Prevalence and Demographics
The 2000 Behavioral Risk Factor Surveillance System (BRFSS) data shows that up to 46% of U.S. women and 33% of U.S. men are presently trying to lose weight (Bish et al., 2005). Currently, the weight-loss charge is practiced most frequently by Hispanic (50%), Caucasian and African-American (45-46%) women, followed by Hispanic (34%), Caucasian (32%), and African-American (31%) men based on data from over 180,000 participants in the BRFSS 2000 (Bish et al., 2005). Weight-loss strategies included eating fewer calories (56% of women, 53% of men). This strategy was most frequently used by non-Hispanic White women between the ages of 30 and 69 years, men over 50 years of age, and those who received a medical recommendation to lose weight (Bish et al., 2005). Physical activity was used by approximately 66% of men and women trying to lose weight. As with the strategy of cutting calories, physical activity was used most frequently by non-Hispanic White women (Bish et al., 2005). Although this combination of strategies is considered the most healthful and effective approach to weight loss, only 20% of respondents combined these in an effective manner by reducing calories and meeting the physical activity requirements of >150 min/week of leisure-time physical activity (Bish et al., 2005). Rather, most individuals choose simply to cut calories in an effort to lose weight.
While dieting was once the purview of adults, young children have now joined the ranks of dieters whereby 46% of 9-11 year olds are dieting (National Eating Disorders Association [NEDA], 2005a). Dieting has also increased among those on the other end of the age range, with 43% of women ages 60-69 years and 30% of women 70 years and older trying to lose weight. This is true for older men as well, with 37% of men in the 60-69 years age range and 26% of men 70 years and older also trying to lose weight (Bish et al., 2005). It is questionable as to the necessity of older people attempting weight loss as mortality related to obesity declines with age and is nearly absent by the time people reach their mid 70s (McTigue, Hess, & Ziouras, 2006). However, McTigue and colleagues recommend a careful evaluation of potential risks and benefits of weight loss for each individual patient. Perhaps more interesting are the findings that over 9% of adult men, nearly 29% of adult women (Bish et al., 2005), and 66% of high school-aged girls with normal BMIs were also trying to lose weight (Calderon, Yu, & Jambazian, 2004). These results are not surprising when 9% of adult men and over 25% of adult women of normal weight, BMI of 18.5 to 25, perceive themselves to be overweight (Paeratakul, White, Williamson, Ryan, & Bray, 2002). Such findings illustrate the problem of people dieting in hopes of meeting a potentially unrealistic ideal when their weight is already at what is considered a healthy level.
Definitions, Health Implications, and Controversies According to the NEDA (2005b) dieting is defined as "Any attempts in the name of weight loss, 'healthy eating,' or body sculpting to deny your body of the essential, well-balanced nutrients and calories it needs to function to its fullest capacity." Although the majority of dieters, 71%, engaged in restrictive eating practices to improve their health (Paeratakul, White et al., 2002), research now questions whether dieting is health-enhancing as commonly believed or whether it is actually detrimental to physical and psychological well-being. The concern regarding the health effects of dieting lies in the difficulty of maintaining weight loss (Elfhag & Rossner, 2005). Weight cycling or yo-yo dieting, whether during childhood or adulthood, can cause ongoing stress on the cardiovascular and renal systems. Weight cycling can also cause vascular damage due to extreme fluctuations in circulating glucose and lipids and contribute to an increase in body fat and obesity (Montani, Viecelli, Prevot, & Dulloo, 2006). Montani et al. make the point that the physical impact of yo-yo dieting also affects people of normal weight who gain and lose weight throughout their lives. Because of the societal appearance standards for women and girls, they are at particular risk for the side effects of weight cycling. Additionally, the reduced calorie intake can cause dieters to receive inadequate nutrients such as calcium, thus increasing their risk for osteoporosis and related injury (NEDA, 2005b). Psychologically, dieting can lead to eating disorders such as binge eating, depression, and lowered self-esteem (Darby, Hay, Mond, Rodgers, & Owen, 2006; Grilo, & Masheb, 2000; NEDA, 2005b). However, the National Task Force on the Prevention and Treatment of Obesity's (2000) review of the literature suggests that these concerns have been exaggerated and makes the point that, "such concerns should not preclude attempts to reduce caloric intake and increase physical activity to achieve modest weight loss or prevent additional weight gain" (p. 2581).
Obesity and overweight are now acknowledged among researchers to be the result of multiple interacting variables including environmental, social, cultural, physiological, genetic, and behavioral variables (CDC, 2006c). In fact, a burgeoning body of literature attests to the unmodifiable influence of genetic endowment on body weight and shape (Montani et al., 2006). Nevertheless, many individuals still look to dieting as a means of altering their physiques. Some time ago, Brownell (1991b) suggested that people operate under a mistaken assumption that body weight and shape is completely under their control—that essentially, through individual effort and personal will, they can achieve any body type that they desire. Despite some recognition of these constraints in professional circles, the message has not trickled down to the lay public, evidenced by the vast and increasing number of dieters in the United States.
One needs only look to the mass media to understand this national obsession with dieting and weight loss. For instance, thinness is equated with both health and beauty in magazines, especially those targeting women and girls (Levine & Harrison, 2003). Similarly, Greenberg and Worrell (2005) found that articles and advertisements about dieting are 10 times more likely to appear in female-focused magazines than in male-targeted magazines. In addition to print media, television also projects a narrow range of desirable body types. Greenberg, Eastin, Hofschire, Lachlan, and Brownell (2003) found that, during the 1999-2000 television season, over 30% of women on television sitcoms were underweight, whereas in the real world, only 5% are underweight. In contrast, only 3% of women on television were obese, whereas in the real world, 25% are obese. Notably, the overweight women presented on television were the subject of jokes and had fewer positive relationships, friends, or love interests. Apart from these portrayals, television—and its celebrities—are also the direct source of information on self-help books, often those that promote dieting and weight loss (Kissling, 1995; Wilson, 2003).
