Via Dr. Steve Parker comes news of a very interesting four-year followup to Shai et al’s two-year diet study comparing Mediterranean (Med), low-carb (LC), and low-fat (LF) diets.
As with previous annoying studies, it is less the study itself that is the problem and more the interpretation and/or reporting of said study.
Study design. 322 “moderately obese” subjects were randomly assigned to one of the following: 1) LF, restricted-calorie; 2) Med, restricted-calorie; or 3) LC, non–restricted-calorie. 272 subjects completed the two-year intervention, with mean weight losses of 3.3 kg, 4.6 kg, and 5.5 kg, respectively. That translates to 7.3, 10.1, and 12.1 pounds.
As with similar diet studies, participants also saw improvements in other health markers like total cholesterol, triglycerides, and LDL:HDL ratio.
The authors’ conclusion in the original study?
Mediterranean and low-carbohydrate diets may be effective alternatives to low-fat diets. The more favorable effects on lipids (with the low-carbohydrate diet) and on glycemic control (with the Mediterranean diet) suggest that personal preferences and metabolic considerations might inform individualized tailoring of dietary interventions.
No (major) complaints here, though had I blogged about the original study, I’d probably have been less than enthusiastic about 12 pounds of weight loss in the moderately obese after two years.
Four-year followup. So the news is about a four-year followup after the original two-year study ended. The researchers were able to check in with 259 of the participants who completed the study (an impressive 95.2%). You can read their letter to the editor in the NEJM for all the details, but here’s what I find the most relevant:
Weight loss after six years? 0.6 kg in the LF group, 1.7 kg in the LC group, and 3.1 kg in the Med group. That’s 1.3, 3.7, and 6.8 pounds. In moderately obese individuals.
The authors’ followup conclusion?
In conclusion, a 2-year workplace intervention trial involving healthy dietary changes had long-lasting, favorable post-intervention effects, particularly among participants receiving the Mediterranean and low-carbohydrate diets, despite a partial regain of weight.
Well, nothing like looking at the bright side of things! Of the health markers, total cholesterol remained slightly improved, but both triglycerides and LDL:HDL ratio showed a similar J curve response (i.e., were moving back towards baseline).
So what’s annoying? Here’s what WebMD had to say:
Overall, the Mediterranean diet led to the most dramatic changes, but people on the other diets also did pretty well. …
Everyone regained some of the weight they had lost in the original study, but all were thinner than when the study first began. …
“Most anyone can lose weight, but keeping it off is the harder part,” says Nancy Copperman, RD. … “The Mediterranean diet seems to be the winner,” she says. “This is a livable diet and has positive physiologic benefits.”
Allison Krall, RD … reviewed the study for WebMD. “The Mediterranean diet won out overall. It is a more balanced diet with more options and choices,” she says. “Finding ways of eating that a person can stick to over the long haul is the key to losing weight and keeping it off because yo-yo dieting is dangerous.”
All I can say is, you gotta be effing kidding me. The most generous interpretation of this study, IMO, is to say that any of these diet interventions was successful at halting weight loss. And per Dr. Sharma, that may be a very good thing.
But in terms of the holy grail of weight loss aka fighting the war on obesity?! I’m sorry, a 4-7 pound weight loss in 6 years in the moderately obese is basically supporting what the HAES folk say:
Long-term follow-up studies document that the majority of individuals regain virtually all of the weight that was lost during treatment, regardless of whether they maintain their diet or exercise program.
Anyways, props to the authors for the followup. But IMO this is further proof that if we are to address lifestyle-mediated weight gain, we probably gotta stop looking at weight loss diets to do so.
BTW, for more on the original study, head over to Evelyn’s (here and here).
Weight Maven is written by Beth Mazur. Beth believes that obesity is more symptom than cause and that the real problem is our Western diet -- especially sugar, refined grains, and industrial oils. Beth writes about nutrition, ancestral health & food policy. And cats! 
Any info on the outliers? Were there any individuals who lost a significant amount of weight and kept it off? If so, what was different about them?
No, there was no info or discussion re outliers, but I would expect that it would be very similar to what Gardner reported in his ATOZ diet study (starting around 46:00). That said, Gardner reported only on change after 1 year, not longer term maintenance.
It’s rare that a letter to the editor receives this much attention. I don’t apply the same standards to letters as I do to lengthy, peer-reviewed articles.
Six years after the original diet assignment, 67% of participants say they stuck with the original diet they were given. They only signed up for a two-year study anyway. Was compliance monitored between years 3 through 6? No. The 33% of participants who admitted going off diet were still lumped into their original assignment cohort for analysis purposes.
I guarantee you that long-term compliance with the assigned diet was low.
When I see these discouragingly small weight losses after six years, I remind myself that they are only averages. Probably 20% of dieters did indeed lose and maintained clinically significant amounts of weight.
-Steve
Steve, I don’t disagree with any of your points. I’ve long said that compliance is the real issue. I think trying to maintain any given weight loss diet in our SAD environment is a bit like trying to stay dry in the middle of lake in a boat with a hole in it. Some may do better than others, but maybe learning to swim is a better strategy than perpetually bailing!