Myles S Faith & J Kevin Thompson. The Health Psychology Handbook: Practical Issues for the Behavioral Medicine Specialist. Editor: Lee M Cohen. Sage Publications. 2003.
Obesity in the United States and other countries is a significant and growing public health problem. The U.S. surgeon general recently released a report referring to the current rates of obesity as a “public health epidemic” (Centers for Disease Control and Prevention, 2002). Obesity among adults, adolescents, and children is increasingly encountered in clinical settings, and an awareness of current etiological models, assessment strategies, and intervention methodologies is necessary for the optimal management of this important health problem. This chapter explores these issues, with an emphasis on the practical strategies that may inform the behavioral management of obesity. It also provides a brief discussion of targeting body image in obesity treatment.
Definitions and Description of Obesity
A consensus has emerged during recent years that one of the best methods for clinically defining obesity is the body mass index (BMI). The formula for BMI is weight in kilograms divided by the square of height in meters (w/h2). This method is also referred to as Quetelet’s index (Garrow & Webster, 1985). BMI can also be computed from pounds and inches: weight (in pounds) divided by height (in inches) times 704.5. BMI cutoffs for weight classification are presented in Table 8.1. Note that a BMI between 25.0 and 29.9 connotes an overweight status, with BMIs greater than 30 indicating obesity.
Despite its widespread use, there are limits to BMI as a measure of weight status (Heymsfield et al., 2000). For instance, it cannot be used as a specific indicator of the level of body fat on the individual, and it is influenced by factors such as age, gender, and exercise status (i.e., sedentary vs. active). Specifically, women tend to have a higher percentage of body fat than do men given the same BMI. In addition, older individuals tend to have a higher percentage of body fat than do younger individuals for a given BMI, and people who work out regularly (especially those who lift weights) may have a lower percentage of body fat than do individuals who are less athletic for a specific BMI.
| Table 8.1 Classification of Overweight and Obesity by BMI | ||
| Obesity Class | BMI | |
| Underweight | < 18.5 | |
| Normal | 18.5-24.9 | |
| Overweight | 25.0-29.9 | |
| Obesity | I | 30.0-34.9 |
| II | 35.0-39.9 | |
| Extreme obesity | III | ≥ 40.0 |
Source: NIH/NHLBI (1998).
Prevalence statistics reported during the past few years reveal an astonishing increase in the level of overweight and obese individuals in the United States (Centers for Disease Control and Prevention, 2002; Mokdad, Bowman, & Ford, 2001; Mokdad, Serdula, & Dietz, 1999, 2000). Currently, 61% of U.S. adults ages 20 to 74 years are either overweight or obese. The level of obesity increased from 12% of the U.S. population in 1991 to 19.8% in 2000. Thus, an estimated 38.8 million adults in the United States met the BMI cutoff of 30.0 for obesity in the year 2000. Table 8.2 displays these data by gender and ethnicity. These data reflect the particularly high levels of obesity in African American and Hispanic populations.
Physical and Psychological Problems Associated with Obesity
Physical Problems
A wide variety of health problems are associated with obesity. It has been estimated that 325,000 deaths could be attributed to obesity each year (Allison, Fontaine, Manson, Stevens, & VanItallie, 1999). An overweight or obese status has been linked to such seemingly disparate health problems as heart disease, cancer, type 2 diabetes, stroke, arthritis, breathing problems (sleep apnea), high blood pressure, high blood cholesterol, gout, and gallstones (Field, Barnoya, & Colditz, 2002). (More information on physical problems relevant to the assessment of obesity can be found later in this chapter.)
Psychological and Psychosocial Problems
A wide variety of psychological problems have been examined as a concomitant of obesity, and perhaps even more research efforts during recent years have focused on the psychosocial “consequences” of obesity. Psychosocial consequences entail the specific interpersonal, social, and occupational problems encountered by the obese individual specifically due to an elevated weight status such as being teased about one’s size, facing societal prejudice against obesity (which may have economic consequences), and encountering physical barriers (e.g., plane seats that are too small).
Somewhat surprisingly, the common assumption that an elevated weight must necessarily be associated with a plethora of psychological problems is not supported by the literature (Faith & Allison, 1996). Early work in the area did not confirm the expected finding that obese individuals were more depressed than nonobese individuals. Recently, however, a nationally representative sample of more than 40,000 people were evaluated via structured interviews for level of depression (Carpenter, Hasin, Allison, & Faith, 2000). Obese women were 37% more likely to have met DSM-IV (Diagnostic and Statistical Manual of Mental Disorders, fourth edition) criteria (American Psychiatric Association, 1994) for major depression during the previous year compared with women of average weight. On the other hand, obese men had a significantly reduced risk for depression. Interestingly, for men, being underweight was associated with greater depression. Research into the reasons behind such a dramatic gender difference in the psychosocial consequences of obesity is an active area of inquiry.
| Table 8.2 Increasing Adult Obesity Prevalence: Obesity Trends (percentages) | ||||
| Sample | 1991 | 1995 | 1998 | 2000 |
| Total | 12.0 | 15.3 | 17.9 | 19.8 |
| Gender | ||||
| Men | 11.7 | 15.6 | 17.7 | 20.2 |
| Women | 12.2 | 15.0 | 18.1 | 19.4 |
| Ethnicity | ||||
| White, non-Hispanic | 11.3 | 14.5 | 16.6 | 18.5 |
| Black, non-Hispanic | 19.3 | 22.6 | 26.9 | 29.3 |
| Hispanic | 11.6 | 16.8 | 20.8 | 23.4 |
Source: Mokdad, Bowman, Ford, Vinicor, Marks, and Koplan (2001).
