Supplement bottles line up on kitchen counters next to the evidence of why they were bought — a skipped meal, a protein bar standing in for dinner, a week where meal prep didn't happen again. The quiet deal millions of people make with themselves: the diet isn't great, but at least the supplements are covering part of the gap.
Seven major supplement categories have been tested against that deal. The evidence is in.
Supplements Cannot Replace a Bad Diet
Most supplements do not compensate for a poor diet. Five of seven major supplement categories — including fat burners, testosterone boosters, BCAAs, omega-3, and collagen — showed zero reliable body composition benefit across large-scale analyses. The two that work, creatine and caffeine, produce small, genuine effects that supplement an existing diet but cannot replace what food provides.
— Clark & Welch 2021 + 6 additional meta-analyses · 7 supplement categories reviewed
Fat burners went first. Across twenty-one studies, every body composition measure returned the same verdict: no reliable effect in any direction. The supplements designed to speed up the metabolic engine produced nothing measurable. Some studies found something worse — a chronic decrease in metabolic rate after supplementation.
The category built to raise your metabolism may actually lower it
Testosterone boosters carried the widest disconnect between what the shelf promises and what the evidence delivered. Ninety percent of products claim to boost testosterone. When the ingredients were checked against published research, 24.8% had any evidence of raising it, and 10.1% may actually decrease it. More than 61% had no human data at all.
BCAAs produced one meaningful body composition result across every trial that measured it. Omega-3 supplements — commonly stacked for muscle recovery — showed an effect on muscle protein synthesis so close to zero the supplement group and the placebo group looked the same. Collagen peptides showed a small lean-mass signal, but at certainty levels rated low to very low.
Five of seven categories. Zero reliable body composition benefit.
Two survived. Creatine added 0.82 kg of fat-free mass — the strongest evidence in the entire supplement landscape — a consistent finding across every age group, both sexes, trained and untrained. But part of that gain may be water retention rather than actual muscle tissue, and the effect appears only alongside resistance training. Caffeine produced a genuine strength boost — real enough to detect across trials, too small to feel on any given set — entirely a performance effect, not a nutritional one.
Both supplements work. Both are small enough that framing either one as nutritional insurance requires ignoring what the numbers actually say.
The creatine finding deserves its caveat out loud: when the strongest supplement evidence on the planet amounts to less than one kilogram of lean mass, and part of that may be stored water, the ceiling for what any supplement can deliver comes into sharp focus.
The deal was always one-sided. The bottles covered the feeling of progress — the ritual of doing something right when the meals didn't happen. The gap they were hired to fill is still open, and it only closes one way: the meals themselves get the attention the bottles were getting.