Only 16% of persons who are classified as overweight and half of those who are classified as obese have sought professional advice to lose weight (Bish, et al., 2005). In fact, 42% created their own version of a diet or used a plan that they had read or heard about from another source (Paeratakul, York-Crowe et al., 2002). Over 90% of college students attempted to lose weight without a formal program and to diet using their own approach (Klesges et al., 1987). These statistics give a sense of the number of people attempting to lose weight through self-help methods rather than professionally led programs. Furthermore, they speak to the importance of practitioners gaining an understanding of various popular self-help approaches to weight loss. While there are many self-help groups that promote weight loss through dieting (e.g., Weight Watchers and Overeaters Anonymous), our discussion will focus on media-based self-help approaches, specifically self-help books.
Starker (1989) takes a historical look at the self-help diet and weight loss industry in the United States, which began nearly a century ago with the publication of Diet and Health With Key to the Calories in 1918. This book was based upon the new science of nutrition, appearing a year after the American Dietetic Association was formed. Although grounded in this new science, its author, physician Lulu Hunt Peters, wrote specifically to appeal to laypersons. As such, she included colorful case examples along with humorous anecdotes and drawings. This formative work, which achieved best-seller status from 1922 through 1926, was followed by Victor Lindlahr's 1940 You Are What You Eat, an expression still common in today's vernacular. In contrast to its predecessor, this book focused on maintaining a "balanced diet" incorporating the appropriate vitamins and minerals, rather than one based simply on calorie-counting. According to Starker (1989), this best-selling book was reprinted in paperback as late as 1971. Another noteworthy self-help text of this early era was dietician Adelle Davis' 1954 Let's Eat Right to Keep Fit. Here, Davis provided readers with detailed nutritional information, criticizing "the typical American diet, with its excess salt and sugar, excessive processing, and its contamination by pesticides" (as cited in Starker, 1989, p. 99). She forewarned that unless preventive steps were taken immediately, Americans could expect to experience, among other disorders, increased rates of cardiovascular disease and cancer. Her insight proved to be prophetic as this is the state that Americans find themselves in today—despite the subsequent surge in self-help titles surrounding these topics.
Self-actualization was the Zeitgeist of the 1960s and 1970s, giving rise to an increased demand for self-help products, particularly in the areas of diet and weight loss. As Starker (1989) explains, "Being slim, healthy, and beautiful, was quite necessary, after all, to self-fulfillment and personal growth" (p. 100). In 1961, Herman Taller (1961), an obstetrician-gynecologist, published Calories Don't Count, perhaps the first of the "miracle" diets, states Starker. The so-called magic of this diet was to encourage the body to burn its own stored fat by consuming polyunsaturated fats. Other physicians followed suit with their own "proven" approaches to weight loss. Dr. Stillman's (1968) program, The Doctor's Quick Weight Loss Diet, was a high protein diet without carbohydrates that reached best-seller proportions in 1968. Still highly popular in the 21st century, Dr. Atkins' Diet Revolution (Atkins, 1972) made its debut in 1972. This book prescribed an initial one-week elimination of carbohydrates for rapid weight loss, and only small amounts thereafter. The goal was for dieters to achieve a state of ketosis in order to burn their own fat. While carbohydrates were denied, dietary fats such as bacon and eggs were permissible. Other popular self-help books appearing during this period included The Save Your Life Diet by psychiatrist David Reuben (1975), based on high fiber intake, The Last Chance Diet by osteopath Robert Linn and author Sandra Lee Stuart (1976) based on liquid protein intake, The Scarsdale Diet by physician Herman Tarnower and author Samm Sinclair Baker (1982), based on high protein intake with reductions in carbohydrates and especially fats, and The Pritkin Program by inventor and heart-attack survivor Nathan Pritkin and author Patrick McGrady, Jr. (1982), based on large consumption of vegetables along with increased exercise.
Self-help books on dieting and weight loss became firmly entrenched in American culture during the 1960s and 1970s. However, the 1980s saw an unprecedented demand for these products. Brownell (1991a) notes that the amount of money spent on diet foods, programs, and books nearly doubled in the 1980s from previous decades. Among the 1980s' bestsellers were Richard Simmon's (1980) Never Say Diet and Jane Fonda's Workout Book (Fondar & Schapiro, 1981). Kissling (1995) notes that this era was marked by numerous celebrity guides to weight loss, with books authored by the likes of Brooke Shields, Victoria Principal, Elizabeth Taylor, and Cher, among others. Regardless of author, Starker (1989) cites a 1984 report that found over 600 self-help books on weight loss in print at the time.
Starker (1989) attributes this trend to the baby boomers' entry into midlife. He remarks that members of this generation "had not yet surrendered their youth, beauty, and physical prowess, and the sudden awareness of the transitory natures of these characteristics only drove them to greater extremes in attempts to retain them" (p. 138). Thus, they sought solutions in the myriad of self-help books on these topics. Starker also connects the increased purchase of these products to changes in the structure of the American family. Specifically, from 1950 to 1980, households with married couples decreased by 55% and single-person households doubled. Starker speculates that the drive for physical self-improvement was spurred by the desire to appear physically attractive to potential dating partners. Titles such as Thin Thighs in 30 Days (Sterling, 1982) and 30 Days to a Flatter Stomach for Women (Burstein, 1982) seem to confirm that weight loss efforts were undertaken for aesthetic, as well as health, reasons.
Brownell (1991b) offers further explanation as to why the self-help industry flourished as it did during the 1980s—particularly in the areas of dieting and weight loss. He states that "The concept of personal responsibility for health is deeply ingrained in our culture" (p. 304). Furthermore, this canon of personal responsibility is more pronounced in capitalistic societies with conservative governments in place. Thus, the 1980s marked a rejuvenation of this tenet in the United States. It also marked a time when cultural standards began to equate attractiveness, especially for women, with extreme thinness (Brownell, 1991a). As such, "This focus on individual responsibility reaches extremes in the search for the perfect body" (Brownell, 1991b, p. 304). Perini and Bayer (1995) concur that ours is a society based on the idea of self-determination in which "Americans have retained the ideal that we can achieve anything as long as we work hard enough for it" (p. 294). These authors note that while this is a concept rooted in masculinity, women in the 1980s began to assimilate the norms of traditional male culture, including individualism and its corollary, self-control. Perini and Bayer also stress that self-help diet books for women perpetuate the idea that weight loss is essential for aesthetic appeal, even if this message is couched in the rubric of health and fitness.