The comorbidity of eating disturbances with obesity has been a controversial area. Obesity is not a diagnosable psychiatric condition, and it is not listed as an eating disorder in the DSM-IV (nor has it been labeled as such in previous editions of the DSM). However, the association of a particular type of eating disturbance—binge eating—has been noted as an associate of obesity for many years (Stunkard, 2002). During recent years, binge eating disorder has been proposed as a new entry in the DSM and is currently included in the DSM-IV as a disorder in need of further study. The issue is relevant for obesity because perhaps 30% to 40% of individuals with binge eating disorder are also obese (Johnson & Torgrud, 1996). In addition, individuals with binge eating disorder have higher levels of depression than do non-binge eating controls (Wadden, Womble, Stunkard, & Anderson, 2002). Recently, Bulik, Sullivan, and Kendler (2002) examined the prevalence of binge eating and obesity in a population-based sample of female twins. Obese women with binge eating disorder had higher levels of health dissatisfaction, major medical disorders, major depression, panic disorder, phobias, and alcohol dependence.
One of the most consistent findings related to body image dissatisfaction is that obese individuals are more dissatisfied with their appearance than are nonobese individuals. One possible reason for the greater dissatisfaction of obese individuals is, of course, that they live in a society that glorifies the antithesis of obesity— a slender, nonfat, “ideal” body (Thompson, Heinberg, Altabe, & Tantleff-Dunn, 1999). Pressure and information related by the media, family, and peers may be a constant reminder of the unacceptable nature of the physicality of obese individuals. For instance, negative verbal commentary in the form of teasing or criticism of appearance is a common experience of obese individuals. Some studies have shown that more than 90% of obese individuals have been teased about their appearance (Thompson & Smolak, 2001). Interestingly, teasing may be a more direct influence on body dissatisfaction than is weight. In fact, Thompson, Coovert, Richards, Johnson, and Cattarin (1995) found that BMI had no direct effect on body image but was mediated through teasing history. That is, only those individuals with an elevated BMI who had been teased developed body dissatisfaction.
Occupational and social discrimination against obese individuals is also widely supported by survey and laboratory studies. Gortmaker, Must, Perrin, Sobol, and Dietz (1993) evaluated more than 10,000 overweight and normal-weight adolescents for 7 years. At the conclusion of the study, overweight females were less likely to be married and also had lower incomes, whereas overweight males were only less likely to be married. Experimental designs have also demonstrated that job applicants’ weight has a powerful effect on factors such as selection for a particular job, promotion, and dismissal (e.g., Pingitore, Dugoni, Tindale, & Spring, 1994; Roehling, 1999).
Models of Obesity
Genetic, metabolic, environmental, and developmental models have been offered to explain the onset and maintenance of obesity. To date, evidence supports each as a potentially relevant factor. There is, however, a great deal of variability across individuals, making unitary models limited in their explanatory value. Given the tremendous increase in the prevalence of obesity over the past 10 years, it is understandable that environmental causes have received considerable examination (Hill & Peters, 1998).
Environmental Factors
Environmental models address factors that promote the intake of excessive energy (calories) as well as those variables that promote a reduction of energy expenditure via exercise. The ready availability of energy-dense foods, advent of larger portion sizes at restaurants, increased use of fast foods, and decreased preparation of meals are environmental factors that may promote weight gain (French, Story, & Jeffery, 2001). In addition, there have been numerous environmental modifications as a result of technological advances that have enhanced the sedentary nature of many individuals’ lifestyles both at home and in the workplace. Computers, video games, and television are often selected to the exclusion of recreational activities that burn calories (Dowda, Ainsworth, Addy, Saunders, & Riner, 2001), whereas physical education is increasingly eliminated or downsized in schools due to liability concerns or increased emphasis on classroom activities to enhance standardized testing scores. Finally, urban planning and sprawl often do not include sidewalks or bike paths to encourage exercise, and residents who perceive their neighborhoods to be unsafe may be reluctant to leave their homes.
Horgen and Brownell (2002) referred to the “toxic environment” as a primary cause of the emerging problem of obesity. Support for this position is enhanced not only by the recent epidemic in the United States but also by the burgeoning problem worldwide, leading the World Health Organization (1998) to declare that a global epidemic exists. Obesity now appears to be increasing in every country that has been surveyed. Migration, urbanization, and affluence, coupled with the “Americanization” of food selections and eating habits in other countries, have been suggested as dominant, large-scale, environmental factors (VanItallie, 1994). In the United States, the pervasiveness of the “toxic environment” is apparent from even a superficial examination of the strategies of food marketing. Fast-food restaurants offer options to “bigger size” one’s meals, inducing even greater consumption of meals already high in fat content.
