Introduction

Bariatric surgery appears to be the only effective long-term treatment for extreme obesity (defined as a BMI 40 kg/m²). Pharmacotherapy (e.g., orlistat, sibutramine) and lifestyle modification typically induce losses of 8–10% initial body weight (1)(2)(3)(4)(5)(6). Weight losses of this magnitude may ameliorate weight-related health and psychosocial difficulties experienced by some overweight or obese individuals but may not be sufficient to improve the health and quality of life of persons with extreme obesity. Furthermore, weight regain after discontinuation of these behavioral and pharmacological treatments is often considerable (5)(6)(7). Consequently, bariatric surgery has become a more popular and viable treatment option for persons with a BMI 40 kg/m² (or those with a BMI 35 kg/m² in the presence of significant obesity-related comorbidities). The number of individuals undergoing bariatric surgery in the United States has swelled from 16,000 in the early 1990s to 103,000 in 2003 (8).

In 1992, the NIH Consensus Development Conference on Gastrointestinal Surgery for Severe Obesity provided guidelines on identifying appropriate candidates for bariatric surgery. In addition to meeting the BMI criteria, the Consensus Development Conference indicated that bariatric surgery may be appropriate for "well-informed and motivated patients with acceptable operative risks" and that individuals "seeking therapy for severe obesity for the first time should be considered for treatment in a non-surgical program" (9). In 2005, a panel of experts published a revised consensus statement that noted that "bariatric surgery candidates should have attempted to lose weight by non-operative means, including self-directed dieting, nutritional counseling, and commercial and hospital-based weight loss programs, but should not be required to have completed formal non-operative obesity therapy as a precondition for the operation" (10). One common interpretation of both of these statements is that patients should have a history of failed attempts with more conservative weight loss interventions before considering bariatric surgery.

Little, however, is known about the dieting history of bariatric surgery candidates. Do most patients have a track record of failed weight loss efforts and turn to surgery as an appropriate last resort? Alternatively, has the increasing popularity of bariatric surgery for extreme obesity been fueled, in part, by individuals with relatively little dieting experience? Information on the dieting history of bariatric surgery patients takes on practical importance as an increasing number of insurance companies currently require patients to provide a documented history of failed weight loss attempts before issuing coverage for bariatric surgery. The present study was undertaken with these issues in mind and with the goal of describing the previous weight loss experiences of a sample of 177 bariatric surgery candidates.

Research Methods and Procedures

Participants

Participants were 138 (78% ) women and 39 (22% ) men (total N = 177) with extreme obesity who sought bariatric surgery at the Hospital of the University of Pennsylvania between May and September 2004. The majority (69.3% ) of participants were European American, 23.9% were African American, and 5.1% were Hispanic. Three described their ethnicity as other. Participants had a mean weight of 140.0 28.2 kg, height of 166.2 9.5 cm, and BMI of 50.5 8.2 kg/m². They reported a mean of 13.6 2.3 years of education. A sizable minority (44.6% ) reported that they were married; 36.7% were single, 12.4% divorced, and 6.2% separated or widowed. These demographic characteristics are similar to those of participants in other studies of this population at this institution (11).

Procedure

All individuals seeking bariatric surgery at the Hospital of the University of Pennsylvania complete a behavioral/psychological evaluation with a mental health professional at the University's Weight and Eating Disorders Program. This evaluation is undertaken to identify any potential psychiatric or behavioral contraindications to surgery (12). In preparation for the evaluation, candidates complete the Beck Depression Inventory-II (13) to assess mood and the Weight and Lifestyle Inventory (WALI)¹ (14). The WALI is a 20-page questionnaire that covers a variety of topics including family history of obesity, eating and activity habits, and psychosocial status and history.

Of particular interest in the present study was participants' weight loss history. Applicants were asked to record each major weight loss effort (e.g., diet, exercise, medication, etc.) that resulted in a weight loss of 10 lbs (4.5 kg) or more, starting with the first effort and continuing to the most recent. For each effort, participants indicated their age and weight at the time, number of pounds lost, and the method used to lose weight. Respondents also rated their accuracy in recalling their weight loss history, using a scale of 1 to 10, on which 1 = not at all accurate and 10 = completely accurate. A previous study demonstrated acceptable test-retest reliability for both the total number of diets (r = 0.77, p < 0.001) and total number of pounds lost with these efforts (r = 0.87, p < 0.001) in participants who reported an accuracy rating of 7 or more (15). Participants also were asked how many times in the previous year they had initiated a diet that lasted 3 days or more, as well as those lasting fewer than 3 days.

