I’ve blogged before about the Look AHEAD Type 2 diabetes study, but mostly focused on the effect of the weight loss on cardiovascular events, which is what the study was designed to measure, as well as comparing the amount of calorie restriction to some reference points. I was looking at it again recently and thought I would examine it as a study of weight loss itself, as it is by far the longest-running study of weight loss that I have seen. According to Dr. Thomas Wadden, the lead author of the study, “This is the largest and longest controlled evaluation of a behavioral weight-loss program to date.” It was also a fairly large study, with 5145 participants.
An important question to ask here is, in what ways are the study participants different from other dieters? For one thing, the intervention group received a lot more support than a typical dieter/someone undertaking a “lifestyle change”. According to the Medscape article, Meaningful, Long-Term Weight Loss Is Possible: Look AHEAD, the relatively high success of weight loss in this study “was likely a result of the intensive, comprehensive nature of the lifestyle intervention“, so there are also differences between the intervention group and the typical person undertaking a “lifestyle change”. Obviously, in both the control and intervention group, they all had Type 2 diabetes, which may have made it more difficult for them to lose weight. Their diabetes was also well-controlled, both in the control and intervention group; participants also had to pass a prestudy treadmill test and have blood pressure below 160/100. (One outcome of the study was that in both the control and intervention groups, there were fewer cardiovascular events than in past studies of people with Type 2 diabetes.) The participants were described as “ethnically diverse”, and they all had a BMI over 25, which of course not all dieters do. Their ages were 45 to 76 years, with an average age of about 58, so they were older than many dieters. This is perhaps the most meaningful difference. Would this age range make them less or more likely to have trouble losing weight than adults younger than 45? People on the younger end would probably have more difficulty than younger adults, people on the older end less difficulty. According to the NIH, men typically begin to lose weight after about age 55 and women after about age 65. According to the researchers quoted in the Medscape article, “Some of the weight loss in the trial participants may be a normal part of aging, as the participants were about age 58 years at the start of the trial.” The fact that the control group also lost weight, so that 36% of the control group were at least 5% below their initial body weight at 8 years (vs 50% of the control group), also suggests that some of the weight loss in this trial was a natural effect of aging. The pattern of weight loss in the control group also suggests an effect of aging: “Participants in the usual-care group lost 0.6% of their initial weight at 1 year and 2.1% of their initial weight at 8 years,” according to the Medscape article. This may give weight loss advocates hope that weight loss is a more effective diabetes treatment than it appears, since the difference between the control and intervention group was not that big; on the other hand, it makes the weight loss intervention look less effective when you compare the intervention group to the control group instead of assuming that the participants’ weight would have stayed the same.
So how well did the lifestyle intervention work? Well, the average weight loss at eight years was 4.7% of body weight. As I touched on in my previous post, that nevertheless required a pretty strict diet and exercise program: 1200-1500 calories per day for people who weighed less than 250 lbs, 1500-1800 for people who weighed more than 250 lbs, and at least 175 minutes of moderate activity a week.
Now, presumably if the researchers thought that they could realistically get the subjects to lose a greater percentage of their body weight and keep it off during the study, they would have done so. [ETA: according to the protocol for the experiment, “The lifestyle intervention… is aimed at achieving and maintaining at least a 7% decrease in weight from baseline.”] And in fact, some participants did lose more. In fact, the average weight loss at one year in was 8.5%, though they’d regained about half of that on average by year four, which was still true at year eight. On the other hand, some participants could not complete the initial 5% weight loss, or if they did had gained it back at 1 year in. According to the Medscape article “Meaningful, Long-Term Weight Loss Is Possible: Look AHEAD”,
In Look AHEAD, at 1 year, 68.0% of participants who received intensive lifestyle counseling vs 13.3% of participants who received usual care lost at least 5% of their initial body weight. At 8 years, these percentages were 50.3% vs 35.7%, respectively.
So the success rate of their weight loss intervention was 50%, long-term (at 8 years)–if we set the cutoff for “success” at a modest 5% of body weight loss. Even with their sights set at 5% weight loss–which most dieters would probably not describe as “meaningful”–and even with the high level of support provided to the intervention group, only about half of the participants were able to keep that much weight off. This is the kind of weight loss amount and success rate that is described by researchers (again, from the Medscape article) as “very successful”:
“The primary finding is that overweight participants with type 2 diabetes could lose weight and were very successful in keeping their weight off at 8 years,” lead author Thomas Wadden, PhD, from the University of Pennsylvania, Philadelphia, told Medscape Medical News. This is very different from most previous studies, which report that by 3 to 5 years, most people have regained any lost weight, he added.
So what about the goals of the typical dieter, who is probably hoping to lose at least 10% of their body weight? The results from this study are not encouraging; in the intervention group, 38% of participants had lost 10% or more of their body weight at 1 year, and about 40% of them had maintained a loss of at least 10% at 8 years. This means that, for people in this study receiving intensive lifestyle counseling, only about 15% both lost 10% or more of their body weight and were still down 10% after 8 years. Sure, that’s better than the (questionably sourced) 95% of diets fail within 5 years*, but that’s in a group that could be expected to lose some weight as part of aging and that was receiving a good deal of support that most dieters do not have access to. And 10% is still a great deal less than many dieters are hoping to lose. For a 5’5″ woman weighing 200 lbs, it would be 20 lbs, which is not quite enough to take her out of the “obese” category.
Let’s do a back-of-the-envelope, speculative calculation and say that people who maintain radical weight loss are a similar fraction of the people who maintain >10% weight loss as people who maintain >10% weight loss are of people who maintain >5% weight loss. In a group of 100 people that was similar to the intervention arm of this study, about 50 would still be down least 5% of their body weight at 8 years, and 30% of those 50, or 15 people, would be down at least 10% of their body weight That would make it 4.5% (15 is 30% of 50; 4.5 is 30% of 15) of people–between 4 and 5 people out of a group of 100–who would be able to maintain radical weight loss for at least 8 years. Not such a small group that you’d be unlikely to have any acquaintances who had done it, but not so far away from the oft-cited 5% success rate/95% failure rate of diets.
It seems logical that the greater percentage of body weight a person loses, the lower the success rate in keeping it off long-term. The Look AHEAD study shows that, 8 years after the initial weight loss, and despite a high level of support and participants being able to expect aging-related weight loss, only about half of participants maintained at least 5% weight loss, and only 15% maintained a weight loss of 10% or more. Radical weight loss almost certainly has a very small long-term success rate, but even with a modest goal of 5% of body weight and a great deal of support, odds are only 50-50 that weight loss will be maintained long term.
*This particular version, at least, never seems to come with a citation, but the lead author of the study, Dr. Wadden, has said that most previous studies report that by 3 to 5 years, most people have regained any lost weight. It seems to be pretty well-accepted in the scientific and health community that weight loss is rarely maintained. The NIH press release for this study also said, “Few, if any, studies of this size and duration have had comparable success in achieving and maintaining weight loss.” Questionably-sourced though it may be, I’ve never seen it contradicted in the form I sometimes see, that includes the additional caveat of “radical weight loss” sometimes defined as the type of weight loss that would make a fat person thin and not just less fat, often defined as at least 30 lbs (this is the weight cutoff for joining the National Weight Control Registry, though their time cutoff is only “at least a year”).