The Spinning of Look AHEAD
On January 13, 2014 the Obesity Society issued a press release1 proclaiming, “Largest U.S. Weight-Loss Study Shows Long-Term Weight Control is Possible with Intensive Weight Lifestyle Intervention (ILI) Eight-year Look AHEAD Results Show 50% of Participants Kept Off 5% or More of Initial Weight Lost.” The subjects of the press release are an article2 by the Look AHEAD Research Group (LARG) which describes the weight loss occurring among subjects in the federally-funded Look AHEAD trial which was prematurely ended for “futility” in 2012 and an accompanying editorial by Michael G. Perri.3
The release states, “The largest United States weight-loss study using diet and exercise alone shows long-term weight loss is possible through intensive lifestyle intervention (ILI) for weight management…Researchers found that ILI produced clinically meaningful weight loss (>5%) over eight years in half of the individuals and believe the intervention can be used for long-term management of obesity-related co-morbid conditions.
The release continues, “Although some weight regain occurred in this study, Look AHEAD’s 8-year trial results stand in contrast to this prior research, with study participants maintaining more than half of their weight loss after 8 years. Further, nearly 40% of participants who lost >10% of initial weight at year 1 maintained this loss at year 8….In addition, weight losses were similar among men and women and across racial-ethnic subgroups indicating that all individuals can benefit from an intensive diet and exercise intervention.” The release quotes with approval Perri’s editorial which states, “The success is likely due to the intensive and comprehensive nature of the lifestyle intervention, and confidence in the study’s findings is underscored by the trial’s excellent rate of retention.”
Sounds good but is it accurate? Let’s take a ‘deep-dive’ into the Look AHEAD trial.
Look AHEAD was a very important clinical trial, maybe one of the most important clinical trials given the prevalence of obesity and type 2 diabetes. With the increases in overweight and obesity not only in the United States but globally, type 2 diabetes is also increasing. Globally, the population with diabetes is expected to double between 2000 and 2030, based on demographic changes, especially the growth in the population over 65 years of age. 4 With type 2 diabetes, there is a risk of progression to cardiovascular disease (CVD) and death. About 2/3 of those with diabetes have CVD as the cause of death.5
While an earlier study6 showed that bariatric surgery could cause enough weight loss to lower rates of myocardial infarction, it had not been demonstrated that less weight loss from behavioral interventions could obtain the same effect. The implications for future development of type 2 diabetes and CVD is significant. A 2013 study found that all of the traditional risks factors for CVD in the diabetic population were improving over the last decade, with the exception of Body Mass Index (BMI) and waist circumference.7 (Body Mass Index is a frequently used measure of body weight and is calculated by dividing weight in kilograms by height in meters squared.)
In the words of the Look Ahead Research Group (hereafter “LARG”), “The purpose of the Look AHEAD trial is to determine whether cardiovascular morbidity and mortality in persons with Type 2 Diabetes can be reduced through an intensive lifestyle intervention aimed at producing and maintaining weight loss. Briefly the intensive lifestyle intervention (I call it über intensive lifestyle intervention, below) includes moderate-intensity physical activity to achieve and sustain at least 200 minutes per week of exercise together with a healthy diet that includes portion-controlled foods. The goal of this intervention is for individuals to achieve and maintain at least 10% loss of body weight. Failure to meet appropriate goals is followed by the option of initiating other ‘toolbox strategies (e.g. medication for weight loss).”8
Look AHEAD was a randomized clinical trial, implemented at 16 centers, involving 5,145 participants with half randomized to the Intensive Lifestyle Intervention (ILI) arm and half to Diabetes Support and Education (DSE) arm which is standard of care for patients with Type 2 Diabetes. According to Wadden et al, “In year 1, participants in this (ILI) group were provided a comprehensive intervention, designed to induce an average loss >7% of initial weight. Individuals were given the goal of losing >10% in order to increase their likelihood of meeting the 7% study-wide goal.9 (Emphasis added.) The primary endpoint of the study was a reduction in the incidence rate of a combination of cardiovascular death (fatal myocardial infarction and stroke), non-fatal myocardial infarction, non-fatal stroke and hospitalization for angina. Secondary endpoints included cardiovascular disease risk factors, diabetes control and complications, general health, hospitalizations, quality of life and psychological outcomes, cost and cost effectiveness.
