News · Fitness & Exercise
Intervals ranked best for belly fat, but more was not better
Pooling 60 trials in 3,308 adults, researchers ranked four kinds of exercise against the fat packed around the organs. Intervals came top, and the dose-response curve turned back on itself past a moderate weekly load.
Based on 60 randomized trials in 3,308 adults, pooled and ranked against each other
- 60 randomized trials in 3,308 overweight or obese adults, pooled into one ranking.
- All four kinds of exercise cut visceral fat, meaning the deep fat around the organs; intervals ranked highest.
- The ranking gap was small: intervals 72.5%, combined training 69.1%, aerobic 59.0%.
- Dose response was U-shaped, best around 1,100 MET-minutes a week, not at the maximum.
- Rankings are probabilities of being best, not proof that one method beats another.
The fat that matters most for metabolic health is not the fat you can pinch. It is the deeper deposit packed around the liver and intestines, called visceral fat, and it responds to exercise even when body weight does not move much.
BMC Sports Science, Medicine and Rehabilitation published a network meta-analysis on the question, meaning a pooling method that can place treatments in order even when they were never tested against each other. It set out to find which kind of exercise shifts visceral fat best, and how much you need.
What was pooled
A total of 60 RCTs involving 3308 overweight or obese adults were included, where RCT means a randomized controlled trial.
That indirect linking buys a ranking no single trial could produce, at the cost of leaning on assumptions about how comparable the trials are.
Four approaches were compared: interval training, aerobic exercise, resistance training, and the two combined.
What it found
The first result is the one to hold onto. All four exercise interventions significantly reduced VAT in overweight or obese adults, writing VAT for that visceral fat.
Nothing failed. The argument is only about order.
The ranking
On the ranking measure, high-intensity interval training (HIIT) had the highest cumulative probability at 72.5%, followed by combined aerobic and resistance training (COM) at 69.1%, then aerobic exercise at 59.0% and resistance training at 49.5%.
Those numbers are probabilities of being the best option, not effect sizes. A gap of three percentage points between the top two is not a meaningful difference between two training methods; it is two methods that could swap places on the next trial.
The dose
This is the more interesting half. Dose-response analysis indicated a nonlinear U-shaped association between total exercise dose and VAT outcomes, with an optimal dose of 1100 MET-min/week.
A MET-minute is a way of combining how hard an activity is with how long it lasts, so that a short hard session and a long easy one can be compared. Around 1,100 of them is roughly four to five hours of brisk walking, or rather less if the work is harder.
The shape is what makes it useful. A straight line would say more is always better. A U-shaped curve says the benefit builds, flattens, and then falls away again, and the authors read it as suggesting that a moderate exercise load is beneficial for reducing VAT.
What this does not settle
The ranking is the weakest part of the paper and will be the most quoted. Rankings of this kind are probabilities, they are sensitive to which trials exist rather than which programs work, and a four-way ordering built from 60 heterogeneous trials should be held loosely.
The turn at the top of the dose curve deserves particular caution. Very high training doses appear in fewer and smaller trials, so the falling-off is estimated from the thinnest part of the data. That the curve bends is a reason not to assume more is always better. It is not evidence that hard training does harm.
And these were overweight or obese adults, in programs lasting months. None of that transfers automatically to someone lean, or to a program measured in years.
The practical shape of it
Every kind of exercise tested moved the fat that matters, which is the finding with the most evidence behind it and the least drama.
If you want the ranking, intervals came first by a nose. If you want the number, a moderate weekly load did more than a punishing one. Both point the same way: the version you will actually keep doing is the version that works.
People also ask
What is visceral fat and why does it get singled out?
It is the fat stored deep in the abdomen, packed around the liver, pancreas and intestines, as opposed to the fat directly under the skin. It is singled out because it tracks more closely with insulin resistance, type 2 diabetes and heart disease than total body weight does. Two people at the same weight can carry very different amounts of it.
How much exercise is 1,100 MET-minutes a week?
A MET-minute combines how hard and how long. Moderate activity such as brisk walking runs around 4 METs, so 1,100 MET-minutes lands near four to five hours a week of that. Vigorous work counts faster, so the same total takes less time. The reason to think in MET-minutes at all is that it lets trials with very different programs be compared on one axis.
Should I switch to interval training then?
The ranking favors it, but not by much and not on firm ground. Interval training scored 72.5% on the ranking measure, combined aerobic and resistance training 69.1%, aerobic alone 59.0% and resistance training 49.5%. Those are close enough that the ordering could shift with a few more trials. The finding that all four worked is more solid than the finding about which worked best.
Why would more exercise stop helping?
The authors report a U-shaped curve rather than a straight line, with the effect strongest around a moderate load and weaker at the extremes. Caution is warranted in reading that: the trials at very high doses are fewer and smaller, so the upturn at the top of the curve rests on thin data. It is a reason not to assume more is always better, not evidence that hard training is counterproductive.
Does this mean exercise beats diet for belly fat?
The study cannot say. Every trial in it compared an exercise program against a control, not against a diet, and diet was not the variable under test. What it establishes is that exercise moves visceral fat on its own, which is worth knowing for anyone whose weight has not shifted.