Genetic Factors
Mapping of the human genome has generated interest in studying the genetic factors for a variety of psychological and physical disorders. Analysis of the genomic regions with linkage to obesity is at a relatively early stage of scientific development (Price, 2002), although molecular studies are identifying an evergrowing list of genes that may confer increased risk for obesity in humans (Rankinen et al., 2002). More traditional work in this area has focused on the analysis of familial/genetic influence by examining monozygotic and dizygotic twins. Specifically, examination of body composition among twins has revealed that monozygotic twins have correlations ranging from .60 to .70, higher than the .20 to .30 observed among dizygotic twins (Price, 2002).
Recent summaries of the work in this area suggest that 67% of the variation in BMI among monozygotic twins may be attributable to genetics (Price, 2002; Tataranni & Ravussin, 2002). This genetic influence may manifest in physiological and behavioral effects such as different resting metabolic rates, reduced physical activity (e.g., fidgeting), and excessive food consumption. Although a “thrifty” gene has long been hypothesized to lie at the roots of the elevated obesity rates in some subgroups (e.g., among the Pima Indians), no such single gene has been detected. Even among such subgroups, however, rates of obesity have increased along the lines of other population groups over the past 50 years, suggesting more than a genetic influence (Price, Charles, Pettitt, & Knowler, 1993). Hence, obesity is most likely a “polygenic trait,” influenced by multiple genes that interact among themselves as well as by environmental inputs (Comuzzie & Allison, 1998).
An interesting question in this area is whether genes have a direct biological effect on obesity and perhaps drive an individually based response to specific environmental factors (e.g., low activity, excessive eating) (Faith, Johnson, & Allison, 1997; Keller, Pietrobelli, Must, & Faith, 2002). Put more clearly, are genes partly responsible for those gaining the most weight in our toxic environment? Bouchard and colleagues (2000) overfed identical twins for 3 months and found that although there were large differences in the levels of weight added across pairs, within-pair gains were similar. Therefore, it is possible that gene-environment interactions exist with respect to weight gain.
Developmental Factors
There is strong support for the prediction of obese status in adulthood from an examination of the weight of the individual during childhood and adolescence (Thompson & Smolak, 2001). Overweight adolescents have a 70% chance of becoming overweight or obese adults. Interestingly, the contribution of the parents’ weight, along with the child’s weight, may depend on the child. For instance, Whitaker, Wright, Pepe, Seidel, and Dietz (1997) looked at the weight of both the parents and the child to determine whether one or both contributed to the development of obesity in the child. They found that very young children (ages 1 to 2 years) who were obese and had an overweight parent had a fourfold risk for adult obesity as compared with overweight children of the same age who had average-weight parents. However, by age 10 years, the effect was independent of the parents’ weight.
Other early developmental factors that have received examination include type of infant feeding (breast vs. bottle), feeding style, and parental control over feeding. Each of these areas is receiving a great deal of attention, and the findings are often inconclusive. However, it appears that, despite the widely held view to the contrary, there is no difference between breast-feeding and bottle-feeding in terms of the later development of obesity (Berkowitz & Stunkard, 2002). Feeding style involves the study of factors such as rapid eating and the vigor of the sucking response during infancy. Findings in this area are intriguing. For example, Stunkard, Berkowitz, Stallings, and Schoeller (1999) found that the infant offspring of overweight mothers had a more vigorous sucking style than did the infants of control mothers. Importantly, the authors also found that they could predict level of fatness at age 1 year by an examination of the sucking style. Finally, parental control is the degree to which a parent, typically the mother, attempts to manage or direct the child’s intake. Birch and colleagues have pioneered work in this area and found intriguing results suggesting that parental control may influence not only weight gain but, in certain circumstances, restriction of intake as well (Fisher & Birch, 2001).
Summary
Prevalence rates of obesity have increased drastically during the past few decades in children, adolescents, and adults as well as among both genders and all ethnicities evaluated. This “epidemic” has generated a wealth of research into associated health and physical conditions as well as strategies to understand causal variables. To date, etiological models are inconclusive, but compelling epidemiological evidence points toward a “toxic” environment that sets the stage for the development and perpetuates the maintenance of obesity.
Assessment and Treatment of Obesity
This section summarizes approaches for the assessment and treatment of obesity in adults. General guidelines and strategies are provided here; however, more detailed descriptions are provided elsewhere (Foreyt & Goodrick, 1992; Kirschenbaum, 1994; Wadden & VanItallie, 1992). In particular, clinicians specializing in this area are encouraged to read the seminal report, Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults (National Institutes of Health/National Heart, Lung, and Blood Institute [NIH/NHLBI], 1998). This report provides the most authoritative, comprehensive, and empirically based guidelines to date for obesity treatment. Most of the guidelines provided herein are adapted from these expert recommendations.
This section is broken down into assessment and treatment. Within treatment, behavioral/lifestyle interventions and brief updates on pharmacological and surgical interventions are discussed. This is followed up with a discussion of the role of body image enhancement in obesity treatment.
Assessment
With approximately half of the U.S. population overweight or obese, it is important to determine when treatment is warranted. Two measurements that are practical for clinical settings are recommended as first-step screening instruments: BMI and the waist circumference. As described previously (Table 8.1), BMI can be used to define degree of obesity. Waist circumference, as determined by a tape measure, is suggested because BMI alone does not provide an informative index of so-called “abdominal fat.” Abdominal fat is important for assessment because higher levels of abdominal fat are an independent predictor of health complications related to obesity (Pi-Sunyer, 1993). Hence, an individual who is “overweight” but has excess abdominal fat may be at greater risk for various diseases (e.g., cardiovascular disease, type 2 diabetes) than is another individual who is “obese” with lower levels of abdominal fat.