According to the recent LARG report in the New England Journal of Medicine (NEJM)10, weight loss was greater in the ILI group than in the DSE (control) group. The ILI produced greater reductions in HbA1C and improvements in fitness and all CVD risk factors except for low HDL. The primary outcome (CVD events and death) was nearly identical in both groups. The NEJM article reports a mean weight loss in the ILI of 8.5% of initial body weight at year one and an average weight loss of 6% by the end of the trial. The DSE group lost very little in the first year but, at the end of the trial, had lost 3.5% of initial body weight. In other words, the mean difference between the two groups was only 2.5%, below the level considered “clinically significant.” Although not specified in either the NEJM or Obesity articles, a simple calculation shows that the ILI arm mean, placebo-adjusted, weight loss at the end of the study was all of 5.7 pounds, reducing the mean BMI by 0.8, from an estimated 35.8 to 34.9, still in the obesity category.11
As the trial showed that ILI could not reduce the rate of CVD events, it was stopped by the National Institutes two years early for futility.12
Contrary to the press release, Look AHEAD did not use “diet and exercise alone.” According to the LARG paper describing the baseline characteristics, “This was a medicated population with a high frequency of medication use”. At the outset, more than 80% of participants were taking one or more glucose lowering drugs. Metformin most frequently used 48%, 14% were on insulin, 47% on antidepressants 15.8% on diuretics, and 28.4% were taking antihypertensive medications. 50% were on lipid-lowering agents, primarily statins.8 High statin use, greater in the control group than in the ILI group, apparently resulted in far fewer deaths and cardiovascular events than was anticipated. This necessitated, in the second year, an adjustment in the age limits of recruitment and later, the addition of additional components to the definition of cardiovascular events.13 “Only about 350 participants in each arm were not on a diabetes medication.14
Look AHEAD specifically encouraged the use of one weight loss drug, orlistat, which is approved by the Food and Drug Administration. According to the Look AHEAD Counselor’s Manual, “Orlistat has been shown to significantly improve the induction and maintenance of weight loss. The use of orlistat is described in a separate chapter of the Lifestyle Counselor’s Manual; review it before discussing the medication with participants. Information is also provided on a video tape that can be shown to participants. (Orlistat costs approximately $100 a month (per patient) but is being donated to Look AHEAD by its manufacturer.)”15 (We do not know how many participants actually took orlistat or for how long.)
Another FDA-approved weight loss drug, sibutramine, was on the market for most of the trial but was not recommended by the Look AHEAD counselors. Medication decisions were made by the participant’s own personal physician. Herbal weight loss product usage was not tracked.
According to the study published in Obesity, 76 participants in ILI and 84 in the DSE group had bariatric surgery. The weight loss from those who had surgery was not counted in the final tallies.2
It’s hard to see Look AHEAD as a “weight loss trial” per se. The trial did not compare diet to physical activity, or diet and exercise to weight loss drugs, or lifestyle intervention to bariatric surgery or low-fat to low-carbohydrate diet. DSE is not considered a weight loss program. Clinical trials are evaluated according to whether they met their primary and/or secondary endpoints. Weight loss was neither a primary or secondary endpoint.
The TOS press release states, ”weight losses were similar among men and women and across racial-ethnic subgroups indicating that all individuals can benefit from an intensive diet and exercise intervention.” 1 Two things are wrong with this sentence: One, the participants were not typical of the whole population. Two, the Intensive Lifestyle Intervention was really an “Über-Intensive Lifestyle Intervention” where the participants received, for free, individual and group sessions, meal replacements, gym memberships, personal trainers and a $100 honorarium for the annual weigh-in.
