A new network meta-analysis published in Nutrients has reordered the standard advice on caffeine dosing for endurance athletes, finding that low-dose caffeine capsules at approximately three milligrams per kilogram of body weight produced the largest mean improvement in time-trial completion time of any administration strategy tested. The headline finding: a 2.2 per cent reduction in time-trial duration against placebo, comfortably ahead of moderate-dose capsules at 1.8 per cent and moderate-dose caffeinated gum at 1.6 per cent.
The review pooled studies that compared caffeine doses and administration routes across cycling and running time-trials of varying length, with the authors using a network design to rank protocols against one another rather than only against placebo. Capsules consistently outperformed gum and chewable formats at matched doses, in line with established pharmacokinetics that have peak plasma caffeine concentration arriving 45 to 60 minutes after ingestion when caffeine is absorbed through the gut. The data also reinforced earlier work that found doses above six milligrams per kilogram returned diminishing performance gains while sharply lifting the risk of adverse effects such as palpitations, anxiety and gastrointestinal disturbance.
For a 70 kilogram marathon runner, the new analysis points to an effective dose of around 210 milligrams of caffeine taken roughly an hour before the gun, which is comfortably inside the range produced by two strong cups of coffee or a single pre-race gel and tablet combination. The mechanism is unchanged from earlier consensus statements: caffeine antagonises adenosine receptors in the central nervous system, lowers perceived exertion, sharpens motor unit recruitment in trained muscles and supports fat oxidation, all of which combine to extend time to exhaustion at a given physiological cost.
Marathon practitioners often layer a second dose into the race itself, with many of the elite fuelling protocols published over the last two years calling for caffeine-loaded carbohydrate gels in the second half of the race, typically one between 20 and 25 kilometres and a second around 35 kilometres. The new meta-analysis does not directly model split-dose race-day strategies, but its finding that low pre-race doses outperform large ones is consistent with race-day protocols that frontload a small priming dose and rely on in-race gels to maintain plasma concentration through the back half of the marathon.
The caveats remain familiar. Individual variability in caffeine response is large and is driven in part by the CYP1A2 gene that codes for the liver enzyme metabolising the drug, so fast metabolisers may need to chase the upper end of the recommended range while slow metabolisers will get more from a conservative dose. Sleep quality on the night before a goal race is also routinely worsened by caffeine taken late in the day, and athletes are still advised to dial in any new caffeine plan well in advance of race day. None of that contradicts the headline finding, but it is a reminder that the optimal dose is the one that works for the runner taking it, and that the new analysis is best read as a starting point rather than a prescription.