Waist circumference is measured by a standard protocol (NIH/NHLBI, 1998) and should be performed by trained staff. Expert guidelines suggest that waist circumference measurements should be taken among persons with BMIs between 25.0 and 34.9. For individuals with BMIs greater than or equal to 35.0, or who are “short” in stature, waist circumference may be minimally informative beyond BMI alone (NIH/NHLBI, 1998). Among adults with BMIs between 25.0 and 34.9, waist circumferences greater than 102 centimeters (40 inches) in men and greater than 88 centimeters (35 inches) in women have been proposed as “high-risk” cutoffs for disease. Elevated waist circumferences are associated with increased risk for type 2 diabetes, dyslipidemia, hypertension, and cardiovascular disease. Table 8.3 summarizes the relative risk of disease as a function of BMI and waist circumference profiles and illustrates how abdominal fat levels can amplify or attenuate the health risks associated with elevated BMI.
| Table 8.3 Risk of Type II Diabetes, Hypertension, and Cardiovascular Disease as a Function of Weight Class and Waist Circumference | ||
| Weight Class | “Normal” Waist Circumference | “High-Risk” Waist Circumference |
| Underweight | — | — |
| Normal | — | — |
| Overweight | Increased | High |
| Obesity-I | High | Very high |
| Obesity-II | Very high | Very high |
| Extreme obesity | Extremely high | Extremely high |
Souce: Adapted from NIH/NHLBI (1998).
Beyond BMI and waist circumference assessment, expert guidelines pinpoint additional clinical markers that may indicate individuals at a “very high absolute risk” of disease. The presence of one or more of these conditions may suggest the need for more aggressive obesity intervention. Table 8.4 summarizes these diseases or target organ damages that should be considered when evaluating obese patients (NIH/NHLBI, 1998).
Treatment
Obesity Treatment Algorithm
The NIH/NHLBI (1998) report provides a concrete algorithm to guide the obesity evaluation-treatment interface.
The heuristic begins with assessment of family history of obesity. If there is no history of obesity, experts recommend 2-year intervals for the assessment of changes in BMI. At 2-year assessments (or more frequent assessments if obesity is familial), health professionals are advised to calculate BMI as well as to measure weight, height, and waist circumference. For patients who are overweight or obese, or who have a high-risk waist circumference (greater than 88 centimeters in females, greater than 102 centimeters in males), a thorough medical assessment is warranted that assesses the conditions listed in Table 8.4. For patients who are obese, are overweight, or have a high risk waist circumference in conjunction with at least two risk factors, development of a treatment plan for weight control and risk factor reduction is warranted. Individuals who meet these criteria but do not want to lose weight may be advised to maintain their current weight or address concurrent risk factors until they are sufficiently motivated to lose weight. Treatment generally encompasses dietary, physical activity, and lifestyle changes, with appropriate assessment of barriers to treatment and periodic weight checks.
Intervention Components
A well-established literature indicates that behavioral/lifestyle approaches to obesity treatment are effective for inducing weight loss, although relapse is very common (Faith, Fontaine, Cheskin, & Allison, 2000; Wadden, Sternberg, Letizia, Stunkard, & Foster, 1989). Enhanced dietary quality is a hallmark feature of behavioral treatment, as patients are traditionally prescribed a “low-calorie diet” that is generally 800 to 1,500 calories per day. Features of the American Heart Association’s Low-Calorie Step I Diet are summarized in Table 8.5 (NIH/NHLBI, 1998).
Reducing total caloric intake by 500 to 1,000 calories per day, assuming no changes in physical activity levels, is expected to induce weight loss of approximately 1 to 2 pounds per week. Reducing total caloric intake by 300 to 500 calories per day (the equivalent of approximately two soft drinks) is expected to induce weight loss of ½ to 1 pound per week, assuming no changes in physical activity. For these reasons, obese adults are typically prescribed to consume 1,000 to 1,200 calories per day (women) or 1,200 to 1,500 calories per day (men) to achieve weight loss.
| Table 8.4 Diseases and Target Organ Damages for Potential Assessment in Obese Adults |
| 1.