While published in 1989, Starker's treatise on self-help predicted that "such body-oriented issues as diet, nutrition, and exercise will remain popular topics for years to come" (p. 145). Indeed, he was correct. Selig-man (1994) speaks to the costs of the 54 million copies of diet books sold in the early 1990s. In 2002, Polivy and Herman note the perennial popularity of self-help books, with diet books outselling those addressing other health-related topics. A 2006 search on Amazon.com for books using the key words "diet" and "weight loss" resulted in 5,726 and 231 relevant results respectively. Included in today's top 10 advice books are The South
Beach Diet and The South Beach Diet Cookbook by Agatston (2003, 2004), a newcomer to the diet book genre titled French Women Don't Get Fat by Guiliano (2004), The Abs Diet by Zinczenko and Spiker (2004), and The Ultimate Weight Solution by television talk show host, Dr. Phil McGraw (2003). The Atkins diet, an instant best-seller upon publication in 1972, and other "low carb" diets are enormously popular as well. Dr. Atkins' diet empire is still strong even after his death in 2003, with Amazon.com listing multiple versions of his diet book, cookbooks, shopping guides, and even CDs.
In spite of the popularity of self-help diet books, there is relatively little empirical evidence of their efficacy for weight loss or weight maintenance, not to mention their safety. This leaves health professionals uninformed and, thus, unable to make sound recommendations as to their use (Dansinger, Gleason, Griffith, Selker, & Schaefer, 2005; Hill & Astrup, 2003). Consumers are equally uninformed, thus possibly incurring harm from following these prescriptions (Brownell & Rodin, 1994). In an effort to remediate this problematic state of affairs, the following section presents an overview of scholarly articles speaking to the effectiveness of popular diets.
Freedman, King, and Kennedy (2001) offer a characterization of popular diets, placing them into three distinct categories. The first is comprised of low-carbohydrate, high-fat and protein diets such as the Atkins, Protein Power, and Life Without Bread diets. The Sugar Busters! and Zone diets would also fall into this category, which we shall abbreviate as Low-Cho diets. The rationale behind these diets is that by greatly decreasing carbohydrate intake, insulin release will be decreased, resulting in greater satiety and weight loss through adipose metabolism, along with a decrease in serum triglyceride levels.
On the opposite end of the spectrum, Freedman et al. (2001) describe very high-carbohydrate, low- to very low-fat, and moderate protein diets, which we shall abbreviate as Low-Fat diets. These are exemplified by the Ornish and Pritikin programs. According to Freedman et al. (2001), these diets have traditionally been thought of more as a prescription to prevent cardiovascular disease rather than as a weight-loss approach for the general public. However, capitalizing on Americans' exponential weight gain, some authors of books promoting these diets have changed both the focus and the title to emphasize weight loss rather than heart disease reduction. These Low-Fat diets recommend eating high fiber and complex carbohydrates until one feels satiated. The fat intake is defined as <10% of total calories for very low-fat diets and 11-19% fat of total calories for low-fat diets. The rationale for these diets is that individuals will eat fewer calories, thus losing weight and body fat.
Lastly, Freedman et al. (2001) describe balanced nutrient reduction diets that are moderate in fat and protein intake while high in carbohydrate intake. We shall designate these as Bal diets, which are exemplified by the Weight Watchers, Jenny Craig, Nutri-Systems, DASH, and USDA Food Guide Pyramid diets. The LEARN Program for Weight Maintenance (Brownell, 2004) also utilizes this dietary approach and recommends that physical activity accompany these nutritional changes. Beyond these recommendations, the LEARN program also employs cognitive-behavioral principles for weight management including self-monitoring, reinforcement, and restructuring of negative thinking patterns. The rationale for these diets is that weight loss will occur when the body is in a state of negative energy balance produced through calorie reduction and concomitant increase in physical activity. The goal of these Bal diets "is to provide the greatest range of food choices to the consumer, to allow for nutritional adequacy and compliance, while still resulting in a slow but steady rate of weight loss (e.g., 1 to 2 lbs/wk)" states Freedman et al. (2001, p. 20S). Finally, Freedman et al. note that Bal diets, particularly in contrast to Low-Cho diets, are generally based on sound scientific principles and have been subjected to the most empirical scrutiny to date.
In the remainder of this section, articles will be reviewed that speak to the efficacy of diets within each of these three categories, either singly or in contrast to each other. Articles reviewed are of three types. The first consists of a single article (Anderson, Konz, & Jenkins, 2000) that predicts outcomes based on a nonclinical computer analysis of dietary protocols. The second consists of literature reviews (Bravata et al., 2003; Freedman et al., 2001; Katz, 2005) and the third consists of treatment outcome studies (Bacon et al., 2002; Dansinger et al., 2005; Gardner et al., 2007).
Computer Analysis In lieu of direct administration of self-help dietary approaches to human participants, Anderson et al. (2000) opted to simulate these diets by entering suggested menu plans into a computer analysis. This analysis aimed to yield information on how closely each diet adhered to USDA Food Pyramid recommendations and to what extent each diet might impact coronary heart disease risk factors. These researchers chose to examine eight popular weight loss diets. Four of these fall into Freed-man et al.'s Low-Cho diet category (i.e., the Atkins, Protein Power, Sugar Busters!, and Zone diets). Three fall into the Low-Fat diet category (i.e., the Pritkin and Ornish diets, along with Dr. Anderson's High Fiber Fitness Plan). Lastly, Anderson et al. (2000) examined the American Dietetic Association's Exchange Diet which might best be described as a Bal diet.
The researchers began by creating menus based on each diet author's recommendations, holding menus at a 2000 calorie limit. Upon analysis, they found that the Atkins diet was the highest in fat, saturated fat, and cholesterol and scored the poorest when compared to the USDA Food Pyramid. Not surprisingly, the Atkins, followed by the Protein Power, diet deviated the most from USDA Food Pyramid guidelines in terms of recommended servings of grains, vegetables, and fruits. Indeed, the Atkins diet stresses the avoidance of such foods. In contrast, the Ornish and Prit-kin diets "most strongly encourage 'eating at the bottom' of the food pyramid" (Anderson et al., p. 584).
