Skip to content

Beta-alanine showed no clear benefit for repeated sprint performance

A meta-analysis found no statistically significant improvement in average sprint performance, peak performance or fatigue across repeated sprints.

By 100HP editorialAbstract-based explanation checked

Based on the published abstract. The full paper may contain additional methods, results and limitations.

The 30-second takeaway

Beta-alanine did not clearly improve repeated sprint ability compared with placebo. The pooled results were nonsignificant for average performance, peak performance and fatigue decrement, meaning the fall in performance across sprints. These findings do not prove there is no effect, and they apply to repeated sprinting rather than exercise performance in general.

Keep in mind

The abstract does not provide individual protocols or evidence-certainty ratings. This abstract-based account cannot fully assess confidence in the pooled results.

THE NUMBERS, WITH CONTEXT

What researchers found

SMD -0.018 (95% CI -0.207 to 0.170)

Mean repeated sprint performance

Beta-alanine versus placebo after at least 2 weeks. Hedges' g uses standard deviation units, based on how spread out the scores are. Zero means no average difference; the confidence interval spans zero.

Front Nutr, 2026 · Original source ↓

Reading the pooled estimate

The analysis accounted for multiple results from the same participant groups. The average-performance estimate was close to zero, but its confidence interval included values on either side of no difference.

Subgroups left uncertainty

Analyses did not detect significant differences in results by total dose, supplementation duration, exercise type or training status. That does not establish that all doses or protocols work equally well.

CHECK THE ORIGINAL

The original publication

No ergogeniceffect of β-alanine on repeated sprint ability: a systematic review and multilevel meta-analysis of randomized controlled trials.

Liang W, Kong D, Wang Y et al.
Front Nutr · 2026

PubMed ID
41971372
Record checked

AI-assisted research and writing. This explains one selected publication; it is not a complete review of everything known. Our approach.