· Identification of patients at very high absolute risk: 1a. o Established CHD § History of myocardial infraction § History of angina pectoris (stable or unstable) § History of coronary artery surgery § History of coronary artery procedures (angioplasty) 1b. o Presence of other atherosclerotic disease § Peripheral arterial disease § Abdominal aortic aneurysm § Symptomatic carotid artery disease § Type 2 diabetes § Sleep apnea 2. · Identification of other obesity-associated diseases: 2a. o Gynecological abnormalities 2b. o Osteoarthritis 2c. o Gallstones and their complications 2d. o Stress incontinence 3. · Identification of cardiovascular risk factors that impart a high absolute risk (patients can be classified as being at high absolute risk for obesity-related disorders if they have three or more of the multiple risk factors listed below): 3a. o Cigarette smoking 3b. o Hypertension: A patient is classified as having hypertension if systolic blood pressure is > 140 millimeters of mercury or diastolic blood pressure is > 90 millimeters of mercury or if the patient is taking antihypertensive agents. 3c. o High-risk LDL cholesterol: A high-risk LDL cholesterol is defined as a serum concentration of > 160 milligrams per deciliter. A borderline high-risk LDL cholesterol (130 to 159 milligrams per deciliter), together with two or more other risk factors, also confers high risk. 3d. o Low HDL cholesterol: A low HDL cholesterol is defined as a serum concentration of < 35 milligrams per deciliter. 3e. o Impaired fasting glucose: The presence of clinical type 2 diabetes (fasting plasma glucose of > 126 milligrams per deciliter or 2 hours postprandial plasma glucose of > 200 milligrams per deciliter) is a major risk factor for CVD, and its presence alone places a patient in the category of very high absolute risk. Impaired fasting glucose (fasting plasma glucose 110 to 125 milligrams per deciliter) is considered by many authorities to be a risk factor for cardiovascular disease. 3f. o Family history of premature CHD 3g. o Age (males > 45 years or females > 55 years or postmenopausal) 4. · Other risk factors: 4a. o Physical inactivity 4b. o High triglycerides (400 to 1,000 milligrams per deciliter = “high”, > 1,000 milligrams per deciliter = “very high”) |
| SOURCE: NIH/NHLBI (1998). |
| NOTE: CHD = coronary heart disease; LDL = low-density lipoprotein; HDL = high-density lipoprotein; CVD = cardiovascular disease. |
Physical activity is another pillar of obesity treatment. Physical activity without concurrent improvements in diet quality will have a less potent effect on weight loss than will changes in diet plus physical activity (NIH/NHLBI, 1998). During recent years, research has focused on the beneficial effects of so-called “lifestyle activity” changes compared with more regimented and structured exercise programs. Lifestyle programs teach individuals how to program physical activity changes into everyday living. Examples include parking farther away in the parking lot so that one needs to walk farther to the office/shop, taking daily walks, cleaning the house regularly, moving while talking on the phone, and taking stairs instead of elevators whenever one has a choice. A study by Andersen and colleagues (1999) compared weight loss among 40 obese women who were randomized to a weight loss group that incorporated a structured aerobic exercise regimen versus a lifestyle activity regimen. Although the two groups lost comparable amounts of weight during the 16-week intervention, the aerobic exercise group regained significantly more weight on average (1.6 kilograms) than did the lifestyle intervention group (0.08 kilograms). Data from the Women’s Healthy Lifestyle Project Clinical Trial (Kuller, Simkin-Silverman, Wing, Meilahn, & Ives, 2001) also confirm that prescriptions for lifestyle physical activity, in conjunction with the aforementioned dietary prescriptions, can prevent excess weight gain and elevations in low-density lipoprotein (LDL) cholesterol among women as they progress from perimenopause to postmenopause.
| Table 8.5 Low-Calorie Step I Diet | |
| Nutrient | Recommended Intake |
| Calories | Approximately 500 to 1,000 calories per day reduction from usual intake |
| Total fat | ≤ 30% of total calories |
| Saturated fatty acids | 8% to 10% of total calories |
| Monounsaturated fats | ≤ 15% of total calories |
| Polyunsaturated fats | ≤ 10% of total calories |
| Cholesterol | < 300 milligrams per day |
| Protein | Approximately 15% of total calories |
| Carbohydrates | 55% or more of total calories |
| Sodium chloride | No more than 100 millimoles per day (approximately 2.4 grams of sodium or approximately 6 grams of sodium chloride) |
| Calcium | 1,000 to 1,500 milligrams |
| Fiber | 20 to 30 grams |
| Source: NIH/NHLBI (1998). | |
On a practical level, increasing daily physical activity can be personally tailored to accommodate individual lives. “Moderate” intensity physical activity is conceptualized as activities that translate to an energy deficit of 150 calories per day, which would sum to 1,000 calories per week (NIH/NHLBI, 1998). The amount of time needed to achieve this deficit each day will depend on the nature of the activity and the individual’s weight. Expert guidelines suggest that initially the obese patient should engage in moderate activity levels for 30 to 45 minutes per day on 3 to 5 days per week (NIH/NHLBI, 1998). According to the surgeon general’s report, most adults should ultimately strive to accumulate at least 30 minutes of moderate-intensity physical activity on most, if not all, days of the week. Table 8.6 illustrates the different ways in which an adult can achieve moderate physical activity levels each day, varying the time spent doing the activity and the strenuousness of the activity.
Behavioral Techniques
Descriptions of the behavioral treatment of obesity can be traced back to classical articles by Ferster, Nurnberger, and Levitt (1962) and Stuart (1967). Between the 1970s and the 1990s, classical experiments tested behavioral treatments for obesity among a range of obese patients. As the methodological rigor of these studies (including sample sizes, treatment lengths, and follow-up lengths) improved over time, so did the efficacy of treatment (Faith et al., 2000; Wadden & Foster, 2000). Whereas intervention studies published in 1974 achieved a mean weight loss of 3.8 kilograms (or approximately 8.5 pounds), those published between 1985 and 1987 achieved a mean weight loss of 8.4 kilograms (or approximately 18.5 pounds). Unfortunately, a cardinal disappointment of these same weight loss studies—and a finding that transcends decades of research—is an excessive rate of recidivism (Wadden et al., 1989; Wilson, 1994).
Specific components of the behavioral treatment of obesity have been outlined in detail elsewhere (Foreyt & Goodrick, 1992; Kirschenbaum, 1994; Wadden & VanItallie, 1992). Critical concepts reviewed by expert guidelines (NIH/NHLBI, 1998) include the following.