As for effects on cardiovascular risk factors, the Atkins diet was the worst and the Ornish diet was the best. The Atkins diet was implicated in increasing risk by virtue of the effects of low soluble fiber levels that would also increase serum cholesterol levels—the only diet in this analysis that contributed to cardiovascular risk for this reason. Based on their calculations, Anderson et al. (2000) state that "long-term use of the Atkins diet would increase serum cholesterol values by ~25%, while long-term use of the Ornish diet would decrease serum cholesterol concentrations by ~32%"(p. 586). They conclude that long-term use of the Ornish diet might decrease risk of heart disease by greater than 60% while long-term use of the Atkins diet might actually increase risk of heart disease by greater than 50%. Unfortunately, many consumers are seduced by the relatively rapid weight loss produced by the Atkins and other Low-Cho diets—weight loss that Anderson et al. (2000) contend is largely a function of water loss rather than fat. Anderson et al. conclude that the potential long-term hazards of Low-Cho diets outweigh any short-term benefits, thus obviating their use as a means of enhancing health. Based on their findings, this group instead recommends diets lower in fat and higher in both carbohydrates and fiber as the best means of reducing cardiovascular disease risk.
Literature Reviews Freedman's group (Freedman et al., 2001) drew similar conclusions based on their review of literature dating from the 1960s. These researchers compared the claims made by popular diets to empirical studies that address the efficacy of various dietary approaches. They found that overweight people on Low-Cho diets eat less and lose weight when allowed to eat ad libitum. Like Anderson and colleagues (2000), Freedman et al. (2001) determined that Low-Cho diets produced weight reductions through a greater loss of body water than fat. However, at the conclusion of the diet, this water weight would be regained. This point is particularly salient given the possibility that individuals may not adhere to this dietary protocol in the long-term. Freedman et al. (2001) found that very few studies lasted long enough to truly address compliance, but dropouts in many of the studies were reported without explanation. They also found that, despite claims by Low-Cho diet proponents that caloric intake is inconsequential, this is not the case. Caloric balance is indeed the chief determinant of weight loss regardless of the diet's composition. Freedman et al. (2001) also warned of adverse effects associated with Low-Cho diets. These include ketosis as well as potential for increased cancer risk due to the lowered intake of fruits and vegetables. Low-Cho diets may also result in a host of side effects including constipation, diarrhea, nausea, fatigue, headache, insomnia, thirst, and halitosis. A final drawback to these diets is that they require supplementation with vitamins, specifically A, E, and B vitamins, as well as minerals and dietary fiber because of the lack of food variety, specifically fruits and vegetables.
In their review, Freedman et al. (2001) also assessed the worth of Low-Fat diets. As with Low-Cho diets, individuals adhering to these lose weight due to caloric reductions. However, in contrast to Low-Cho diets, weight reductions here are the result of a loss of body fat rather than water. As in Anderson et al.'s (2000) computer-based study, Freedman and colleagues (2001) found cardiovascular benefits associated with this dietary composition. These include a decrease in low-density lipoproteins and, in some cases, a decrease in plasma triglyceride levels and a reduction in blood pressure. Freedman et al. (2001) found few adverse effects for Low-Fat diets but noted that the American Heart Association recommends that persons with insulin-dependent diabetes mellitus avoid very low-fat diets. Another concern with these diets is that they require supplementing with vitamins E and B12 as well as zinc due to the limited animal protein sources. Individuals on Low-Fat diets typically reported having more than enough to eat and ate less when allowed to eat ad libitum. The high levels of dietary fiber likely contribute to feelings of satiety and thus may aid in compliance. One caveat, though, is that many of the studies of Low-Fat diets involved clinical populations who the authors admit may be more motivated to follow these relatively stringent protocols than members of the general population.
Finally, Freedman et al. (2001) spoke to Bal diets, which produce weight loss because, by reducing fat, a greater proportion of calories are removed. Specifically, "when dietary fat decreases from 34-36% to less than 30%, caloric intake significantly decreases and results in significant body weight reduction" (p. 21S). Unlike Low-Cho and Low-Fat diets, Bal diets are nutritionally sound with all food groups included at some level. Health-wise, improvements are seen in blood lipid levels and blood pressure, and no adverse effects were detected. Individuals on these diets generally reported being satiated and rated the diet highly in terms of palatability, although, once again, long-term compliance for the general public has yet to be assessed. Freedman et al. (2001) conclude that the best dietary approach for preventing weight gain, accomplishing weight loss, and maintaining this loss is one that is moderate in fat and low in calories. Such a diet—perhaps more aptly called a lifestyle—would incorporate a large amount of fruits, vegetables, and whole grains as well as low-fat dairy products. In concert with Anderson et al.'s (2000) conclusions, this diet would do best to reduce the risk of chronic disease. Finally, like Anderson's group, Freedman and colleagues (2001) discourage the use of Low-Cho diets, which, unfortunately, the general public has seized upon in recent years.
In a subsequent literature review, again dating from the 1960s, Katz (2005) reached many of the same conclusions as did Freedman et al. (2001) regarding the efficacy—and potential hazards—of various popular dietary approaches. That is, the evidence he gathered indicates that sustained weight loss is accomplished via caloric restriction, rather than dietary composition itself. While Low-Cho diets produce notable initial weight loss, this is attributable to loss of water and muscle protein, both undesirable effects. Katz (2005) also determined that "the more rapid the initial weight loss, in general the greater and more rapid the subsequent weight gain" (p. 66). That is, the methods used to achieve rapid initial weight loss are typically unsustainable. Supporting this statement, Katz noted high attrition and recidivism rates in studies of Low-Cho diets.
Additionally, Katz (2005) discussed the adverse side effects of Low-Cho diets, classifying them in terms of problems associated with high fat intake (e.g., worsened serum cholesterol levels), high protein intake (e.g., worsened renal functioning), and low carbohydrate intake. As to this last component, Katz (2005) listed 11 known ill effects of inadequate carbohydrate intake ranging from depression to increased cancer and cardiovascular disease risk. Thus, these diets appear damaging on many fronts. Katz (2005) concluded that more research must be done to identify a sustainable dietary approach that promotes both health and weight control. However, for the time being, he avers that "diets rich in fruits, vegetables, and whole grains; restricted in animal fats and trans fat from processed foods; limited in refined starches and sugar; providing protein principally from lean sources; and offering fat principally in the form of monounsaturated and polyunsaturated oils are linked to good health" (p. 74). Finally, Katz was equally adamant about the short-comings and ill effects of Low-Cho diets that the populace seems to prefer.
