reported using non-prescription medications or supplements, including aspirin, antihistamines, and vitamins or minerals. However, data on use of these items may not have been complete, as patients were not systematically questioned about their use, and some patients may only have reported use of prescription medications. 26 Chapter IV: Results T a b l e 3 . C o e x i s t i n g hea l t h cond i t i ons i n obese m e n w i t h O S A ( n = 4 9 ) 1 H e a l t h C o n d i t i o n N u m b e r o f S u b j e c t s 2 % None identified Hypertension: - currently treated (13); history of hypertension, or medical exam suggestive, but no treatment currently (7) Accident-related injuries Coronary artery disease Elevated cholesterol Respiratory diseases: asthma (2); obstructive or restrictive lung disease (4) Arthritis or other chronic joint problem Other: - Gout (3); hiatus hernia or gastro-esophageal reflux (3); allergic rhinitis (2); fatty liver (1); hepatomegally (2); mild glucose intolerance (1) ; heart murmur or heart valve dysfunction (2) ; ankle edema (1); previous skin cancer (1); migraine headaches (1); diverticular disease (1); former drug and alcohol abuse (1); reactive polycythemia (1); fibromyalgia (1); cold urticaria (1); irritable bowel syndrome (1); shingles (1); myasthenia gravis, bulbar type (1) 16 20 7 6 6 6 5 17 33 41 14 12 12 5 10 34 'Note that patients with diabetes, hypothyroidism, or psychiatric illnesses were excluded from participating in the study Note that some subjects had more than one of the conditions listed 27 Chapter IV: Results T a b l e 4. M e d i c a t i o n s used b y obese m e n w i t h O S A (n=49) M e d i c a t i o n s U s e d Prescription Medications: - None - Anti-hypertensives - Lipid-lowering agents - Pain killers - Broncho-dilators/anti-inflammatory inhalers - Anti-anginal - Gastric acid inhibitors - Other: anti-inflammatory nose drops (1), diuretic (1), supplementary oxygen (1) Non-prescription Medications: - Aspirin - Antihistamines - Vitamin or mineral supplements N u m b e r o f Sub jec ts % 23 47 13 26 4 8 4 8 3 6 3 6 3 6 3 6 4 8 2 4 2 4 Subject demographic characteristics are presented in Table 5. About one-third of subjects lived within greater Vancouver; the others lived in geographically diverse areas of British Columbia and the Yukon. Almost all were white, and presently married or living with a partner. Over half were employed, while almost one quarter were unemployed, on medical leave or on a disability pension. Subject occupations were varied (see Table 6). The most prevalent single subject occupation was truck or bus driver. 28 Chapter IV: Results T a b l e 5. D e m o g r a p h i c cha rac te r i s t i c s o f obese m e n w i t h O S A (n=49) C h a r a c t e r i s t i c s N u m b e r o f Sub jec ts % Place of residence: - Greater Vancouver 19 39 - Central Interior/Cariboo 8 16 - North Coast/North West 8 1 6 - Vancouver Island/Gulf Islands 5 10 - Kootenays 3 6 - Fraser Valley 2 4 Sunshine Coast 2 4 Southern Interior 1 2 - Yukon 1 2 Ethnicity: - White 41 84 - Non-white: - Asian 3 6 - First Nations 3 6 - East Indian 2 4 Marital Status: - Married/with partner 41 84 - Single 7 14 - Divorced/separated 1 2 Employment Status: - Employed 30 61 - On disability pension or medical leave 8 16 - Retired 7 14 - Unemployed 3 6 - Student 1 2 29 Chapter IV: Results Table 6. Occupations of obese men with O S A (n=49) Occupation Number of % Subjects Professional/technical 18 37 Bank manager (2); businessman, marketing (1); computer supervisor (1); court director (1); lawyer (1) ; mathematician (1); park ranger (1); photographer, internet publisher (1); physician (1); police officer (2) ; professional engineer (1); pulp mill technician (1); research chemist (1); school principal (1); school teacher (2). Non-professional, trades 29 59 Assistant manager, tire store (1); bar tender (1); bus boy (1); bus or truck driver (9); construction contractor (1); construction worker (1); cook (1); dairy foreman (1); farmer or rancher (2); fisherman (1); labourer (1); logger (1); maintenance superintendent (1); meat cutter (1); mechanic (1); mill worker (2); supervisor, liquor store (1); train conductor (1); welder (1) Other college student (1); unknown (1) 2 4 To examine whether subjects varied by occupation category in age, anthropometric findings, or OSA severity, unpaired t-tests were used to compare subjects in professional-technical versus other types of occupations. Subjects in both groups were similar in age (50 +13 vs 50 + 11 years, P=.85). However, subjects with professional/technical occupations were less obese (BMI 33 + 5 vs 39 ± 6, P=.001), had smaller waist circumferences (111 ± 12 vs 121 + 14 30 Chapter IV: Results cm, P=. 02) and neck circumferences 43 + 3 vs 47 + 4 cm, P=.000), and tended to have less severe OSA (RDI, 34 + 23 vs 47 + 29, P=.08; minimum Sa0 2 , 78 + 12 vs 65+18, P=.003) than those with other types of occupations. The subjects' smoking status and reported alcohol intake are summarized in Tables 7 and 8. About one quarter of the subjects were current smokers; a similar proportion were lifetime non-smokers, and about half were former smokers. Reported alcohol intake was variable. About one third of subjects reported abstaining from alcohol. Complete alcohol intake histories were not available; however, during the interview segment of the study, several subjects mentioned previous problems with alcohol dependency. Five subjects appeared to be very heavy drinkers, consuming more than 14 drinks per week. T a b l e 7. S m o k i n g status o f obese m e n w i t h O S A (n=49) S m o k i n g S ta tus N u m b e r o f Sub jec ts % Smoker 13 26 Former smoker: 24 49 - Quit < 5 years ago (11) - Quit >5 and < 10 years ago (3) - Quit > 10 years ago (10) Non-smoker 12 24 31 Chapter IV: Results T a b l e 8. A l c o h o l i n t ake repo r ted to p h y s i c i a n b y obese m e n w i t h O S A (n =49) I n take ca tego ry N u m b e r o f Sub jec ts % Alcohol abstainer 18 37 Light drinker (< 7 drinks per week, or intake reported as "seldom") 20 41 Moderate drinker (>7&<14 drinks per week or intake reported as "modest") 4 8 Heavy drinker (> 14 drinks per week) 5 10 Intake level unspecified 2 4 3. OSA Treatment Status The OSA treatment recommendations and outcomes for all subjects are presented in Table 9. The most common physician treatment recommendation was CPAP, recommended to about three-quarters of study participants. Several months following the initial sleep study, more than one third of these patients had either not pursued CPAP or had encountered problems with its use and were not using it regularly. Nine of 11 patients who had been advised to pursue treatments other than CPAP (including lifestyle interventions) had not returned to the clinic; therefore, their treatment outcomes were unknown. Only two patients attending follow-up visits reported achieving weight loss in the months following their diagnosis with OSA. 32 Chapter IV: Results Table 9. O S A treatment recommendations and outcomes 1 for obese men with O S A (n=49) Treatment Recommendations and Outcomes Number of % Subjects CPAP recommended as primary treatment: 38 78 - CPAP reported to be in use (22, 58%) - did not attend any follow-up appointments (7, 18%) - reported difficulties in using CPAP (4,11%) - has agreed to try CPAP, but has not yet obtained machine (3, 8%) - pursuing laser surgery (1,3%) - achieved some weight loss (amount not documented), did not pursue CPAP (1,3%) Other treatment options recommended as 11 22 primary treatment: - no treatment outcome documented (9, 82%) - pursuing laser surgery (1) - achieved weight loss of 4 kg (1) 'As assessed by chart review, conducted > 6 months following diagnostic sleep study 33 Chapter IV: Results 4. Food Records and Questionnaires Thirty-two subjects completed and returned written questionnaires and food records. The questionnaires from one subject could not be used, as he had initiated CPAP therapy prior to completing them. Thus, usable questionnaires were obtained from 31 (63%) subjects. Respondents and non-respondents to the written questionnaires were almost identical in BMI (36 + 6 vs 36 + 7, unpaired t-test, P=991). Respondents appeared to have a lower mean RDI than non-respondents (36 + 25 vs 50 + 29), although these differences were not significant (unpaired t-test, P=. 131). However, respondents were significantly older (53 + 12 vs 43 + 11 years, unpaired t-test, P=.013). a) Food Intake Food Records Analyses were conducted on 30 food records, as one subject's food record was insufficiently detailed to be analyzed. An additional subject had provided insufficient detail on one of three days of his food record; his intakes from the two remaining days were averaged. Subjects' reported mean daily intakes of energy and macronutrients are summarized in Table 10. Reported energy intake was highly variable; however, the majority of subjects reported intakes that would theoretically allow weight maintenance or loss. About one third of subjects (12,40%) had reported intakes within ten percent of Canadian reference standards for males 25-74 (Health and Welfare Canada, 1990), while half of subjects (15, 50%) reported intakes more than ten percent below the reference standard. Only three subjects reported intakes greater than ten percent above the reference standard. On average, energy intake was fairly evenly distributed between day and night, although there was considerable individual variability. 34 Chapter IV: Results About two-thirds of subjects (21, 70%) had fat intakes higher than the recommended 30% of calories (Canadian Consensus Conference on Cholesterol, 1998; National Cholesterol Education Program, 1994; Kraus, Deckelbaum, & Ernst, 1996). Thirteen of these subjects had fat intakes in excess of 35% of calories. Most subjects also had protein intakes well in excess of recommendations (Health and Welfare Canada, 1990). Alcohol consumption was reported on food records by eight of the 30 subjects (27%). According to their medical charts, six of eight were habitual alcohol consumers, drinking some alcohol each week. Of the 22 subjects who did not include alcohol on their food records, 15 had indicated to their physician that they either abstained from alcohol, or drank infrequently. The remaining seven had reported usual consumption of 1-7 drinks per week. T a b l e 10. M e a n d a i l y ene rgy a n d m a c r o n u t r i e n t i n take o f obese m e n w i t h O S A (n=30) I n take M e a n + S D R a n g e Energy (kcal) 2274 + 736 1123 -4825 (kJ) 9518 + 3081 4700- 20,195 Distribution: Percent consumed < 1700 hours 55+13 32 -89 Percent consumed > 1700 hours 45+13 11 -68 Protein: - grams/day 101 + 50 37-•307 - % calories 18 + 7 10 -45 Fat: - grams/day 86 + 42 31 - 227 - % calories 33 + 7 17 -44 Carbohydrate: -grams/day 274 + 75 145 -482 - % calories 50 + 9 26 -66 Alcohol: - % calories 2 + 4 0- 17 35 Chapter IV: Results Estimated daily servings from the food groups of Canada's Food Guide (CFG) are summarized in Table 11. As with calculated intake, mean intakes of servings from the food groups varied considerably. Collectively, subjects' diets appeared adequate in grain products and meats and alternatives, marginal in vegetables and fruits, and low in milk products. About one half (16, 53%) of subjects appeared to be consuming more servings of meats and alternatives than recommended, while half or more of subjects had reported consumption of vegetables and fruits (16, 53%) and dairy products (26, 87%) below the minimum intake recommendation. T a b l e 11 . M e a n d a i l y se rv ings f r o m each food g r o u p o f C a n a d a ' s F o o d G u i d e ( C F G ) consumed by obese men w i t h O S A (n=30) F o o d G r o u p C F G In take R e p o r t e d R a n g e R e c o m m e n d a t i o n s ( M e a n + S D ) G r a i n P r o d u c t s 1 542 6.5 ±2.4 .3 -12 F r u i t s a n d Vege tab les 5-10 4.7 + 2.2 1.3-9 M e a t a n d A l t e r n a t i v e s 2-3 2.9+1.1 .7 -5 M i l k P r o d u c t s 2-3 1.0+9 0-3.5 Note that sweet baked goods (e.g., pies, cakes) were not included Subjects had recorded food intake on both weekdays (n=71, 79%) and weekends (n=T9, 21%). For those subjects who recorded intake on both weekdays and weekends (n=10), intake was similar between the two types of days (2337 + 1002 vs 2230 + 576 calories, paired t-test, P=76). To determine whether reported food intake varied by occupation category, unpaired t-tests were used to compare subjects with professional/technical versus other types of occupations. No significant differences were found in energy intake (2288 + 947 vs 2256 + 537 36 Chapter IV: Results calories, P-.91), or percentages of energy consumed as protein (18 + 8 vs 18 + 5, P=.81), fat (31 + 9 vs 34 + 6, P=.43), or carbohydrate (50+10 vs 49 + 8, P=.67 ). Bivariate correlations among food intake variables are presented in Table 12. Total energy intake was associated positively with the percentage of energy consumed as fat, negatively with the percentage of energy consumed as carbohydrate, and not associated with the percentage of energy consumed as protein. Daily servings of grain products, meats and alternatives, and milk products were associated with energy intake, but daily servings of vegetables and fruits were not. Associations between percentage of energy consumed in the evening and most of the food intake variables were weak; however evening intake was associated with lower intake of milk products. There were some correlates among macronutrient and food group variables. As expected, there was a relationship between the percentage of energy consumed as protein and the number of daily servings of meats and alternatives and milk products. Fat intake was inversely associated with carbohydrate intake. Intake of milk products was significantly correlated with daily servings of grain products and meats and alternatives. Correlations between daily servings of vegetables and fruits and other food intake variables were weak. 37 Chapter IV: Results T a b l e 12. C o r r e l a t i o n m a t r i x f o r food i n take v a r i a b l e s f o r obese m e n w i t h O S A ( n = 3 0 ) + Energy Intake % Calories >5pm % Calories From Protein % Calories From Fat % Calories From CHO C F G Grain Products C F G Veg.& Fruits C F G Meats & Alt. C F G Milk Products Energy Intake 1.0 % Calories >5pm -.05 1.0 % Calories From Protein -.05 -.26 10 . % Calories From Fat .55** .26 -.22 1.0 % Calories From CHO -.49** -.15 -.10 -.78*** 1.0 C F G Grain Products .50** -.23 .15 .08 .05 1.0 C F G Veg. & Fruits -.25 -.19 .20 -.21 .27 -.13 1.0 C F G Meats & Alt. .45* -.28 .42* .44* -.59** .18 .12 1.0 C F G Milk Products 54** -.38* .44* .02 -.13 .53** -.14 .41* 1.0 ++Pearson correlation coefficients are shown; all significance levels are 2-tailed *p<0.05 **p<0.01 38 Chapter IV: Results Correlations between food intake variables and age, BMI, and measures of OSA severity are presented in Table 13. Age was associated with percentage of energy consumed as carbohydrate and inversely associated with percentage of energy consumed as fat. As well, a significant association between daily servings of meats and alternatives and RDI was found. There were no other significant associations between these variables. Table 13. Correlat ion matrix for food intake variables, age, B M I , and measures of O S A severity + Energy % % % % CFG CFG CFG CFG Intake Calories Calories Calories Calories Grain Veg. & Meats & Milk >5pm From From From Products Fruits Alt. Products Protein Fat CHO Age -.33 .01 .28 -.39* .40* .09 .38* -.35 -.08 BMI .00 -.02 .14 .05 -.07 .08 .09 .01 .06 RDI .01 -.18 .22 -.02 -.07 .29 .07 .39* .18 Min. 0 2 .14 .27 -.18 .09 -.01 -.12 -.22 -.31 -.24 Sat. +Pearson correlation coefficients are shown; all significance levels are 2-tailed *p<0.05 **p<0.01 Verbal alcohol and coffee consumption history As part of the brief verbal food frequency questionnaire used to augment food records, all 49 study subjects were interviewed about their usual alcohol consumption. To examine whether there was agreement between alcohol intake estimated by clinic respirologists and the investigator, estimated weekly intakes were compared. Ten subjects had described their alcohol intake to their respirologist in words rather than quantities so these intakes could not be directly compared to the data collected by the investigator. For the 39 subjects for whom there was complete quantitative data, there was good agreement between intakes reported to the two 39 Chapter IV: Results sources (r=.88, p=.000). For 36 subjects (92%), discrepancies between intake reported to the two sources were small (less than two drinks per week for 33 subjects, and 3-6 drinks per week for three subjects). Substantial discrepancies in reported intakes existed for three subjects, ranging from 7-26 drinks per week. In two of these cases, higher intakes had been reported to the investigator than to the respirologist, and in one case the higher intake had been reported to the respirologist. Because a complete set of alcohol intake data was obtained by the investigator, these data were used in correlational analyses to examine whether alcohol intake was related to degree of obesity. There was a relatively weak inverse correlation between reported alcohol intake and BMI (r=-.24, P=. 10). Subjects reported a mean coffee intake of 3.6 + 3.8 cups per day (range 0-18 cups). Ten subjects (20%>) reported drinking no coffee, while 20 (41%) drank from 1-3 cups per day, and 19 (39%) drank four or more cups. Coffee intake was not significantly correlated with other food intake variables or measures of OSA severity. Summary In summary, subjects reported variable levels of energy intake, with most reporting intakes that would theoretically be associated with weight maintenance or loss. Prevalent nutritional issues included high intakes of meats and alternatives and fat, and low intakes of vegetables and fruits and dairy products. Reported intake of alcohol was variable, with extremes ranging from no consumption to several drinks per day. There was good agreement between alcohol intake reported to respirologists and the investigator. Coffee intake varied greatly, and was not associated with other food intake variables. No food intake variables were associated 40 Chapter IV: Results with BMI, and most associations between food intake variables and measures of OSA severity were weak. b) Eating Behaviour Subject Binge Eating Scale (BES) and Three Factor Eating Questionnaire (TFEQ) scores are summarized in Table 14. Although BES scores were variable, most respondents had scores in the lower end of the possible range, associated with lesser likelihood of binge eating problems. No subjects had scores in the range suggestive of binge eating disorder (BED). Similarly, on the disinhibition sub-scale of the TFEQ, few subjects had scores suggestive of binge eating. Scores on the hunger sub-scale of the TFEQ were variable, although about two-thirds of subjects had scores in the range associated with non-binge eaters. On the restraint subscale of the TFEQ, about one-third of subjects had low scores suggestive of low intention to control food intake. About half of subjects had moderate scores, while less than a quarter had high restraint scores. 41 Chapter IV: Results T a b l e 14. B i n g e E a t i n g S c a l e a n d T h r e e F a c t o r E a t i n g Q u e s t i o n n a i r e scores o f obese m e n w i t h O S A (n=30-31) Q u e s t i o n n a i r e M e a n + S D R a n g e Binge Eating Scale (possible range 0-46)1: 9.2 + 6.2 0 -23 - < 17 (little or no problem with binge eating): 27 (87%) - 18-26 (moderate problem): 4 (13%) - > 27 (severe problem; likely Binge Eating Disorder): 0 (0%) Three Factor Eating Questionnaire2: Restraint subscale (possible range 0-21)3: 7.5 + 4.3 0 -16 < 5 (low restraint): 9 (30%) > 5 & < 13: 16(53%) > 13 (high restraint): 5(17%) Disinhibition subscale (possible range 0-16)4: 7.0 + 2.9 2 -16 < 8 (unlikely binge eating problem): 23 (77%) 9 (possible binge eating problem): 2 (7%) > 10 (likely binge eating problem): 5 (17%) Hunger subscale (possible range 0-14) 5.8 + 3.7 0 -13 < 7 (range associated with non-binge eaters): 21 (70%) > 8 (range associated with binge eaters): 9 (30%) Cut-off points defined as per (Gormally, Black, Daston, et al., 1982; Marcus, Wing & Lamparski, 1985) 2 Note that n=30 for TFEQ (an invalid questionnaire was received from one subject) 3 Cut-off points defined as per Lawson, Williamson, Champagne, Delany, Brooks, Howat, et al., 1995 4 Cut-off points defined using data for obese binge eaters in Adami, Gandolfo, Bauer, et al., 1995 42 Chapter IV: Results To examine whether subjects differed in their mean scores on BES scale items addressing eating behaviour versus cognitions related to eating, scores on scale items addressing these two dimensions of binge eating were compared (Table 15). No significant differences were found. Table 15. Comparison of mean scores on Binge Eat ing Scale items reflecting eating behaviour versus cognitions for obese men wi th O S A (n=31) + Items reflecting eating behaviour 1 Items reflecting cognitions 2 P + B E S Scale Item Scores (Mean + SD) .20 + .13 .18 +.18 .42 Items 2,4, 5, 7-13; note that higher scores reflect eating behaviours associated with Binge Eating Disorder 2 Items 1, 3, 6, 14-16; note that higher scores reflect cognitions associated with Binge Eating Disorder + Comparisons were made using student's paired t-test To determine whether BES and TFEQ scores varied by occupation category, unpaired t-tests were used to compare subjects with professional/technical versus other types of occupations. No significant differences were found in TFEQ sub-scale scores (restraint, 6.8 + 4.0 vs 7.8 + 4.4, P=.54; disinhibition, 6.5 + 2.4 vs 7.6 + 3.3, P=.33; hunger, 4.8 + 3.5 vs 6.8 + 3.7, P=.14); however the subjects with professional/technical occupations had significantly lower binge eating scores (6.1 + 5.7 vs 11.9 + 5.8, P=.01). Bivariate correlations among BES and TFEQ scores are shown in Table 16. BES scores were significantly correlated with all TFEQ sub-scale scores, although the relationship between BES and TFEQ restraint scores was relatively weak. Most of the TFEQ sub-scale scores were significantly correlated with each other. 43 Chapter IV: Results T a b l e 16. C o r r e l a t i o n m a t r i x f o r B i n g e E a t i n g S c a l e ( B E S ) a n d T h r e e F a c t o r E a t i n g Q u e s t i o n n a i r e Scores ( T F E Q ) (n=30-31) + B i n g e E a t i n g S c a l e T h r e e F a c t o r E a t i n g Q u e s t i o n n a i r e : R e s t r a i n t D i s i n h i b i t i o n H u n g e r B E S 1 T F E Q : R e s t r a i n t .43* 1 T F E Q : D i s i n h i b i t i o n .75*** 351 1 T F E Q : H u n g e r . 