Caffeine gum was not invented for sport: it was developed for soldiers who needed alertness back quickly and without water. In recent years it has migrated into race packs — convenient, nothing to wash it down with, easy to carry on the course and spit out afterwards. Demand keeps growing: more and more runners are hearing about the format and looking for it.

On 9 September 2026 Sports Medicine — Open published an updated meta-analysis of this format. It is useful because it states the size of the effect honestly, not merely its existence.

Why gum should work faster in the first place

The difference is not in the dose but in the route of absorption. Kamimori and colleagues (International Journal of Pharmaceutics, 2002) gave 84 healthy men caffeine in gum or in capsules at 50, 100 and 200 mg and took blood 16 times over 29 hours.

Time to peak concentration with gum was 44–80 minutes, with capsules 84–120 minutes.
The absorption rate constant for gum was 3.21–3.96 h⁻¹ versus 1.29–2.36 h⁻¹ for capsules.
For Cmax and total area under the curve there was no difference: allowing for the fact that around 85% of the caffeine is released from the gum base, relative bioavailability for the 100 and 200 mg doses came to 87–90%.

In other words, gum delivers the same amount of caffeine, but earlier — part of it goes through the lining of the mouth, bypassing the stomach. When there are 20 minutes left before the gun, that matters.

What the meta-analysis showed

Miraftabi, Berjisian, Pickering, Lopez-Samanes and Girard selected 25 studies with 390 participants and fitted three-level models.

Overall effect: SMD = 0.195 (95% CI 0.115–0.275; p = 0.001) with very low heterogeneity (I² = 4%). By the accepted thresholds this is a trivial effect size — statistically significant, but practically tiny.

By type of work:

aerobic endurance — a small significant effect (p = 0.002), the best result in the sample;
anaerobic work (p = 0.005) and vertical jump (p = 0.008) — trivial but significant;
strength and strength endurance — no effect (p = 0.372).

Then comes the most useful part: the moderators.

You have to chew for 10 minutes. At 10 minutes the effect is significant (p = 0.001), at 5 minutes it is not (p = 0.091). That follows directly from the mechanism: caffeine leaves the gum base gradually, and a short chew simply does not give it time to be released.

It works in trained people. In athletes the effect is significant (p = 0.018), in the merely physically active it is not (p = 0.676).

And it does not work in people who drink a lot of coffee. There is a significant effect at low (p = 0.003) and moderate (p = 0.013) habitual caffeine intake, and none at high intake (p = 0.290).

Dose matters, timing does not. Meta-regression found a significant relationship with dose (p = 0.031), while the moment of intake relative to the start of the work had no effect on the result (p = 0.683) — which is logical given how fast the absorption is.

What not to believe too firmly here

The quality of the evidence is low. The authors say it outright: most of the studies carry some concerns or a high risk of bias, and confidence under GRADE ranges from low to very low.

The samples are tiny. 390 people across 25 studies averages out at 15–16 participants each — a typical size for a crossover in sports science, but too few for reliable subgroup conclusions.

A trivial effect on average may be zero for you personally. An SMD of 0.195 is a shift of about a fifth of a standard deviation. Over a race distance it gets lost among the weather, the pacing and last night's sleep.

Some subgroups could not be checked for publication bias because there were too few studies.

The authors' own conclusion sounds accordingly: the improvements are statistically significant but trivial, with limited practical relevance — although in particular competitive situations even a gain that small may give an edge.

How to apply this

Treat gum as a delivery route, not as a separate supplement. The ISSN position on caffeine (Guest and colleagues, 2021) is 3–6 mg per kg of body mass; gum does not change the dose, it changes the speed. A detailed look at doses is in the article on caffeine before a race.
Chew for 10 minutes, not 5. This is the single parameter of use that separates a working protocol from a non-working one in the data.
Do not count on it if you drink 4–5 cups of coffee a day. In the high habitual intake subgroup there is no effect. If you want your sensitivity back, you have to cut your intake in advance, not on race day.
One piece is usually 100 mg. A 70 kg runner needs about two to reach the bottom of the ISSN range. Count the total dose together with your gels: many gels already contain 50–100 mg of caffeine, and your race nutrition plan is worth checking against that total.
It is most convenient where there is no water: the last minutes in the start pen, the late kilometres of a trail, the bike leg before the run in a triathlon.
Do not expect a gain in strength work — in that domain no effect was found at all.

The bottom line

Gum delivers the same caffeine faster: peak concentration at 44–80 minutes versus 84–120 for capsules, with comparable overall bioavailability.
The meta-analysis of 25 studies and 390 participants: an overall effect of SMD 0.195 — statistically significant but trivial; aerobic endurance looks best of all.
The effect is there with 10 minutes of chewing and absent with 5; present in trained athletes and absent in the merely active; present at low and moderate coffee intake and absent at high.
The moment of intake relative to the start does not affect the result, while the dose does; confidence in the evidence under GRADE runs from low to very low.

Sources: Miraftabi H., Berjisian E., Pickering C., Lopez-Samanes A., Girard O. “Fast-Acting Caffeine Strategy: A Systematic Review and Meta-Analysis of the Ergogenic Effects of Caffeine Chewing Gum on Physical Performance”. Sports Medicine — Open, 2026;12. DOI: 10.1186/s40798-026-01104-y · Kamimori G.H., Karyekar C.S., Otterstetter R. et al. “The rate of absorption and relative bioavailability of caffeine administered in chewing gum versus capsules to normal healthy volunteers”. International Journal of Pharmaceutics, 2002;234(1-2):159-167. DOI: 10.1016/S0378-5173(01)00958-9 · Guest N.S., VanDusseldorp T.A., Nelson M.T. et al. “International society of sports nutrition position stand: caffeine and exercise performance”. Journal of the International Society of Sports Nutrition, 2021;18:1. DOI: 10.1186/s12970-020-00383-4