Bravata and colleagues (2003) reviewed the literature on Low-Cho diets, searching for empirical articles testing their effects between 1966 and 2003. These researchers also determined that weight loss was a result of caloric restriction as well as duration of the diet, rather than carbohydrate restriction per se. In contrast to Freedman et al. (2001) and Katz's (2005) literature reviews, Bravata's examination of empirical evaluations of Low-Cho diets failed to yield notable ill effects on health status. These researchers state "lower-carbohydrate diets were not associated with adverse effects on serum lipid levels, fasting serum glucose levels, or blood pressure" (p. 1847). Nevertheless, Bravata et al. (2003) stopped short of endorsing the widespread use of these diets. Rather, they enumerated a number of methodological weaknesses in the studies they reviewed.
First, they comment that relatively few studies reported on the metabolic variables on which they found no ill effects. In addition, these studies generally lacked any long-term follow-up, thus they were unable to determine both efficacy and harm that might be incurred over time. Bravata et al. (2003) also observe that most articles reported results only for participants who had completed the study, thus effects of the dietary interventions may have been overstated. On a related note, adherence to dietary protocols was often not assessed even among participants who completed the studies. For the most part, studies failed to include measures of exercise behavior, thus the impact of any changes in physical activity could not be assessed. Finally, Bravata et al. (2003) make an excellent point in light of obesity patterns in the United States. That is, information on race and ethnicity of participants in the studies reviewed was often lacking. Presumably, the majority of participants in these studies were White, non-Hispanic Americans, thus limiting the extent to which findings could be generalized to diverse groups in this country and abroad. Freedman et al. (2001) voice a similar concern, noting that the existing studies primarily examined dietary interventions among adults, with little research examining their effects on children and adolescents. This is worrisome given the increasing numbers of children and adolescents who embark on diets to lose weight.
Treatment-Outcome Studies Because of the paucity of data on increasingly popular diets, Dansinger et al. (2005) set out to empirically evaluate the effects of the Atkins, Zone, Weight Watchers, and Ornish on weight loss and cardiovascular risk reduction, as well as adherence. The first two of these diets are of the Low-Cho variety while the Weight Watchers reflects the Bal diets, and the Ornish is representative of the Low-Fat diets. Participants were adult men and women of any age who were between 27 and 42 BMI and had at least one cardiovascular disease risk factor. Participants were randomly assigned to the four dietary conditions, which then met as a group for one hour on four occasions during the 2 months of active intervention. Each group was led by a dietician and physician who presented the rationale for the diet along with written materials and the official diet cookbook. They also provided diet-specific advice, reinforced positive dietary changes, and addressed barriers to adherence. Participants also received monthly telephone calls inquiring about adherence and assessing changes in medications, hospitalizations, and any adverse effects. Thus, the intervention in this study most closely resembled self-help as an adjunct to therapy rather than a pure self-help intervention. Individuals enrolled in this study received no monetary compensation for their participation.
Dependent variables were assessed at pretest, posttest (2 months), and follow-up (6 and 12 months). The researchers found that, in a year's time, all four diets resulted in a modest weight loss, with no significant difference among them. At year's end, all four diets also produced modest improvements in some, but not all, cardiovascular risk factors with some variation among the diets. Furthermore, no diet resulted in a significant worsening of any risk factor over the course of the year. However, adherence to dietary protocols did seem to be a problem, particularly for those assigned to the Ornish and Atkins diets, which only 50 and 53% of participants, respectively, completed. Attrition was still high among the other groups, with only 65% of participants in both the Zone and Weight Watchers diets completing the study. The two most common explanations for discontinuation was that the diet was too difficult to follow or that it was not meeting participants' expectations in terms of amount of weight lost. Even among those who completed the study, self-monitoring records revealed a clinically meaningful adherence level among only 25% of participants in each condition, with adherence diminishing as time progressed. Noteworthy is the fact that weight loss was significantly related to adherence levels in this study. Thus, the authors concluded that dietary adherence may be more salient than the diet itself in terms of producing weight loss and improvements in cardiovascular risk. While no adverse effects for any diet were encountered, Dansigner et al. (2005) did caution that their study was limited in its ability to elucidate long-term safety risks.
More recently, Gardner et al. (2007) conducted a similar study comparing four diets, the Atkins, Zone, and Ornish diets, but using the LEARN program, another Bal diet, in place of the Weight Watchers diet that Dansinger et al. (2005) had examined. This study also spanned 12 months, but participants here were restricted to premenopausal women ages 25 to 50 with BMIs between 27 and 40. Participants were randomized in blocks and assigned to one of the four experimental conditions. Again, the intervention more closely resembled self-help as an adjunct to therapy as participants attended one-hour classes once per week for 8 weeks. Classes were led by a dietician who covered approximately one eighth of the material in each self-help diet book during each meeting. Although the LEARN program is meant to be navigated across 16 weeks, it was covered in this accelerated time frame to match the intervention span of the other three groups. Additional contact with participants included e-mail and telephone prompts for appointments and additional contact throughout the study. In contrast to Dansinger et al., (2005) Gardner et al. (2007) offered participants increasing monetary incentives for completing assessments at 2, 6, and 12 months.
Gardner et al. (2007) aimed to assess the effects of these interventions on weight loss, which they dubbed their primary outcome. They also aimed to assess their effects on 11 metabolic measures, which they referred to as secondary variables. As in the Dansinger et al. (2005) study, all four diets assessed here resulted in weight loss over the course of a year. At 2- and 6-month assessments, weight loss for the Atkins diet was significantly greater than that for all other groups. This is consistent with reports of rapid initial weight loss on Low-Cho diets. However, by 12 months, the weight loss produced by the Atkins diet was significantly better than only the Zone diet. The weight loss produced by all diets was modest at best, with the Atkins resulting in the greatest loss, an average of 4.7 kg after one year. Modest improvements were also detected on the metabolic measures after one year, with the Atkins diet faring significantly better than the Zone in terms of BMI and triglycerides, better than the Ornish in terms of HDL and diastolic blood pressure, and better than all three other diets in terms of systolic blood pressure. In contrast to the Dansinger et al. (2005) study, far fewer individuals discontinued their participation throughout the course of Gardner et al.'s (2007) intervention. In all four diet groups, over 85% of participants attended over 75% of their assigned classes and retention at 12 months ranged from 76 to 88%, with no significant differences among groups. The authors add, however, that "adherence to the 4 sets of dietary guidelines varied within each treatment group and waned over time, especially for the Atkins and Ornish diets" (p. 976).
