Part One: What does the evidence reveal? Can diets work?

This two-part post is going to examine the evidence surrounding the question on everyone’s mind with the beginning of each new year: Do diets work?

It is widely believed they do and certainly the weight loss industry has been rigorously promoting that belief among both consumers and the medical community. The health dangers of obesity have been similarly promoted as being so great that weight loss is seen as imperative. What may be surprising, is that for some forty years, research has failed to support the effectiveness of weight loss interventions, namely diet and exercise, for the treatment of obesity. But each time a major review of the growing body of this evidence is released, it rarely reaches popular discourse. Belief in weight loss continues, while disregarding overwhelming evidence to the contrary.

We’ll begin with the latest meta-analysis of weight loss clinical trials recently published in the Journal of the American Dietetic Association. What makes this systematic review noteworthy and educational is that it provides a perfect illustration of why it’s important to look at the actual evidence and not shape our beliefs or make health decisions based on interpretations made for us in an abstract or by the media. The conclusions that are stated often bear little resemblance to the actual findings!

Weight management meta-analysis

The authors** of this meta-analysis were managers at HealthPartners Health Behavior Group and Kaiser Permamente’s Care Management Institute — who also financed the study — and it was led by Marion J. Franz, MS, RD, a respected diabetes educator and consultant with Nutrition Concepts by Franz, Inc. They said that their objective was to “determine what types of weight-loss interventions contribute to successful outcomes.”

Increasingly more insurer performance measures, disease case management by insurers, and clinical guidelines recommend weight loss interventions.

The question of which weight loss interventions are effective and what are realistic long-term outcomes are important for health professionals counseling patients, as well as for the general public, wrote the authors.

So, they stated, they reviewed all weight loss interventions reported in the medical literature — diet alone, diet and exercise, exercise alone, meal replacements, very-low-energy diets, and weight-loss medications (orlistat and sibutramine) — that used weight loss as the primary outcome and had at least 12 months of follow-up in order “to determine treatment interventions that contribute to sustained weight loss maintenance.”

Searching PubMed, they identified 80 studies that met their criteria. Given the many more weight loss trials that have been conducted and are known to obesity researchers, a closer look finds that only studies published from January 1997 through August 2004 were included in this meta-analysis. They pooled the data for each type of intervention. The main caution they noted was that most of the studies reported weight-loss outcomes based on the participants who completed the studies and attrition (drop-out) rates were exceedingly high, averaging 31%.

The reported findings. Very-low-energy diet studies resulted in a dramatic weight loss followed by rapid and substantial weight regain, they reported. Exercise alone or just providing weight loss advice, regardless of the method or intensity, did not result in successful weight loss. Interventions incorporating reduced caloric intakes “seem to produce the most encouraging short-term results,” they concluded, and were associated with “moderate weight loss at 6 months. Although there is some regain of weight, weight loss can be maintained.”

Overall, they said, “in studies extending to 48 months, an average 3 to 6 kg of weight loss was maintained [6 -13 pounds after 4 years], with none of the groups experiencing weight regain to baseline.”

A closer look, however, shows that exceedingly few studies included in their review were long-term weight loss studies (2 or more years follow-up) or able to demonstrate any maintenance of weight loss — the definition of success. In fact, only three studies in their entire review extended to 4 years follow-up to support their conclusions:

· Only one diet study (Manning et al. 1998) was long-term, which had a 4 year follow-up after a 26 week intervention period. It compared diet to dexfenfluramine among 132 diabetics. With an unknown number completing the diet arm, the average weight loss was 1.7 kg at the 4-year follow-up, but there was a steady regain trajectory which was not followed to its conclusion.

· There was only one diet and exercise trial that was long-term included in this review, (DPP 2002) with a follow-up of 4 years. It was a 24 week intervention on diabetics comparing weight loss with metformin. But it also had a regain trajectory, showing a 4 kg average weight loss at the 4-year follow-up mark.

· None of the exercise or meal replacement trials were long-term studies (at most 1.5 and 1 year, respectively).

· Only one very low calorie diet (VLCD) was long-term (>2 years follow-up). In this study, only 25 people had completed it and they were also on a regain trajectory. Overall, the VLCD studies showed regain trajectories, with the longer the follow-up the greater the regain, (averaging 80-90% regain after 2 years).

It’s not known what any of these trajectories would have finally shown had the studies been continued, rather than discontinued before the weight rebounds stopped. Even so, few would likely argue that these demonstrate ‘weight maintenance.’

Conclusions. The authors concluded:

Despite the limitations, this systematic review provides a set of results that represents the longest-term studies of weight loss and maintenance available in the literature and places them in a context that is meaningful to practitioners....At approximately 6 months, weight loss begins to plateau across nearly all interventions, but with continued professional support such as was provided in the clinical trials, weight loss can be maintained.

