Hobson · Amino acids 2012 · meta-analysis · n=360 participants (15 studies)

Effects of β-alanine supplementation on exercise performance: a meta-analysis.

Cited 300 times in the scientific literature.

Level 1 - systematic review of randomized trials

Meta-analysis of comparative intervention studies

PubMed 22270875 · doi:10.1007/s00726-011-1200-z · record verified 2026-08-29

What was done

A meta-analysis was conducted across 15 published manuscripts meeting strict exclusion criteria to evaluate the ergogenic effects of β-alanine (BA) supplementation. The analysis included 57 outcome measures from 23 exercise tests across 18 supplementation regimes, comprising 360 total participants (174 in the BA group and 186 in the placebo group). Outcomes were analyzed by exercise type (capacity vs. performance) and exercise duration (<60 s, 60–240 s, and >240 s).

What was found

BA significantly improved exercise outcomes compared to placebo (P = 0.002), with a median effect size (interquartile range) of 0.374 (0.140–0.747) for BA versus 0.108 (-0.019 to 0.487) for placebo. The median improvement was 2.85% (-0.37 to 10.49%) with a median total supplemented dose of 179 g of BA. - Exercise capacity significantly improved with BA compared to placebo (P = 0.013), whereas exercise performance did not (P = 0.204). - Performance in exercise lasting 60–240 s (P = 0.001) and >240 s (P = 0.046) was significantly improved with BA, while exercise lasting <60 s showed no benefit (P = 0.312).

Why it matters

This review provides quantitative evidence that β-alanine supplementation offers a modest (~2.85%) performance benefit, primarily in high-intensity exercise tasks lasting longer than 60 seconds and specifically for capacity rather than performance tests.

Limits

The total sample size was small (360 participants across 15 studies). Included studies featured substantial heterogeneity in dosing regimes (18 regimes) and testing protocols (23 tests). Participant characteristics, training status, safety/side effects, and study quality were not reported in the abstract.

Cited by