Gardner et al. (2007) concluded that concerns about the adverse effects of the Atkins diet were not confirmed, at least within the 12 months that their study entailed. They conceded that questions remain as to the source of weight loss—i.e., was it attributable to the composition of the diet itself, and if so, what dietary component? They acknowledged that questions about the long-term effects of these diets remain. While the authors extolled the external validity of their findings, one must remember that self-help diet books were used as an adjunct to therapy in a research setting where individuals were prompted and paid for their participation. The monetary incentives in this study might account for the much greater retention rates than those in Dansinger et al.'s (2005) study, where no such incentives existed. Such conditions vastly differ from those in which millions of Americans purchase these books and attempt to negotiate them on their own.
A final empirical study to be examined here is one that compared use of the LEARN Program for Weight Control manual to use of a manual espousing a nondieting approach to healthy eating, nutrition, physical activity, social support, and body acceptance (Bacon et al., 2002). Participants were obese (BMI > 30) women ages 30-45 years who had a chronic history of dieting. They were randomly assigned to either the dieting or nondieting condition, each of which entailed attending 24 weekly 90-minute sessions. Following this active treatment phase, participants could attend monthly aftercare sessions for 6 more months in which no new material was presented. The dieting group was led by a dietician, whereas the nondieting group was led by a counselor. Once again, the interventions resembled self-help as an adjunct to therapy rather than pure self-help.
Results revealed that only in the dieting group did significant weight loss occur, most of this happening by mid-treatment with no significant reductions thereafter. A significant reduction in BMI also occurred for the dieting group. In stark contrast, the nondieting group experienced virtually no change in these variables across time. However, both groups significantly improved in total cholesterol, LDL, triglycerides, and systolic blood pressure at 12 months. These changes occurred at mid-treatment for the dieting group and following aftercare for the nondieting group. Unfortunately, HDL values significantly worsened for both groups, more so for the nondieting group. Bacon et al. (2002) measured activity levels throughout their study, finding that there was a significant increase in energy expenditure for the nondieting group at 12 months, contrasting with a significant decrease for the dieting group.
Unlike many other studies assessing the effects of dietary approaches, Bacon et al.'s (2002) study included an assessment of psychological variables: eating disorder pathology, depression, and self-esteem. On the eating disorder measures, both groups improved significantly and similarly over time on some subscales, but the nondieting group significantly improved on a greater number, with the dieting group actually faring worse over time on cognitive restraint. The researchers, however, framed this as a successful outcome because this is a quality that dieting approaches attempt to imbue. Although both groups improved over time, the nondieting group exhibited significantly better scores on body image avoidance and disinhibiton than the dieting group. Furthermore, the nondieting group significantly improved over time and had significantly better scores on rigid control, whereas the dieting group's initial significant improvements disappeared by year's end. Both groups improved on depression and did not differ from each other on this variable. Regarding self-esteem, the nondieting group made a significant improvement although this did not occur until the 12-month assessment. The dieting group exhibited a significant improvement at the end of the 6-month treatment program, but this effect was not sustained at 12 months.
Last, but certainly not least, Bacon et al. (2002) assessed attrition levels throughout their study, finding significant differences between the two groups. While 33 and 41% of participants in the dieting group dropped out of the study at mid- and posttreatment, respectively, only 8% of participants dropped out of the nondieting group at mid-treatment, with no additional dropouts at posttreatment. Among those who remained in the study, significantly fewer participants (67%) in the dieting group attended sessions during the latter half of treatment than participants (76%) in the nondieting group. Similar proportions of participants did participate in the optional aftercare sessions, 50 and 53% for the dieting and nondieting groups, respectively. Participants also had the opportunity to complete a self-evaluation questionnaire. The researchers reported striking differences between the groups on the items, "I feel like I have failed the program" and "This program has helped me feel better about myself," with 38 and 35% of the dieting group endorsing the "failure" item at mid- treatment and 12-months, respectively, compared to 5 and 7% of the nondieting group at these junctures. On the "feel better" item, 51 and 78% of the dieting group endorsed this item at mid-treatment and 12 months compared to 93% of the nondieting group at both time periods.
Based on their findings, Bacon et al. (2002) urge practitioners to consider recommending nondieting interventions, at least to their larger women clients who have a history of failed dieting attempts. In this study, such an approach resulted in similar improvements in metabolic fitness variables as did a dieting approach. Furthermore, the nondieting approach resulted in greater improvements in psychological health, fewer attributions of failure, and seemingly related higher levels of adherence.
Generally speaking, self-help programs are less costly than professionally led psychological and medical interventions. This advantage is particularly meaningful for persons attempting to lose weight as most insurance programs do not cover weight-loss treatments (Downey, 2002). Furthermore, weight is inversely correlated with income in the United States. Thus, individuals classified as obese or overweight are likely to have fewer economic resources at their disposal. Fabricatore, Wadden, and Foster (2005) recognize that people of color are less likely to access health care even though they have higher rates of obesity and overweight than Caucasians in the United States. This same point can be made about American women, who have greater rates of obesity than their male counterparts but who have fewer financial means for treatment.