The wording appears to suggest that this paper provides evidence that healthcare providers have the ability to ensure long-term weight loss maintenance in their patients. But it didn’t. Given that insurance plans, however, might use this conclusion as evidence to justify P4P (pay-forperformance) measures for doctors, both doctors and laypeople have an interest in learning what is known about the effectiveness of counseling by doctors and long-term weight loss maintenance.

P4P: Doctors’ obligatory roles as weight loss counselors. First, is there any evidence that dietary and lifestyle counseling — increasingly required of doctors by insurers — is effective for long-term weight loss of their ‘overweight’ and ‘obese’ patients, or even for those with specific life-threatening chronic diseases?

· No. Researchers at Tufts England Medical Center in Boston, Massachusetts, for example, examined 46 clinical trials of dietary counseling for weight loss, involving nearly 6,400 dieters. Regardless of the programs — the intensity of the interventions, number and frequency of support meetings, individual or group counseling, caloric or other dietary restrictions, inclusion of exercise, use of electronic media such as the internet, or level of participation — all produced an average weight loss of about 6 percent during the first year..... But a steady regain every month after that. By the end of five years, all participants had fully regained their weight.

· No. The scientific reviews of the evidence on dietary counseling conducted by the U.S. Preventive Services Task Force in 1996 and 2003 concluded there is insufficient evidence to recommend routine counseling by healthcare professionals to promote a healthy diet in adults. Even the evidence for intense interventions among highly-motivated high-risk groups still demonstrated only modest short-term changes, with no information on actual health benefits and insufficient evidence to recommend widespread dietary counseling.

· No. The USPSTF could not determine the balance of benefits and potential harms of counseling with behavioral interventions to promote weight loss and concluded that the evidence is insufficient to recommend for or against the use of counseling of any intensity and behavioral interventions to promote sustained weight loss in overweight adults.

While weight loss was the primary outcome measure for a successful weight loss intervention, even these authors recognized that effectiveness cannot be determined or claimed unless weight loss is maintained. Obesity researchers and the research universally define weight loss success as being weight loss maintained for at least five years.

· The expert panel of the FTC reviewed the evidence for their 2003 report and found that one of the top bogus weight-loss claims was for permanent weight loss. According to Dr. Susan Yanovski: “Unfortunately, as we all know, weight regain after weight loss is the rule rather than the exception...There are no known supplements, devices, programs that give you a permanent alteration in your body’s metabolism, and there is no way that lost weight will be maintained.” In their scientific analysis, the expert committee said that according to the National Academy of Science, Food and Nutrition Board, “many programs and services exist to help individuals achieve weight control. But the limited studies paint a grim picture: those who complete weight-loss programs lose approximately 10 percent of their body weight only to regain two-thirds of it back within 1 year and almost all of it back within 5 years.”

· In fact, the FTC charged five of the largest commercial diet companies (Weight Watchers, Jenny Craig, Diet Center, Nutri/System and Physicians Weight Loss Center with, among other things, deceptive advertising for making unsubstantiated claims about weight loss and long-term maintenance. [Read what happened here.]

· Remember that $2 million clinical trial conducted by Stanford University researchers who tested the four top commercial diets? Among 311 healthy women and after one year of dieting, all of the diets resulted in modest weight loss at best, and all dieters were already steadily rebounding, with the biggest losers rebounding the most dramatically. The participants were still eating notably fewer calories (359 to 500 kcal/day less), even while they were regaining the weight lost. This study, however, was too short to be able to demonstrate long-term effectiveness. It’s like other studies of commercial diets: none have reported 5-year results.

As we consider this HealthPartners-Kaiser meta-analysis, the most important question to ask ourselves is: Did these researchers ask the right question?

Is healthcare about appearances and helping people lose weight to look slim — let alone get into a smaller dress/pant size for a brief time — or is it about improving health and longevity? If medicine is about health, then the evidence of health risks and benefits of obesity and weight loss are the imperative considerations. Without that evidence, weight is nothing more than a false surrogate endpoint.

In Part Two, we’ll look at three reviews that did examine that evidence: one of the most renowned and comprehensive review completed in 1991 by professors and experts on behavioral treatments for obesity and eating disorders; the recent done this spring by researchers at the University of California, Los Angeles; and one by the lead author, Franz, herself.

© 2007 Sandy Szwarc

* N. P. Pronk is executive director, HealthPartners Health Behavior Group

A. L. Crain is a statistician at HealthPartners Research Foundation

J. VanWormer is a program evaluation consultant at HealthPartners Health Behavior Group

J. L. Boucher was director, Health Programs and Performance Measurement, HealthPartners Health Behavior Group during this study (currently, director of education at Minneapolis Heart Institute Foundation)

T. Histon is director of weight management initiative at Kaiser Permanente’s Care Management Institute

W. Caplan is director of clinical development at Kaiser Permanente’s Care Management Institute

J. D. Bowman is manager of knowledge and information at A. Kaiser Permanente Innovation