6 8 * * * .50** 71*** \ + Pearson correlation coefficients are shown; all significance levels are 2-tailed *p<0.05 **p<0.01 ***p<0.001 1 p=0.06 Bivariate correlations between eating behaviour scores and age, BMI, measures of OSA severity, and food intake are presented in Table 17. Eating behaviour scores were weakly correlated with age, RDI, and most food intake variables. BES and TFEQ disinhibition scores were correlated with BMI; associations between TFEQ hunger and TFEQ restraint scores and BMI were not significant. Although most eating behaviour scores had significant negative correlations with minimum Sa0 2 , these significant associations were no longer present when partial correlational analyses were performed controlling for BMI. 44 Chapter IV: Results T a b l e 17. C o r r e l a t i o n m a t r i x f o r ea t i ng b e h a v i o u r scores a n d age, B M I , measu res o f O S A seve r i t y , a n d food i n take (n=30-31) + Binge Eating Scale Three Factor Eating Questionnaire: Restraint Disinhibition Hunger Age -.09 .05 .03 .23 BMI .71** .36' .39* .23 RDI .16 -.13 .17 .21 Min. 0 2 Saturation -.52** -.03 -.36' -.41* Energy Intake -.09 -.351 -.18 -.04 % Calories From Protein .20 .16 .361 .38* % Calories From Fat -.01 -.24 -.31 -.22 % Calories From CHO -.14 .28 .16 .10 +Pearson correlation coefficients are shown; all significance levels are 2-tailed *p<0.05 **p<0.01 1 trend toward significance p=.05-.07 45 Chapter IV: Results Subject responses to the NES scale are presented in Table 18. While about half of respondents reported waking with little appetite and therefore beginning eating later in the day, considerably fewer reported other behaviours suggestive of NES, which include eating through the evening without enjoyment, feeling tense following such eating episodes, and having difficulty falling asleep. No subjects met all of the criteria for NES. Table 18. Night Ea t ing Syndrome Scale responses of obese men with O S A (n=31) Night Ea t ing Syndrome Scale Number of % Subjects Scale items: 1. W a k e up wi th little appetite 16 52 2. Usually begin eating later in the day 15 48 3. Ea t ing through the evening without enjoyment 6 19 4. Feeling tense and upset after eating through the evening 3 10 5. Frequent difficulty going to sleep 8 26 A l l scale items present 0 0 In summary, subjects' overall BES and TFEQ disinhibition scores were not high; however they were significantly correlated with BMI, indicating increasing propensity for binge eating with increasing relative weight. Subjects with professional/technical occupations had lower binge eating scores than those with other types of occupations. Most TFEQ hunger scores were not in the range associated with binge eating. TFEQ restraint scores suggested that subjects varied in their intention to restrict intake. BES and most TFEQ subscale scores were associated with one another. TFEQ and BES scores were not significantly correlated with measures of OSA severity, or most food intake variables. No subjects had the full set of experiences suggestive of NES. 46 Chapter IV: Results c) Physical Activity Although 31 subjects completed the physical activity questionnaire, some completed it incorrectly or did not fully complete all sections. Subjects had the most difficulty with the section involving estimation of hours per day spent engaging in activities of varying intensity. In addition, a few subjects were unable to estimate blocks walked or stairs climbed per day; some offered descriptive terms rather than numbers (e.g., "lots", "few"). When possible, subjects were contacted and responses clarified. Complete and valid responses to individual questions were obtained from 26 or more subjects. Responses are summarized in Table 19. 47 Chapter IV: Results T a b l e 19. P h y s i c a l ac t i v i t y , as assessed b y the P a f f e n b a r g a r q u e s t i o n n a i r e , f o r obese m e n w i t h s leep a p n e a (n=31) Q u e s t i o n n a i r e C o m p o n e n t s M e a n + R a n g e V a l i d N SD Blocks walked per day (12 blocks = 1 mile): 15+ 16 1 -72 29 - <6: 11(38%) - > 6 & < 12: 8(28%) - >12: 10(35%) Walking pace: - Casual: 6(21%) - Average: 14(48%) - Fairly brisk: 9(31%) - Brisk: 0(0%) Stairs climbed per day: 77 + 68 0-240 26 - < 50 or "few": 12(43%) - 50-100: 9(32%) - >100 or "lots": 7(25%) Sweat-producing activities (times/week): 2 + 3 0 -10 31 - Never: 13(42%) - 1-2: 8(26%) - 3-4: 6(19%) - >4: 4(13%) Sweat-producing activities (n= 18)1: - Walking (9,29%) - Work (9,29%): Driving truck (1), working as park ranger (1), yard work (5), work around the house (1); "house cleaning, I have seven children" (1) - Cycling (3,10%) - Weight lifting (1,3%) - Jogging (1, 3%) 1 Some subjects identified more than one activity, therefore, number of responses exceeds number of respondents 48 Chapter IV: Results About two-thirds of subjects reported walking a mile or less per day, and most reported walking at an average or casual pace; none reported walking briskly. Almost half were infrequent stair climbers, and a similar proportion reported never engaging in sweat-producing activity. Of the subjects who did report participating in sweat-producing activity (n=18), the most frequent types reported were walking and work-related activity (usually work around the house). Few subjects mentioned any other types of activities. To determine whether the reported number of blocks walked or hours per week of sweat-producing activity varied by occupation category, unpaired t-tests were used to compare subjects with professional/technical versus other types of occupations. No significant differences were found (blocks walked, 18 + 11 vs 13 + 19, P=.36; hours per week of sweat-producing activity, 1.4+ 1.8vs2.3 + 3.0,P=.28). Subject estimations of daily time spent in activities of varying intensity are presented in Table 20. Despite the relative infrequency with which subjects reported participating in sweat-producing activities in the previous question, in this section subjects reported spending an average of one hour per day in vigorous activity, which was stable from weekdays to weekend days. Time allocation for the other activity categories varied significantly from weekdays to weekend days. On weekends, subjects reported sleeping more, and participating in more moderate intensity and less light intensity activities. 49 Chapter IV: Results T a b l e 20 . T i m e spent e n g a g i n g i n ac t i v i t i es o f v a r y i n g in tens i ty by obese m e n w i t h O S A (n=27) A c t i v i t y C a t e g o r y H o u r s / d a y w e e k d a y (mean + S D ) H o u r s / d a y w e e k e n d (mean + S D ) P + V i g o r o u s 1 0.9+1.3 1.0+1.2 0.472 M o d e r a t e 2 2.3 ± 1.9 3.8 + 3.1 0.005** L i g h t 3 13.0 + 2.8 11.1 + 3.7 0.005** S l e e p i n g 7.8+1.3 8.1 + 1.2 .028* Digging in the garden, strenuous sports, jogging, chopping wood, heavy carpentry, bicycling on hills, etc. 2 House work, light sports, walking, yard work, lawn mowing, painting, household repairs, light carpentry, bicycling on level ground, etc. 3 Sitting, office work, driving a car, eating, personal care, etc. + Comparisons were made using student's paired t-test Statistically significant at P<0.05 Statistically significant at P<0.01 The final questions addressed changes in activity over the previous year (see Table 21). About three-quarters of subjects felt their activity level hadn't changed, while all but one of the remaining subjects felt that their exercise level had decreased. Subjects cited health-related and scheduling reasons for exercising less. T a b l e 2 1 . P e r c e i v e d changes i n p h y s i c a l a c t i v i t y o v e r the p rev ious y e a r o f obese m e n w i t h O S A (n=31) P e r c e i v e d A c t i v i t y C h a n g e N u m b e r o f % (over past y ea r ) Sub jec ts No change 24 77 Less activity 6 19 Reasons cited: - Health-related (3): chronic pain (1); asthma, carpal tunnel, back pain (1); too tired (1) - Time-related (2): busy schedule (1), busy with committee work (1) - Combination of factors (1): busy schedule, no energy (1) More activity 1 3 50 Chapter IV: Results Bivariate correlational analyses examining the relationships between physical activity parameters (blocks walked and hours per week spent engaging in sweat-producing activities) and age, BMI, measures of OSA severity, eating behaviour and food intake variables are presented in Table 22. Physical activity was not significantly correlated with age, BMI, measures of OSA severity, food intake, or most measures of eating behaviour. Although there were inverse correlations between sweat-producing activity and BES, TFEQ disinhibition, and TFEQ restraint scores, only the relationship with disinhibition scores was significant. T a b l e 22 . C o r r e l a t i o n m a t r i x f o r p h y s i c a l ac t i v i t y , age, B M I , measu res o f O S A seve r i t y , food i n take v a r i a b l e s , a n d ea t i ng b e h a v i o u r f o r obese m e n w i t h O S A (n=27-30) + D a i l y B l o c k s W a l k e d S w e a t - p r o d u c i n g A c t i v i t y ( t imes /week) A g e .01 -.03 B M I -.18 -.10 R D I .05 .01 M i n i m u m O x y g e n S a t u r a t i o n .09 .04 E n e r g y I n take .11 -.01 P e r c e n t C a l o r i e s F r o m P r o t e i n .24 -.23 P e r c e n t C a l o r i e s F r o m F a t .01 .13 P e r c e n t C a l o r i e s F r o m C H O -.01 .03 B E S S c o r e -.14 -.21 T F E Q D i s i n h i b i t i o n -.06 . 4 9 * * T F E Q H u n g e r -.15 -.22 T F E Q R e s t r a i n t .15 .19 + Pearson correlation coefficients are shown; all significance levels are 2-tailed **p<0.01 51 Chapter IV: Results In summary, responses to questionnaire components assessing walking, stair climbing, and regular participation in sweat-producing activities indicated that a substantial proportion of subjects were not very physically active. Of those who did report engaging in sweat-producing activity, most identified walking and working (either around the house or at their jobs) as sweat-producing activities. Few subjects participated in other types of physical activity. Daily blocks walked and hours per week of sweat-producing activity was similar in subjects with professional/technical versus other types of occupations. Subjects had experienced some difficulty completing the section of the questionnaire assessing daily participation in activities of varying intensity. In this section, a greater proportion of subjects indicated that they participated in vigorous exercise on a daily basis. In this section of the questionnaire, subjects also indicated that they were more active on weekends than on weekdays. Most subjects did not perceive that their physical activity level had changed in the previous year. Of those subjects who did perceive a change in physical activity, all but one perceived themselves to be exercising less. 52 Chapter IV: Results d) Psychological Variables i) Depression and anxiety scores Zung Self-rated Depression Scale (ZSRDS) and Zung Self-rated Anxiety Scale (ZSRAS) scores (total and by scale item) are presented in Tables 23 and 24. About one third of respondents (11, 35%) had ZSRDS and ZSRAS scores suggestive of depression (10, 32%), anxiety (9,29%), or both (8,26%). For both scales, items receiving highest scores reflected common OSA symptoms (fatigue, sleep disruption, and impaired task performance). Mean scores for scale items clearly related to these common OSA symptoms were significantly higher than for those items with no known association with OSA symptoms (see Table 25). 53 Chapter IV: Results T a b l e 23 . Z u n g S e l f - R a t e d D e p r e s s i o n S c a l e ( Z S R D S ) scores f o r each sca le i tem f o r obese m e n w i t h O S A (n=31) Z S R D S M e a n R a n g e Total score1 47+12 25-71 Scores on individual scale items2: 19. I feel that others would be better off i f I were dead. 1.2 ±0.5 1.0-3,0 3. I have crying spells or feel like it. 1.2 ±0.4 1.0-2.0 8. I have trouble with constipation. 1.4 ±0.7 1.0-3.0 7. I notice that I am losing weight. 1.6 ±-0.9 1.0-4.0 9. My heart beats faster than usual. 1.6±0.7 1.0-3.0 1. I feel downhearted, blue and sad. 1.6±0.8 1.0-4.0 17. I feel that I am useful and needed. (R) 1.6 ±0.8 1.0-4.0 18. My life is pretty full. (R) 1.7 ±0.9 1.0-4.0 13. I am restless and can't keep still. 1.7 ±0.9 1.0-4.0 15. I am more irritable than usual. 1.7±0.8 1.0-4.0 14. I feel hopeful for the future. (R) 1.7 ±0.9 1.0-4.0 5. I eat as much as I used to. (R) 1.9 ±0.9 1.0-4.0 20. I still enjoy the things I used to. (R) 2.0 ±1.0 1.0-4.0 11. My mind is as clear as it used to be. (R) 2.0 ± 1.9 1.0-4.0 6. I enjoy looking at, talking to, and being with attractive women 2.2 ±1.0 1.0-4.0 (or men). (R) 16. I find it easy to make decisions. (R) 2.2 ± 1.1 1.0-4.0 2. Morning is when I feel the best. (R) 2.3 ± 1.3 1.0-4.0 10. I get tired for no reason. 2.4 ±1.2 1.0-4.0 4. I have trouble sleeping through the night. 2.5 ±1.2 1.0-4.0 12. I find it easy to do the things I used to. (R) 3.1 ±0.9 1.0-4.0 1 Scores > 50 are considered indicative of depression 2 Possible scores on each item range from 1-4; agreement with statements results in higher scores, except for items followed by (R), indicating reverse scoring. Higher scores more reflective of depressive symptoms. 54 Chapter IV: Results T a b l e 24. Z u n g Se l f - r a t ed A n x i e t y Sca le ( Z S R A S ) scores f o r each sca le i tem f o r obese m e n w i t h O S A (n=31) Z S R A S M e a n R a n g e Total score1 42+12 29-74 Scores on individual scale items : 18. My face gets hot and blushes. 1.2 + 0.4 1-2 2. I feel afraid for no reason at all. 1.2 + 0.5 1-3 6. My arms and legs shake and tremble. 1.2 + 0.5 1-3 12. I have fainting spells or feel like it. 1.2 + 0.6 1-4 4. I feel like I am falling apart and going to pieces. 1.3 ±0.6 1-3 11. I am bothered by dizzy spells. 1.4 + 0.8 1-4 20. I have nightmares. 1.4 + 0.5 1-2 14. I get feelings of numbness and tingling in my toes* 1.4 + 0.8 1-4 10. I can feel my heart beating fast. 1.5 + 0.8 1-4 15. I am bothered by stomach aches or indigestion. 1.6 + 0.9 1-4 3. I get upset easily or feel panicky. 1.7 + 0.9 1-4 1. I feel more nervous or anxious than usual. 1.8 + 1.0 1 -4 17. My hands are usually dry and warm. (R) 1.8+1.1 1-4 13. I can breathe in and out easily. (R) 1.8+1.1 1-4 5. I feel that everything is all right and nothing bad will happen. (R) 2.0+1.1 1-4 16. I have to empty my bladder often. 2.1 + 0.9 1-4 7. I am bothered by headaches, neck, and back pains. 2.1+1.1 1-4 9. I feel calm and can sit still easily. (R) 2.3+1.2 1-4 8. I feel weak and get tired easily. 2.4+1.0 1-4 19. I fall asleep easily and get a good night's rest. (R) 2.5+1.1 1-4 Scores > 45 are considered indicative of anxiety 2 Possible scores on each item range from 1-4; agreement with statements results in higher scores, except for items followed by (R), indicating reverse scoring. Higher scores more reflective of symptoms of anxiety. 55 Chapter IV: Results T a b l e 25 . C o m p a r i s o n o f m e a n scores on Z u n g Se l f - r a ted D e p r e s s i o n Sca le ( Z S R D S ) a n d Z u n g Se l f - r a ted A n x i e t y sca le ( Z S R A S ) i tems re f l ec t i ng vs . not re f l ec t i ng c o m m o n O S A s y m p t o m s f o r obese m e n w i t h O S A (n=31) Sca le I tems re la ted to O S A 1 I tems not re la ted to O S A 2 P + ZSRDS 2.5+ .8 1.6 +.5 0.000"* ZSRAS 1.9+ .6 1.6 +.5 0.007" 1 ZSRDS items 4, 10-12; ZSRAS items 7, 8, 19 2ZSRDS items 1, 3, 6, 8, 13, 14, 17-19; ZSRAS items 1-6, 9, 10, 14-18 + Comparisons were made using student's paired t-test ** Statistically significant at P<0.01 "* Statistically significant at PO.001 ii) Self-esteem scores Respondents' mean score on the Rosenberg Self-Esteem scale (RSES) was 18 + 5 (lower scores reflect higher self-esteem; possible scoring range is 0-40). Scores were distributed as follows: one third of respondents had scores from 10-14, one third had scores from 15-19, and the remaining third had scores from 20-31. iii) Associations among and between variables Bivariate associations among psychological variables are presented in Table 26. A l l three psychological variables were strongly correlated with each other. 56 Chapter IV: Results T a b l e 26 . C o r r e l a t i o n m a t r i x f o r p s y c h o l o g i c a l v a r i a b l e s f o r obese m e n w i t h O S A (n=30-3 1 ) + Z u n g Se l f - r a ted Z u n g Se l f - r a t ed R o s e n b e r g Se l f -D e p r e s s i o n Sca le A n x i e t y S c a l e Sco res E s t e e m Sco res Scores Z u n g Se l f - r a ted 1 D e p r e s s i o n S c a l e Scores Z u n g Se l f - r a ted .69*** A n x i e t y S c a l e Sco res R o s e n b e r g Se l f - .75*** E s t e e m Sco res +Pearson correlation coefficients are shown; all significance levels are 2-tailed ***P<0.001 To determine whether psychological scores varied by occupation category, unpaired t-tests were used to compare subjects with professional/technical versus other types of occupations. No significant differences were found (ZSRDS, 44 + 14 vs 48 + 11, P=.44; ZSRAS, 39+ 11 vs 43 + 11, P=.35; RSES, 17 + 5 vs 18 + 5,P=.33). Bivariate associations between psychological variables, age, BMI, measures of OSA severity, food intake, eating behaviour and physical activity are presented in Table 27. Associations between psychological variables and age, OSA severity, and food intake were weak. Notably, depression and self esteem scores were significantly associated with BMI. There were also several significant correlations between psychological and eating behaviour scores. In particular, BES scores were strongly associated with all three psychological variables examined. Associations between TFEQ restraint scores and psychological variables were not significant. The only significant association between psychological scores and a physical activity measures was an inverse association between self-esteem scores and frequency of sweat-.75 *** 57 Chapter IV: Results producing activity, indicating higher self-esteem with increased reported sweat-producing activity. T a b l e 27 . C o r r e l a t i o n m a t r i x f o r p s y c h o l o g i c a l v a r i a b l e s , age, B M I , measures o f O S A sever i t y , food i n take , ea t i ng b e h a v i o u r a n d p h y s i c a l ac t i v i t y f o r obese m e n w i t h O S A (n=28-31) + Z u n g Se l f - r a ted D e p r e s s i o n S c a l e Scores Z u n g Se l f - r a t ed A n x i e t y Sca le Scores R o s e n b e r g Se l f -es teem Sco res A g e -.13 -.35* -.12 B M I .37* .23 .38* R D I -.11 -.19 -.17 M i n i m u m O x y g e n S a t u r a t i o n -.22 -.09 -.04 E n e r g y I n take -.16 -.23 -.01 % C a l o r i e s F r o m P r o t e i n -.04 -.01 -.02 % C a l o r i e s F r o m F a t -.07 .18 .15 % C a l o r i e s F r o m C H O -.14 -.35' -.22 B E S S c o r e 77*** 4 9 * * .58** T F E Q : R e s t r a i n t .36* .30 .25 T F E Q : D i s i n h i b i t i o n 64*** .21 .42* T F E Q : H u n g e r .50** .08 .33 B l o c k s W a l k e d -.18 .05 -.25 S w e a t - p r o d u c i n g A c t i v i t y ( t imes/week) -.32 -.01 -.37* + Pearson correlation coefficients are shown; all significance levels are 2-tailed *p<0.05 **p<0.01 ***p<0.001 1 trend toward significance p=.05-.O7 58 Chapter IV: Results To examine the influence of BMI on associations seen between psychological and eating behaviour variables, partial correlation coefficients were computed, controlling for BMI (Table 28). Independent of BMI, relationships between binge eating and psychological variables persisted, as did associations between depression scores and other eating behaviour variables. In this analysis, restraint scores were once again not significantly associated with any of the psychological variables. T a b l e 28. C o r r e l a t i o n s be tween p s y c h o l o g i c a l a n d ea t i ng b e h a v i o u r v a r i a b l e s , c o n t r o l l i n g f o r B M I , f o r obese m e n w i t h O S A (n=30-31) + B i n g e E a t i n g S c a l e T h r e e F a c t o r E a t i n g Q u e s t i o n n a i r e : H u n g e r Sco res T h r e e F a c t o r E a t i n g Q u e s t i o n n a i r e : D i s i n h i b i t i o n Sco res T h r e e F a c t o r E a t i n g Q u e s t i o n n a i r e : R e s t r a i n t Sco res Z u n g Se l f - r a t ed D e p r e s s i o n S c a l e Scores .42* .56** .21 Z u n g Se l f - r a ted A n x i e t y S c a l e Sco res .46* -.01 .11 .23 R o s e n b e r g Se l f -es teem Sco res .45* .22 .30 .13 +Partial correlation coefficients are shown; all significance levels are 2-tailed *p<0.05 **p<0.01 ***p<0.001 In summary, about a third of subjects had scores suggestive of depression, anxiety, or both. ZSRDS and ZSRAS scale items receiving the highest scores reflected common OSA symptoms. Depression, anxiety, and self-esteem scores were strongly correlated with each other. No occupational differences in scores were observed. There were significant associations 59 Chapter IV: Results between psychological scores and BMI, and between psychological scores and many of the eating behaviour scores, in particular the BES. Relationships between psychological scores and eating behaviour scores persisted in analyses controlling for BMI. Correlations between psychological scores and TFEQ restraint scores were weak. 60 Chapter IV: Results 4. Interview Data A l l 49 subjects participated in the interview component of the study. Forty-six interviews were conducted in the sleep laboratory immediately following the overnight sleep study, and the remaining three interviews were conducted by telephone (n=2), or in person (n=T) within a few days of the sleep study. The interview was planned to take about 30 minutes to complete; actual interview duration ranged from 20-60 minutes. The interview had several sections: symptom profile, weight history, history of weight loss attempts, perceived relationship between OSA symptoms and food intake, nocturnal eating history, and life events and stressors. a) Symptom Profile Each interview opened with the question, "What brought you to the sleep clinic?". In response, subjects stated that they had attended the clinic because of the complaints or recommendations of others (28, 57%), various OSA symptoms (16, 33%), or both (4, 8%); one subject could not recall why he had come. Next, subjectswere asked an open question about their symptoms, "Can you describe any symptoms you were having?". The most prevalent symptoms mentioned were poor sleep or sleep-related breathing difficulties (16, 33%), daytime fatigue (13, 27%), and snoring (13, 27%). Snoring was mentioned either as a single symptom (7, 14%), or as a symptom co-existing with sleep-related breathing difficulties or daytime fatigue (6,12%). Two subjects mentioned headache, while individual subjects mentioned lack of energy, rapid heart beat, morning sore throat, and "blood too thick". One subject could not identify any symptoms. In response to the follow-up probe, "And were there any other symptoms?", about half of subjects (22,45%) could not identify any. Of the 27 subjects who did respond, 21 (78%) mentioned poor sleep or sleep-related breathing difficulties, daytime fatigue, and/or snoring. 