Intermittent 16:8 fasting doesn't work for weight loss, UCSF study claims
A new UC San Francisco study is challenging popular dietary dogma by suggesting a common type of intermittent fasting does not work for weight loss. The randomized clinical trial tested the 16:8 method of time-restricted eating and found it was no more effective for losing weight than eating throughout the day.
Time-restricted eating (TRE) is a form of intermittent fasting where, instead of interspersing whole days of fasting across a week or month, a person’s daily caloric intake is limited to a short window of time within a given 24-hour period. The most common TRE method is known as the 16:8 plan, restricting everyday eating to a consistent eight-hour window.
UC San Francisco researcher Ethan Weiss first became interested in TRE after reading some of the earliest research on the subject from Salk University biologist Sachin Panda. The idea of TRE initially arose through mouse experiments. When the animals’ food intake was restricted to a small window every day, they ended up thinner and healthier than animals eating the same volume of calories across a longer period of time.
Weiss was so convinced of the benefits of TRE from these early investigations that he began doing it himself. But, as a scientist, Weiss was uneasy. There was very little rigorous clinical study in humans showing TRE worked. So he recruited PhD student Dylan Lowe to helm a human clinical trial to find out whether TRE, particularly the 16:8 plan, actually led to increased weight loss.
Over the last couple of years a few preliminary human clinical trials have investigated the effects of TRE. Most of the research so far has been promising, finding the dietary strategy may indeed promote weight loss, alongside a number of other metabolic benefits.
However, many of these cursory human trials saw weight loss from TRE correlating with a drop in daily caloric intake. The majority of subjects who shift to TRE seem to also ultimately eat less. On top of that, the majority of prior clinical trials on the topic did not utilize a control group, making it difficult to measure the true effect of the dietary intervention.
“Each study asks one or two very specific questions, and at best (and if designed properly), will only answer those specific questions,” says Lowe, explaining why his new research seems to offer contradictory results to prior TRE studies. “This study asked a different question than previous studies and is just one more piece to the larger puzzle.”
So, the specific focus of this new trial was to test the effect of a TRE intervention, in real-world conditions with no other behavioral or dietary recommendations. Plus, the trial would utilize a robust control arm.
The trial followed 116 overweight or obese adults, who were randomly split into TRE and control groups. The TRE group was simply directed to continue eating as normal, but only during the window of 12 pm to 8 pm. The particular time-frame was chosen based on the assumption compliance to the regime would be best if the window was not too disruptive to a routine social life.
Of course, blinding is impossible in a dietary study such as this, so the control group would clearly know they were receiving no intervention. Weiss says, to avoid this inability to blind the trial, they used some creative language when communicating with the control cohort. To give the impression of an intervention being studied, the control group was dubbed the "continuous meal timing group."
Twelve weeks later and the results were in – the TRE group had successfully lost an average of 2 lb (0.94 kg), but the control group had also lost weight, an average of 1.5 lb (0.68 kg).
“This is important,” notes Weiss. “If this had been an uncontrolled study, we'd be saying there was a significant decrease in weight. But there was that pesky control arm. We know that people in clinical trials [tend] to lose weight no matter what you do. Including the control arm was important.”
Interestingly, the results not only showed no marked difference in weight loss between the two groups, but no difference across a number of metabolic markers. Unexpectedly, the only notable difference was the TRE group seemed to lose significantly more lean muscle mass than the control. This is admittedly a novel finding and needs much more future validation, but Weiss suggests it does raise concerns.
“Focusing just on weight loss may lead one to assume benefit when there could be harm,” says Weiss. “The lean mass example is a good one. Losing 1 kg (2.2 lb) of mass =/= a kg of fat. Indeed, if one loses 0.65 kg (1.4 lb) of lean mass and only 0.35 kg (0.8 lb) fat mass, that is an intervention I’d probably pass on.”
As with all clinical trials, there are caveats. It is unclear whether the 12 pm to 8 pm timeframe selected for the trial was an ideal window for harnessing the metabolic advantages of TRE. The researchers note earlier, and shorter, windows may be more suited to TRE, such as a six-hour window spanning 9 am to 3 pm.
Another concern with the study is the fact it did not directly measure the participant’s caloric intake, so it is unclear whether the TRE group ate less than the control. The researchers do suggest their mathematical modeling of energy intake changes suggests there was little difference in caloric intake between the two groups. Plus, it is noted that this particular model has been independently validated as more accurate that self-reported energy intake measures.
So what does all this mean? At the very least Weiss suggests people should ignore any nutrition clinical trial without a robust control group. And while Weiss himself has stopped his personal TRE routines based on this research, he does not claim this study is definitive proof intermittent fasting doesn’t work. Different durations and eating windows may be more effective that what his trial investigated, for example.
“… there is so much left to learn here,” Weiss notes. “We are going to stay at it and try to answer some of the obvious next questions. Really this is just the beginning.”
The new study was published in the journal JAMA Internal Medicine.