Self-Monitoring of Eating Habits and Physical Activity. Recording in detail what one eats every day can provide revealing insights into the environmental, emotional, and/or interpersonal circumstances that may prompt intake of particular foods. Keeping such recordings with standard paper-and-pencil methods or with novel palm computers can bring patterns of eating and inactivity to consciousness and help make them the target of treatment.
Stress Management. Stress can prompt overeating in certain obese individuals, especially those with binge eating tendencies (Faith, Allison, & Geliebter, 1997). Teaching obese individuals coping strategies to deal with stress without eating can be an important component of treatment. Exercise can be an effective stress management technique that also confers benefits for weight management.
Stimulus Control. Unhealthy food selections and/or overeating can be stimulated by foods that are more readily accessible and conveniently ready for eating. For example, homes that do not keep an ample supply of fresh fruits and vegetables for snacking are less conducive to weight control than are homes in which these foods are readily available. Therefore, behavioral interventions focus on barriers to restructuring the environment in ways that stimulate healthy eating and physical activity.
Contingency Management. Rewards for behavioral changes can be an effective component of treatment. Rewards can be tangible gifts for oneself or praise from family members and friends (i.e., social support).
Cognitive Restructuring. Positive self-statements that challenge defeating self-statements can be effective for teaching obese patients to cope with treatment “failures” (e.g., periods of weight regain) or other unexpected challenges during treatment. These matters can be especially important for obese patients lacking in self-acceptance or self-esteem due to limited weight loss or weight regain (Wilson, 1996). Recent advances in treatment are focusing on promotion of more realistic weight loss expectations to promote better psychological well-being and perhaps even weight loss.
Pharmacological Therapy
Detailed reviews of the drug treatment literature have been published elsewhere (e.g., Haddock, Poston, Dill, Foreyt, & Ericsson, 2002). There are currently two prescription medications that are approved by the Food and Drug Administration (FDA) for obesity treatment: sibutramine (Meridia) and orlistat (Xenical). Sibutramine is an anorexiant (or appetite suppressant) that exerts its effect by inhibiting the reuptake of norepinephrine, dopamine, and serotinin in the brain. Controlled clinical trials document the clinical effectiveness of sibutramine for weight loss (Berube-Parent, Prudhomme, St-Pierre, Doucet, & Tremblay, 2001; Fanghanel, Cortinas, Sanchez-Reyes, & Berber, 2001; Wirth & Krause, 2001). On the other hand, a major concern surrounding sibutramine is the potential for increased heart rate and blood pressure (Bray et al., 1996). For these reasons, individuals with a history of high blood pressure, coronary heart disease, congestive heart failure, arrhythmias, or stroke should not take sibutramine. All patients taking the medication are encouraged to have their blood pressure monitored regularly (NIH/NHLBI, 1998).
Orlistat exerts its effects by inhibiting pancreatic lipase and thereby decreasing fat absorption. The effectiveness of orlistat for inducing long-term (i.e., 2-year) weight loss compared with controls was documented in two sets of multisite studies: one conducted in Europe (Rossner, Sjöström, Noack, Meinders, & Noseda, 2000) and one conducted in the United States (Davidson et al., 1999). Orlistat can also have beneficial effects on blood glucose levels and diabetes-related risk factors (Heymsfield et al., 2000). The fact that orlistat’s principal mechanism of action does not involve brain neurotransmitters represents a distinct advantage over sibutramine. On the other hand, orlistat has its own adverse effects, including a decrease in absorption of fat-soluble vitamins, soft stool and anal leakage in some individuals, and a possible link to breast cancer (NIH/NHLBI, 1998).
In sum, there are currently two FDA-approved drug options for obese individuals who may need additional help to achieve weight loss. However, these pharmacological agents can have adverse effects and so require careful medical supervision if they are to be taken. Expert guidelines propose these alternatives as possible options only after a patient has tried 6 months of behavioral intervention, including diet and exercise, without success. Even then, these alternatives are recommended only for patients with BMIs greater than or equal to 30 without obesity-related risk factors or diseases or for patients with BMIs greater than or equal to 27 with concomitant obesity-related risk factors or diseases (NIH/NHLBI, 1998).
Surgical Approaches
Surgical approaches to obesity are generally recommended only for the most severe cases in which other behavioral and pharmacological approaches have failed. These approaches are recommended only for individuals with BMIs greater than or equal to 40 or for individuals with BMIs greater than or equal to 35 plus comorbid medical conditions (NIH/NHLBI, 1998). Current surgical approaches include so-called “gastric restriction” (vertical gastric banding) and “gastric bypass” (Roux-en Y), which can achieve sizable weight loss and improvements in comorbidities. The most striking data documenting these effects come from the Swedish Obesity Study, initiated in 1987, that has prospectively followed 1,000 obese patients who underwent surgical treatment compared with 1,000 obese patients who received conventional nonsurgical intervention. A decade after the treatment was initiated, long-term results provided compelling data favoring surgical treatment. Surgical patients lost an average of 28 kilograms (or approximately 61 pounds) compared with 0.5 kilograms (or approximately 1 pound) lost by nonsurgical controls. Compared with the control group, surgical patients had a 32-fold reduction in 2-year diabetes incidence and a 5-fold reduction at 8-year follow-up (Torgerson & Sjöström, 2001). Surgical patients were also more likely to show reduced time on sick leave and disability pension (Narbro et al., 1999).