61 Chapter IV: Results One subject mentioned weight gain in association with these symptoms. Two subjects mentioned dizziness, while individual subjects mentioned clogged nose upon waking, aching legs, wheezing, dry mouth and nasal passages, and shortness of breath. Symptoms were further clarified through a series of directed questions about common OSA symptoms: fatigue, falling asleep during the day, unrefreshing sleep, and snoring. Responses are summarized in Table 29. The majority of subjects reported fatigue, falling asleep during the day, and unrefreshing sleep; all reported snoring. In response to a final probe about additional symptoms, a few subjects responded, mentioning headache, loud daytime breathing, and changes in mood. 62 Chapter IV: Results T a b l e 29 . S y m p t o m p ro f i l e o f obese m e n w i t h O S A (n=49) S y m p t o m s N u m b e r o f Sub jec ts % Fatigue: - Yes 38 78 - No 8 16 - Sometimes 1 2 - Unsure (used to feeling this way) 2 4 Falling asleep during the day: - Yes 31 63 - No 8 16 - Sometimes 3 6 - No, but could 7 14 Unrefreshing sleep: - Yes 36 74 - No 10 20 - Sometimes 1 2 - Unsure (used to feeling this way) 2 4 Snoring: - Yes 49 100 Other: - Headache 2 4 - Loud daytime breathing 1 2 - Ill-tempered, unable to get things done 1 2 - Headache, memory loss, depressed 1 2 The two final questions in this section of the interview addressed symptom onset. In response to the first of these questions, "Was it you who first noticed the symptoms, or was it someone else?", more than half of the subjects (28, 57%) stated that someone else had noticed 63 Chapter IV: Results the symptoms, usually the spouse/partner, and a further eight subjects (16%) stated that symptoms had been identified by both themselves and others. Only about a quarter of subjects (13, 27%) stated that they had noticed the symptoms on their own. Most of these subjects (10 of 13) were married or living with a partner. The second question in this section was, "When did these symptoms start to bother you or interfere with your life?". Six subjects were unable to respond, either because symptoms were long-standing and they couldn't recall their onset, or because they didn't perceive symptoms to bother them or interfere with their lives. For the 43 respondents, duration of symptoms was variable, ranging from 5 months to 20 years. The mean duration of symptoms was 6.3 + 5.8 years. In summary, in response to open questions about presenting symptoms, subjects identified a variety of symptoms, the most common being sleep disturbances, sleep-related breathing difficulties, fatigue, and snoring, although relatively few subjects mentioned each of these symptoms. Weight gain was only mentioned by one subject, in response to a follow-up probe. In response to directed questions about common OSA symptoms, it was determined that all subjects snored, and about two-thirds or more subjects experienced fatigue, falling asleep during the day, and/or unrefreshing sleep. For about three-quarters of subjects, symptoms were noticed by, or in combination with others. The point at which symptoms had become problematic was variable, ranging from a few months to several years. b) Weight History The weight history section of the interview addressed weight changes since onset of OSA symptoms, perceived weight at various life stages, and periods of dramatic weight gain. 64 Chapter IV: Results In response to the first question, "Since the symptoms of OSA started to bother you or interfere with your life, has your weight changed?", almost all subjects (36, 84%) stated that they had gained weight, while four subjects (9%) stated their weight had been stable; three (7%) stated that they had lost weight. Six subjects (12%) could not respond to this question, as they were unable to define the point at which symptoms had started to bother them, or they didn't perceive that the symptoms did bother them. Subjects reported being at their current weight for a mean duration of 1.8 + 2.4 years, and this period of weight stability ranged from zero (i.e., weight still changing) to a maximum of ten years. Of the 36 subjects who had reported weight gain since symptom onset, one subject attributed his weight gain of 22.7 kg to previous prednisone therapy. The mean reported weight gain for the remaining subjects was 16.9+15.4 kg; weight gain ranged from 2.3 to 77.0 kg. Weight gain was less than 10 kg for 14 subjects (40%), from 10 to 20 kg for 11 subjects (31%), and 20 kg or more for 10 subjects (29%). This weight gain had occurred over 5.3 + 4.8 years; the weight gain period ranged from 2 months to 18 years. Correlational analyses examining the association between weight gain since onset of OSA symptoms, and potentially related variables including age, BMI, measures of OSA severity, food intake, alcohol intake, eating behaviour, physical activity, and psychological variables for the subjects with complete data on these variables are presented in Table 30. Weight gain was significantly correlated with BMI, and binge eating and depression scores. No other significant correlations were present. 65 Chapter IV: Results Table 30. Correlation matrix for weight gain since onset of OSA symptoms, and age, BMI, measures of OSA severity, food intake, alcohol intake eating behaviour, physical activity and psychological variables for obese men with OSA who had gained weight since symptom onset (n=22-30) + Weight gain since onset of OSA symptoms Age -.38 BMI .44* RDI .17 Minimum Oxygen Saturation -.28 Energy Intake -.01 % Calories From Protein -.03 % Calories From Fat .15 % Calories From CHO -.27 Alcohol intake -.07 BES Score .53** Three Factor Eating Questionnaire: Restraint Score .18 Three Factor Eating Questionnaire: Disinhibition Score .30 Three Factor Eating Questionnaire: Hunger Score .14 Blocks Walked -.04 Sweat-producing Activity (times/week) -.24 Zung Self-rated Depression Scale Score .47* Zung Self-rated Anxiety Scale Score .35 Rosenberg Self-esteem Score .16 + Pearson correlation coefficients are shown, all significance levels are 2-tailed *p<0.05 **p<0.01 66 Chapter IV: Results The next three questions addressed perceived weight at various life stages: as a child, as a teenager, and as an early adult (early 20s). Responses are summarized in Table 31. The majority of subjects perceived that they had not been overweight in childhood or as teenagers. Several subjects reported that they started to gain weight in their early 20's, at which point about half of subjects perceived themselves to be overweight. A few subjects who mentioned their actual weights as teenagers (3) or in early adulthood (4), who had not perceived themselves to be overweight, had BMIs indicative of overweight (> 27). Table 31. Perceived body weight at various life stages of obese men with OSA (n=49) Perceived Body Weight Life Stage Overweight1 Not Overweight2 N O . % N O . % Childhood 10 20 39 80 Teen-aged years 11 22 38 78 Early adulthood (early 20s) 23 47 26 53 1 Described by subjects as "slightly heavy", "chunky", "heavy", "chubby", "pudgy", "overweight", "obese", or " a bit husky" 2 Described by subjects as "thin", "skinny", "normal", "average", "not fat", "no fat at all", "muscular, not fat", "big strong kid", "light", "pretty good", "fairly slim", "lean", or "didn't look fat" The final set of questions in this section focused on periods of dramatic weight gain. In response to the first question, "Have there been any times in your life when you have experienced a dramatic weight gain?", 20 subjects (41%) replied 'yes', while the remaining 29 (59%) replied that they had experienced weight gain, but they had considered it gradual or not dramatic. The subjects who had perceived a dramatic weight gain reported a mean weight gain of 23.9+20.2 kg; weight gain ranged from 7.9 to 77.0 kg. This weight gain occurred over 3+5 years; the period of weight gain ranged from one month to 16 years. 67 Chapter IV: Results When asked, "What do you think caused this weight gain?", subjects provided an array of responses (see Table 32). For the 20 subjects with dramatic weight gain, all but one felt their weight gain was attributable to identifiable factors, the most predominant being quitting smoking, a combination of lifestyle factors, reduced activity, and dietary factors. For the 29 subjects reporting gradual weight gain, about one-third (9, 31%) were unsure why they had gained the weight. The majority of remaining subjects, like the subjects with dramatic weight gain, attributed their weight gain to lifestyle factors. T a b l e 32 . P e r c e i v e d causes o f d r a m a t i c w e i g h t g a i n f o r obese m e n w i t h O S A (n=20) P e r c e i v e d C a u s e o f W e i g h t G a i n N u m b e r o f Sub jec ts % Unsure 1 5 Quit smoking: mentioned on its own (5) or in combination with other lifestyle factors (beer drinking and evening snacking (1); less active 7 35 job(l)) A combination of lifestyle factors: lifestyle change, less activity, more alcohol, eating habit changes (1); job change, eating habits changed (1); worked in a logging camp, more food (1); new, less active job, rich food (1) Less activity: low exercise tolerance (1); accident-related injuries (2); athletic injury (1) Dietary factors: post- diet or amphetamines (1); eating rich food on 6 week vacation (1) Other: stress (1); previous prednisone therapy (1) 4 2 2 4 20 20 10 10 68 Chapter IV: Results T a b l e 33 . P e r c e i v e d causes o f g r a d u a l o r n o n - d r a m a t i c we igh t g a i n f o r obese m e n w i t h O S A (n=29) P e r c e i v e d C a u s e o f W e i g h t G a i n N u m b e r o f Sub jec ts % Unsure 9 31 Less activity: no time for activity (11; less exercise over time, back injury (1); less activity, more business travel (1); more sedentary job (1); knee injury (1); lack of exercise (1); quit sports, 7 24 got lazy (1) Quit smoking: mentioned on its own (51, or in combination with taking new job involving travel 6 21 (1) Dietary factors: love of food 0 ) ; binge cycle (11; pop, chocolate bars and ice cream (1); my diet and alcohol intake (1) 4 19 A combination of lifestyle factors: affluent lifestyle (1); life easier, not working as hard (1) 2 9 Other: Immigrated to Canada 1 3 In summary, most subjects reported weight gain since symptom onset; however, the amount of weight gain reported was variable. Only about one-quarter of subjects had perceived themselves to be overweight as children or as teenagers, whereas in early adulthood, about half of subjects perceived themselves to be overweight. Less than half of subjects perceived that they had experienced a dramatic weight gain in their adult life; all other subjects stated they had experienced gradual or non-dramatic weight g a i ^ The majority of subjects, whether they had experienced a dramatic or gradual weight gain, felt they could attribute their weight gain to identifiable factors, usually related to lifestyle. 6 9 Chapter IV: Results The questions in the next section of the interview addressed frequency of weight loss attempts, weight loss methods used, most weight ever lost, degree of weight maintenance following weight loss attempts, and weight loss attempts since onset of OSA symptoms. This section of the interview began with the questions, "Have you ever tried to lose weight?", and "How often have you attempted weight loss?". More than two-thirds of subjects (34, 69%) had attempted never or rarely attempted weight loss, while 15 subjects had attempted weight loss with some regularity (see Table 34). Table 34. Frequency of weight loss attempts by obese men wi th O S A (n=49) Frequency of attempts Number of Subjects % Never 10 20 Rarely2 Fairly often3 Frequently4 24 7 8 49 16 14 No weight loss attempts in adulthood 2 "rarely"; "occasionally"; "every few years"; no more than 5 attempts in total 3 1-2 times per year; 6-10 attempts in total 4 "constantly"; "lots"; 3 or more times per year 70 Chapter IV: Results When asked, "What methods have you used to try to lose weight?", subjects provided an array of responses (see Table 35). More than two-thirds mentioned eating habit changes. Fewer subjects mentioned specific diet plans, exercise, fasting or liquid diets, or diet pills. T a b l e 35. M e t h o d s used to a t tempt we igh t loss b y m e n w i t h O S A (n=39) M e t h o d s U s e d N u m b e r o f Sub jec ts % Eating habit changes: - ate less (general portion control) - ate less of specific foods (e.g., fat, sugar, alcohol) - healthier eating (ate less of specific foods as well as adopting healthier eating practices, e.g., ate more fruit and vegetables, drank more water, ate breakfast) T o t a l Specific diet plans: - Commercial programs (Weight Watchers, NutraSystem, LifeStyles Program, unspecified) - Fad diets (Scarsdale, milk/banana, high protein, low carbohydrate) - Other (diet from hospital nutritionist, wife's diabetic diet, diet sheet and calorie counting, unspecified) T o t a l Other: - Exercise - Fasting or liquid diets - Diet pills 15 6 28 6 5 5 16 13 7 4 38 15 18 72 15 13 13 41 33 18 4 T o t a l 24 62 71 Chapter IV: Results Next, the subjects who had attempted weight loss were asked, "What is the most weight you ever lost?". One subject stated that he had made weight loss efforts but had never weighed himself. For the other subjects, mean weight loss was 12.8+9.5 kg; weight loss ranged from .7 to 45 kg (see Table 36). When asked, "What happened to your weight following these weight loss attempts?", all but one subject responded that they had either regained the weight (29, 76%) or had regained all of the weight plus more (8, 21%); one subject stated that he had made recent weight loss efforts and it was too soon to tell what would happen to his weight. T a b l e 36 . M o s t we igh t eve r lost b y obese men w i t h O S A w h o h a d a t tempted we igh t loss (n=39) W e i g h t L o s s C a t e g o r y N u m b e r o f % Sub jec ts < 5 kg 10 26 >5and<10kg 8 21 >10and<20kg 14 36 >20kg 7 18 The final questions in this section of the interview examined weight loss efforts since the onset of OSA symptoms. The first question in this section was, "Since the symptoms of OSA started to bother you or interfere with your life, have you attempted weight loss?". Two subjects who did not perceive their symptoms to bother them did not respond to this question. Of the remaining 47 subjects, 30 subjects (64%) said they had attempted weight loss since symptom onset. Of these subjects, 22 (73%) said they had been able to lose weight, although none mentioned any change in OSA symptoms with the weight loss. Only six of these subjects 72 Chapter IV: Results reported any degree of weight maintenance, and no subjects had maintained their weight loss for more than one year. The amount of weight lost by these six subjects ranged from 4.5 to 13.6 kg. In summary, the majority of subjects had attempted weight loss. Subjects reported using a variety of weight loss methods, eating habit changes being the method mentioned most often. Although most subjects reported losing weight in response to past weight loss efforts, invariably they regained all (or all plus more) of weight lost. Since the onset of OSA symptoms, about two-thirds of subjects had attempted weight loss. Although many of these subjects had initial success, only about a quarter of them had achieved any degree of weight maintenance, and no subjects had maintained their weight loss for more than a year. d) Perceived Relationship Between OSA Symptoms and Food Intake In this section of the interview, subjects were asked, "Do you think being tired affects the type or amount of food that you eat?". Seventeen subjects reported that being tired had affected their food intake, while 19 felt it hadn't; 13 were unsure or could not respond because they did not feel fatigued. Subjects who felt fatigue had affected food intake stated that they ate more, made different food choices, or were less motivated to lose weight or think about food intake (see Table 37). 73 Chapter IV: Results Table 37. Perceived effects of fatigue on food intake for obese men with O S A (n=20) Perceived Effects Number of Subjects % Ate more: eat more (3); more evening snacking (2); eat more often (1); more sitting, more eating (1); feel better after eating (1); eat more, eat to stay awake (1) 9 45 Made different food choices: would eat easier foods (2); more likely to eat prepared or fast foods (2); not eating right foods (1) 5 5 Less motivated: no energv to lose weight CD; no willpower (1); when tired, don't think about eating (1) 3 15 To examine whether perceptions of the effect of fatigue on food intake were related to the amount of weight gained since symptom onset, subjects who had gained weight since symptom onset and had an opinion about whether fatigue had affected their food intake (n=32), were divided into two groups on the basis of whether they felt it had (n=T4) or had not (n=T8) affected their food intake. Those subjects who perceived fatigue had affected their food intake had gained significantly more weight since symptom onset (20 + 13 vs 11 + 7 kg, unpaired t-test, P=.03). In summary, about one-third of subjects perceived that fatigue had affected their food intake. These subjects had gained significantly more weight than subjects who had not perceived fatigue to affect their food intake. d) Nocturnal Eating History This section of the interview focused on nocturnal binge eating (NBE). In response to the first question in this section, "Do you ever get up in the night to eat?", seven subjects 74 Chapter IV: Results responded 'yes'. Of these, four described eating behaviour suggestive of binge eating (large amounts of food consumed at a sitting, accompanied by feelings of loss of control over eating). Three of these four subjects reported binge eating nocturnally at least once per week. Their characteristics are described in Table 38. A l l three subjects had moderate to severe OSA. Al l had experienced weight gain in recent years; for two subjects this weight gain was dramatic (>20 kg over a two year period). A l l had a degree of awareness of their nocturnal eating behaviour. Frequency of nocturnal binge eating ranged from once per week to nightly. For two subjects, nocturnal binge eating had commenced or escalated with the onset of OSA symptoms; for one, both OSA symptoms and nocturnal bingeing were long-standing. Two subjects reported consuming readily available foods, while one prepared cooked meals. Two of three subjects had experienced traumatic life events at onset of OSA symptoms. Unfortunately, two of these three subjects were non-respondents to written questionnaires. The one subject who did complete the questionnaires had not recorded any N B E episodes on his food record and his reported food intake appeared to be very restrictive. His TFEQ restraint score was high, perhaps reflecting a current attempt at controlling food intake. Both BES and TFEQ disinhibition scores were among the highest reported by study subjects, indicating some propensity for binge eating problems. His perceived hunger score was moderate. Like other study subjects, he did not meet all of the criteria for the Night Eating Syndrome (NES). Scores on the ZSRDS and ZSRAS were elevated, indicating symptoms of depression and anxiety. His RSES score was moderate. 75 Chapter IV: Results T a b l e 38 . C h a r a c t e r i s t i c s o f n o c t u r n a l b inge eaters (n=3) C h a r a c t e r i s t i c s C a s e 1 C a s e 2 C a s e 3 A g e 39 36 51 B M I 29.4 45.6 52.8 R D I 28.4 92.7 28.3 W e i g h t h i s t o r y Described himself as thin until age 27 (BMI 22). Gained 21.8 kg over past 12 years. Weight stable for past two years. Stated that he started gaining weight as a teenager; described himself as obese by early 20s. Gained 22.7 kg over past two years. Described himself as "not fat at al l" as a child and teenager. From early adulthood, gradual weight gain of 22.7 kg (over 25 years). In 1988, BMI was 37, started to gain weight (BMI 46 until 5 months ago). Further gain of 20 kg over past 5 months. P e r c e i v e d reasons f o r we igh t g a i n Smoked marijuana from early teens - quit 1.5 years ago. "Eating went along with it, I wonder if it had a bearing on my weight." Stated he had no idea why he gained. Weight gain over past 10 years -chronic pain and decreased activity due to accident. Weight gain over past few months -quitting smoking. L e v e l o f awa reness o f n o c t u r n a l ea t i ng b e h a v i o u r "Half asleep but would remember it" Fully aware "Sort of awareness. Aware, but not enough to stop you." F r e q u e n c y o f Almost daily Once/week Almost daily n o c t u r n a l b inge ea t i ng 76 Chapter TV: Results Table 39. Additional characteristics of nocturnal binge eaters (n=3) Characteristics Case 1 Case 2 Case 3 Duration of NBE About 5 years About 2 years Long-standing Nocturnal binge eating prior to onset of OSA symptoms? "No, not prior to symptoms. This is a change. I wasn't a habitual snacker before." "Yes, sometimes, but now it's less controlled." Both nocturnal eating and OSA symptoms long-standing ("since childhood"). Foods consumed "Leftovers from the fridge, whatever is available" "I'll cook meals... eggs, rice, bread..." "Fast foods. Now I'm trying to eat fruit juices and salads." "I eat but don't assimilate food. I'm constantly staffing myself, but still hungry." Life events at symptom onset Life event mentioned, getting remarried, was a positive one for him. Mentioned various significant life events occurring in recent years, including losing his job, having to lower standard of living, and relocating away from family, friends. Reported traumatic childhood (frequent moves, changing caregivers, living with abusive aunt); bed wetting until age 25. Reported food intake, eating behaviour, and psychological scores Non-respondent to written surveys. Non-respondent to written surveys. Reported Food Intake: 1265 calories: 25% protein, 31% fat, 45% CHO (no nocturnal eating was recorded on food records) Eating Behaviour Scores: BES: 23 TFEQ Hunger: 8 TFEQ Disinhibition: 11 TFEQ Restraint: 14 Psychological Scores: ZSRDS: 64 ZSRAS: 56 RSES: 20 77 Chapter IV: Results e) Life Events and Stressors The final section of the interview addressed life events and stressors at symptom onset, and perceived effects of these on eating habits. The section concluded with questions about current stressors. The first question in this section was, " I'd like you to think back again to when the symptoms started to bother you or interfere with your life. Can you think of any major events or changes taking place in your life at that time?". The six subjects who could not recall the point at which symptoms started to bother them, or stated that symptoms did not bother them, could not respond to this question. Of the 43 respondents, 15 (35%) could not recall major life events or changes. The remaining 28 provided an array of responses, the most predominant being a job change (5,18%), family or relationship change (5,18%), and being involved in an accident (5, 18%). Subject responses are summarized in Table 40. Most respondents had gained weight since symptom onset, including 24 of 28 (86%) subjects who had identified life events, and 12 of 15 (80%) subjects who had not. The amount gained was almost identical between subjects who had and had not identified life events or changes at onset of OSA symptoms (17 + 16 vs. 16 + 15 kg, unpaired t-test, P=.99). 78 Chapter IV: Results T a b l e 40 . M a j o r l i fe events o r changes at onset o f s y m p t o m s f o r obese m e n w i t h O S A (n=43) L i f e E v e n t s o r C h a n g e s N u m b e r o f % Sub jec ts None identified 15 35 Family or relationship changes: remarried, happier (1); took custody of four nephews (1); marriage stress, impotence (1); surprise birth of third son (1); young son disrupted sleep (1) 5 12 Accident: had accident, stopped working (2); car accident (2); broke neck in accident (1) 5 12 Job change: new job (2); job change involving shift work (1); business changes (1); more responsible job (1) 5 12 Relocated: mentioned alone (1), or in association with loss of job (1) or reduction in activity level (1); immigrated to Canada (1) 4 7 Quit smoking: mentioned alone (2) or in association with job change (1) 3 7 Change in living situation: moved in with friends, a disaster (1); had full-time job and girlfriend, good 2 5 food, good wine (1) Other: developed coronary artery disease (2); getting older and more tired (1); stock market crash, financial 4 9 change (1) 79 Chapter IV: Results The next question in this section was (referring again to the point at which symptoms had started to bother subjects or interfere with their lives), "What would you say was causing you the most stress at that time?". Subject responses are summarized in Table 41. Of 43 respondents, eight (19%) could not identify any sources of stress, and one could not recall, as OSA symptoms had been long-standing. For the remaining subjects work-related and family/relationship stresses were predominant. Only one subject mentioned health-related stress, chronic pain resulting from an accident. T a b l e 4 1 . Sou rces o f stress at s y m p t o m onset f o r obese m e n w i t h O S A (n=43) S o u r c e s o f St ress N u m b e r o f % Sub jec ts None identified, or couldn't recall 9 21 Work/employment: job stress (10); business stress (3); unemployment (1); dispute with Worker's Compensation Board (1); coworker conflict (1); working too much (1); keeping logging truck on the 18 42 road(l) Family/relationship: raising children (1); being away from family (1); custody battle (1); fighting with girlfriend (bothered by symptoms) (1); wife's drinking problem (1); son's sleeping pattern (1) 6 14 Both work and family/relationship: raising children and job stress 1 2 Financial 4 9 Other: - traffic in Arizona during winter vacation (1) - flashbacks to Vietnam (1) - heavy drug use (1) - living in an isolated area (1) - pain due to accident (1) 5 12 80 Chapter IV: Results In response to a follow-up probe, "And were there any other sources of stress?", most subjects (32, 74%) could not identify any, while nine subjects mentioned work (3), family/relationship (5) or financial stress (1). One subject, on disability due to an accident, mentioned "sitting around not working", and one subject mentioned "lousy sleep". Next, subjects were asked, "Do you think that these stresses or events in your life affected your eating habits?". Twenty subjects (47%) thought they had, while 14 subjects (33%) thought they had not; the remaining subjects were either unsure or had not identified events or stresses. The subjects who felt their eating habits had been affected felt they ate more, changed their eating habits, or changed their focus on eating or appearance in response to stress (see Table 42). Most respondents had gained weight since symptom onset; 18 of 20 subjects (90%) who felt stress had affected their eating habits, and 12 of 14 (86%) who felt it hadn't. However, those subjects who felt stresses or events had affected their eating habits had gained significantly more weight than those who felt their eating habits had not been affected (22 +14 versus 9 + 6 kg, unpaired t-test, P=.003). 