Only one in five of the individuals screened (5,145 out of27,000) were selected largely because they met the physical activity and motivational characteristics considered necessary to achieve the 10% initial and 7% final weight losses.8 Participants in Look AHEAD were recruited with the expectation that they could meet these metrics: Lose 10% of body weight and maintain the loss, change dietary intake: 1200-1500kcal/day if <250 lbs,, 1500-1800 kcal/day > 250 lbs, and < 30% calories from fat. They were expected to gradually increase their physical activity to 200 minutes a week. Participants were to keep daily records of their food intake, physical activity and other behaviors. 88.9% of the women and 86.2% of the men in the study had at least one prior intentional weight loss of greater than 5%.16
As part of the screening, participants had to pass the Graded Exercise Test. According to Brancati, “The Graded Exercise Test (GXT) excluded participants most likely to develop CVD events: The study included GXT as an inclusion criterion based on concerns about safety and liability related to initiating an exercise program in adults with type 2 diabetes at high risk for CVD. The GXT effectively excluded some higher-risk patients (e.g. those with prevalent symptomatic CVD) who demonstrated electrocardiographic or blood pressure abnormalities during the test. The prospect of XT may have also discouraged some individuals with exercise-induced symptoms from attempting to enroll in the trial.13 Eleven percent of those taking the GXT were excluded from the trial.17 Ribisi et all agreed, “ Look AHEAD had a graded exercise test and excluded the very unfit and those with a limited ability to sustain a regular exercise program which was a major segment of the intervention…The results of this study many not be representative of all individuals with type 2 diabetes in this age range because of specific inclusion/exclusion criteria. Since only overweight/obese individuals were included, this will bias the trend toward greater comorbidities and lower fitness than would be expected. Conversely, each participant also had to meet a minimum fitness level of 4.0 MET, which excluded the very unfit and those with a limited ability to sustain a regular exercise routine, which was a major aspect of the subsequent behavioral intervention.”18
Participants were healthier than expected. Look Ahead was not a cohort study where some 50% of eligible individuals in a community are enrolled. This was a type of trial which “typically forgo representativeness in favor of motivation to participate fully in study interventions”.13 “Overall, they (the participants) are healthier than diabetic individuals in the National Health and Nutrition Examination Survey with regard to glucose, HbA1C, and lipid levels and are less likely to smoke.”19
Study participants were required to participate in a 2-week run-in period that included successful monitoring of diet and physical activity. Look Ahead participants had a high usage of weight control practices at the outset. About 60% used increasing fruit and vegetable consumption, cutting sweets and junk food, and eating less carbohydrate foods. 41% reported self-weighing at least once a week.16
Can the results be generalizable to the entire population as the TOS press release states? The commentators say no. Perri’s accompanying editorial states, “ it is unknown whether the patterns of long-term weight changes observed by Look AHEAD are generalizable to younger adults (ages 21-44 years) and to individuals who do not have diabetes.”3 The LARG wrote in the NEJM paper, “We recruited patients with type 2 diabetes who were motivated to lose weight through lifestyle intervention and who could successfully complete a maximal fitness test at baseline. Thus, the results cannot be generalizable to all patients with type 2 diabetes.”10 A leading researcher in Look AHEAD, Dr. Rena Wing, told the FDA Advisory Committee, (referring to both the Look AHEAD trial and the Diabetes Prevention Program), “these are efficacy trials, not effectiveness trials, so that we are trying to select a group of participants appropriate for efficacy trials.”20 Efficacy trials are designed to see if the intervention has the desired effect, “does more good than harm under optimum conditions”. “Effectiveness” trials, on the other hand, are designed to see if the intervention works under real-world conditions, i.e. does the patient get better in typical usage.22
The Über-Intensive Lifestyle Intervention
The Centers for Medicare and Medicaid Services decided in 2012, following a recommendation from the US Preventive Services Task Force to cover, for Medicare beneficiaries, intensive behavioral counseling for adult obesity. Coverage is limited up to 21 visits a year.23 The Look AHEAD trial ILI arm is vastly more “intense”.
For the first year, participants were urged to attend up to twice (42) the meetings than Medicare covers. After year 1, participants were encouraged to attend 2 visits per month year to end of the trial or up to 224 for the duration. The visits included group and individuals sessions. This was augmented by a second contact by phone or email. Participants were paid a $100 honorarium for attendance at annual weigh-ins.24
As noted above, the weight-loss drug orlistat was provided at no charge to those not reaching their weight loss goal. Meal replacements were also provided at no charge for the first 4 months with participants encouraged to replace two meals and one snack daily with liquid shakes and meal bars (up to 120 meal replacements. Gym memberships, personal trainers, rented home-exercise equipment, a more intensive diet option or clothing were also available at no charge up to $100 a month to the participants if they lost less than 5% of their initial body weight. 25 During years 2-8, all sites offered monthly group meetings, one Refresher Group and one National Campaign, challenging participants to reach a specific goal for which they would receive a prize. Those providing the counseling included registered dieticians, psychologists, exercise therapists, all of whom were certified. These personnel were trained in cognitive behavioral therapy, and incorporated elements of problem-solving, motivational interviewing, and cultural tailoring.