At the same time, surgery-related complications can be common and should be monitored. Side effects noted in one surgical study (Pories et al., 1995) included Vitamin B deficiency (39.9% of patients), depression (23.7% of patients), gastritis (13.2% of patients), and dehydration/malnutrition (5.8% of patients).
Targeting Body Image in Obesity Treatment
Body image disparagement, or dissatisfaction with one’s body, is one of the more common psychological hardships of obesity (Faith & Allison, 1996; Friedman & Brownell, 1995). This is readily understandable given the numerous prejudices and forms of discrimination that confront obese individuals (Puhl & Brownell, 2001). To this end, consideration of a patient’s body image both at the beginning and over the course of treatment is an important consideration. There are at least two ways in which body image issues might be built into treatment programs: (a) body image as an outcome measure index to be targeted in addition to other traditional health risk factors and (b) body image and size acceptance as a potential mediator of weight change.
Body Image as a Treatment Outcome. Assessing body image changes is justified based on the frequency with which obese individuals report dissatisfaction with their bodies. A detailed list of validated body image measures for obese individuals is available elsewhere (Thompson et al., 1999). Results from behavioral interventions indicate that body image is significantly enhanced when patients lose weight (Foster, Wadden, & Vogt, 1997; Sarwer, Wadden, & Foster, 1998). Hence, clinicians can anticipate improvement in body image as an additional benefit of weight loss. Given the challenges of achieving weight loss, other investigators have attempted to develop cognitive-behavioral interventions that target psychological well-being and body image without necessarily inducing weight loss. Many of these treatments stem from the so-called “anti-diet” framework. Examples include Ciliska’s (1990) program, which attempted to help patients “reestablish normal eating, improve self-esteem, and learn to deal with negative messages about our body shape in order to be more accepting of ourselves” (p. 49), and Polivy and Herman’s (1992) “Undieting” program.
In one of the most comprehensive programs to date, Rosen and colleagues (Rosen, 1996; Rosen, Orosan, & Reiter, 1995) developed a treatment that targeted negative body image among obese women. Individuals were assigned to either a no-treatment control group or a cognitive-behavioral intervention that included exploration of the social consequences of obesity, factors that cause and maintain negative body image, stress management surrounding physical appearance, cognitive restructuring surrounding assumptions of physical appearance, and body image exposure in stressful social situations. Results indicated that the treatment group showed significant improvements on most psychological and body image measures as compared with the control group. Details of this treatment protocol are provided elsewhere (Rosen et al., 1995); however, these data indicate that body image enhancement can be a viable treatment goal for certain obese patients.
| Table 8.7 Sample Questions to Probe Patient Weight Loss Expectations |
| 1.· What is your target weight loss goal? Why?
2. What would you consider to be an acceptable weight loss goal? Why? 3. What would you consider to be a disappointing but somewhat acceptable weight loss goal? Why? 4. What would you consider to be your dream weight loss goal? Why? 5. How would you feel about a weight loss of [10%]? Why? |
| Source: These questions were modeled after Foster, Wadden, Vogt, and Brewer (1997). |
Body Image and Size Acceptance as Possible Mediators of Weight Loss. Studies by Foster and colleagues revealed that many obese patients hold vastly unrealistic expectations about the amount of weight loss they can achieve through current intervention approaches (Foster, Wadden, Phelan, Sarwer, & Sanderson, 2001; Foster, Wadden, Vogt, & Brewer, 1997). Obese individuals enrolling in a university-based weight loss program were asked to define the amounts of weight loss they defined as a goal or target weight loss, acceptable but not particularly happy weight loss, and disappointing weight loss. Results indicated that most patients brought unrealistic weight loss goals to treatment. (See Table 8.7) Patients defined a 32% reduction in weight as their target goal. A 25-kilogram (or approximately 55-pound) weight loss was considered to be an acceptable but not particularly happy weight loss, and a 17-kilogram (or approximately 37.5-pound) weight loss was considered to be a disappointment. At the end of treatment, none of the patients had achieved his or her “dream” weight, 9% had attained a “happy weight,” 20% had attained a “disappointing” weight, 24% had attained an “acceptable” weight, and 47% had lost less weight than what they defined as a disappointing weight. To put these numbers into context, behavioral interventions currently strive to induce weight loss of approximately 8% to 10%. Although a 10% weight reduction is less than what patients might desire, it can confer important health benefits (NIH/NHLBI, 1998).
Based on these data, Foster, Wadden, Vogt, and Brewer (1997) concluded, “It seems more appropriate to help patients accept more modest weight loss outcomes rather than attempt to devise treatments to increase weight loss” (p. 85, emphasis in original). Indeed, Cooper and Fairburn (2001) called for novel behavioral treatments that assess patients’ baseline treatment expectations and that use cognitive strategies to promote more realistic treatment goals. The underlying hypothesis is that helping patients to adopt more realistic treatment expectations will ultimately promote better treatment outcomes given that frustrations associated with modest weight loss will be reduced. This hypothesis awaits empirical evaluation.