81 Chapter IV: Results T a b l e 42 . P e r c e i v e d effects o f l i fe events o r stresses at t ime o f s y m p t o m onset on ea t i ng hab i t s f o r obese m e n w i t h O S A (n=43) E f f ec t s on E a t i n g H a b i t s N u m b e r o f Sub jec ts % None identified 14 33 Unsure or N/A (no stresses/events identified) 21 Ate more to compensate: ate more, nothing to do (3); ate more, nobody here to control me (1); if nervous or bored, 1 eat (1); ate to compensate (1); eat to get over anxiousness (1); get depressed, eat more (1); work changes, bored, less active, ate more (1); if you are down, food might help (1); ate more, sitting around (1); eat more when stressed (1); 12 28 Changed eating habits: never home, don't eat at normal times (1); more travel, less healthy food (1); ate more, erratic schedule (1); drank more alcohol (1); more alcohol, eat appetizers when out drinking (1) 12 Changed focus: less focus on appearance (in secure relationship) (1); less concerned about eating habits (1); other things on my mind, ate only when hungry (considered this to be a positive effect of stress) (1) 82 Chapter IV: Results The final questions in this section of the interview focused on current sources of stress. In response to the first question, "What would you say is causing the most stress for you now?", eight subjects (16%) could not identify a current source of stress. As at symptom onset, work/employment, family/relationship, and financial stress were frequently identified. Among the current work/employment stresses, five subjects mentioned the stress of not working or difficulties performing work-related tasks. Several subjects now cited health and weight concerns that had not been identified as sources of stress at symptom onset (see Table 43). Table 43. Cur ren t sources of stress for obese men wi th O S A (n=49) Sources of Stress Number of % Subjects None identified 8 16 Work/employment: not working, nothing to do (2); unemployment (1); work (9); work (am tired, often late) (1); work changes (1); work, longer hours (1); difficulty performing work tasks (1); not doing what I want to, career-wise (1); business (2); business conflict with sons (1) 20 41 Health or weight: my health (OSA) (1); others telling me to do something about my health (1); falling asleep all the time (1); growing old (1); waiting for angioplasty (1); shingles (1); my weight (2); smoking - if I quit, I will eat (1) 9 18 Family/relationship: disagreements with wife over housing (1); mother-in-law (1); being split up from my girlfriend (1); home life (1); marriage, relationship (1); recent deaths in the family (1) 6 12 Financial: financial stress (4); financial concerns for 6 12 future (2) 83 Chapter IV: Results The final question in this section was a follow-up probe, "Is there anything else that is causing you a lot of stress now?". Only about a quarter of subjects could identify additional sources of current stress. Subject responses are summarized in Table 44. Table 44. Addi t ional sources of current stress for obese men wi th O S A (n=49) Sources of Stress Number of Subjects % None identified 38 78 Family/relationship: relatives living with us (I): separated from wife (1); remarriage (1); wife manic-depressive (1) 4 8 Work/employment: administering large contracts at work (1) 1 2 Health or weight: pressure to quit smoking (1); my weight, pissed off I can't move, feel useless (1) 2 4 Financial: financial stress 2 4 Other: everything, nothing in particular (1); people mad at me for no reason (1) 2 4 In summary, more than half of subjects identified life events and changes at onset of OSA symptoms; however, there was no difference in weight gain between subjects who had and had not identified such events or changes. More than three-quarters of subjects identified sources of stress at symptom onset, the most common being work/employment, family/relationship, and financial. Subjects who perceived that life events, changes, or stresses had affected their eating habits had gained significantly more weight since onset than those who had not perceived such a relationship. Current sources of stress were similar to those of symptom onset, with the exception of health or weight related stress, which was more prevalent. 84 Chapter IV: Results 7. Feedback From Subjects Four subjects offered feedback, either in writing or in person, about their participation in the study. Three subjects expressed thanks for being involved in the study, stating that they felt it would assist them in their weight loss efforts. One subject reported that he found the written surveys offensive, as he felt they had suggested that obese people lack personal control. A l l four of these subjects completed all aspects of the study. 85 Chapter V: Discussion D I S C U S S I O N 1. Overview The purpose of this study was to examine several variables linked to body weight in obese men with newly diagnosed OSA, to establish a profile of this patient population for use in developing health intervention strategies. In this chapter, study results are discussed in relation to existing literature on OSA, obesity, and study variables. Next, study limitations, implications of findings, and directions for further research are discussed. Finally, study conclusions are presented. 2. Subject Characteristics Study subjects were predominantly middle-aged with upper body obesity, common findings of other studies of obese men with OSA (Grunstein et al., 1993; Flemons, Whitelaw, Brant, & Remmers, 1994; Shinohara, Kihara, Yamashita, et al., 1997). However, associations between anthropometric measures and measures of OSA severity were weaker than in other studies, possibly reflecting the homogeneity of the study population (all with OSA and obesity). Other studies identifying associations between abdominal and neck girth and OSA severity (Flemons et a l , 1994; Grunstein et al., 1993; Shinohara et al., 1997) have involved patients of highly variable weight and/or OSA status. Statistically, it is recognized that linear relationships between two variables can go undetected if the study sample contains a restricted range of observations (Glenberg, 1988). The lack of correlation between measures of OSA severity and most variables in the current study may not rule out the existence of such relationships. These findings may also reflect the weight status homogeneity of the study population. In addition, available measures of OSA severity do not completely reflect how individuals are affected by OSA. Presenting 86 Chapter V: Discussion symptoms of individuals at a given level of RDI can be highly variable (Kripke et al., 1997). Although RDI has been positively associated with symptoms and conditions known to coexist in OSA in some instances (Coy et al., 1996; Flemons et al., 1994; Grunstein, et al., 1993), in others it has not (Borak et al., 1994; Guilleminault, 1994; Kripke et al., 1997). The upper body obesity seen in subjects of this and other studies of OSA is a known correlate with cardiovascular disease risk (Reeder et al., 1997). It has been recently shown to have occupational, social and psychological associations as well. Rosmond et al. (1996) found that WHR, independent of BMI, smoking status, and alcohol intake, was associated with unemployment, work dissatisfaction, low socioeconomic status, low levels of exercise, and divorce, in a large cohort of middle-aged men. The same investigators have also reported a link between WHR and symptoms of depression and anxiety (Rosmond et al, 1996). It is unknown what proportion of subjects in these studies may have had OSA. Nonetheless, it appears that men with upper body obesity may have common attributes that need to be considered in planning health intervention strategies. About 40% of study subjects appeared to have hypertension, which is in the range of 40-60% observed in other OSA studies (Coy et al., 1996; Grunstein et al., 1995; Partinen & Telakivi, 1992). The prevalence rate of hypertension in OSA is higher than the 25% estimated for the general adult population (Burt et al., 1995). There are indications that lifestyle changes can contribute significantly to blood pressure reduction. In a four-year randomized trial of patients with hypertension (Elmer et a l , 1995), a lifestyle intervention program was found to result in long term changes to several lifestyle variables, and blood pressure reduction. A recent study of diet and hypertension (Appel et al., 1997) concluded that qualitative dietary changes (increased fruit, vegetable, and low fat dairy product consumption) could augment existing 87 Chapter V: Discussion dietary recommendations to significantly reduce blood pressure in the absence of weight loss. It is believed that hypertension is the strongest contributor to OSA mortality (Lavie et al., 1995). Individuals with OSA appear to be a logical target for lifestyle interventions such as these, to augment medical management of hypertension. Many of the more prevalent coexisting health conditions of subjects in the current study have been observed in other studies of OSA, including coronary artery disease, elevated lipids, and respiratory conditions other than OSA (Cetel & Guilleminault, 1994). Several subjects also reported accident-related injuries and arthritis. Many of these coexisting conditions could have been contributory to low physical activity in these subjects. It isn't clear whether the workplace or motor vehicle accidents reported by seven subjects had.resulted in inactivity, weight gain, and onset of OSA symptoms, or if undiagnosed OSA was a factor in accident occurrence. It is recognized that individuals with OSA are at increased risk for driving accidents (Stoohs et al., 1995; Wu & Yan-Go, 1996). There do not appear to be studies examining the link between OSA and workplace accidents; however, such a relationship would not be surprising. Workplace accidents have other correlates including stressful life events, problem drinking, and job dissatisfaction (Webb, Redman, Hennrikus, Kelman, Gibberd, & Sanson-Fisher, 1994), suggesting that individuals with specific characteristics may be more prone to accidental injury. The subgroup of obese men with OSA who have had accidents may be facing a similar array of issues. For the subjects using medication, their medication profile appeared to reflect the common conditions associated with OSA. However, it was noted that about half of subjects reportedly used no medication at all. It has been observed that many individuals with OSA may delay seeking medical attention because they don't perceive themselves to have an illness (Borak 88 Chapter V: Discussion et al., 1994), which would be supported by their low rates of prescription drug use. Conclusions can't be drawn about these subjects' apparently low usage of non-prescription medications or vitamin or mineral supplements, due to the possible incompleteness of these data. About two-thirds of study subjects lived in various communities outside the Lower Mainland. This geographical diversity has implications for designing health intervention programs. Programs based at the hospital would be of limited use to the majority of clinic patients, particularly if regular follow-up visits were to be included. Rather than focusing only on patients during their visits to the sleep program at the hospital, perhaps efforts need to be made to connect them to community-based or work-site sponsored health promotion programs. It is known that communities can differ vastly in health behaviour, and it has been suggested that community focussed programs may offer more effective ways to change individual health behaviour than efforts aimed at individuals (Diehr, Koepsell, Cheadle, et al., 1993). Almost all study subjects were Caucasian. It is not certain whether this is a representative sample of obese men with OSA, or whether Caucasian men are more likely to seek medical attention for OSA symptoms. OSA is clearly less common in ethnic or racial groups with low rates of obesity (Kripke et al., 1997), which may help to explain why few Asians were recruited into the study, despite their prevalence in the local population. Almost all study subjects were married or living with a partner. This is in contrast with findings of the Swedish obesity study (Grunstein et al., 1995), in which a community sample of men with high likelihood of OSA were found to have significantly higher rates of divorce than obesity-matched men with low likelihood of OSA. Sleep clinic populations may be biased towards patients with spouses, as it is frequently these individuals who witness nocturnal apneas and encourage their spouses to seek medical attention (Grunstein et al., 1995). Although it 89 Chapter V: Discussion might seem positive that spouses were actively involved in these subjects' health, it may have signified that the subjects themselves did not have insight into the issues facing them, or motivation to take action. In men, spousal participation in weight control programs has been inversely associated with their weight loss success (Jeffery et al., 1984). About one in five of the study subjects were unemployed or receiving disability payments. It is unknown to what degree OSA symptoms may have contributed to their unemployment, although impaired work performance is a recognized symptom of OSA (Grunstein et al., 1995). Relevant to the current study is that unemployment has been linked to dramatic weight gain, and to undesirable health behaviour prior to employment loss, including smoking, alcohol consumption, and low physical activity (Morris, Cook, & Shaper, 1992). The range of occupations represented by study subjects illustrates that OSA can occur throughout society. However, study results indicated that subjects with professional/technical occupations were less obese, had less severe OSA, and lower binge eating scores than those with other types of occupations. The reasons for this are unknown. However, population studies have shown an inverse association between obesity, education level, and socioeconomic status (Flegal, Harlan, & Landis, 1988; Rosmond, Lapidus, & Bjorntorp, 1996), factors that are likely to be linked to occupation category. It is also possible that the individuals in the current study with professional/technical occupations were more skilled at navigating the health care system, and sought treatment for their OSA at an earlier stage. Interestingly, no differences emerged between occupational groups in reported food intake, most eating behaviour measures, or any of the psychological scores. Still, these two groups may have different support and educational needs. 90 Chapter V: Discussion The high prevalence of truck or bus drivers in the current study sample was likely indicative of the priority given within the sleep program to assessing and treating professional drivers with OSA, due to the recognized adverse affects of the disorder on driving performance (Stoohs et al., 1995; Wu & Yan-go, 1996). Sleep-disordered breathing is known to be common in commercial truck drivers (Stoohs et al., 1995). The reasons why aren't completely clear; however, the degree of obesity appears to be a primary contributing factor (Stoohs et al., 1995). Low levels of physical activity and erratic eating and sleeping patterns may be contributory. In planning health intervention programs for obese individuals with OSA, consideration should be given to the lifestyle issues facing this patient subgroup. The subjects' reported smoking status compares to national smoking statistics (Statistics Canada, 1995), which indicate that a substantial proportion of middle aged men are former smokers. Several subjects mentioned smoking cessation as a cause of their dramatic weight gain, raising the possibility that smoking cessation may have indirectly contributed to their OSA onset. However, weight gain was reported by almost all subjects, whether they had recently quit smoking or not, so the weight gain was unlikely to be due to this single factor. Despite common perceptions of dramatic weight gain with smoking cessation, several large studies have shown the average weight gain to be only 4-5 kg (Flegal, Troiano, Pamuk, et al., 1995; Gray, Cinciripini, & Cinciripini, 1995; Stamler, Rains-Clearman, Lenz-Litzow, Tillotson, & Grandits,1997; Swan & Carmelli, 1995). With concurrent lifestyle modification it appears possible to prevent this weight gain (Talcott, Fiedler, Johnson, et al., 1995). However, about 10% of quitters gain much more weight than average (Swan & Carmelli, 1995). These individuals appear to have different baseline characteristics from those who gain lesser amounts of weight: they are younger, are of lower SES, are less physically active, drink more coffee, and 91 Chapter V: Discussion are heavier smokers (Swan & Carmelli, 1995). Following smoking cessation, they are more likely to report increased liquor and candy consumption, decreased coffee consumption, and having to diet to keep weight low (Swan & Carmelli, 1995). The similarities in post-smoking cessation weight gain observed in identical twins (Eisen, Lyons, Goldberg, & True, 1993; Swan & Carmelli, 1995) suggests that genetic factors are a primary contributor to propensity for weight gain following smoking cessation. It is possible that obese men with OSA who experience escalation of weight with smoking cessation may be genetically predisposed to weight gain. Considering their cardiovascular risk profile, smoking cessation must be encouraged; however, patients may need strategies for minimizing weight gain post- smoking cessation. The alcohol intake reported by study subjects was variable, although more than a third reported abstaining from alcohol. Although population surveys in the US have found a similar proportion of non-drinkers (Eisen, et al., 1993; Midanik & Clark, 1994), a Canadian population health survey (Statistics Canada, 1995) found only 15% of middle-aged men to be reported alcohol abstainers or current non-drinkers. OSA is known to be more common amongst alcoholics, and symptoms often persist with alcohol abstention (Aldrich, Shipley, Tandon, & Kroll, 1993). It is unclear if this finding is reflective of permanent physiological changes related to alcoholism, and whether coexisting factors such as degree of obesity and health habits such as smoking or low physical acitivity may be contributory (Aldrich et al., 1993). The high prevalence of alcoholism in OSA may signal an increased tendency for other types of addictive behaviour including binge eating. Histories of alcohol addiction are relatively common in obese binge eaters (Spitzer et al., 1993; Yanovski et al., 1993). 92 Chapter V: Discussion Another alcohol-related issue is the degree to which subjects' alcohol intake may have had a bearing on their weight. Low correlations between reported alcohol intake and both BMI and weight gain since onset of OSA symptoms would suggest that alcohol was not a major factor in either weight status or recent weight gain. However, these findings may have been influenced by the large proportion of former drinkers in the study, whose food records would not have reflected the contribution that previous heavy alcohol intake may have made to weight gain. The issue of whether alcohol intake affects body weight is unresolved in the literature (Eisen, Lyons, Goldberg, et al., 1993; Istvan, Murray, & Voelker, 1995); however, a number of studies suggest that alcohol intake has a negligible effect on body weight (Eisen, Lyons, Goldberg, et al., 1993; Istvan, Murray, & Voelker, 1995), consistent with findings of the current study. The review of subjects' clinic charts several months following their overnight sleep study revealed that a substantial proportion of them appeared to face obstacles in the initiation and use of CPAP therapy. Obstacles related to CPAP use are recognized (Reeves-Hoche, Meek, and Zwillich, 1994). Although increased propensity for weight loss with CPAP use has been observed (Borak et al., 1994; Loube et al., 1997; Sullivan & Grunstein, 1994), in the current study weight loss was not mentioned at follow-up visits in the charts of subjects reportedly using CPAP. However, patient weights at these visits were not consistently recorded, nor was compliance with CPAP systematically measured, so conclusions can't be drawn from this observation. The chart review also revealed that patients who choose not to use CPAP, or for whom CPAP was not recommended, were unlikely to receive clinic follow-up. Given that these types of patients are generally advised to pursue lifestyle measures such as weight loss, they are an obvious target for ongoing nutrition and lifestyle counselling. Also, knowing that many weight 93 Chapter V: Discussion loss interventions are unsuccessful (Garner and Wooley, 1991), these patients may not achieve symptomatic improvement; therefore, reconsideration of treatment approaches might be required at some point in the future, and this might be more likely i f they were to remain connected to the clinic. 