Moderate caffeine dose beat low dose in time trials

Caffeine is one of the most used pre-exercise aids, and ergogenic effects have been reported across a wide dose range of 2-9 mg·kg-1. The trouble is inconsistency: reported time-trial performance has ranged from about -3% to +16%, so it has been unclear how much anyone actually needs to take. Because higher doses may increase adverse effects without clear additional benefit, the practical question is the shape of the dose-response rather than whether caffeine works at all. This review set out to compare low, moderate and high doses directly against placebo on a single, hard outcome.

The authors ran a systematic review and meta-analysis of randomised, placebo-controlled trials found in PubMed, Embase and the Virtual Health Library. Eligible trials involved healthy adults aged 18-59 who acutely ingested oral anhydrous caffeine before an aerobic time-trial test, and only time-to-completion outcomes were accepted. Doses were grouped as low (≤3 mg·kg-1), moderate (4-6 mg·kg-1) and high (>6 mg·kg-1). Results were pooled as standardised mean differences with 95% confidence intervals under random-effects models, and risk of bias was assessed with the Cochrane tool. Forty-eight studies covering 689 participants met the criteria.

Both low and moderate doses significantly reduced time-trial completion time compared with placebo. The low-dose pooled effect was an SMD of -0.27 (95% CI -0.44 to -0.11; p = 0.001), described in the paper as spanning 1.3-3 mg·kg-1. The moderate-dose effect was larger at -0.52 (95% CI -0.77 to -0.28; p < 0.0001) and was characterised as more consistent. Subgroup analysis showed similar benefits from moderate doses in both trained and highly trained participants. No study using more than 6 mg·kg-1 reported a time-to-completion outcome, so the high-dose group could not be analysed at all.

This is presented as the first meta-analysis focused specifically on aerobic time-trial completion time, which narrows the outcome compared with earlier syntheses that mixed power, time-to-exhaustion and work-done measures. It supports the direction of travel in sports nutrition towards smaller doses, since even the low band beat placebo, while suggesting the moderate band is the more dependable choice. It does not settle whether going higher adds anything, because no eligible trial tested that. Open questions include habitual intake, individual variability behind that -3% to +16% spread, and tolerance of side effects.

The pooled picture is that caffeine's endurance benefit does not require large amounts, and that the moderate band delivered the steadier effect in these trials. Effect sizes were modest rather than transformative, and were measured in laboratory-style time trials rather than races. Trained and highly trained participants responded similarly to moderate doses, so being fitter did not appear to remove the effect. The evidence base has nothing to say about doses above 6 mg·kg-1 for this outcome, so pushing higher is not supported by these data either way.