The availability of self-help resources in this area may also be advantageous as they eliminate the stigma of presenting oneself to a professional for assistance. Indeed, Fabricatore et al. (2005) report that health care professionals' attitudes toward obese patients are as negative today as they were 40 years ago. However, present-day practitioners are less likely to express this bias in an explicit manner. Thus, these authors suggest that patients' reluctance to seek formal weight loss treatment may stem from factors other than overt hostility on the part of practitioners. For instance, patients may encounter a more diffuse lack of empathy. In addition, they may be advised that their weight is problematic (in terms such as "fat" and "obese," which evoke negative emotional reactions) but offered little guidance for remedying this situation. Non-interpersonal elements of the health care setting may also deter persons from seeking professional care such as when chairs and medical equipment are too small for the patient or when scales are in plain view of other patients. All of these elements may conspire to evoke shame and embarrassment among persons of weight, leading them to seek private solutions for weight loss among the countless self-help books on this topic.
While the first chapter in this text lists enhanced self-efficacy among the psychological benefits of self-help therapies, Polivy and Herman (2002) suggest that this relationship is somewhat muddled when it comes to dieting and weight loss. Self-efficacy refers to confidence that one can successfully manage a task—in this case, lose weight. Polivy and Herman (2002) note that such efficacy, or confidence, is often based on a record of success or competence—in other words, some performance accomplishment. Bandura (1977) defined self-efficacy as the conviction that one can successfully execute the behavior required to produce the outcomes. Herein lies the problem with respect to weight loss. Polivy and Herman note that the "pernicious aspect of dieting is not that diets eventually fail, but that they do not fail before first succeeding" (p. 683). Indeed, Katz's (2005) literature review indicated that any diet that restricts calories will produce weight loss in the short term. However, the means used to achieve this initial weight loss is intrinsically unsustainable, with weight gain soon to follow. Low-Cho diets are the epitome of this problem because they tend to produce the largest initial weight losses, and they appear to be the least sustainable, perhaps due to their host of side effects.
Polivy and Herman lament that such initial success leads dieters to develop an unrealistic sense of self-efficacy. That is, if they keep trying—or try harder the next time—they will indeed achieve substantial and permanent weight loss. They speculate that such individuals persist in their dieting efforts because they attribute their inevitable failure to themselves rather than the inadequacy of the diet. Bacon et al.'s (2002) self-evaluation findings, reported in the previous section, poignantly illustrate this effect. Polivy and Herman suggest that if dieters do invoke an external attribution, it may be that this particular diet was "not for me" (p. 683), so they move on to the next self-help program that purportedly holds the key to lasting weight loss. Dieters certainly do not develop such notions in a social vacuum. Polivy and Herman assert that the diet industry thrives due to repeat customers. That is, "promoters of the diet in question have a vested interest in blaming the dieter rather than the diet" (p. 681). Dieters who fail to achieve (and maintain) the unrealistic weight loss that these programs promise—and the generalized happiness that such weight loss will certainly bring—can redeem themselves simply by trying harder next time.
The diet industry certainly seems to be one in which capitalistic motives usurp those of a therapeutic nature. Seligman (1994) accuses the industry with creating a public that is "discontent, even despairing, about their bodies, and willing—even eager—to spend a substantial portion of their earnings in the belief that they can and should become much thinner than they are" (p. 180). Wilson (2003) echoes this sentiment, stating that advertising strategies are devised to foster insecurity and inadequacy among recipients. This, of course, is followed by the unwavering promise that purchasing a certain product will ameliorate this distress. Seligman (1994) vehemently calls for an end to this practice, instigating for truth in advertising. For instance, he maintains that self-help diet books should disclose that long-term weight loss is an unlikely outcome. Instead, Brownell (1991b) observes that books make claims like "Lose Up to a Pound a Day and Never Gain it Back" (p. 305). Freedman et al. (2001) list "Not a single adverse reaction" as one of the "outrageous" claims found in Bantam Book's Protein Power and "Sugar is toxic!" (p. 40S) as an outrageous statement contained in Ballantine Book's Sugar Busters! Seligman (1994) goes so far as to say that those who author and promote these books are approaching a violation of the profession's "do no harm" oath (p. 197). Rosen, Glasgow, and Moore (2003) have long insisted that ineffective self-help programs are not necessarily benign—that these can actually lead to worsening of the problem. Katz (2003) warns that fad diets should be "presumed guilty" (p. 33) regarding their impact on health and that it is the responsibility of their authors to prove their program's efficacy, including lack of harm.
Unfortunately, numerous self-help diet books are created by individuals who are not even bound by professional ethics in the health-related fields. Starker (1989) states that "Anyone—even persons without an iota of nutritional training can design, develop, publish, and promote a diet ... all it takes is an idea and the ability to string some words together" (p. 4).
Kissling (1995) is particularly critical of self-help books on dieting and weight loss authored by celebrities who have no qualified expertise in this area. As such, messages and recommendations contained in these books may run counter to sound scientific evidence. For instance, these books often categorize foods as either good or bad—even "forbidden" (p. 215). In turn, ingestion of such foods is seen as abusive, weak, and/or sinful. Readers are counseled to be hypervigilant, lest they experience lapses in their efforts. Kissling (1995) quotes one book as warning "a dieter must be on National Guard duty twenty-four hours a day" (p. 214). Putterman and Linden (2004) cite a body of research linking this approach, known as dietary restraint or restrained eating, to a number of deleterious physical and psychological states. In fact, they acknowledge that dietary restraint may be etiologically involved in the onset of eating disorders.
"Looks do count" is another common theme that runs throughout celebrity-authored self-help books (Kissling, 1995, p. 213). That is, weight loss is to be pursued for aesthetic purposes, thus reinforcing unrealistic societal beauty standards and the idea that self-worth is necessarily contingent on body weight and shape, particularly for women. In their own examination of self-help books, Perini and Bayer (1995) encountered this same theme espoused by authors, even those of the non-celebrity variety. This focus on appearance also extended to muscularity in that books encouraged women to engage in strength training "because muscles are sexy!" as one book proclaimed (p. 300). Putterman and Linden (2004) produced empirical evidence showing that when weight loss is driven by appearance versus health concerns, female dieters report greater use of unhealthy practices (e.g., eliminating whole food groups) and a greater number of the aforementioned lapses in restraint. In addition, women participants who dieted to improve their appearance, rather than their health, experienced greater body dissatisfaction and lower self-esteem. Putterman and Linden (2004) further remark that drastic approaches to weight loss are also the least likely to work, setting dieters up to try ever harder, only to fail again and again.
