Does time-restricted eating (otherwise known as ‘intermittent fasting’) work? Like most things, the answer is ‘it depends’. When a person asks me if s/he should add TRE into their dietary routine, I ask a simple question: are you the type of person who can go several hours without eating and then move into a steady state of consumption? Or are you the type of person who will have an all-out binge once you enter your eating window? The honest answer to this question will likely determine your outcome of TRE.
Enter science. In a recent paper released on TRE, the researchers concluded that time-restricted eating was not significantly more effective for weight loss than the control group. The control group was allowed to eat three set meals per day + snacking, and the TRE group had an 8-hour eating window where participants were allowed to eat ‘ad-libitum.’ In other words, the control group was allowed to eat what they wanted, in whatever quantity they wanted, as long as it fell within the set eating window. I’m not even going to get into the issue with telling and overweight or obese person s/he can consume as much as s/he wants of whatever s/he wants for a significant period. I am more concerned about simplifying group conclusions and believing we can extrapolate results at the individual scale.
The study was well done, but here is the issue with studies: The nuance that matters on an individual scale is rarely assessed. When it comes to weight loss, you are going to get lots of noise and little signal. The noise (as you can see below) concentrates on the median. Why do I see the vast majority of participant results as ‘noise’?
Because the vast majority of individuals will not succeed with any diet and exercise strategy, regardless of the potential efficacy. Compliance is not standard, and variance will be high. For instance, any two individuals in the TRE group will be seen as equal participants in the study. But what is ever equal about two people? Person A may join the TRE arm of the study and: eat junk food, not exercise, drink, smoke, and possibly lie about his or her actions. Person B may: eat well, exercise, and not binge himself sick during the eating window. This isn’t an issue that can be easily controlled, and researchers will suggest that the point of ‘averages’ is that outliers on both sides of the coin come out as a wash. Therefore, the average result tells the story. But does it?
Retake a look at the image. What I want to know is who the outliers from the TRE group in both the weight loss and weight gained sections of the plot? What is shared between the TRE participants who saw great success and the participants who gained weight? If you look closely, there are significant groupings of TRE participants on both sides. While both the time-restricted eating and the control participants converge in the middle, TRE participants show the best and worst results on the edges.
Why is this important to note? A significant group of people succeeded with a TRE intervention, and a considerable amount who did quite poorly. Knowing the commonalities within each contrasting group is where the rubber meets the road. Knowing this, we can say ‘if you are like person A, you are more likely to succeed with intervention ‘X.’ If you are like person B, you will most likely fail with this intervention.’ This goes for all diet and exercise interventions. It is less helpful to know the average result because ordinary people don’t stick to diet and exercise interventions very well. A person will carry an element of shame in not sticking to the study’s outline, so s/he (given the opportunity) will lie about his or her actions. You’re left with a very unhelpful average, but there could be powerful information within the outlier groups if you look deeply enough.
Here’s an analogy. Imagine a highly successful entrepreneur ran a 12-month long course on small business creation for 1000 people. The course is intensive and requires 10 hours per day of dedicated work. After a year, course members launch their small business idea built from the strategies learned in the course. Chances are their will be a handful of course participants (let’s say, 50) who see incredible success. An equal amount who absolutely fall flat, and 900 individuals who start mediocre small businesses. Would you conclude that the course leader was ineffective and his methods were worthless, or would you be interested in what was common between the 50 people who absolutely crushed it? Chances are, those 50 people actually showed up, listened intently, and did the work.
I don’t blame researches for the challenge of dietary interventions. It is an almost impossible task. I also understand that the point of focusing on averages is to put forth conclusions for the broad population. But if you’re in the broad population, nothing short of surgery will work for you. By definition being in the average population aggregates you into the demographic of people who continuously struggle to take care of themselves. I believe some responsibility should be taken in how researchers choose to interpret the results. It looks like a form of scientism to boil results down to averages when we could be focusing on what is common between those who see great success in any intervention. Science needs nuance. Even some of those involved in the study say that they are moving away from TRE (a method of dietary control that they used themselves) due to the results of the study. This is the epitome of scientism. ‘I do this thing, and it seems to give me great results so I’ve done it for many years, but now according to this highly limited study it is meaningless, therefore, I must stop.’