As part of the behavioral evaluation, weight loss history was further assessed with an aided-recall approach to determine the total number of diets participants had tried, regardless of whether they lost 10 lbs or more with the effort. Using a supplemental questionnaire, the mental health professional asked participants whether they had used any of the approaches listed in each of the 11 categories shown in Table 1 and the number of times they had done so. This aided recall was used in an effort to capture all of participants' weight loss efforts, even if they had been short-lived or unsuccessful. The evaluators periodically met as a group to discuss any questions about how previous diets should be categorized.

Table 1. - Categories of diets assessed by the aided recall.

Full table

Statistical Analyses

Means (SDs) and frequencies were computed for demographic and weight loss history variables. One-way ANOVAs were used to compare participants by gender and/or ethnicity with regards to continuous variables. ² analyses were used to compare men and women with regard to previous weight loss attempts in each diet category. Paired samples t tests were used to compare the number of diet categories tried and the total number of diets reported with respect to values obtained by the WALI vs. aided recall.

Results

Demographic Characteristics

As expected, men were significantly (p < 0.001) taller (174.7 10.4 vs. 163.8 7.7 cm) and heavier than women (158.3 31.6 vs. 134.9 25.0 kg). Men and women, however, did not differ significantly in BMI, age, or number of years of education.

Weight Loss History Assessed by the WALI

Of the 177 participants who completed the weight loss history items on the WALI, 83 were confident that their responses were accurate (as indicated by a score of 7 or more on the recall item). Responses of these 83 participants were selected for further analysis. (These 83 individuals did not differ from excluded participants on any demographic variable of interest.) They reported a mean total of 4.7 2.9 diets, on each of which they lost 10 lbs or more. Over their lifetime, participants reported losing a mean total of 61.1 41.3 kg in these attempts, equal to 12.4 9.7 kg per diet. Despite these efforts, their weight increased from a mean of 89.4 27.4 kg at the time of their first diet (at age 21.2 10.1 years) to 144.5 30.8 kg at the time they were evaluated for bariatric surgery (at age 43.0 11.0 years).

Among these 83 participants, the number of efforts that resulted in a loss of 10 lbs or more ranged from 0 (N = 5) to 10 (N = 7). (The WALI inquires about a maximum of 10 diets.) As expected, age at the start of each diet increased steadily across the 10 possible efforts, from 21.2 10.1 years at the first effort (N = 69) to 42.1 10.6 years at the last effort (N = 10). Weight similarly increased with each diet. The mean starting weight at the first effort was 89.4 27.4 kg, which increased to 122.6 18.1 kg at the 10th effort. Figure 1 shows the mean change in weight over time in 56 persons who reported at least four successful weight loss attempts.

Figure 1.

Weight and dieting histories of 56 bariatric surgery candidates, all of whom had engaged in a minimum of four diets, as reported on the WALI.

Full figure and legend (63K)

In the previous year, participants reported starting 3.0 3.4 diets on their own that lasted >3 days. They reported beginning an additional 2.8 5.3 diets that lasted 3 or fewer days. Reported diets in the past year did not differ by gender, race/ethnicity, age, or BMI.

Weight Loss History Assessed by Aided Recall

As expected, responses to the supplemental questionnaire administered by the mental health professional revealed that participants had engaged in significantly more weight loss efforts than those reported on the WALI alone. Thus, the 83 participants who reported a mean total of 4.7 2.9 successful diets on the WALI indicated, when queried, that they had attempted an average of 6.0 1.7 of the categories of weight loss methods listed in Table 1. They reported trying to lose weight a total of 14.6 9.1 times using these approaches. Table 2 shows the percentage of persons who reported trying one or more approaches in each of the 11 categories, as determined by aided recall (using the supplemental questionnaire) compared with the WALI. Thus, the aided recall revealed a significant history of brief, failed weight loss attempts, beyond the numerous successful diets reported on the WALI.

Table 2. - Percent of participants (N = 83) who reported they had tried the following categories of diets, as determined by the WALI vs. aided recall.