Despite Kissling's (1995) critique of celebrity-authored self-help books, works continue to be published that rehash some of these same troubling themes. For instance, the back cover of talk show host Dr. Phil McGraw's The Ultimate Weight Loss Solution (2003) presents exaggerated claims in that it promises "permanent weight loss" with "immediate results." Messages about weight are moralistic in tone as exemplified by pejorative terms such as "fatties" and "couch potatoes." While the book touts good health as a consequence of weight loss, it simultaneously conveys the importance of weight loss for physical appearance; e.g., "You'll love what you start seeing in the mirror" (p. 207). It also places the blame for failure to lose weight squarely on the shoulders of the individual. McGraw (2003) refers to the statement "Because obesity runs in my family, I just can't lose weight" as a "handy excuse or justification for not losing weight" (p. 73) rather than acknowledging the empirically documented genetic limitations to weight loss. Along these lines, McGraw (2003) makes the statement, "Overweight people simply eat more calories, more fat, and more carbohydrates than normal-weight people do" (p. 181), which clearly contradicts scientific evidence regarding the influence of genetics on energy expenditure regulation, lipid metabolism, and other elements of weight and weight loss (Moreno-Aliaga, Santos, Marti, & Martínez, 2005).
An inherent limitation of the self-help genre is that it fails to address etiological factors beyond the individual's own psychological makeup. The self-help literature on dieting and weight loss presents perhaps the most glaring example of this critique. In this case, individuals must contend with both biological and sociocultural barriers to weight loss. In his aptly titled article, "Dieting and the Search for the Perfect Body: Where Physiology and Culture Collide," Brownell (1991a) cites early evidence for genetic influences on weight—which consequently places confines on the extent of weight loss one might realistically achieve. Research over the ensuing decades has reinforced this contention (e.g., Moreno-Aliaga et al., 2005). That is, the human body is not infinitely malleable because of biological forces beyond one's control. Brownell and colleagues (e.g., Horgen & Brownell, 2002) have also spoken of sociocultural barriers to weight loss—what they term, a "toxic environment" in which high-fat, high-calorie food is inexpensive, easily accessible, and heavily advertised. These authors raise the possibility that obesity is actually a normal response to this abnormal environment.
Despite the recognition that biological and sociocultural factors influence individuals' body weight and shape, Brownell (1991b) states that "Efforts on the management of obesity have focused almost entirely on changes the individual can make to lose weight" (p. 306). The focus on individual responsibility exemplifies the duty model of health care articulated by Winkler (1986) in which persons are expected to engage in positive health behaviors, with their resultant health status in their own hands. This lies in contrast to the rights model in which the broader social system is expected to create policies and environments which facilitate individuals' health status. Influenced by Winkler's (1986) writings, McFadden and Evans (1998) examined a random sample of articles describing obesity interventions from four leading behavior therapy and/or clinical psychology journals seeking to what extent these addressed macrolevel (e.g., cultural processes and social constructions) relative to microlevel (e.g., individual knowledge level and behavior) issues. As predicted, these articles largely attributed excess weight to individual behavior, and the interventions described therein reflected this perspective. Like Brownell
(1991b), McFadden and Evans (1998) deemed this problematic. As such, they advised against an overreliance on the duty model of health, which dictates that individuals have considerable control over their well-being. Such an emphasis may result in victim-blaming rather than recognizing the many social constraints on behavior change. That is, a bias toward personal responsibility may obscure the sociocultural factors that hamper individual choice. Evans (2005) subsequently stated that the target for treatment should never be the individual behavior alone. Rather, he contends that the individual must be considered in cultural context.
In his article "Personal Responsibility and Control Over Our Bodies: When Expectations Exceed Reality," Brownell (1991b) also warns against victim-blaming in the area of dieting and weight loss. McLellan (1995) argues that self-help books on this and other topics inherently do just that. Therefore, if weight loss is not achieved, the reader is at fault. McLellan (1995) laments that placing the onus solely on the individual does a major disservice, especially to those whose condition is a function of social oppression. Likewise, Evans (2005) remarks that some populations experience double jeopardy when their problems are a function of historical injustice and exploitation, but the proposed solutions rest solely upon individual behavior change. Kissling (1995) contends that self-help books that equate fitness, beauty, and self-esteem certainly contribute to "a climate of oppression by encouraging women to participate in the objectification of their own and other bodies" (p. 215). Perini and Bayer (1995) speak to this issue as well, stating that "The body is first and foremost to be understood as an object which is to be looked at and desired" (p. 299), with this theme rampant among the self-help books that these authors reviewed.
The language of self-blame is often incorporated in self-help books for diet and weight loss, as evidenced by the use of phrases such as letting one's self go (Kissling, 1995). Modern-day self-help books may not invoke overtly pejorative terms such as those found in books written in the early 1900s. For instance, Starker (1989) describes one book that used female case studies who were given "humorous" names such as "Ima Gobbler, Mrs. Sheesasite, Mrs.Weyaton, and Mrs. Knott Little" (p. 95). Nevertheless, the vocabulary used in more contemporary self-help guides still communicates that "the non-ideal female body is sick, lost, and disordered" (Perini & Bayer, 1995, p. 297). According to these authors, female readers are prompted to identify themselves as the problem in need of changing rather than question the sociocultural context in which they live.
Earlier in this chapter, we spoke of the cultural canon of personal responsibility as fueling the popularity of self-help books, with women increasingly buying into this notion, especially in the area of bodily transformation. Perini and Bayer (1995) are critical of this conception of the self-made man as it only applies to White men of economic means in our society. In essence, women, especially those of color and lower socioeconomic status, who embrace this Protestant ethic face greater societal barriers in their quest for success and are more likely to fall short in their efforts. When applied to overweight and obesity, this ideology sees these conditions as resulting from a lack of self-discipline, hard work, and moral character rather than resulting from dysfunctional social systems. Working from this presumption, Quinn and Crocker (1999) set out to explore the relationship between Protestant ethic ideology and psychological well-being for women who identified as "normal" weight or "overweight." In both correlational and experimental studies, these researchers found that self-perceived overweight women who embraced the Protestant ethic exhibited the poorest psychological well-being. Thus, messages about personal control espoused by the dieting self-help industry may propagate psychological distress and disordered attempts to lose weight.
Was this article helpful?