Weight, body composition, and adherence

Three different numbers

Weight is what a bathroom scale reports: body fat, lean tissue, water, stored carbohydrate (glycogen), and what is in the gut. A two-kilogram drop in week one is often not two kilograms of body fat.

Body composition splits that mass, usually into fat mass and fat-free mass (muscle and other non-fat tissue), when a trial uses a dual-energy X-ray scan (DEXA), a similar scan, or at least waist circumference. Many papers never do that. “Lost weight” is then allowed to stand in for “lost body fat,” which it is not.

Adherence is whether people actually kept the assigned eating window or fast-day calendar, and whether they stayed in the study. A plan that works only when staff watch every meal is a different experiment from a window people keep at home. People who quit because they could not keep the hours are left out of the final average, so that average can look better than the same plan would look if everyone had stayed.

What was typically studied

Weight is the default main outcome in time-restricted eating (TRE), alternate-day fasting (ADF), and 5:2 trials. Body fat and waist appear when someone had a scanner or a tape. Lean mass is measured less often and less carefully. Hunger, binge-restrict cycles, sleep, and mood decide whether a window survives a bad month, but trials report those less often than they matter.

The usual enrollment is adults with overweight or obesity, for weeks to a few months, in groups of tens to a few hundred people. In 2025, Zhila Semnani-Azad, Frank Hu and colleagues reviewed 99 randomized trials (6,582 adults) comparing intermittent fasting, continuous calorie cutting, and unrestricted eating. Those trials still last weeks to about a year. They are not a decade of follow-up, and they are not a sample of everyone who wants to get leaner.

Almost no consumer-facing fasting trial is designed to answer whether this prevents disease or keeps the weight off for years. What happens after people stop the assigned hours is rarely the number in the headline.

What trials tend to find on the scale

Compared with eating whenever you want, intermittent-fasting styles usually produce a modest drop in body weight, typically a few kilograms or a few percent of body weight over weeks to about three months. Part of that comes from quieter intake: many people skip evening snacks without counting calories.

Compared with ordinary calorie restriction (eat less, no special clock), daily time-restricted eating is often similar. Semnani-Azad’s 2025 review found time-restricted eating and continuous restriction both beat unrestricted eating. Differences between those two were small, especially in trials that lasted 24 weeks or more. Alternate-day fasting was the only intermittent style with a small extra average drop versus continuous restriction in shorter work, on the order of about 1.3 kilograms. That extra difference mostly faded in the longer trials.

A window can help compared with no plan, and still not outrun eating less on a regular schedule. Headlines usually quote only one of those two findings. A kilogram on a group average is not a promise to you. Some people lose more. Some eat extra on feast days or loosen the window and lose little, and the average does not show either group.

Body fat, lean mass, and the first-week drop

When composition is measured, body fat usually falls along with weight. Lean mass can fall too. In 2026, Kaijun Xing, Yannan Zhang and colleagues pooled 28 randomized trials by age. Fat-free mass was down under both daily windows and intermittent energy restriction, on the order of about a kilogram in the pooled estimates, with more certainty in some age groups than others. Not every time-restricted-eating subset shows a lean-mass drop. Researchers do not all agree, and scanners are imperfect.

That is not the same as saying fasting melts muscle. Eating less energy than you use does not automatically spare lean tissue. Protein when you eat still matters, and so does some lifting if that is already part of your life, and so does not treating a tighter window as a reason to under-eat. Exercise and fasting covers training. The numbers here are about weight and the tissue that was measured, not a workout plan.

Early movement on the scale is often stored carbohydrate and water. Glycogen holds water, so a longer overnight gap or a first hard week can drop the number before much body fat has moved. That is ordinary physiology, not a special “whoosh” that proves the plan is working, and it is not a number to chase week after week.

Staying with the hours is the practical result

A window that people keep beats a stricter calendar they abandon. In several reviews, the share of people who leave intermittent-fasting groups is similar to the share who leave ordinary diet groups. That does not make dropout unique to fasting, and it does not prove that everyone will stick with 16:8.

At home the eating hours often slide later, feast days overshoot the gap created by fast days, assigned fast days get skipped, and the last two hours of a daily fast become a negotiation. Clinic kitchens and daily check-ins hide that drift. Ordinary life does not.

Staying with the hours is also why longer trials look quieter than eight-week ones. Weight often looks better in short studies. By six months, the extra gap that made “fasting beat dieting” in the short work shrinks, which is what most diet studies show after the research staff stop watching. A pattern you can still recognize in three months, the subject of Building a sustainable fasting rhythm, is closer to that evidence than a perfect two weeks.

Limits (read these as part of the result)

The people in these trials are mainly adults with extra weight. Pregnancy, underweight, eating-disorder history, and complex disease are often excluded, so a mean drop in that group is not a goal for everyone. See Who should be careful with fasting.

Weeks are not years, and regain after the assigned window ends is poorly reported. A bathroom scale is not a DEXA scan, waist is not visceral fat, and a home Bluetooth scale or a “body fat percent” app is not a research instrument.

Calories still explain most of the weight change. Eating less because evening snacks stopped is not the same as the clock uniquely burning body fat. Time-restricted eating does not fix an eating window full of ultra-processed food or too little protein. Several trials controlled when people ate more tightly than what they ate.

Dizziness, sleep disruption, binge-restrict cycles, and medicine issues still appear in real life. Trials that exclude high-risk groups will not show those stories. When to stop a fast still applies, including when the scale is moving the way you wanted.

What this does not mean

Intermittent fasting is not a superior method for losing body fat, and longer hours are not a higher dose. 18:6 will not automatically beat 16:8 on the scale, and alternate-day fasting is not the only version that counts. You do not need to weigh every morning, and a flat week is not a failed plan, because water, hormones, meals, and training all move the number, and a lower weight is not always better.

This timer should not become a diet app. History on this phone or computer records fasts. It does not record kilograms, and we will not add a weight log to chase this literature. If weight is a medical topic for you, take it to a clinician rather than to a stage name or a share card.