Full table

Table 3 shows responses to the aided recall of all 177 participants assessed by the medical health professionals. It reveals that self-directed weight loss efforts were the most popular approach, used by 94% of patients (see Table 3). Commercial programs also were reported by the vast majority of patients. Approximately seven in 10 patients reported using exercise, Food and Drug Administration (FDA)-approved weight loss medications, or over-the-counter (OTC) weight loss medications. Slightly more than one-half of the participants reported using liquid meal replacements, nutritional counseling, or a physician supervised program. Approximately 5% reported previous surgery for weight loss.

Table 3. - Percent of participants (N = 177) who reported they had tried the following categories of diets, as determined by the aided recall.

Full table

A greater percentage of women than men reported experience with self-directed diets [ ²(1) = 7.22, p = 0.016] and commercial programs [ ²(1) = 5.21, p = 0.03] . There were no other significant gender differences with regards to weight loss history as assessed during the behavioral/psychological evaluation.

Discussion

The great majority of candidates for bariatric surgery at our institution had made multiple efforts to lose weight using conventional methods. This finding was suggested by the results of the WALI, which inquired only about efforts that had resulted in a loss of 10 lbs (4.5 kg) or more and by findings from a structured interview, in which candidates were queried about their use of methods from 11 different classes of interventions. The 83 candidates who were confident in the accuracy of their responses to the WALI reported a mean total of 4.7 weight loss efforts on which they had lost a total of 61.1 kg. Despite these efforts, their weight increased from an average of 89.4 kg at the time of their first diet (at age 21.2 10.1 years) to 144.5 30.8 kg at the time they sought surgery at our institution (at age 43.0 11.0 years). This weight gain occurred in the face of multiple additional, but unsuccessful, efforts to lose weight that were captured by the aided recall. These 83 candidates reported an average of approximately 10 additional efforts, using principally self-directed or commercial weight loss programs. In the total sample of 177 individuals, respondents reported over 15 previous efforts to lose weight, using methods from an average of six of the 11 classes of interventions.

The most commonly reported methods used to lose weight were self-directed diets and commercial programs, which were endorsed by the vast majority of participants. Nearly one-half of the 83 participants, confident in the accurancy of their weight history (as detailed in the WALI), described losing in excess of 10 lbs with OTC medications. Almost one-third of individuals had used FDA-approved medications to successfully lose weight, and nearly one-quarter had utilized very low calorie diets or hospital-based programs.

Insurance coverage for more conservative weight loss treatments is rare. Thus, dieters spend billions of dollars out of pocket each year in trying to lose weight (16). The price, when the cost of personal time and treatment or product fees is considered, is prohibitive for many individuals. It is likely no coincidence that the most commonly reported types of diets are typically the least expensive. In a recent review, the price of three months of weight loss treatment in the form of organized self-help or commercial programs ranged from a minimum of $0 (Overeaters Anonymous), to moderately priced Weight Watchers ($167), up to $2000 for 12 weeks of participation in Health Management Resources or OPTIFAST (17), two medically-based very-low-calorie diet programs. These programs cost upwards of $3500 for 6 months (18).

The present results suggest that for individuals with extreme obesity who seek bariatric surgery, long-term maintenance of weight loss has been all but impossible. The inability to maintain weight lost through multiple previous dieting efforts is likely a large motivating factor in the decision to undergo bariatric surgery. Results also suggest that these individuals, on average, have made several efforts to lose weight in the previous year, with minimal success. Participants reported starting three diets (3.0 3.4) in the past year that lasted for >3 days and an additional 2.8 5.3 diets lasting 3 or fewer days. Thus, the obligatory recommendation, made by many insurance companies, that bariatric surgery be delayed until the candidate can demonstrate experience with more conservative weight loss treatment options may be unwarranted.

Conversely, for some patients, a complete lack of dieting experience may raise concerns with regards to postsurgical outcomes. Bariatric surgery, although a powerful tool in ameliorating several obesity-related health conditions, is not without significant medical risks. Furthermore, 10% to 20% of patients fail to lose >40% excess weight (19). Unsuccessful outcomes are often attributed to patients' inability to follow the postoperative dietary recommendations (19)(20)(21)(22)(23). It may be that those individuals with a relative lack of experience with weight loss behaviors are less prepared to make the postoperative behavioral and dietary changes necessary for a successful long-term result. One study reported that more dieting experience was correlated with greater loss of excess weight 1 year postoperatively (24). The assessment of dieting history in this study (24) was cursory, however, based on responses to two multiple choice and two free response questions.

