Our Longevity Diet

Continuing our theme on intermittent fasting studies done on human subjects, today we look at a study called Alternate-day fasting in nonobese subjects: effects on body weight, body composition, and energy metabolism, published in January 2005. In this study, eight men and eight women practiced alternate day fasting — quite literally — they ate one day, then fasted from midnight that day to the following midnight. The study lasted three weeks.

Before describing the results, let me jump to the author’s conclusion:

Conclusions: Alternate-day fasting was feasible in nonobese subjects, and fat oxidation increased. However, hunger on fasting days did not decrease, perhaps indicating the unlikelihood of continuing this diet for extended periods of time. Adding one small meal on a fasting day may make this approach to dietary restriction more acceptable.

This is almost exactly the type of diet Isabel and I follow, excepting the timing of the fast. I can just visualize those subjects waiting for midnight, then eating a big meal just before sleeping. Knowing they would fast again the following day, they probably ate a lot again just before going to sleep at the end of their eating day. So they slept with a full stomach — but were awake (and hungry) during all the difficult hours of the fast. As I’ve stated before, making this regime comfortable requires careful selection of the timing for your fast!

So, what were the results? Well, the subjects lost, on average 2.5% of their initial body weight, even though they were not restricted in how much they could eat on their eating days. In fact they were told they would need to eat ‘twice as much’ as normal to maintain their weight. Hunger, not surprisingly, did not decrease over time. I don’t know why anyone would expect that it might. Hunger doesn’t increase either — one is either hungry or not — with persistent intermittent fasting you get used to being hungry at times, but knowing food is coming soon makes it more bearable.

Measures of resting metabolic rate, respiratory quotient, glucose and ghrelin levels were unchanged, excepting that respiratory quotient decreased on the last day, which involved fasting longer than the usual 24 hours (the researchers obviously did not want the inconvenience of needing to test the subjects at midnight to get fasted results, so they had to fast until 7:00 AM the next morning as well). The only big change, besides the weight loss, was in insulin, which was down an average of 57% after the 31 hour terminal fast.

Obviously, this study was highly flawed in design. There was no measurement of what or how much the subjects ate on their eating days. Subjects even reported eating more than usual on their last eating day because they knew they needed to fast longer than usual afterwards! That alone undoubtedly affected all of the measurements taken.