3. Food Intake and Questionnaire Data a) Food Intake The subjects' reported energy intake varied considerably, which has been observed previously in OSA (Pasquali, et al., 1990). Highly variable intake has also been observed in obesity not specifically associated with OSA (Andersson & Rossner, 1996). Variable intakes could be related to differing energy requirements, or weight instability. Given the relative homogeneity of subjects in the present study, with respect to health status, degree of obesity and physical activity level, it would seem unlikely that their widely disparate reported energy intakes could be accounted for by metabolic differences. Subjects' weight histories did reflect weight instability, with most subjects having reported recent weight gain. Although that may help to explain the very high energy intakes reported by a few subjects, it does not explain the very low intakes reported by many. The most likely explanation may be that, in light of their confirmed diagnosis with OSA, some subjects were making efforts to reduce intake. Although subjects were advised not to alter their food intake during the period of food intake recording, this may have been an unrealistic expectation, considering that weight control is a primary treatment recommendation in OSA (Aubert-Tulkens et al., 1989; Browman et a l , 1984; Cetel & Guilleminault, 1994; Kryger, 1989; Loube et al., 1994; Wittels & Thompson, 1990). Accurate quantification of intake in newly diagnosed OSA may be very difficult to achieve, as study participants are likely to be influenced by treatment 94 Chapter V: Discussion recommendations. However, it is an encouraging sign that some patients may be making efforts at controlling food intake in response to news of their diagnosis. Subjects' mean energy intake (about 2300 calories) was somewhat lower than the 2900 calories reported in Pasquali's study of obese individuals with OSA (1990). Subjects in that study were more severely obese than in the current study, which may help to explain their higher intake. The mean energy intake reported in the current study was similar to that reported by a large sample of middle-aged men in the third National Health and Nutrition Examination Survey (NHANES III) in the US (Briefel et al., 1995). NHANES III researchers also assessed dietary underreporting. Although men were found to have lower overall rates of dietary underreporting than women, underreporting was more prevalent in overweight men than it was in lean men. Other studies of dietary underreporting have had similar findings (Forbes, 1993; Klesges, Klesges, Haddock, & Eck, 1995). The fact that most subjects in the current study had gained weight recently, despite reported intakes which theoretically should have been associated with weight maintenance or loss, is evidence for a degree of dietary underreporting. The current study assessed daily intake distribution, to determine whether subjects were inclined to consume a disproportionate amount of their intake in the evening. The finding that intake was fairly evenly distributed between day and night, was similar to findings for middle-aged men in the Nutrition Canada survey (Bureau of Nutritional Sciences, 1973). It appears that most subjects did not have a unique daily intake distribution characterized by evening or nocturnal eating. It is recognized that unconscious nocturnal bingeing would go unrecorded; however, interview findings suggested that nocturnal bingeing was a rare phenomenon; most subjects did not have awareness of nocturnal bingeing, nor had they or their spouses found evidence suggesting such eating episodes had taken place unconsciously. 95 Chapter V: Discussion Although the true energy intake of the subjects is in question, food intake records were useful in identifying qualitative dietary concerns, including high fat and protein intake, and low intakes of vegetables and fruits, and milk products. A study of specificity of dietary underreporting in obesity concluded that nutritious foods are less likely to be underreported than fatty foods or foods rich in simple sugar (Heitmann & Lissner, 1995). Therefore, it is unlikely that low reported intakes of vegetables and fruits and milk products were due to dietary underreporting. However, it is possible that dietary fat intakes were even higher than reported. Many subjects were consuming more fat than the 30% of calories recommended (Canadian Consensus Conference on Cholesterol, 1998; National Cholesterol Education Program, 1994; Kraus, et al., 1996). This reported level of fat intake is of concern, due to its association with chronic disease, particularly cardiovascular disease (Subar, Ziegler, Patterson, Ursin, & Graubard, 1994), for which individuals with OSA are already at risk (Lavie, et al., 1995). Fat intake also appears to be contributory to obesity (Klesges et a l , 1992), although efforts at reducing body weight through fat intake reduction have not been overly encouraging (Lissner & Heitmann, 1995). If there is a role for fat intake reduction in obese men with OSA, it may be in the prevention of weight gain, rather than in obesity treatment. On a positive note, about one third of subjects had reported fat intakes below 30% of calories, although it is unknown whether these low fat intakes were reflective of long-term intake, or whether they represented transient food intake changes in response to OSA diagnosis. The low intake of vegetables and fruits reported by many subjects, while not unique to this study population (Krebs-Smith, Cook, Subar, Cleveland, & Friday, 1995), is of concern because of its association with cardiovascular disease risk (Serdula, Byers, Mokdad, Simoes, Mendlein, & Coates, 1996). Low vegetable and fruit intake may also be a marker for negative 96 Chapter V: Discussion health behaviours, including sedentary lifestyle, smoking, and high intake of fat and alcohol (Serdula et al., 1995; Subar et al., 1994). Obese men with OSA need to be encouraged to change their diets to include more vegetables and fruits. Most subjects' intake of milk products was well below the recommended 2-3 daily servings recommended (Health and Welfare Canada, 1992). Increased consumption of low fat dairy products in these patients would contribute to overall nutritional intake, and in conjunction with other dietary improvements, may contribute to blood pressure reduction (Appel et al., 1997). The finding that subjects reported consuming similar amounts on weekends and weekdays was somewhat surprising. Population studies have found reported food intake to vary according to day reported; generally reported intake is higher on weekends than it is on other days of the week (Andersson & Rossner, 1996; Briefel et al., 1995). Admittedly, current study findings were limited in that relatively few subjects had recorded both weekday and weekend intake to allow the two types of days to be compared. However, many subjects were not working or had occupations with variable schedules, so their intake patterns may not have been typical of those seen in people with regular Monday to Friday schedules. Many of the relationships observed between food intake variables in the current study have also been observed in population intake studies, including the positive association between fat and energy intake, negative association between fat and carbohydrate intake, and a decline in reported fat intake with age (Subar et al., 1994). Although an inverse association between vegetable and fruit intake and fat intake was observed as in other studies (Serdula et a l , 1996; Subar et a l , 1994), in this study the relationship was relatively weak. However, it does appear 97 Chapter V: Discussion that subjects' reported food intake exhibited many similar patterns to those reported in population studies. In the current study, coffee intake was not associated with OSA severity. In a study of self-reported automobile accidents in OSA (Wu & Yan-Go, 1996), coffee drinking was found to be inversely associated with accident risk, suggesting that it may have a role in counteracting OSA symptoms. The previously discussed inconsistent relationship between measures of OSA and its symptoms is an obstacle to assessing whether individuals with more severe OSA symptoms drink more coffee. However, a substantial proportion of subjects in the current study did appear to be heavy coffee drinkers. In the Health Professionals Follow-up Study, about ten percent of a large cohort of middle aged men reported drinking four or more cups of coffee per day (Aldoori, Giovannucci, Stampfer, Rim, Wing, & Willett, 1997), compared to about 40% in the current study. Therefore, it is possible that coffee intake was being used by the subjects in the current study to diminish OSA symptoms. The alcohol intake histories obtained by the investigator were generally in agreement with histories taken by clinic respirologists, indicating that most subjects were consistent in their reporting of alcohol consumption. Although there are validity problems with all methods of estimating alcohol intake, assessing average daily/weekly consumption, the method used in the current study, is believed to be as good as other available methods (Eisen et al., 1993). The few subjects who had reported widely discrepant intakes all appeared to be heavy drinkers. Reasons for the differences reported could have included difficulty estimating usual intake, actual changes in intake from clinic visit to overnight sleep study, and intentional distortion of information to either respirologists or the investigator. 98 Chapter V: Discussion b) Eating Behaviour According to BES score cut-off points (Gormally et al., 1982; Marcus, et al., 1985), less than 10% of subjects in the current study appeared likely to have binge eating problems. This is in contrast to studies of obese persons enrolled in weight loss programs, in which prevalence of binge eating has been estimated at 25-50% (Adami, Gandolfo, Bauer, & Scopinaro, 1995; Gormally, et al., 1982; Marcus et al., 1985). However, weight control program participants have been suggested to represent a more distressed population (Brownell, 1993). Also, these studies have involved a disproportionate number of female subjects; women are known to have higher rates of binge eating than men (Yanovski, 1993). Few studies have assessed binge eating prevalence in obese non-patient community samples; however, it appears that binge eating problems are less prevalent in such populations (Bruce & Wilfley, 1996; Ferguson & Spitzer, 1995). Considering their low BES scores, subjects in the current study appear to represent an obese population distinct from the weight control program-seeking populations so frequently described in the literature. This may have positive implications for treatment; both weight control (Brownell, 1993; Ferguson & Spitzer, 1995) and alcoholism treatment literature (Tucker & Gladsjo, 1993) suggest that program seekers have characteristics that make them less successful in their behaviour change efforts. The BES assesses both behavioural and cognitive components of binge eating (Gormally et al., 1982). Theoretically, individuals could exhibit the behavioural aspects (e.g., episodes of rapid consumption of large amounts of food) and not the cognitive aspects (e.g., guilt, feelings of loss of control, mental distress) of binge eating. In such individuals, BES scores might not be elevated, yet weight could be escalating due to overeating episodes. In the current study, this did 99 Chapter V: Discussion not appear to be the case, as subject scores were not significantly different for the two types of scale items. Subjects' relatively low TFEQ disinhibition scores provided additional evidence that binge eating was not prevalent in this study population. BES and TFEQ disinhibition scores were significantly correlated, which has been observed previously (Marcus et al., 1985). The TFEQ disinhibition score has been suggested to be a measure of binge eating (Lawson et al., 1995; Stunkard & Messick, 1985). In the current study, only a few subjects had disinhibition scores in the range reported by obese binge eaters (Adami et al., 1995; Brody et al., 1994). In reference to the use of BES and TFEQ disinhibition scores to assess likelihood of binge eating problems, it should be noted that the diagnostic criteria for binge eating disorder (BED) (American Psychiatric Association, 1994) were established several years after the BES and TFEQ were developed. Although these scales assess behaviour and cognitions associated with BED, they are not specific to the diagnostic criteria of the disorder (see Appendix R); a scale for this purpose does not appear to be available. However, individuals determined via clinical interview to meet the diagnostic criteria for BED have been shown to have significantly higher BES and TFEQ disinhibition scores than those without BED (Brody, et al., 1994). In the current study, and in other studies using BES and TFEQ, elevated scores on these scales are suggestive, but not diagnostic, of BED. The TFEQ hunger scale assesses perceived hunger. Hunger scores are known to be linked to disinhibition scores (Adami et al., 1996; Bjorvell, Rossner, & Stunkard, 1986; Stunkard and Messick, 1985), although the hunger construct appears to make a small, but independent contribution'to the prediction of binge eating (Lowe & Caputo, 1991). Binge-eaters have been shown to have higher hunger scores than non-binge eaters (Adami et al., 1995). Hunger scores 100 Chapter V: Discussion have also been shown to diminish with successful weight loss efforts (Bjorvell, Aly, Langius, & Nordstrom, 1994; Pekkarinen, Takala, & Mustajoki, 1996). In the current study, most subjects had hunger scores lower than those typically reported by obese binge eaters (Adami et al., 1995), offering more evidence that binge eating was not prevalent in this study population. The relatively low scores observed can obviously not be explained on the basis of successful weight loss efforts, as most subjects had gained weight recently. In the current study, BES and TFEQ disinhibition scores were correlated with BMI, suggesting greater propensity for binge eating with increasing relative weight. A link between binge eating severity and BMI has been frequently (Brody et a l , 1994; Marcus et al., 1985; Spitzer et al., 1993; Stunkard & Messick, 1985), but not consistently (Adami et al., 1995; Grisset & Fitzgibbon, 1996) observed in other study populations. In obese patient populations where binge eating severity does correlate with BMI, the BMI has potential for use as a screening tool for likely binge eating behaviour. Binge eaters are more likely to experience relapse in weight following weight loss efforts (Bruce & Wilfley, 1996), and have increased psychopathology compared to non-binge eaters (Mitchell & Mussell, 1996; Yanovski et a l , 1993). For obese men with OSA, the issues facing those with more severe obesity are likely to be more complex. Different approaches to intervention with these individuals may be required. The TFEQ restraint score is a measure of intention to restrict food intake, which may or may not be successful (Ogden, 1993). Obese individuals making weight loss efforts have been found to have higher scores than those not attempting weight loss (Bjorvell et a l , 1986; French et al, 1994); scores have also been shown to be correlated with weight loss in those who are intentionally losing weight (Bjorvell et a l , 1986; Bjorvell et a l , 1994). In the current study, subjects' TFEQ restraint scores varied, indicating variable intention to restrict food intake. 101 Chapter V: Discussion Differences in restraint scores help to explain the wide variation in subjects' reported energy intakes; an inverse relationship was observed between restraint scores and reported intake. Although all subjects would have been advised to pursue weight loss as a component of OSA treatment, on the basis of their restraint scores, it appeared that less than 20% had a strong intention to do so. Taken together, BES and TFEQ scores suggest that binge eating was not prevalent amongst study subjects, nor was a strong intention to control food intake. The BES appears to be a potentially useful tool in assessing eating behaviour in this population, as BES scores were correlated with TFEQ sub-scale scores, BMI, and measures of psychopathology. Interestingly, BES scores were linked to BMI, and reported food intake was not. Of the TFEQ sub-scales, the restraint scale appears to assess a component of eating behaviour not addressed by the BES, intention to control food intake. The restraint scale may be of use in assessing outcomes of weight control intervention programs for this population. Subject responses to the NES scale suggest that NES, a syndrome consisting of evening hyperphagia, difficulties falling asleep, and morning anorexia, was not an issue in this study population. Studies have estimated 10-15% of obese individuals to have NES (Kuldau & Rand, 1986; Stunkard, Berkowitz, Wadden, Tanrikut, Reiss, & Young, 1996); however, subjects in these studies were weight control program participants or individuals self-selected on the basis of their binge eating behaviour, likely not representative of all obese individuals. Although no subjects had all manifestations of NES, individual scale responses offer some additional descriptive data on their eating issues. About half of subjects reported waking with little appetite and starting to eat later in the day. It is unknown if this eating pattern is specifically linked to OSA, and whether it may be contributory to obesity in these men. This 102 Chapter V: Discussion eating pattern does not seem to be a response to late evening eating; very few subjects reported evening eating as a problem, c) Physical Activity Subjects' responses to questionnaire items on walking, stair-climbing, and sweat-producing activities suggested that that many of them were sedentary. Although this finding is not unique to OSA, it is of concern. Physical activity appears to offer some protection against several chronic conditions obese individuals with OSA are at risk for, including hypertension, coronary artery disease, diabetes, and depression (Crespo, Keteyian, Heath, & Sempos, 1996; Paffenbarger, Lee, & Leung, 1994). Low physical activity may also be contributory to weight gain (Wing, 1995). In the current study, almost half of subjects reported never engaging in sweat-producing activities. In Canada's Health Promotion Survey (Health and Welfare Canada, 1993), about 30% of men aged 25-64 reported no leisure time physical activity, and in NHANES III, about 17% of men aged 40-59 reported no leisure time physical activity (Crespo, et al., 1996). The higher rate of physical inactivity reported in the current study may reflect the obesity status of study subjects; obese individuals tend to be less-physically active than lean individuals (Foreyt, Brunner, Goodrick, St. Joer, & Miller, 1995; Health and Welfare Canada, 1993). It could also be related to their coexisting health conditions. Some subjects perceived their routine daily activities (e.g., doing house work, driving a truck) to be sweat-producing, which helps to illustrate how physically challenging activities of daily living can be in the presence of obesity and untreated OSA. In light of this, it is not surprising that very few subjects reported participating in recreational sports. 103 Chapter V: Discussion There was a discrepancy in the data obtained about participation in vigorous activity. On the section of the questionnaire asking respondents to identify hours per day spent in activities of varying intensity, on average, subjects indicated that they spent an hour per day engaging in vigorous activity. Their responses to direct questions in another section of the questionnaire cast some doubt on the accuracy of these data. Subjects appeared to have had no problems responding to the direct questions, whereas several had difficulty with categorical activity estimations. For successful use, more detailed instructions may have required. This questionnaire has been used without reported difficulty in the Harvard Alumni Study (Paffenbarger et al, 1978). However, it is unknown whether the longitudinal nature of that study offered an opportunity to refine instructions for use. Few significant correlations were observed between physical activity measures and measures of food intake and eating behaviour. A significant inverse relationship was observed between sweat-producing activity and TFEQ disinhibition scores. However, the fact that a similar relationship was not observed between sweat-producing activity and BES scores suggests that caution should be used in speculating that physical activity may be inversely related to binge eating in this population. Most subjects reported that their level of physical activity had not changed substantially in the previous year, suggesting that decreased physical activity was not a major contributor to the recent weight gain many subjects had experienced. However, chronically low physical activity may have contributed to weight gain occuring over a longer time period. d) Psychological Variables In the current study, symptoms of depression and/or anxiety were found in about a third of subjects. Several other (Borak et al., 1994; Cheshire et al., 1992; Millman et al., 1989; Platon 104 Chapter V: Discussion & Sierra, 1992), but not all (Cassel, 1993) studies of psychopathology in OSA have had similar findings. The prevalence of depression in the current study was somewhat less than the 41% observed by Millman et al. (1989) using the ZSRDS, which may be due in part to the greater severity of OSA of subjects in that study. OSA symptoms obviously contributed to elevated depression and anxiety scores in study subjects, as scores on scale items reflecting these symptoms were higher than for items not reflecting them. Millman et al. (1989) made the same observation. In addition to providing possible evidence of depression and anxiety, ZSRDS and ZSRAS scores appear to serve as an index of OSA symptoms, which cannot be reliably predicted on the basis of RDI (Borak et al., 1994; Guilleminault, 1994; Kripke et al., 1997). Individuals with severe symptoms may have difficulty carrying out recommended lifestyle changes. Since there is evidence that treatment of OSA with CPAP diminishes adverse symptoms and reduces depression scores (Borak et al., 1994; Millman et al., 1989; Platon & Sierra, 1992), it may be advantageous to delay lifestyle interventions in those with elevated scores until OSA treatment is established. As expected, there were strong correlations among psychological variables in the study. However, a strong association between psychological variables and binge eating scores observed in the current study has not been previously demonstrated in OSA. This association has been widely demonstrated in studies of eating behaviour and psychopathology in obesity (Brody et al., 1994; Bruce & Wilfley, 1996; Rosmond et al., 1996; Yanovski et al., 1993). In the current study, both binge eating and depression scores were associated with degree of obesity. Considering their propensity for binge eating and psychiatric comorbidity, severely obese individuals with OSA appear more likely than moderately obese individuals to represent an obesity treatment challenge. 