An eight-hour eating window, supervised exercise and their combination did not produce a statistically significant additional reduction in liver fat compared with usual care in a new randomised trial. That is the primary finding of the TEMPUS study published in Nature Communications on October 14, 2026.
The result is more specific than saying fasting never helps or exercise does not work. It concerns one measured outcome, a twelve-week period and a particular comparison in adults with obesity. Understanding the comparator is essential before turning the headline into advice.
Four groups, not a treatment-versus-nothing comparison
The trial randomly assigned 187 adults with obesity to four groups: usual care, time-restricted eating, supervised exercise, or both interventions. All 187 were included in the primary intention-to-treat analysis. Usual care included Mediterranean-diet and physical-activity education, so the comparison group was not simply told to remain inactive and change nothing.
Time-restricted eating used an eight-hour eating window. The primary outcome was the change in intrahepatic fat fraction, measured using magnetic resonance imaging after twelve weeks. In plain language, the researchers were assessing fat within the liver, not merely a change on the bathroom scale.
The trial's outcome assessors and statisticians were blinded to group allocation, although the people delivering the interventions were not. That distinction is common in lifestyle research: participants can usually tell whether they are attending supervised exercise or following an eating window.
What the primary result supports
The reported between-group comparisons did not show a statistically significant advantage for time-restricted eating, exercise or their combination over usual care on the primary liver-fat measure. Nor did the trial establish a significant difference among the intervention groups on that outcome.
That finding should be reported directly, even when it is less dramatic than a breakthrough promise. It prevents us from assuming that combining two plausible lifestyle approaches must automatically deliver a larger effect on every target.
However, a result that does not reach statistical significance is not proof that all possible differences are exactly zero. The estimates have uncertainty. Nor does one primary outcome tell the whole story about every possible effect on fitness, eating habits, weight or longer-term health.
Why the outcome and timeframe matter
Liver fat, body weight and everyday energy are different outcomes. A study designed around one cannot be treated as a complete verdict on all three. Before accepting a claim about a diet or exercise programme, ask what was actually measured and which comparison produced the result.
Twelve weeks can answer a useful short-term question without establishing what happens over years. The participants also had specific eligibility criteria. The study does not automatically describe every age group, every health condition or people following a different kind of eating schedule.
Likewise, a supervised exercise programme is not identical to any activity someone calls exercise at home. The details of a research intervention matter when comparing it with another study or with ordinary life. An appealing label alone is not enough to assume the same intervention has been tested repeatedly.
Put it beside other research without mixing the results
Time-restricted eating has been studied using different populations, schedules and outcomes. An older NIA report, for example, discusses a separate study of eating-window timing and weight-related outcomes. That is background, not a replication of the October TEMPUS liver-fat result.
Different findings can reflect different questions rather than a simple contradiction. To compare studies responsibly, keep the population, comparator, intervention and outcome in view. Otherwise a weight-loss result from one experiment can be used to imply a liver-fat result that another did not show.
This is also why a meal schedule should not be treated as a universally appropriate self-directed treatment. People using medicines or managing health conditions may need individual advice before changing eating patterns. The news is evidence to understand, not a prescription to copy a research protocol.
The useful next question
Future research can clarify who responds, whether a longer intervention changes the comparison and how different goals should be assessed. Those are open questions, not findings we can add to the present paper.
For now, the advance is a clearer test of a popular claim. In this trial, the additional interventions did not demonstrate superiority over active usual care on the primary liver-fat outcome. Reporting that boundary accurately is part of progress, even when the answer is less simple than a promise about fasting and exercise.