Case Study
The case of “Virginia M.” is a useful one that illustrates the many facets of human obesity and reminds us that obese individuals are no different from nonobese individuals in many respects. Obese individuals lead complicated and busy lives, and they desire professional success and achievement as well as gratifying interpersonal relationships. Obese individuals have a genetic constitution, however, that makes weight gain easier when they engage in behaviors that favor excess energy intake (overeating high-fat foods) and reduced energy expenditure (being sedentary).
Virginia’s background fits this description reasonably well. A 37-year-old divorced attorney, Virginia is a professionally active woman who “lives on the run” and perceives there to be limited time to lead a healthier life. Weighing 200 pounds (or approximately 91 kilograms) and measuring 5 feet 5 inches in height, her BMI is 33.3, putting her in the Obesity-I category. Virginia has a waist circumference of 92 centimeters, putting her at elevated risk for obesity-related complications (according to the guidelines reviewed herein). Obesity runs on the maternal side of her family but not on the paternal side. Although Virginia says that she feels fine, a recent medical examination suggested slightly elevated total and LDL (“bad”) cholesterol and blood pressure. Based on the treatment algorithm reviewed herein, Virginia would benefit from weight loss.
When Virginia was asked what her “ideal” weight would be, she indicated 185 pounds. Such a 15-pound weight loss (from 200 to 185 pounds) would represent an 8% weight reduction, which is reasonable given current behavioral interventions. Fortunately, her expectations do not appear to be unrealistically high, and this can be used clinically to enhance her motivation to make changes. If Virginia had defined her ideal weight to be 160 pounds, it might have been necessary to explore more attainable goals before beginning treatment.
Virginia sought behavioral therapy, and a treatment program was developed that targeted healthier eating, increased physical activity, and reduced job stress. One of the most critical themes in her treatment was that of building structure into her life. Structure was operationalized as planning and preparing appropriate meals ahead of time (something Virginia avoided previously), planning ways in which to increase lifestyle physical activity that would be compatible with her active professional goals, and joining and working out in a gym. To modify her diet, Virginia began keeping a food record and self-monitored her dietary intake several days each week. Keeping this record revealed important patterns (e.g., snacking on high-fat foods during afternoons when she skipped lunch, eating fast foods when she was feeling lonely during the evenings). Based on these insights, she began to structure lunches more regularly, even when she was busy. She also began to find substitute activities during the evening to replace snacking (e.g., movies, walks, social events for singles). In terms of physical activity, she began 30 minutes of fast-pace walking 3 days per week, and this was gradually increased to 5 days per week before she joined a gym. She also began lessons in relaxation training and began reading books on time management to better balance her daily activities.
To date, Virginia has lost nearly 10 pounds and continues toward her goal. She recognizes that “treatment” will require lifelong lifestyle modification. Given that Virginia’s job is central to her self-image, she will continue to work hard; however, she recognizes the potential health complications of her weight and will be especially attentive to cardiovascular disease and type 2 diabetes symptoms. These serious medical conditions cannot be discarded. At the same time, Virginia better prioritizes personal happiness and relationships. She is increasingly recognizing that beauty, like body weight, can come in different sizes.
Collectively, these data speak to the utility of assessing patients’ expectations about weight loss goals. Health care professionals should be candid about the types of weight loss that patients can expect, given the current data, and may wish to address weight loss goals that seem unattainable. At the same time, clinicians should be sensitive to patients’ motivations and should collaboratively establish attainable weight loss goals that will foster a sense of mastery and self-efficacy. Table 8.7 illustrates questions that clinicians may wish to ask as part of an assessment of weight loss expectation.
Conclusions
Obesity is one of the most pressing public health disorders confronting Western societies. It is a multifaceted problem in terms of its causes and consequences. Obesity is determined by genetic and environmental factors that are poorly understood. It seems apparent that one’s genetic background will affect weight gain in response to eating. Certain individuals inherit a propensity to resist weight gain, whereas others readily put on weight in response to modest elevations in daily caloric intake. On a broader environmental level, changes appear to be occurring that put the entire population at increased risk for obesity. Indeed, the population prevalence of obesity has increased dramatically during the past several decades.
The diagnosis and treatment of obesity starts at the simplest level with assessment of BMI, although this is hardly the place to stop. Because obesity is associated with many health complications, it is important that these be assessed especially among the heaviest individuals. High blood pressure, elevated total and LDL cholesterol, and type 2 diabetes are just a few of the health complications that must be considered. Psychosocial factors, such as poor self-esteem and poor body image, also deserve consideration. Because body image and obesity are so strongly intertwined, it is important that the overall life satisfaction of obese individuals be assessed.
Behavioral treatments that target lifestyle changes compatible with everyday life, promote healthier food choices and reduced fat intake, and increase physical activity most days of the week are generally the first intervention strategy. Pharmacological agents can be used to assist with weight loss, although potential side effects need to be considered carefully. Each of the two current FDA-approved drugs, sibutramine and orlistat, has its own side effects that render it prohibitive for many obese adults. Surgical treatments have encouraging outcome results for severe obesity. For those morbidly obese individuals who have obesity-related complications and cannot lose weight through other prescribed methods, surgery may be a consideration when evaluated judiciously.
The high recidivism rate for obesity treatment indicates that many obese individuals will not achieve their desired weight loss. This underscores the importance of helping obese individuals to achieve personal satisfaction, self-respect, and healthy interpersonal relationships through the highs and lows of weight management.