105 Chapter V; Discussion 4. Interview Data a) Symptom Profile The subjects' symptom profile was variable, although most were experiencing common OSA symptoms to some degree. Interestingly, very few subjects mentioned weight escalation or changes in mood, suggesting that most didn't connect these symptoms to OSA. In most cases, symptoms didn't come immediately to mind when subjects were asked an open question about symptoms. To obtain an accurate symptom profile in this study population, directed questions appeared to be required. It has been suggested that individuals adapt to OSA symptoms due to the insidious nature of the disorder, and therefore, may not recognize that they have an illness (Cassel, 1993). The fact that most subjects had attended the clinic primarily because of the urgings of family members and friends suggests that they may have not had personal insight into how they were being affected by OSA. This lack of insight may have affected their ability to follow through on treatment recommendations such as CPAP therapy and lifestyle changes. b) Weight History The majority of patients had gained weight since symptom onset. Although the amount of weight gain varied, mean weight gain was about 15 kg over a five year period. Additionally, about half of subjects reported that they had experienced a dramatic weight gain at some point in their life, on average 24 kg over a three year period. Longitudinal population studies suggest that weight gains of these magnitudes are unusual. In the First National Health and Nutrition Follow-up study (Williamson, 1993), about 6% of middle-aged men were found to gain 15 kg or more over ten years. Similarly, in the British Regional Heart Study (Walker, Wannamethee, Whincup, & Shaper, 1995), approximately 7% of participants gained more than 10% of body weight over 106 Chapter V: Discussion five years. Studies with one to three year follow-up periods also suggest that major weight gains are rare (Garn & Pilkington, 1984; Meltzer & Everhart, 1995). Although substantial weight gain may be relatively unusual, it appears to be more common in obesity (Garn & Pilkington, 1884; St. Jeor et al., 1995; St. Jeor et al., 1997). Obese weight gainers are also prone to weight fluctuation (St. Jeor et al., 1997). Therefore, it is possible that some of the weight gain reported in the current study may have represented weight fluctuation (i.e., rebound weight gain from weight previously lost). Both major weight gain and weight fluctuation have been linked to upper body obesity, and cardiovascular disease risk (Higgins, D'Agostino, Kannel, & Cobb, 1993; St. Jeor et al., 1995; Walker et al., 1995). Weight gainers and fluctuators have also been shown to have a diminished sense of well-being, greater levels of perceived stress, and lower eating self-efficacy (Foreyt et al., 1995), factors that may hinder their ability to make positive lifestyle changes. These subjects' propensity for weight gain could be due to several factors, working independently or in concert. First, a change in metabolism could accompany OSA onset. This seems unlikely in light of metabolic studies of this population (Richman, 1994; Ryan et al., 1995). Second, physical activity level, and thus total daily energy expenditure, could be decreased. Physical activity data did not suggest a recent decline in physical activity for most of the men, making it unlikely that recent weight escalation was due to this factor. However, subjects did appear to be chronically inactive, a factor which may have contributed to weight increases over a longer period of time. Third, changes in dietary habits could have caused the weight gain. This seems to be very likely based on the subjects' perceptions of why they had gained weight, although it was not corroborated by food intake data. A lack of association between reported energy intake and weight gain has been a common finding in studies 107 Chapter V: Discussion examining the link between intake and weight gain (French et al., 1994; Gerace & George, 1996; Jorgensen, Sorensen, Schroll, & Larsen, 1995), and may be related to methodological challenges in studies of free-living individuals (Williamson, 1996). Although lifestyle factors seemed likely to be responsible for the weight gain experienced by most of the subjects, the triggers for eating habit and activity changes appeared to vary. Such triggers included smoking cessation, employment changes, and impairment to activity caused by accidental injuries or disease. These factors have been previously associated with weight gain in men (Morris et al., 1992; Swan & Carmelli, 1995). Few subjects perceived themselves as overweight as children or teenagers. This finding is not surprising. Although childhood obesity is a significant predictor of adult obesity, overall, childhood weight is not a good predictor of adult weight (Power, Lake, & Cole, 1997). The adult onset of most subjects' obesity suggests that lifestyle, and not just genetic factors, may have been contributory. The apparent surge in prevalence of overweight in early adulthood in these subjects is supported by other literature suggesting that weight gain at this time of life is common (Meltzer & Everhart, 1995; Williamson, 1993). Since there is little evidence that weight gained in early adulthood is lost (Grinker, Tucker, Vokanas, & Rush, 1995), prevention efforts need to target younger men. c) History of Weight Loss Attempts About 30% of the subjects in the current study reported making weight loss attempts fairly often or frequently. This is in contrast to Canada's Health Promotion Survey (Health and Welfare Canada, 1993), in which 49% of overweight men reported current weight loss efforts. Similarly, in a large telephone survey in the US (Serdula et al., 1994), 60% of obese men reported current weight loss efforts. Obese weight fluctuators, such as many study subjects in 108 Chapter V: Discussion the current study appeared to be, have been found to be more likely to diet than individuals, either obese or lean, who keep weight stable (St. Jeor, Brunner, Harrington, et al., 1995). Considering this, it is curious that a greater proportion of subjects in the current study didn't report frequent weight loss efforts. The finding that eating habit changes were the most predominant weight loss strategy used is consistent with that of other studies of weight loss practices (Health and Welfare Canada, 1993; Serdula et al., 1994; St. Jeor et al., 1995); men seem unlikely to join organized weight loss programs or use commercial weight loss aides. Notably, only about one third of subjects who had made weight loss efforts reported using exercise, and no subjects reported using the recommended combination of eating habit changes and exercise. In contrast, the telephone survey on weight control practices found that about half of men reported using the combination of eating habit changes and exercise (Serdula et al., 1994). Although many study subjects had achieved weight loss through weight loss efforts, either before or since OSA symptom onset, almost all had subsequently regained weight lost. Similar findings have been observed in general obesity (Garner & Wooley, 1991), and in obesity associated with OSA (Strobel & Rosen, 1996). Of concern, was the tendency for some patients to gain more weight than they had lost through their weight loss efforts, also a common finding in general obesity literature (Garner & Wooley, 1991). d) Perceived Relationship Between OSA Symptoms and Food Intake Subjects' perceptions of the effects of fatigue on their eating habits were an important predictor of their weight gain since OSA onset. Those subjects who felt that fatigue had affected their intake perceived that they had either eaten more, or different types of foods in response to 109 Chapter V: Discussion their fatigue, or that they had felt less inclined to control intake. These perceptions offer additional evidence about the causes of weight gain in this population. e) Nocturnal Eating History Nocturnal binge eating (NBE) appeared to be a relatively rare phenomenon in study subjects, and hence, was unlikely to be a major cause of weight gain for most of them. As with NES, there is the possibility that the prevalence of N B E in these men is higher than reported, as the study relied on self-report data. However, most of these men were not living alone, and it is somewhat unlikely that this behaviour would go completely undetected by their significant others. By virtue of their inclusion in the study, all subjects with N B E were obese with confirmed OSA. A l l had moderate to severe OSA, history of dramatic weight gain, and some degree of awareness of their nocturnal eating episodes. Cases of N B E in the literature have involved patients with a variety of sleep or psychiatric diagnoses (Eveloff & Millman, 1993; Schenck et al., 1993; Spaggiari, Granella, Parrino, Marchesi, Mel i i , & Terzano, 1994). In these cases, weight escalation was common, although level of awareness of eating behaviour varied: Based on the small numbers of patients with N B E who are identified through sleep disorders programs (Schenck et al., 1993), it would seem that this eating disorder is rare, which is supported by findings of the current study. In addition, the behaviours described appeared to be distinct from those of the night eating syndrome (NES). f) Life Events and Stressors Whether subjects identified life events or stressors in association with symptom onset had no bearing on their reported weight gain. However, subjects who perceived events or stresses to have affected their eating habits gained substantially more weight than those who did 110 Chapter V: Discussion not perceive such a relationship. This suggests that personal reaction to the events or stresses was an important weight gain predictor. Similar conclusions have been drawn in studies examining the effects of stress on eating (Greeno & Wing, 1994). Specific life events can have markedly variable effects on individuals (Ezoe & Morimoto, 1994; Greeno & Wing, 1994); therefore, individual perceptions are considered an important component of stress measurement. The types of life events and stressors reported most often by subjects in the current study, work/employment, family/relationship, and financial, are quite typical for the general population (Ezoe & Morimoto, 1994). Notably, few subjects mentioned health or weight concerns at symptom onset, whereas at the time of the study several mentioned these as sources of stress, suggesting an increasing impact of OSA symptoms on well-being over time. 5. Feedback from Subjects The fact that some subjects expressed thanks for the opportunity to discuss weight-related issues with the investigator offers some evidence for the therapeutic value of listening. Although no advice had been given, subjects perceived they had been helped. 6. Study Limitations When interpreting study findings, study limitations should be considered, including the lack of a control group, reliance on patient self-report, and limitations of written questionnaires used. Obtaining an appropriate study control group for this patient population is difficult. Two options exist. The first, use of obesity-matched men from the general population, was not considered for this study because these men could have had undiagnosed OSA, which could only be ruled out on the basis of expensive and invasive polysomnographic testing. The second option, use of obesity-matched men assessed in the sleep laboratory who were found not to have 111 Chapter V: Discussion clinically significant OSA, was also considered unacceptable because such patients could have had sleep dysfunction and eating behaviour abnormalities preceding clinically significant OSA. In absence of a viable control group option, comparative data from other studies of men were used. Fortunately, many health and nutrition studies of middle-aged men are available. Much of the data collected in this study relied on self-report. The quality of such data is dependent in part on respondents' confidence that honest answers will not have undesirable consequences for them. The personal contact between each subject and the investigator, including a clear explanation of how study results would be used, may have helped to promote honesty in reporting. Self-report data are also dependent on memory acuity and personal insight into behaviours. Although impaired memory can occur in OSA (Cheshire et al., 1992), most subjects did not appear to have difficulty providing information that relied on memory. The personal interview provided an opportunity to probe and clarify responses when required. The limitations of self-report food intake data have been previously discussed (see page 95). Self-reported weights were also used in the study. Subjects used their current, measured weight as a reference point to reflect back on their weight at various life stages, making it unlikely that they would intentionally misrepresent retrospective weights reported. However, it is possible that weights weren't recalled accurately. Validation studies have focused on current, rather than retrospective self-reported weights. In these studies, self-reported weight has been found to correlate reasonably well with measured weights (Stewart, 1981; Stunkard & Albaum, 1981), although about 15% of overweight men have been found to under-estimate their weight by > 4.5 kg (Rowland, 1990). There were some limitations associated with the use of standardized written questionnaires in the study, including the incomplete response, questionnaire shortcomings, and 112 Chapter V: Discussion the possible influence of OSA diagnosis on responses. The response, 63%, while incomplete, is not unusually low for mailed questionnaires (Perkin, 1992). Respondents and non-respondents were similar in BMI and OSA severity. Although respondents were significantly older than non-respondents, associations between age and most study variables were weak, suggesting that this bias in the study population did not present a major problem^ Although the questionnaires were not specifically developed for use with obese men with OSA, with the exception of one section of the physical activity questionnaire, most respondents seemed to be able to complete the questionnaires appropriately. Only one subject commented negatively about questionnaire content. In planning the study, it was tempting to consider changing some of the eating behaviour questionnaire scale items to update language used, eliminate phrases that appear to dictate normative behaviour (e.g., "I don't have difficulty eating slowly in the proper manner"), and insert items with potential relevance to OSA. However, making such changes would have made results difficult to compare to existing literature, which was the main purpose of using the questionnaires. It is possible that questionnaire responses were influenced by OSA diagnosis. As discussed, some subjects may have restricted food intake in response to confirmation of diagnosis. However, it appears less likely that other questionnaire scores were influenced, as most scale items focused on longer term behaviours and beliefs. Despite study limitations, study results do appear to clarify many previously unexplored descriptive characteristics of this patient population. 7. Generalizability of Results Study results would apply to obese men with OSA. They would not necessarily apply to mildly obese men with OSA, or to obese women with OSA. As with any study involving a clinic 113 Chapter V: Discussion population, it should be noted that such populations may differ from non-patient community-based populations. 8. Implications of Findings Study findings have several implications for the timing and nature of lifestyle interventions in obese men with OSA, and for the role of the dietitian working with this patient population. Considering the lengthy sleep program waiting list and the propensity for weight gain in men with undiagnosed OSA, it appears logical to connect patients to lifestyle counselling at the point of referral to the program, rather than waiting until diagnosis is confirmed. This would offer an opportunity to assess whether patients can achieve weight loss, which may assist respirologists in making treatment recommendations at diagnosis. Also, it would offer patients an opportunity to take some action during the waiting period. Even if some of these patients turned out not to have OSA, lifestyle counselling would be warranted on the basis of their obesity and cardiovascular risk profile. Study findings have several implications for the nature of lifestyle interventions to consider for this group. Interventions should focus on weight gain prevention, target those men who are most likely to benefit, provide both education and support, and be accessible to participants in their local community. Weight gain prevention would be a great accomplishment in this population, and might be achieved via qualitative dietary changes and physical activity. It appears that for most obese individuals, significant and sustained weight loss is difficult to achieve (Garner & Wooley, 1991). Toward that end, it might be prudent not to limit the focus to patients who are already severely obese, but to identify slightly overweight patients (e.g., BMI 26-30) and intervene 114 Chapter V: Discussion before they become obese. Also, it may be better to focus on weight maintenance and lifestyle improvements, than to aggressively pursue weight loss that is very unlikely to be permanent. Preventative health strategies should be focused on younger men with the disorder, who are more susceptible to weight gain, and are at greater overall health risk than older men with the disorder (Lavie et al.,1995). Another group to focus on are the patients for whom lifestyle measures are the only OSA treatment recommendation. For these patients, weight maintenance or gradual weight loss may prevent the need for future OSA therapy. Careful consideration needs to be given to management strategies for severely obese men with the disorder, as they are more likely to have a complex array of problems than men who are not as obese. An interdisciplinary team approach may be desirable. It may also be prudent to delay lifestyle interventions for these men until they can be established on CPAP therapy. In formulating intervention strategies, it is essential to consider that many patients may not be at the point where they feel lifestyle counselling would be useful. In this study, there was some evidence that subjects did not see their OSA symptoms as a disease, or make a connection between their weight and OSA symptoms. Health-behaviour theories suggest that individuals must recognize a problem before they can take steps to address it (AbuSabha & Achterberg, 1997; Sandoval, Heller, Wiese, & Childs, 1994). Patients need to be made aware of available services but not forced to use them. With successful CPAP therapy and symptomatic improvement, some patients may be more amenable to lifestyle changes. The nutritional and lifestyle issues common to this population should be considered in formulating educational content for them. Although this group has identifiable health education needs, it would be short-sighted to focus on education in absence of the support that they also appear to need. There may be therapeutic value in listening to patient's weight-related issues. 115 Chapter V: Discussion Considering the geographical diversity of the OSA population, it is critical to link patients to community-based lifestyle programs. It may be useful to compile a list of pertinent community resources available throughout the province, including nutrition counselling services, physical activity programs, and support groups for individuals with OSA. A dietitian working with obese men with OSA has many potential roles, including providing lifestyle counselling, developing educational resources, identifying and linking patients to appropriate community resources, implementing new approaches to lifestyle intervention, and educating colleagues about the nutrition-related implications of OSA. 9. Directions for Further Research The current study raises a few areas for additional research. One possibility is to extend the current study protocol to other obese populations with OSA. During this study, it became apparent that substantial numbers of obese women and mildly obese men attend the lab. It would be interesting to see how obese women with the disease present compared to the obese men in the current study. By extending the current study to include less obese men, it could be determined if trends based on BMI established in the current study apply across a wider range of relative weights. The second possibility for research is to implement and evaluate a lifestyle intervention program designed in consideration of current study findings. An.additional possiblity is to do more indepth examination of the few individuals with NBE, to gain a greater understanding of factors contributing to that eating disorder. 10. Conclusion This study has helped to enhance knowledge of variables potentially related to body weight in obese men with OSA. It confirms that weight escalation is common in this population, and lends support for the hypothesis that lifestyle factors are a major contributor to weight gain. 116 Chapter V: Discussion Although binge eating problems appear to be rare, tendency toward binge eating is linked to both BMI and psychopathology in these patients. This study revealed considerable heterogeneity in this patient group, despite their relative homogeneity in anthropometric and health characteristics. The apparent triggers for weight gain in these patients vary, as do their specific food intake and eating behaviour issues, and degree of psychopathology. Although most of these patients could likely benefit from dietary improvements and increased physical activity, the timing and nature of lifestyle interventions for them need to take their individual symptoms, OSA treatment recommendations, specific eating issues, educational needs, level of motivation, and geographical location into account. / 117 References REFERENCES AbuSabha R, Achterberg C. Review of self-efficacy and locus of control for nutrition- and health-related behaviour. J Am Diet Assoc 1997; 97:1122-1132. Aldrich MS, Shipley JE, Tandon R, Kroll PD, Brower KJ. Sleep disordered breathing in alcoholics: Association with age. Alcohol Clin Exp Res 1993; 17:1179-1183. Adami GF, Gandolfo P, Bauer B, Scopinaro N. Binge eating in massively obese patients undergoing bariatric surgery. Int J Eat Disord 1995; 17:45-50. Aldoori WH, Giovannucci EL, Stampfer M J , Rimm EB, Wing AL , Willet WC. A prospective study of alcohol, smoking, caffeine, and the risk of duodenal ulcer in men. Epidemiol 1997; 8:420-424. American Psychiatric Association. Diagnostic and Statistical Manual-Fourth Edition (DSM-IV). Washington DC: American Psychiatric Association; 1994. Andersson I, Rossner S. The Gustav Study: Repeated, telephone-administered 24-hour dietary recalls of obese and normal-weight men - energy and macronutrient intake and distribution over days of the week. J Am Diet Assoc 1996; 96:686-692. Appel LJ , Moore TJ, Obarzanek E, Vollmer W M , Svetkey LP, Sacks F M , Bray GA, Vogt T M , Cutler JA, Windhauser M M , Lin PH, Karanja N. A clinical trial of the effects of dietary patterns on blood pressure. New Engl JMed 1997; 336:1117-1124. Aubert-Tulkens G, Culee C, Rodenstein O. Cure of sleep apnea syndrome after long-term nasal continuous positive airway pressure therapy and weight loss. Sleep 1989; 12:216-222. Basiotis PP, Welsh SO, Cronin FJ, Kelsay JL, Mertz W. Number of days of food intake records required to estimate individual and group nutrient intakes with defined confidence. JNutr 1987; 117:1638-1641. Bjorvell H, Rossner S, Stunkard A. Obesity, weight loss and dietary restraint. Int J Eat Disord 1986; 5:727-734. Bjorvell H, Aly A , Langius A, Nordstrom G. Indicators of changes in weight and eating behaviour in severely obese patients treated in a nursing behavioural program. IntJObes 1994; 18:521-525. Borak J , Cieslicki J, Szelenberger W, Wilczak-Szadkowska H, Koziej M , Zielinski J. Psychopathological characteristics of the consequences of obstructive sleep apnea prior to and three months after CPAP. Psychiatra Polska 1994; 28:33-44. Bradley T, Brown I, Brossman R, et al. Pharyngeal size in snorers, non-snorers, and patients with obstructive sleep apnea. NEngl JMed 1986; 315:1327-1331. 