At the very least, a relative absence of weight loss experience warrants further inquiry by the treating health professional(s) and should be taken into account when evaluating the risks and benefits of the procedure. This may be particularly true for younger patients, those with a lower BMI, and those without significant comorbidities. Nevertheless, the specific relationship between dieting history and postoperative outcome has received little empirical attention.

The present investigation has several notable strengths. The data characterize the weight loss histories of men and women presenting for bariatric surgery. Results support the notion that this sample of individuals has attempted multiple different types of diets, on a number of occasions, over a long span of time. Despite several successful weight loss attempts, these individuals have had tremendous difficulty maintaining weight losses, which likely plays a role in motivating them ultimately to seek bariatric surgery.

This study also has several limitations. Data were retrospective, calling into question the accuracy of participants' reported histories as they were asked to record dieting efforts that occurred several years, even decades, prior. When including only those with an accuracy rating 7, the sample was reduced from 177 to 83, which may not be representative of the larger patient population. Additionally, weight loss histories were self-reported and potentially influenced by demand characteristics. Patients may strive to convey that they have tried every possible weight loss approach before; thus, bariatric surgery is not only a last resort but also their only recourse.

More research is needed to characterize the weight loss histories of individuals seeking bariatric surgery. Validity could be enhanced with larger, prospective studies or corroboration of retrospective recall with physician weight records. Additionally, demand characteristics could be reduced by assessing weight loss histories anonymously, outside the context of a required behavioral evaluation before surgery. As noted above, the relationship between weight loss experience and postoperative outcome is unknown. There is some evidence to suggest that weight cycling is associated with clinically significant decrements in eating self-efficacy (25) and weight regain after obesity treatment (26). Thus, dieting experience may be especially important in long-term outcome.

Finally, the effects of surgery are so immediately potent that the potential relationship between dieting history and postoperative outcome may only become evident when weight losses begin to plateau, 2 years after surgery. Future studies are warranted to determine whether extensive dieting experience may enhance postoperative weight loss because patients are well practiced or, conversely, if preoperative weight cycling could adversely affect postoperative outcome.

Notes

¹ Nonstandard abbreviations: WALI, Weight and Lifestyle Inventory; FDA, Food and Drug Administration; OTC, over-the-counter.