The Uber-Intensive Lifestyle Intervention was committed to create a strong rapport between the staff and participants. This was true for the DSE participants as well but not for weight loss but retention in the group analysis. According to the LARG, “The DSE Committee worked under the premise that if the DSEI participants had a ‘perceived benefit’ from these sessions and formed a closer bond with the study staff, their commitment would be strengthened and retention in annual outcome assessments would be enhanced. However, from the study perceptive, a key aim was to produce a difference in weight and fitness between the participants in the two study arms; a goal which was achieved after 1 year.” 26
The Outcomes of Look AHEAD
Look AHEAD did not meet its primary nor secondary endpoints. While weight loss was greater in the Über-Intensive Lifestyle Intervention group than in the DSE group, this did not translate into a reduction the rate of cardiovascular events. 15.6% of participants in U-ILI and 16.2% in DSE died or had one of the specified cardiovascular events.11
14% of participants in U-ILI gained above their baseline weight. 32% of the U-ILI group did not lose at least 5% of initial weight in the first year (the most intense) and just over one-third (34.5%) of these individuals achieved this goal by year 8.3
The U-ILI group failed to meet 10% weight loss goal in first year and 7% goal by the end of the trial. The four year report9 included the percentages reaching the 7% goal but the eight year report did not provide the 7% figure. The TOS press release, Obesity article and editorial all indicate that these results provide hope for managing the comorbidities of obesity. A key indicator are reductions in medications and the start of the use of insulin but the eight year report2 did not include these factors. Hopefully, subsequent publications will report on these factors.
Overall, the U-ILI group lost 4.7% of initial body weight, the DSE group 2.1%, the differential (2.6%) being barely one BMI unit.
How does a 4.7% loss of initial body weight compare to alternatives?
According to Bray27, eliminating the placebo-adjusted calculation, weight-loss drugs such as phentermine showed a -6.4% annual reduction in initial body weight, orlistat a -6.5% reduction, lorcaserin a -10.2% reduction. For bariatric surgery, at 3 years post-surgery, the percent weight loss for participants who underwent Roux-en-Y-Gastric Bypass surgery from baseline was 31.5% and 15.9% weight loss from baseline for Laparoscopic Adjustable Gastric Band in the NIH-funded LABS study.28 An evaluation of a combined health professional-commercial weight loss program in the United Kingdom examined outcomes of a 12 week program by four BMI categories, 30 <, 31-34.9, 35-39.9 and 40>. Percent loss from initial body weight was 3.7%, 4.0%, 4.0% and 3.9%, respectively.29
Nevertheless, the Obesity editorial claims the reductions in 50% of the U-ILI group are a great success, and “likely due to the intensive and comprehensive nature of the lifestyle intervention.” Was the intensive and comprehensive nature of the lifestyle intervention also responsible for the other 50% that did not achieve a 5% reduction in body weight? Can one own the successes and not the failures? The editorial goes on to state that confidence is gained by the study’s high retention rate. It should be kept in mind that the retention rate refers to those who came in for their $100 check at the annual weigh-in. It should not be confused with adherence to the recommended dietary and physical activity recommendations. As Dr. Wing told the FDA Advisory Committee, “in other words, 95 percent are coming to the (annual) visits, but clearly, many of them are not adhering to the diet and exercise prescription.” 30
What are implications for future obesity treatment?
What are the clinical implications? Should über- or intensive or simple lifestyle changes be recommended as the first line treatment when they offer so little? Is holding out intensive or simple lifestyle change only providing “false hope” to those living with overweight and obesity?
It seems the overweight and obese persons with type 2 diabetes should go on lipid-lowering products as soon as possible. Their high usage in the DSE group caused a remarkable drop in the expected cardiovascular event rate.
Second, regarding sub-groups, in this trial (as with the Diabetes Prevention Program) the most improvement is seen in the older populations. There are a variety of reasons for this, including more time to devote to the dietary, higher health care concerns and physical activity but it may also have been due to aging, and other health problems. The progressive loss of muscle and gain in fat (sarcopenia obesity) may complicate treatments for this group.5
Third, there is a very interesting trajectory in the two groups. The U-ILI group lost most of their weight in the first year and gradually regained. The DSE group lost very little in the first year but have seen regular improvements in weight loss since then. The loss is less than considered clinically meaningful but the differences between the two groups is interesting. The Obesity paper acknowledges that the DSE participants could have intentionally lost weight or it could be due to unintentional activity, aging or diabetes (which, of course, would also be true of the U-ILI group).
The NEJM article states that this study represents that the “weight loss achieved in the intervention group is representative of the best that can be achieved by current lifestyle approaches”10. If so, it’s a pretty meager showing for a highly-motivated population receiving an Über-Intensive-Lifestyle Intervention. These results should be the starting point for a broad re-assessment of behavioral lifestyle interventions (whether über-, or intensive or moderate intensity). A ‘scientific society’ should confront the data and lead the discussion for re-assessment instead of putting the proverbial lipstick on a pig.
- http://www.obesity.org/news-center/look-ahead-8-year-results.htm, accessed March 17, 2014.
- Look AHEAD Research Group, Weight Losses with an Intensive Lifestyle Intervention: The Look AHEAD Study, Obesity (2014), 22, 5-13, PubMed.
- Perri MG, Effects of behavioral treatment on long-term weight loss: Lessons learned from the Look AHEAD trial. Obesity 2014 Jan:22(1):3-4 Pub Med.