118 References Bradley TD. Pathophysiological interactions of obesity and sleep apnea. IntJ Obes 1994; 18(suppl 2):43(Abstract). Briefel RR, McDowell M A , Alaimo K, Caughman CR, Bischof AL , Carroll M D , Johnson CL. Total energy intake of the US population: the third National Health and Nutrition Examination Survey, 1988-1991. Am J Clin Nutr 1995; 62 (suppl.):1072S-1080S. Bri l l PA, Kohl HW, & Blair SN. Anxiety, depression, physical fitness, and all-cause mortality in men. JPsychosom Res 1992; 36:267-273. Brody ML , Walsh T, Devlin MJ. Binge eating disorder: Reliability and validity of a new diagnostic category. J Consult Clin Psychol 1994; 62:381-386. Browman CP, Sampson M G , Yolles SF, Gujavarty KS , Weiler SJ, Walslben JA, Hahn P M , Mitler M M . Obstructive sleep apnea and body weight. Chest 1984; 85:435-436. Brownell K D , Wadden TA. The heterogeneity of obesity: fitting treatments to individuals. Behav Ther 1991; 22:153-177. Brownell K. Whether obesity should be treated. Health Psychology 1993: 12;339-341. Bruce B, Wilfley D. Binge eating among the overweight population: A serious and prevalent problem. J Am Diet Assoc 1996; 96:58-61. Bureau of Nutritional Sciences, Health Protection Branch. Food consumption patterns report: A report from Nutrition Canada. Ottawa: Department of National Health and Welfare, 1973. Burt VL , Whelton P, Roccella EJ , Brown C, Cutler JA, Higgins M , Horan M J , Labarthe D. Prevalence of hypertension in the US adult population. Hypertension 1995; 25:305-313. Canadian Consensus Conference on Cholesterol: Final report. Can Med Assoc J1988; 139(suppl): 1-8. Cassel W. Sleep apnea and personality. Sleep 1993; 16(suppl):S56-S58. Cetel M B , Guilleminault C. Obstructive sleep apnea syndrome. In: Chokroverty S, eds: Sleep Disorders Medicine. Boston, MA: Butterworth-Heinemann; 1994:199-217. Charuzi I, Lavie P, Peiser J, Peled. Bariatric surgery in morbidly obese sleep-apnea patients: short- and long-term follow-up. Am J Clin Nutr 1992:55:594S-6S. Cheshire K, Engleman H, Deary I, Shapiro C, Douglas NJ. Factors impairing daytime performance in patients with sleep apnea/hypopnea syndrome. Arch Int Med 1992; 152:538-541. Coy TV, Dimsdale JE, Ancoli-Israel S, Clausen JL. The role of sleep-disordered breathing in essential hypertension. Chest 1996: 108: 890-895 119 References Crespo CJ, Keteyian SJ, Heath GW, Sempos CT. Leisure-time physical activity among US adults. Arch Int Med 1996; 156:93-98. Cuetter AC. Sleep apnea and the Kleine-Levin syndrome. Military Med 1985; 150:286-288. Davies RJO, Stradling JR. The epidemiology of sleep apnoea. Am J Respir Crit Care Med 1995; 152:711-716. Eisen SA, Lyons M J , Goldberg J, True WR. The impact of cigarette and alcohol consumption on weight and obesity. Arch Int Med 1993; 153:2457-2463. Elmer JJ, Grimm R, Laing B, Grandits G, Svendsen K, Van Heel N, Betz E, Raines J, Link M, Stamler J , Neaton J. Lifestyle intervention: Results of the treatment of mild hypertension study (TOMHS). Prev Med 1995: 24:378-388. Eveloff SE, Millman RP. Sleep related eating disorder as a cause of obstructive sleep apnea. Chest 1993; 104:629-630. Ezoe S, Morimoto K. Quantitative assessment of stressors and stress reaction: a review. Jpn J IndHealth 1994; 36;397-405. Ferguson K J , Spitzer RL. Binge eating disorder in a community-based sample of successful and unsuccessful dieters. Int J Eat Disord 1995; 18:167-172. Flegal K M , Harlan WR, & Landis JR. Secular trends in body mass index and skinfold thickness with socioeconomic factors in adult young men. Am J Clin Nutr 1988; 48:544-551. Flegal K M , Trojano RP, Pamuk ER, Kuczmarski RJ, Campbell SM, et al. The influence of smoking cessation on the prevalence of overweight in the United States. New Engl J Med 1995; 333:1165-1170. Flemons WW, Whitelaw WA, Brant R, Remmers JE. Likelihood ratios for sleep apnea clinical prediction rule. Am J Respir Crit Care Med 1994; 150:1279-1285. Forbes GB. Diet and exercise in obese subjects: Self-report versus controlled measurements. Nutr Rev 1993; 51:296-300. Foreyt JP, Brunner RL, Goodrick K, St. Joer ST, Miller GT. Psychological correlates of reported physical activity in normal-weight and obese adults: the Reno diet-heart study. IntJObes 1995; 19 (suppl 4):S69-S72. Foreyt JP, Brunner RL , Goodrick GK, Cutter G, Brownell K D , St. Jeor ST. Psychological correlates of weight fluctuation. Int J Eat Disord 1995; 17:263-275. French SA, Jeffery RW, Wing RR. Food intake and physical activity: A comparison of three measures of dieting. Addic Behav 1994; 19: 401-409. 120 References Ganong, WF. Review of Medical Physiology. Los Altos, California: Lange Medical Publications; 1977: 487. Garn SM, Pilkington JJ. Comparison of three-year weight and fat change distributions of lean and obese individuals. Ecology of Food and Nutrition 1984; 15:7-12. Garner D M , Wooley SC. Confronting the failure of behavioral and dietary treatments for obesity. Clin Psychol Rev 1991; 11:729-780. Gerace TA, George VA. Predictors of weight increases over 7 years in fire fighters and paramedics. Prev Med 1996; 25:593-600. Glenberg A M . Learning From Data: An Introduction To Statistical Reasoning. San Diego, Ca: Harcourt Brace Jovanovich; 1988: 439. Gormally J , Black S, Daston S, Rardin D. The assessment of binge eating severity among obese persons. Addic Behav 1982; 7:47-55. Gray CL, Cinciripini P M , Cinciripini LG. The relationship of gender, diet patterns, and body type to weight change following smoking reduction: A multivariate approach. J Substance Abuse 1995; 7:405-423. Gray DS, Bray GA, Bauer M , Kaplan K, Gemayel N, Wood R, Greenway F, Kirk S. Skinfold thickness measurements in obese subjects. Am J Clin Nutr 1990; 51:571-577. Greeno C G , Wing RR. Stress-induced eating. Psychol Bulletin 1994: 115:444-464. Grinker JA, Tucker K, Volonas PS, Rush D. Body habitus changes among adult males from the Normative Aging Study: Relations to aging, smoking history, and alcohol intake. Obes Res 1995; 3:435-446. Grissett NI, Fitzgibbon ML. The clinical significance of binge eating in an obese population: Support for BED and questions regarding its criteria. Addictive Behaviours 1996; 21:57-66. Grunstein RR, Stenlof K, Hedner J, Sjostrom L. Impact of obstructive sleep apnea and sleepiness on metabolic and cardiovascular risk factors in the Swedish Obese Subjects (SOS) study. Int J Obes 1995; 19: 410-418. Grunstein RR, Stenlof K, Hedner J, Sjostrom L. Impact of self-reported sleep-breathing disturbances on psychosocial performance in the Swedish Obese Subjects (SOS) study. Int J Obes 1995; 18: 635-643. Grunstein R, Wilcox I, Yang T, Gould Y, Hedner J. Snoring and sleep apnea in men: association with central obesity and hypertension. Int J Obes 1993; 17:533-540. 121 References Guilleminault C. Clinical features and evaluation of obstructive sleep apnea. In: Kryger M H , Roth T, Dement WC, Eds. Principles and Practice of Sleep Medicine, 2 n d Edition. Philadelphia: WB Saunders, 1994; 667-677. Harman E, Block A. Why does weight loss improve the respiratory insufficiency of obesity? Chest \9U; 90:153-159. Health and Welfare Canada. The Canadian Nutrient File. Ottawa: Bureau of Nutritional Sciences, 1985. Health and Welfare Canada, Scientific Review Committee. Nutrition Recommendations. Ottawa: Supply and Services Canada, 1990. Health and Welfare Canada. Canada's Food Guide To Healthy Eating. Ottawa: Supply and Services Canada, 1992. Health and Welfare Canada, Stephens T, Fowler Graham D, eds. Canada's Health Promotion Survey 1990: Technical report. Ottawa: Minister of Supply and Services Canada, 1993. Heitmann BL, Lissner L. Dietary underreporting by obese individuals—is it specific or non-specific? BMI 1995; 311:986-989. Higgins M , D'Agostino R, Kannel W, Cobb J. Benefits and adverse affects of weight loss. Ann Intern Med 1993; 119:758-763. Istvan J, Murray R, Voelker H. The relationship between patterns of alcohol consumption and body weight. Int J Epidemiol 1995; 24:543-546. Jeffery RW, Bjornson-Benson W M , Rosenthal BS, Lindquist RA , Kurth CL, Johnson SL. Correlates of weight loss and its maintenance over two years of follow-up among middle-aged men. PrevMed 1984; 13:155-168. Jorgensen L M , Sorensen TIA, Schroll M , Larsen S. The influence of dietary factors on weight change assessed by multivariate graphical models. IntJObes 1995; 19:909-915. Klesges RC, Klesges L M , Haddock CK, Eck LH. A longitudinal analysis of the impact of dietary intake and physical activity on weight change in adults. Am J Clin Nutr 1992; 55:818-822. Klesges RC, Eck L H , Ray JW. Who underreports dietary intake in a dietary recall? Evidence from the second National Health and Nutrition Examination Survey. J Consult Clin Psychol 1995; 63:438-444. Krauss R M , Deckelbaum RJ, Fisher E, et al. Dietary guidelines for healthy American adults. A statement for health professionals from the Nutrition Committee, American Heart Association. Circulation 1996; 94:1795-1800. 122 References Krebs-Smith SM, Cook A, Subar AF, Cleveland L, Friday J. US adults fruit and vegetable intakes, 1989 to 1991: A revised baseline for the Healthy People 2000 Objective. Am JPubl Health 1995; 85:1623-1629. Kripke DF, Ancoli-Israel S, Klauber MR, Wingard DL, Mason WJ, Mullaney DJ. Prevalence of sleep disordered breathing in ages 40-64: A population based survey. Sleep 1997; 20:65-76. Kryger MH. Management of obstructive sleep apnea: Overview. In: Kryger M H , Roth T, Dement WC, eds. Principles and Practice of Sleep Medicine. Philadelphia, Pa: WB Saunders and Company; 1989:584-590. Kuldau JM. The night eating syndrome and bulimia in the morbidly obese. IntJEat Disord 1986;5:143-148. Lansky D, Brownell KD. Estimates of food quantity and calories: errors in self-report among obese people. Am J Clin Nutr 1982; 35:727-732. Lavie P, Herer P, Peled R, Berger I, Yoffe N, Zomer J, Rubin E. Mortality in sleep apnea patients: A multivariate analysis of risk factors. Sleep 1995; 18:149-157. Lawson OJ, Williamson DA, Champagne C M , DeLany JP, Brooks ER, Howat P M , Wozniak PJ, Bray GA, Ryan DH. The association of body weight, dietary intake, and energy expenditure with dietary restraint and disinhibition. ObesRes 1995; 3:153-161. Lissner L, Heitmann BL. Dietary fat and obesity: Evidence from epidmiology. Europ J Clin Nutr 1995;49:79-90. Loube DI, Loube A A , Mitler M M . Weight loss for obstructive sleep apnea: The optimal therapy for obese patients. J Am Diet Assoc 1994; 94:1291-1295. Loube DI, Loube AA , Erman M K . Continuous positive airway pressure treatment results in weight loss in obese and overweight patients with obstructive sleep apnea J Am Diet Assoc 1997; 97:896-897. Lowe MR, Caputo GC. Binge eating in obesity: Toward the specification of predictors. IntJEat Disord 1991; 19:49-55. Marcus M D , Wing RR, Lamparski D M . Binge eating and dietary restraint in obese patients. Addictive Behaviors 1985; 10:163-168. Meltzer A A , Everhart JE. Self-reported substantial 1-year weight change among men and women in the United States. ObesRes 1995; 3 (suppl):123s-134s. Midanik LT, Clark WB. The demographic distribution of US drinking patterns in 1990: Description and trends from 1984. Am JPubl Health 1994; 84:1218-1222. 123 References Millman RP, Fogel BS, McNamara M E , Carlisle CC. Depression as a manifestation of obstructive sleep apnea: Reversal with nasal continuous positive airway pressure. J Clin Psychiatry 1989; 50:348-351. Mitchell JE, Mussell MP. Comorbidity and binge eating disorder. Addictive Behaviours 1996; 20:725-732. Morris JK, Cook DG, Shaper AG. Non-employment and changes in smoking, drinking, and body weight. Brit Med J1992; 304:536-541. National Cholesterol Education Program. Second report of the expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel II). Circulation 1994; 89:1333-1445. Ogden J. The measurement of restraint: Confounding success and failure? Int J Eat Disord 1993; 13:69-76. Paffenbarger RS, Wing AL , Hyde RT. Physical activity as an index of heart attack risk in college alumni. Am J Epidemiol 1978; 108:161-178. Paffenbarger RS, Lee IM, Leung R. Physical activity and personal characteristics associated with depression and suicide in American college men. Acta Psychiatr Scand 1994; 377 (suppl): 16-22. Partinen M , Telakivi T. Epidemiology of sleep apnea syndrome. Sleep 1992; 15 (suppl 6):Sl-4. Pasquali R, Colella P, Cirignotta F, Mondini S, Geradi R, Buratti P, et al. Treatment of obese patients with obstructive sleep apnea syndrome (OSAS): effect of weight loss and interference of otorhinolarlygoiatric pathology. Int J Obes 1990; 14:207-217. Pekkarinen T, Takala I, Mustajoki P. Two year maintenance of weight loss after a VLCD and behavioural therapy for obesity: Correlation to a the scores of questionnaires measuring eating behaviour. Int J Obes 1996; 20:332-337. Perkin J. Design and use of questionnaires in research. In: Monsen ER, Ed. Research: Successful Approaches. Chicago: American Dietetic Association, 1992;120. Platon M J , Sierra JE. Changes in psychopathological symptoms in sleep apnea patients after treatment with nasal continuous positive airway pressure. Int JNeuroscience 1992; 62:173-195. Power C, Lake JK, Cole TJ. Body mass index and height from childhood to adulthood in the 1958 British birth cohort. Am J Clin Nutr 1997; 66:1094-1101. Reeder, BA , Senthilselvan A , Despres J, Angel A , Liu L, Wang H, Rabkin S. The association of cardiovascular disease factors with abdominal obesity in Canada. Can Med Assoc J1997; 157 (suppl):S39-S45. 124 References Reeves-Hoche M K , Meek R, Zwillich CW. Nasal CPAP: an objective evaluation of patient compliance. Am Rev Respir Dis 1994; 149:149-154. Richman R. Obstructive sleep apnea and resting metabolic rate in obese women. Int J Obes 1994; 18(suppl 2): 14( Abstract). Rosenberg M. Conceiving the Self. New York: Basic Books; 1979. Rosmond R, Lapidus L, Marin P. Mental distress, obesity and body fat distribution in middle-aged men. Obes Res 1996; 4:245-251. Rosmond R, Lapidus L, Bjorntorp P. The influence of occupational and social factors on obesity and body fat distribution in middle-aged men. Int J Obes 1996; 20:599-607. Rosmond R, Lapidus L, Marin P, Bjorntorp P. Mental distress, obesity and body fat distribution in middle-aged men. Obes Res 1996; 4:245-252. Rowland ML. Self-reported weight and height. Am J Clin Nutr 1990; 52:1125-1133. Rubinstein I, Colapinto N, Rotstein LE , Brown IG, Hoffstein V. Improvement in upper airway function after weight loss in patients with obstructive sleep apnea. Am Rev Respir Dis 1988; 138:1192-1195. Ryan CF, Love LL, Buckley PA. Energy expenditure in obstructive sleep apnea. Sleep 1995; 18:180-187. Sandoval W M , Heller K E , Wiese WH, Childs DA. Stages of change: A model for nutrition counselling. Top Clin Nutr 1994; 9:64-69. Schenck CH, Hurwitz, TD, O'Connor K A , Mahowald MW. Additional categories of sleep-related eating disorders and the current status of treatment. Sleep 1993; 16:457-466. Schenck CH, Hurwitz TD, Bundlie SR, Mahowald MW. Sleep-related eating disorders: Polysomnographic correlates of a heterogeneous syndrome distinct from daytime eating disorders. Sleep 1991; 14:419-431. Scucchi S, Montera P, Spagnoli S, Di Pietro F, Maceli F, Lanti F, Massignan C, Savioli M. Disinhibition, anxiety and meal pattern in obese people. Int J Obes 1993; 17 (suppl): 5 (Abstract). Serdula M K , Byers T, Mokdad A H , Simoes E, Mendlein J M , Coates RJ. The association between fruit and vegetable intake and chronic disease risk factors. Epidemiol 1996; 7:161-165. Serdula M K , Williamson DF, Anda RF, Levy A, Heaton A, Byers T. Weight control practices: Results of a multistate telephone survey. Am J Public Health 1994; 84:1821-1824. 125 References Shinohara E, Kihara S, Yamashita S, Yamamane M , Nishida M , Arai T, Kotani K, Nakamura T, Takemura K, Matsuzawa Y. Visceral fat accumulation as an important risk factor for obstructive sleep apnea in obese subjects. J IntMed 1997; 241: 11-18. Smith PL, Gold, AR, Meyers DA, Haponik EF, Bleecker ER. Weight loss in mildly to moderately overweight patients with obstructive sleep apnea. Ann IntMed 1985; 103:850-855. Spaggiari M C , Granella F, Parrino L, Marchesi C, Mell i I, Terzano M G . Nocturnal eating syndrome in adults. Sleep 1994; 17:339-344. Spitzer RL, Rodin J. Human eating behaviour: a critical review of studies in normal weight and overweight individuals. Appetite 1981; 2:293-329. Spitzer RL, Yanovski S, Wadden T, Wing R, Marcus M D , Stunkard A, Devlin M , Mitchell J , Hasin D, Home RL. Binge eating disorder: Its further validation in a multisite study. IntJ Eat Disord 1993; 13:137-153. Stamler J , Rains-Clearman D, Lenz-Litzow K, Tillotson JL, Grandits GA. Relation of smoking at baseline and during trial years 1-6 to food and nutrient intakes and weight in the special intervention and usual care groups in the Multiple Risk Factor Intervention Trial. Am J Clin Nutr 1997; 65(suppl): 374S-402S. Statistics Canada, National Population Heath Survey: Alcohol intake by age and sex. Ottawa: Statistics Canada; 1995. Statistics Canada, National Population Heath Survey : Smoking by age and sex. Ottawa: Statistics Canada; 1995. Stewart AL. The reliability and validity of self-reported weight and height. J Chron Dis 1982; 35:295-309. St. Jeor ST, Brunner RL, Harrington M E , Scott BJ , Cutter GR, Brownell K D , Dyer AR, Foreyt JP. Who are the weight maintainers? Obes Res 1995; 3 (suppl):249s-259s. St. Jeor ST, Brunner RL, Harrington M E , Scott BJ , Daugherty SA, Cutter GR, Brownell K D , Dyer AR, Foreyt JP. A classification system to evaluate weight maintainers, gainers, and losers. J Am Diet Assoc 1997; 97:481-488. Stoohs RA, Bingham LA , Itoi A, Guilleminault C, Dement WC. Sleep and sleep-disordered breathing in commercial long-haul truck drivers. Chest 1995; 107: 1275-1282. Strobel RJ, Rosen RC. Obesity and weight loss in obstructive sleep apnea: A critical review. Sleep 1996; 19:104-115. Stunkard AJ , Grace WJ, Wolff HG. The night-eating syndrome. A pattern of food intake among certain obese patients. Am J Med 1955; 19:78-86. 126 References Stunkard AJ. Eating patterns and obesity. Psychiatric Quarterly 1959; 33:284-292. Stunkard AJ , Albaum JM. The accuracy of self-reported weights. Am J Clin Nutr 1981; 34:1593-1599. Stunkard AJ , Messick S. The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger. JPsychosom Res 1985; 29:71-83. Stunkard A, Berkowitz R, Wadden T, Tanrikut C, Reiss E, Young L. Binge eating disorder and the night-eating syndrome. Int J Obes 1996; 20: 1-6. Subar AF, Ziegler R G , Patterson BH, Ursin G, Graubard B. US dietary patterns associated with fat intake: The 1987 National Health Interview Study. Am JPubl Health 1994; 84:359-366. Sugerman HJ. Pulmonary function in morbid obesity. Gastroenterol Clin North Am 1986; 90:225-237^ Sullivan CE , Grunstein RR. Continuous positive airway pressure ventilation. In: Kryger M H , Roth T, Dement WC, eds.Principles and Practice of Sleep Medicine. Philadelphia, Pa: WB Saunders and Company; 1994:694-705. Suratt P M , McTier RF, Findley LJ , Pohl SL, & Wilhoit SC. Effect of very-low-calorie diets with weight loss on obstructive sleep apnea. Am J Clin Nutr 1992; 6:182S-4S. Swan GE & Carmelli D. Characteristics associated with excessive weight gain after smoking cessation in men. Am J Public Health 1995; 85:73-77. Talcott GW, Fiedlr ER, Pascale RW, Klesges RC, Peterson AL , Johnson RS. Is weight gain after smoking cessation inevitable? J Consulting Clin Psychol 1995; 63: 313-316. Telch CF, Agras WS. Obesity, binge eating and psychopathology: are they related? Int J Eat Disord 1994; 15:53-61. Tucker JA, Gladsjo JA. Help-seeing and recovery by problem drinkers: Characteristics of drinkers who attended alcoholics anonymous or formal treatment or who recovered without assistance. Addictive Behaviours 1993; 18:529-542. Van Moffaert M M . Sleep disorders and depression: the chicken and egg situation. J Psychosom Res 1994; 35 (suppl):9-13. Walker M , Wannamethee G, Whincup PH, Shaper AG. Weight change and risk of heart attack in middle-aged British men. Int J Epidemiol 1995; 24:694-703. Webb GR, Redman S, Hennrikus DJ, Kelman GR, Gibberd RW, Sanson-Fisher RW. The relationships between high-risk and problem drinking and the occurrence of work injuries and related accidents. J Stud Alcohol 1994; 55:434-446. 127 References Williamson DF. Descriptive epidemiology of body weight and weight change in US adults. Ann IntMed 1993; 119:646-649. Williamson DF. Dietary intake and physical activity as "predictors" of weight gain in observational, prospective studies of adults. Nutr Rev 1996; 54 (suppl):S101-S109. Wittels E H , Thompson S. Obstructive sleep apnea and obesity. Otolaryngol Clin North Am 1990; 23:751-760. Wing RR. Changing diet and exercise behaviours in individuals at risk for weight gain. Obes Res 1995; 3 (suppl):277s-282s. Wu H, Yan-Go F. Self-reported accidents involving patients with obstructive sleep apnea. Neurology 1996; 46: 1254-1257. Yanovsky SZ, Nelson JE, Dubbert BK , Spitzer RL. Association of binge eating disorder and psychiatric comorbidity in obese subjects. Am J Psychiatry 1993; 150:1472-1479. Yanovski SZ. Binge eating disorder: current knowledge and future directions. Obes Res 1993; 1:306-324. Young T, Palta M , Dempseu J, Skatrud J, Webber S, Badr S. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993; 328:1230-1235. Zung WWK. A rating instrument for anxiety disorders. Psychosomatics 1971; 12:371-379. Zung WWK. A self-rating depression scale. Arch Gen Psychiatry 1965; 12:63-70. 128 Appendix A. Rationale for methodologies used Data collection methodologies: Description, rationale DESCRIPTION RATIONALE Anthropometric Measures a) BMI b) Neck girth c) Waist-hip ratio • These measures are used to describe obese OSA patients in various studies • Skinfold measures were not performed due to difficulties achieving accuracy with an obese population3 Food Intake Assessment 3-day food records Food record forms will be developed to clearly outline various time periods in each day, to assist in the determination of daily intake distribution Eating Behaviour Scales a) Binge Eating Scale0 b) Night Eating Syndrome Scale4' c) 3-Factor Eating Questionnaire6 Reasonable for subjects to complete Considered to be a valid measure of energy and macronutrient intake of a group*5 A binge eating disorder is one possible explanation for weight gain in OSA Has been used in obesity0 NES has been speculated to be related to nocturnal binge eating, which has been reported in individuals with OSA Has been used in obesity^ Examines 3 dimensions of eating behaviour (hunger, disinhibition, restraint) Physical Activity Measure Paffenbarger Questionnaire1 Has been widely used in obesity, eating disorders*"*1 • Short, easily administered • Provides a gross measure of usual participation in physical activities of various intensities Has been used in a large epidemiological study of men (Harvard Alumni Study*3 a Gray et al., 1990 b Basiotis et al., 1987 c Gormally & Black, 1982 d Bjorvell et al., 1986 e Stunkard & Messick, 1985 1 Paffenbarger et al., 1978 f French, Jeffery, & Wing, 1994 S Kuldau, 1986 h Scucchi et al., 1993 129 Appendix A. Rationale for methodologies used Data collection methodologies: Description, rationale, continued DESCRIPTION RATIONALE Psychological Scales a) Zung Depression Seals' * Short, easily administered, widely used, have been used , in OSA and obesity' b) Zung Anxiety Scale c) Rosenberg Self-Esteem Scale"1 • Short, easily administered, widely used • Has been used in obesity0 Subject Interviews • Useful for verifying information obtained from food records and written instruments and exploring issues not easily captured by written questionnaires; offers an opportunity to obtain unanticipated subject feedback or insights • Structured to provide a standardized approach to interviewing subjects j Zung, 1965 mRosenberg, 1979 kZung, 1971 nTelch& Agras, 1994 'Millman, Fogel, & McNamara, 1989 130 Appendix G. Health history and demographic profile form S U R V E Y O F M E N W I T H O S A : H E A L T H H I S T O R Y A N D D E M O G R A P H I C P R O F I L E F O R M P A T I E N T N A M E A D D R E S S T E L E P H O N E D A T E O F S L E E P S T U D Y Study cr i ter ia : Confirmed OSA Not yet on OSA therapy Speaks English No modified diet used No D M No hypothyroid No diagnosed psychiatric disorders No psychotropic or other medications with impact on study variables _ Other Suitable to approach? Yes No Accepted Declined Reason Char t screen: Date of birth Height Neck girth Waist circumference RDI: preliminary Smoking status: Hypertension Cardiovascular disease Other medical conditions Current medications Alcohol Weight Age BMI Hip circumference Final WHR current quit lifetime non-smoker Married/with partner Single Divorced Employed Occupation Not employed Other Comments 136 Appendix H. Food record forms and instructions (3<JX. cooked") 9 I'.T.*. * *& •: j- ; ' •' .'' . •• N : \ .-'\ • 1 s / . 1 -.1 ' .-.' 1 • " '.' t * . . . * V. -'• V • . . " i .'•I r • i> •. • . -' i .