References

1. Wing, RR. (1998) Behavioral approaches to the treatment of obesity Bray, GA Bouchard, C James, WPT eds. Handbook of Obesity 855–873. Marcel Dekker, Inc New York.
2. Wadden, T. A., Butryn, M. L., Byrne, KJ. (2004) Efficacy of lifestyle modification for long-term weight control. Obes Res. 12: 151S–162S. | PubMed | ISI |
3. Wadden, TA. (1993) Treatment of obesity by moderate and severe caloric restriction: results of clinical research trials. Ann Intern Med. 119: 688–693. | PubMed | ISI | ChemPort |
4. Wing, RR. (1992) Behavioral treatment of severe obesity. Am J Clin Nutr. 55: 545S–551S.
5. Davidson, M. H., Hauptman, J., DiGirolamo et al. (1999) Weight control and risk factor reduction in obese subjects treated for 2 years with orlistat: a randomized controlled trial. JAMA 281: 235–242. | Article | PubMed | ISI | ChemPort |
6. James, W. P., Astrup, A., Finer, N., et al. (2000) Effect of sibutramine on weight maintenance after weight loss: a randomized trial: STORM Study Group: Sibutramine Trial of Obesity Reduction and Maintenance. Lancet 356: 2119–2125. | Article | PubMed | ISI | ChemPort |
7. Wadden, T. A., Sternberg, J. A., Letizia, K. A., Stunkard, A. J., Foster, GD. (1989) Treatment of obesity by very-low-calorie diet, behavior therapy and their combination: a five-year perspective. Int J Obes. 51: 167–172.
8. Steinbrook, R. (2004) Surgery for severe obesity. (1992) N Engl J Med. 350: 1075–1079. | Article | PubMed | ISI | ChemPort |
9. Gastrointestinal surgery for severe obesity: National Institutes of Health Consensus Development Conference Statement. Am J Clin Nutr. 55: 615S–619S. | PubMed |
10. Buchwald, H. (2005) Bariatric surgery for morbid obesity: health implications for patients, health professionals, and third-party payers. J Am Coll Surg. 200: 593–604. | Article | PubMed | ISI |
11. Sarwer, D. B., Cohn, N. I., Gibbons, L. M., et al. (2004) Psychiatric diagnoses and psychiatric treatment among bariatric surgery candidates. Obes Surg. 14: 1148–1156. | Article | PubMed |
12. Wadden, T. A., Sarwer, D. B., Womble, L. G., Foster, G. D., McGuckin, B. G., Schimmel, A. (2001) Psychosocial aspects of obesity and obesity surgery. Surg Clin North Am. 81: 1001–1024.
13. Beck, A. T., Steer, RA. (1993) BDI Beck Depression Inventory Manual Harcourt Brace & Company San Antonio, TX.
14. Wadden, T. A., Phelan, S. (2002) Behavioral assessment of the obese patient Wadden, TA Stunkard, AJ eds. Handbook of Obesity Treatment 186–226. Guilford Press New York.
15. Wadden, T. A., Bartlett, S., Letizia, K. A., Foster, G. D., Stunkard, A. J., Conill, A. (1992) Relationship of dieting history to resting metabolic rate, body composition, eating behavior, and subsequent weight loss. Am J Clin Nutr. 56: 203S–228S. | PubMed | ChemPort |
16. Melanson, K., Dwyer, J. (2002) Popular diets for treatment of overweight and obesity Wadden, TA Stunkard, AJ eds. Handbook of Obesity Treatment 249–275. Guilford Press New York.
17. Tsai, A. G., Wadden, TA. (2005) Systematic review: an evaluation of major commercial weight loss programs in the United States. Ann Intern Med. 142: 56–66. | PubMed |
18. Wadden, T. A., Osei, S. (2002) The treatment of obesity: an overview Wadden, TA Stunkard, AJ eds. Handbook of Obesity Treatment 229–248. Guilford Press New York.
19. Latifi, R., Kellum, J. M., De Maria, E. J., Sugerman, HJ. (2002) Surgical treatment of obesity Wadden, TA Stunkard, AJ eds. Handbook of Obesity Treatment 339–356. Guilford Press New York.
20. Brolin, R. E., Robertson, L. B., Kenler, H. A., Cody, RP. (1994) Weight loss and dietary intake after vertical banded gastroplasty and Roux-en-Y gastric bypass. Ann Surg. 220: 782–790. | Article | PubMed | ISI | ChemPort |
21. Hsu, L. K. G., Benotti, P. N., Dwyer, J., et al. (1998) Nonsurgical factors that influence the outcome of bariatric surgery. Psychosom Med. 60: 338–346. | PubMed | ISI | ChemPort |
22. Pories, W. J., Caro, J. F., Flickinger, E. G., Meelheim, H. D., Swanson, MS. (1987) The control of diabetes mellitus (NIDDM) in the morbidly obese with the Greenville Gastric Bypass. Ann Surg. 206: 316–323.
23. Sarwer, D. B., Wadden, T. A., Fabricatore, AN. (2005) Psychosocial and behavioral aspects of bariatric surgery. Obes Res. 13: 639–648. | PubMed | ISI |
24. Ray, E. C., Nickels, M. W., Sayeed, S., Sac, HC. (2003) Predicting success after gastric bypass: the role of psychosocial and behavioral factors. Surgery 134: 555–564. | Article | PubMed | ISI |
25. Foster, G. D., Sarwer, D. B., Wadden, TA. (1997) Psychological effects of weight cycling in obese persons: a review and research agenda. Obes Res. 5: 474–488. | PubMed |
26. Elfhag, K., Rossner, S. (2005) Who succeeds in maintaining weight loss? A conceptual review of factors associated with weight loss maintenance and weight regain. Obes Rev. 6: 67–85. | Article | PubMed | ISI | ChemPort |

Acknowledgments

This work was supported, in part, by the National Institute of Diabetes, Digestive, and Kidney Disease Grant K23-DK60023-04.