- IDF diabetes atlas: global estimates of the prevalence of diabetes for 2011 and 2030, Diabetes Res Clin Pract 2011 Dec;943:311-21 PubMed
- Cetin DC, Nasr G, Obesity in the elderly: More complicated than you think. Clev Cl J Med 81(1) 2014:51-61 PubMed
- Sjöström, et al, Effects of Bariatric Surgery on mortality, NEJM 2007 Aug 23;357(8):741-52 PubMed
- Wong ND, et al, Trends in cardiovascular risk factor among US adults with type 2 diabetes from 1999 to 2010: Comparison by prevalent cardiovascular disease status, Diab Vasc Dis Res 2013 Nov;10(6):505-13. PubMed
- LARG, Baseline Characteristics of the randomized cohort from the Look AHEAD (Action for Health in Diabetes) study, Diab Vasc Dis Res. 2006 Dec;3(3):202-15 PubMed
- Wadden TA, et al, Four-Year Weight Losses in the Look AHEAD Study: Factors Associated with Long-Term Success. Obesity 2011 Oct;19(10):1987-98 PubMed
- LARG, Cardiovascular effects of intensive lifestyle intervention in type 2 diabetes, NEJM 2013 Jul 11: 369(2):145-54 PubMed
- Calculation was Initial weight 100.54kg X 2.6% reduction = 2.6 kg = 5.7 lbs)
- NIH Press Release http://www.nih.gov/news/health/oct2012/niddk-19.htm, accessed Feb.13,2014.
- Brancati, FL et al, Midcourse correction to a clinical trial when the event rate is underestimated: the Look AHEAD (Action for Health in Diabetes) Study, Clinical Trials, 2012; 9 (1):113-124 PubMed.
- Wing, R, Transcript, FDA Endocrinologic and Metabolic Drugs Advisory Committee, March 28, 2012 at p. 160, http://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/EndocrinologicandMetabolicDrugsAdvisoryCommittee/UCM303355.pdf, accessed Feb.13, 2014.
- Look AHEAD Counselor’s Manual. https://www.lookaheadtrial.org/public/dspMaterials.cfm, Ch. 25, p.7, accessed Feb. 13, 2014
- Raynor HA, et al, Weight loss strategies associated with BMI in overweight adults with type 2 diabetes at entry into the Look AHEAD Action for Health in Diabetes) trial, Diabetes Care 2008 Jul;31(7): 1299-304 PubMed
- Wing, op cit, at p. 176.
- Ribisi PM, et al, Exercise Capacity and Cardiovascular Characteristics of Overweight and Obese Individuals with Type 2 Diabetes, Diabetes Care 2007 Oct;3010):2679-84 PubMed
- LARG. Reduction in Weight and Cardiovascular Disease Risk Factors in Individuals with Type 2 Diabetes, Diabetes Care, 30(6):1374-1383. PubMed
- Wing, op cit, at p. 152.
- Wing, op cit, at p. 169.
- Glasgow, RE et al, Why Don’t We See More Translation of Health Promotion Research to Practice? Rethinking the Efficacy-to-Effectiveness Transition Am J Pub Health, 2003, Aug;93(8):1261-1267 PubMed.
- Centers for Medicare and Medicaid Services Decision Memo on Intensive Behavioral Counseling for adult obesity http://www.cms.gov/medicare-coverage-database/details/nca-decision-memo.aspx?&NcaName=Intensive%20Behavioral%20Therapy%20for%20Obesity&bc=ACAAAAAAIAAA&NCAId=253&, accessed February 15, 2014.
- Wing, op cit at p172.
- Look AHEAD Lifestyle Counselor’s Manual, https://www.lookaheadtrial.org/public/dspMaterials.cfm, Ch. 25, p. 7, accessed Feb. 15, 2014.
- Wesche-Thobaben, LARG, The Development and description of the comparison group in the Look AHEAD trial, Clinical Trials 2011, 8:320-329, PubMed.
- Bray, GA, Why do we need drugs to treat the patient with obesity? Obesity, 2013 May; 21(5):893-9. PubMed.
- Courcoulas AP et al, Weight Change and Health Outcomes at 3 years After Bariatric Surgery Among Individuals with Severe Obesity JAMA 2013;310(22):2416-2425. PubMed.
- Stubbs, RL, et al, Service evaluation of weight outcomes as a function of initial BMI in 34,271 adults referred to a primary care/commercial weight management partnership scheme. BMC Research Notes 2013, Apr 24;6: 161. PubMed.
- Wing, op cit. at p. 172.