- • • • * $ * - " . / ' * i V\ . • • • V * V . .. ' . * t <• A V I Z I T h i c k slice of cooked meat or tfiaf will f i r into the box above weighs flpprox. I O L . 2 slicee of cooked meat or fish this size will wei^ h appro*. 2 138 Appendix H. Food record forms and instructions M wmm^m A K HI thick piece of cheese thi&* 5iz.e weighs "^pprox lot. t h i c k p a t t y o f r a w m a r t t h i s s i z e Weighs flpprox.4o2i. ( -Joi,. cooked) t h i c k p ^ t t y of raw m e a t t h i s s i z e w e t a h s appro*. 3oz. it'io-L. Cooked") A raw drumstick this s i re Weighs appro* 4 or. CooKed ^ with skin removed, it yields ^pprox. Z o z . tff 139 meat. Appendix H. Food record forms and instructions Sample Food Record Your Name ~/o A n ^Si^vi > ^ A Date (dav/month/vearl /<? f h J r/,j/u Day of Week M/fi rJ. Time Food and Drink Items Amount Hour and Minute Use a separate line for each item. Describe in detail, like a recipe. Describe accurately, e.g., gm, oz, tsp, cup IS c\m 1 / h^r> / ^c^~kJrJ / / O o SY\r( f^nrt irj/na i. sj JO £tnt J LuncA ' / J 1 / 4/ 1 1 4 0 Appendix H. Food record forms and instructions Food Record Your Name Date (day/month/year). Day of Week j Time IHour and I Minute Food and Drink Items Amount Use a separate line for each item. Describe in detail, like a recipe. Describe accurately, e.g., gm, oz, tsp, cup t t I I • 1 141 Appendix I. Verbal food frequency questionnaire Food Record Cross Check Subject Date Milk and Milk Products Breads, Cereals Frequency of consumption, amount, type... Milk Frequency of consumption, amount, type... Bread Cheese Cereal Yogurt Rice Cream Pasta Ice Cream Meat, Fish, Poultry, Alternates Fruits and Vegetables Frequency of consumption, amount, type... Beef/Pork Frequency of consumption, amount, type... Fruit Processed Meats Fruit juice Chicken Vegetables Fish Peanut Butter Other Eggs Frequency of consumption, amount, type... Alcohol Legumes, Beans Sugar Fats and Oils Honey, jam Frequency of consumption, amount, type... Butter/Margarine Candy, chocolate Oil Sweet baked goods Salad dressing Beverages (pop, fruit drinks, coffee, tea) Gravies, sauces 142 Appendix J. Three Factor Eating Questionnaire Survey of Men with Obstructive Sleep Apnea NAME DATE A. Eating Habits Part I Please circle whether the statements below are true (T) or false (F) for you. 1. When I smell a sizzling steak or see a juicy piece of meat, I find it very difficult to keep from eating, even if I have just finished a meal. T F 2. I usually eat too much at social occasions, like parties and picnics. T F 3. I am usually so hungry that I eat more than three times a day. T F 4. When I have eaten my quota of calories, I am usually good about not eating any more. T F 5. Dieting is so hard for me because I just get too hungry. T F 6. I deliberately take small helpings as a means of controlling my weight. T F 7. Sometimes things just taste so good that I keep on eating even when I am no longer hungry. T F 8. Since I am often hungry, I sometimes wish that while I am eating, an expert would tell me that I have had enough or that I can have something more to eat. T F 9. When I feel anxious, I find myself eating T F 10. Life is too short to worry about dieting. T F 11. Since my weight goes up and down, I have gone on reducing diets more than once. T F 12. I often feel so hungry that I just have to eat something. T F 13. When I am with someone who is overeating, I usually overeat too. T F 14. I have a pretty good idea of the number of calories in common food. T F 15. Sometimes when I start eating, I just can't seem to stop. T F 16. It is not difficult for me to leave something on my plate. T F 17. At certain times of the day. I get hungry because I have gotten used to eating then. T F 18. While on a diet, if I eat food that is not allowed. I consciously eat less for a period of time to make up for it. T F 19. Being with someone who is eating often makes me hungry enough to eat also. T F 20. When I feel blue, I often overeat. T F 21. I enjoy eating too much to spoil it by counting calories or watchinq my weight. T F 22. When I see a real delicacy, I often get so hungry that I have to eat right away. T F 23. I often stop eating when I am not really full as a conscious means of limiting the amount that I eat. T F 24. I get so hungry that my stomach often seems like a bottomless pit. T F 25. My weight has hardly changed at all in the last ten years. T F 143 Appendix J. Three Factor Eating Questionnaire 26. i am always hungry so it is hard for me to stop eating before i finish the food on my plate. T F 27. When 1 feel lonely. 1 console myself by eating. T F 28. 1 consciously hold back at meals in order not to gain T F weiqht. 29. 1 sometimes get very hungry late in the evening or at night. T F 30. 1 eat anything 1 want, any time 1 want. T F 31. Without even thinking about it, I take a long time to eat. T F 32. I count calories as a conscious means of controlling my weight. T F 33. I do not eal some foods because they make me fat. T F 34. 1 am always hungry enough to eat at any time. T F 35. 1 pay a great deal of attention to changes in my figure. T F 36. While on a diet, if 1 eat a food that is not allowed, 1 often then splurge and eat other high calorie foods. T F Part II Please answer the following questions by circling the number above the response most appropriate to you 37. How often are you dieting in a conscious effort to control your weight? 1 2 3 4 rarely sometimes usually always 38. Would a weight fluctuation of 5 lb. affect the way you live your life? 1 2 3 4 not at all slightly moderately very much 39. How often do you feel hungry? 1 2 only at sometimes mealtimes between meals often between meals 4 almost always 40. Do your feelings of guilt about overeating help you to control your food intake? 1 2 3 4 never rarely often always 41. How difficult would it be for you to stop eating halfway through dinner and not eat for the next four hours? 1 2 3 4 easy slightly moderately very difficult difficult difficult 42. How conscious are you of what you are eating? 1 2 3 4 not at all slightly moderately extremely 43. How frequently do you avoid "stocking up" on tempting foods? 1 2 3 4 almost never seldom usually almost always 144 Appendix J. Three Factor Eating Questionnaire 44 How likely are you lo shop for low calorie foods? 1 2 3 4 unlikely slightly moderately very likely unlikely likely 45. Do you eat sensibly in front of others and splurge alone? 1 2 3 4 never rarely often always 46. How likely are you to consciously eat slowly in order to cut down on how much you eat? 1 2 3 4 unlikely slightly likely moderately very likely likely 47. How often do you skip dessert because you are no longer hungry? 1 2 3 4 almost never seldom at least almost once a week every day 48. How likely are you to consciously eat less than you want? 1 2 . 3 4 unlikely slightly likely moderately likely very likely 49. Do you go on eating binges although you are not hungry? 1 2 3 4 never rarely sometimes at least once a week On a scale of 0 to 5, where 0 means no restraint in eating (eating whatever you want, whenever you want) and 5 means total restraint (constantly limiting food intake and never 'giving in'), what number would you give yourself? 0 eat whatever you want, whenever you want it 1 usually eat whatever you want, whenever you want it 2 often eat whatever you want, whenever you want it 3 often limit food intake, but often 'give in' 4 usually limit food intake, rarely 'give in' 5 constantly limiting food intake, never 'giving in' To what extent does this statement describe your eating behaviour? 'I start dieting in the morning, but because of any number of things that happen during the day, by evening l.have given up and eat what I want, promising myself to start dieting again tomorrow.' 1 2 3 4 not like me little like me pretty good describes me description of me perfectly 145 Appendix K. Binge Eating Scale Part III Below are groups of numbered statements. Read all of the statements in each group and mark on this sheet the one that best describes the way you feel. #1 1. I don't feel self-conscious about my weight or body size when I'm with others. 2. I feel concerned about how I took to others, but it normally does not make me feet disappointed with myself. 3. I do get self-conscious about my appearance and weight which makes me feel disappointed in myself. 4. I feel very self-conscious about my weight and frequently, I feel intense shame and disgust for myself. I try to avoid social contacts because of my self-consciousness. #2 1. I don't have any difficulty eating slowly in the proper manner. 2. Although I seem to "gobble down" foods, I don't end up feeling stuffed because of eating too much. 3. At times, I tend to eat quickly and then, I feel uncomfortably full afterwards. 4. I have the habit of bolting down my food, without really chewing it. When this happens I usually feel uncomfortably stuffed because I have eaten too much. #3 1. I feel capable to control my eating urges when I want to. 2. I feel like I have failed to control my eating more than the average person. 3. I feel utterly helpless when it comes to feeling in control of my eating urges. 4. Because I feel so helpless about controlling my eating I have become very desperate about trying to get in control. #4 1. I don't have the habit of eating when I'm bored. 2. I sometimes eat when I'm bored, but often I'm able to "get busy" and get my mind off food. 3. I have a regular habit of eating when I'm bored, but occasionally, I can use some other activity to get my mind off eating. 4. I have a strong habit of eating when I'm bored. Nothing seems to help me break the habit. #5 1. I'm usually physically hungry when I eat something. 2. Occasionally. I eat something on impulse even though 1 really am not hungry. 3. I have the regular habit of eating foods, that I might not really enjoy, to satisfy a hungry feeling even though physically, I don't need the food. 4. Even though I'm not physically hungry, I get a hungry feeling in my mouth thai only seems to be satisfied when I eat a food, like a sandwich, that fills my mouth. Sometimes, when I eat the food to satisfy my mouth hunger, I then spit the food out so I won't gain weight. #6 1. I don't feel any guilt or self-hate after I overeat. 2. After I overeat, occasionally 1 feel guilt or self-hate. 3. Almost all the time I experience strong guilt or self-hate after I overeat. 1 4 6 Appendix K. Binge Eating Scale #7 1. I don't lose total control of my eating when dieting even after periods when I overeat. 2. Sometimes when I eat a "forbidden food" on a diet, I feel like i "blew it"-and eat even more. 3. Frequently, I have the habit of saying to myself, "I've blown it now, why not go all the way" when I overeat on a diet. When that happens I eat even more. 4. I have a regular habit of starting strict diets for myself, but I break the diets by going on an eating binge. My life seems to be either a "feast" or "famine". #8 1. I rarely eat so much food that I feel uncomfortably stuffed afterwards. 2. Usually about once a month, I eat such a quantity of food, I end up feeling very stuffed. 3. I have regular periods during the month when I eat large amounts of food, either at mealtime or at snacks. 4. I eat so much food that I regularly feel quite uncomfortable after eating and sometimes a bit nauseous. #9 1. My level of calorie intake does not go up very high or go down very low on a regular basis. 2. Sometimes after I overeat, I will try to reduce my caloric intake to almost nothing to compensate for the excess calories I've eaten. 3. I have a regular habit of overeating during the night. It seems that my routine is not to be hungry in the morning but overeat in the evening. 4. In my adult years, I have had week-long periods where I practically starve'myself. This follows periods when I overeat. It seems I live a life of either "feast" or "famine". #10 1. I usually am able to stop eating when I want to. I know when "enough is enough". 2. Every so often, I experience a compulsion to eat which I can't seem to control. 3. Frequently, I experience strong urges to eat which I seem unable to control, but at other times I can control my eating urges. 4. I feel incapable of controlling urges to eat. I have a fear of not being able to stop eating voluntarily. #11 1. I don't have any problem stopping eating when I feel full. 2. I usually can stop eating when I feel full but occasionally overeat leaving me feeling uncomfortably stuffed. 3. I have a problem stopping eating once I start and usually I feel uncomfortably stuffed after I eat a meal. 4. Because I have a problem not being able to stop eating when I want. I sometimes have to induce vomiting to relieve my stuffed feeling. #12 1. I seem to eat just as much when I'm with others {family, social gatherings) as when I'm by myself. 2. Sometimes, when I'm with other persons, I don't eat as much as I want to eat because I'm self-conscious about my eating. 3. Frequently, I eat only a small amount of food when others are present, because I'm very embarrassed about my eating. 4. I feel so ashamed about overeating that I pick times to overeat when I know no one will see me. I feel like a "closet eater". 147 Appendix K. Binge Eating Scale # 13 1. I eat three meals a day with only an occasional between meal snack. 2. I eat three meals a day, but I also normally snack between meals. 3. When I am snacking heavily, I get in the habit of missing regular meals. 4. There are regular periods when I seem to be continually eating, with no planned meals. #14 1. I don't think much about trying to control unwanted eating urges. 2. At least some of the time, I feel my thoughts are pre-occupied with trying to control my eating urges. 3. I feel that frequently I spend much time thinking about how much I at or about trying not to eat any more. 4. It seems to me that most of my waking hours are pre-occupied by thoughts about eating or not eating. I feel like I'm constantly struggling not to eat. #15 1. I don't think about food a great deal. 2. I have strong cravings for food but they last only for brief periods of time. 3. I have days when 1 can't seem to think about anything else but food. 4. Most of my days seem to be pre-occupied with thoughts about food. I feel like I live to eat. #16 1. i usually know whether or not I'm physically hungry. I take the right portion of food to satisfy me. 2. Occasionally, I feel uncertain about knowing whether or not I'm physically hungry. At these times it's hard to know how much food I should take to satisfy me. 3. Even though I might know how many calories I should eat. I don't have any idea what is a "normal" amount of food for me. 148 Appendix L. Night Eating Syndrome scale Part IV Yes No 1. Do you usually wake up wilh little appetite? 2. Do you usually begin eating later in the day? 3. Do you often eat through the evening without enjoyment? 4. Do you feel tense and upset after eating through the evening? 5. Do you often have difficulty going to sleep? 149 Appendix M. Paffenbarger Physical Activity Questionnaire B. Physical Activity 1. Walking How many blocks do you walk each day? (12 blocks equal one mile) What is your usual pace? casual or strolling (less than 2 mph) average or normal (2 to 3 mph) fairly brisk (3 to 4 mph) brisk (4 mph) 2. Stair-climbing How manv stairs do vou climb UD each dav? 3. On a typical weekday and a weekend day, how much timt (The total for each day should add to 24 hours): Stairs/day »do you spend on the Weekday (hours per day) following activities? Weekend day (hours per day) A. Viaorous Activity (diqqinq in the garden, strenuous sports, jogging, chopping wood, heavy carpentry, bicycling on hills, etc. B. Moderate Activity (housework, liaht sports, walking yard work, lawn mowing, painting, household repairs, light carpentry, bicycling on level ground, etc.) C. Liaht activity (sitting, office work, driving a car, eating, personal care, etc.) D. Sleeping Total 24 hours 24 hours 4. At least once a week do you engage in any regular activity such as brisk walking, jogging, bicycling, etc., long enough to work up a sweat or get out of breath? No Yes How many times a week? Activity 5. Has your physical activity changed much in the past one year? No Yes Change Since Why? 150 Appendix N, Zung self-rated anxiety scale C. How You Feel Part 1 Please tick the appropriate column indicating how often each of the following statements has applied to you during the past 2 weeks: None or a little of the time Some of the time Good part of the time Most of the time 1. I feel more nervous and anxious than usual. 2. 1 feel afraid for no reason at all. 3. 1 get upset easily or feel panicky. 4. 1 feel like I'm falling apart and going to pieces. 5. 1 feel that everything is all right and nothing bad will happen. 6. My arms and legs shake and tremble. 7. 1 am bothered by headaches, neck and back pains. 8. 1 feel weak and get tired easily. 9. 1 feel calm and can sit still easily. 10.1 can feel my heart beating fast. 11.1 am bothered by dizzy spells. 12.1 have fainting spells or feel like it. 13.1 can breathe in and out easily. 14.1 get feelings of numbness and tingling in my toes. 15.1 am bothered by stomach aches or indigestion. 16.1 have to empty my bladder often. 17. My hands are usually dry and warm. 18. My face gets hot and blushes. 19.1 fall asleep easily and get a good night's rest. 20:1 have nightmares. Total Score: ( ) 151 Appendix O. Zung self-rated depression scale Part 2 Please tick the appropriate column indicating how often each of the following statements has applied to you during the last 2 weeks: None or a little of the time Some of the time Good part of the time Most of the time 1. I feel down-hearted, blue and sad. 2. Morning is when I feel the best. 3. 1 have crying spells or feel like it. 4. 1 have trouble sleeping through the night. 5. 1 eat as much as 1 used to. 6. 1 enjoy looking at, talking to, and being with attractive women (or men). 7. 1 notice that 1 am losing weight. 8. 1 have trouble with constipation. 9. My heart beats faster than usual. 10.1 get tired for no reason. 11. My mind is as clear as it used to be. 12.1 find it easy to do the things 1 used to. 13.1 am restless and cant keep still. 14.1 feel hopeful about the future. 15.1 am more irritable than usual. 16.1 find it easy to make decisions. 17.1 feel that I am useful and needed. 18. My life is pretty full. 19.1 feel that others would be better off if I were dead. 20. I still enjoy the things I used to. Total Score: ( ) Please turn over. There is one more page... 152 Appendix P. Rosenberg self-esteem scale Part 3 Please indicate Ihe appropriate answer per item, depending on whether you strongly agree, agree, disagree, or strongly disagree with it. Slrongly agree Agree Disagree Strongly disagree 1. On the whole, I am satisfied with myself. 2. At times I think 1 am no good at all. 3. ! feel that 1 have a number of good qualities. 4. 1 am able to do things as well as most other people. 5. 1 feel 1 do not have much to be proud of. 6. i certainly feel useless at times. 7. 1 feel that I'm a person of worth, at least on a equal plane with others. 8. 1 wish 1 could have more respect for myself. 9. All in all, 1 am inclined to feel that 1 am a failure. 10. I take a positive attitude toward myself. Thank you for participating. If you have questions or comments, please speak to the investigator. 153 Appendix Q. Subject interview guide Survey of Men with OSA: Interview Guide Name Date I. Sleep apnea I'd like to start with some general questions about your sleep apnea... ' A. What brought you to the sleep clinic? B. Can you describe any symptoms you were having? 1. And were there any other symptoms? Probe... 2 . Were you tired? 3. Were you falling asleep during the day? 4. Were you finding that sleep wasn't helping you to feel more alert? 5. Were you snoring? C. Other... D. Was it you that first noticed the symptoms, or was it someone else? E. When did these symptoms really start bothering you or interfering with your life? II. Weight History / Concept of threshold weight I'd like to ask you some questions about your body weight and eating habits... A. Since the symptoms (insert symptoms mentioned) really started bothering you (insert time frame), has your weight changed? Describe B. How long have you been at your present weight? 154 Appendix Q. Subject interview guide C. How would you describe your weight: 1. as a child? 2. as a teenager? 3. in earlier adult hood? D. Have there been any times in your life that you experienced dramatic weight gain? E. If yes: 1. How much weight did you gain? Over what period of time? 2. What do you think caused this weight gain? III. Dieting history A . Have you ever tried to lose weight ? B. If yes: 1. How often have you attempted weight loss? 2. What methods have you used to try to lose weight? 3. What is the most weight you ever lost? A. What happened to your weight following the weight loss attempts? 5. Have you tried to lose weight since the symptoms of sleep apnea (insert symptoms described) really started to bother you? 6. If yes: a) Have you been able to lose weight? b) Have you been able to keep the weight off? c) If yes: (1) How much weight have you kept off? (2) For how long? 155 Appendix Q. Subject interview guide IV. Relationship between sleep apnea and food intake A. Do you think being tired affects the amount you eat or the types of foods you eat? Describe. V. Nocturnal Eating A. Do you ever get up in the night to eat? 1. If yes: a) How often? b) Are you fully aware of what you are doing when you eat in the night? c) For how long have you been night eating? d) Did you do it before the sleep apnea symptoms really started to bother you? e) What do you eat at these times? VI. Life events/Stressors A. I'd like you to think back again to when the symptoms (insert symptoms you described) really started to bother you (insert time frame). Can you think of any major events or changes taking place in your life around that time? Describe. B . What would you say was causing you the most stress in your life at that time? C. Was there anything else causing you a lot of stress at that time? D. Do you feel that these stresses or events in your life affected your eating habits? Describe. E. What would you say is causing the most stress for you now? 1. Is there anything else causing you a lot of stress now? 156 Appendix R. Diagnostic criteria for binge eating disorder (BED) D S M - I V Diagnostic cri teria for binge eating disorder ( B E D ) 1 1. Recurrent episodes of binge eating. An episode of binge eating is characterized by both of the following: a) Eating, in a discrete period of time (e.g., within any 2-hour period), an amount of food that is definitely larger than most people eat in a similar period of time under similar circumstances; and b) A sense of lack of control over eating during the episode (e.g., a feeling that one cannot stop eating or control what or how much one is eating). 2. The binge eating episodes are associated with at least three of the following: a) Eating much more rapidly than normal b) Eating until uncomfortably full c) Eating large amounts of food when not feeling physically hungry d) Eating alone because of being embarrassed by how much one is eating e) Feeling disgusted with oneself, depressed, or feeling very guilty after overeating 3. Marked distress regarding binge eating. 4. The binge eating occurs, on average, at least 2 days a week for 6 months. 5. The binge eating is not associated with the regular use of inappropriate compensatory behaviours (e.g., purging, fasting, excessive exercise) and does not occur exclusively during the course of anorexia nervosa or bulimia nervosa. 1 Source: American Psychiatric Association. Diagnostic and Statistical Manual-Fourth Edition (DSM-IV). Washington DC: American Psychiatric Association; 1994 157 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0099305/manifest

Comment

Related Items