Eight Hours or Nine? What Eating Window Trials Actually Show

Two studies published weeks apart this year seemed to say opposite things about time restricted eating, the practice of fitting all food into a fixed daily window. One headline said an 8 hour window was best for weight loss. Another said a 9 hour window was no better than seeing a dietitian. Read closely, they do not contradict each other. They answer different questions, and together they give a fairly clear picture of what an eating window can and cannot do.

The case for eight hours

The first study, published in Nutrition & Diabetes on September 12, pooled individual data from five time restricted eating trials lasting three to six months. Participants logged every meal and drink with timestamps in a smartphone app, which let researchers compare not just the window people were assigned but the window they actually kept.

The pooled group was 124 adults, two thirds women, with a mean age of 47. Before the trials they ate across about 15 hours a day. Ninety five were assigned to an 8, 10 or 12 hour window and 29 to a control arm with habitual eating or brief nutrition advice. Those assigned to 8 hours lost 1.66 kilograms more than those on 12 hours and 2.25 kilograms more than controls. Against a 10 hour window, the difference was 1.02 kilograms, a trend that did not reach statistical significance.

Looking at the windows people actually kept, each hour shaved off was linked to about 0.27 kilograms of extra weight loss. That association weakened once the authors adjusted for major confounders, which they attribute to the small sample. Blood pressure, fasting glucose and cholesterol did not differ by window length.

The case against expecting more

The second study, published in Diabetologia, asked a harder question over a longer time. Evelyn Parr, Leonie Heilbronn and colleagues in Adelaide and Melbourne randomized 247 adults at high risk of type 2 diabetes, with a mean age of 56 and an average body mass index of 32.9. Half followed a self selected 9 hour eating window that had to start no earlier than 6 a.m. and end by 7 p.m., with no other dietary advice. The other half received individualized guidance from a dietitian based on Australian dietary guidelines. The trial lasted 12 months.

The main outcome was HbA1c, a measure of average blood sugar. At four months, the eating window was not worse than dietitian advice, the trial's formal goal, but not better either: the difference was 0.02 percentage points. At 12 months, the result drifted to a 0.05 point difference with an upper bound past the prespecified margin of 0.10, so noninferiority could no longer be claimed. Baseline HbA1c averaged 5.8%, and the absolute changes in both groups were small.

Weight is where the comparison becomes instructive. Both groups lost about 2.5 kilograms by four months and kept it off at 12 months. The eating window group shortened its daily window by about 2.5 hours, which cut intake by roughly 400 kilojoules, or 95 kilocalories, a day. About 92% chose a window of roughly 10 a.m. to 7 p.m., which means a late breakfast and an early dinner. They reported better adherence than the dietitian group, but that did not translate into better health markers.

How the two results fit

The pooled analysis compared eating windows with each other and with doing little. In that setting, a tighter window means fewer eating occasions, and fewer eating occasions usually mean fewer calories, so a dose response is unsurprising. The Australian trial compared an eating window with a real alternative: a dietitian helping people eat better. Against that comparison, the window's advantage disappeared.

The window is a tool for eating less, and a reasonably easy one, since people in the Australian trial found it simple to follow and did not have to learn anything new about food. It is not a metabolic switch.

What to take from it

For someone choosing a strategy, the useful finding is that the two approaches tied. A person who finds rules about time easier than rules about food can expect similar results from a 9 hour window as from conventional advice, at least on the scale. A person hoping the clock itself will repair blood sugar should look at the 12 month HbA1c result, which offers no support.

The appeal of time restricted eating was always its simplicity, and on that it delivers. The claim that it beats ordinary dieting has now been tested for a year against a real comparison and did not hold. For another case where a popular swap was tested properly, see our look at the two year trial of diet soda against water.