REF-749 · Mattioni Maturana, Zadon, Cocks, Nieß, Thijssen & Dawson, 2026 · Sports Medicine, online 12 September · Paper

In previously inactive adults, interval training reaches 80% of its VO₂max gain in about 11 minutes a session and flattens by ~23; steady moderate training needs ~52 minutes for its 80% and had not flattened at the longest dose studied

A dose–response meta-analysis of 69 RCTs

Astrong evidenceMeasured on non-athletes and recreational peopleWritten before October 2026, being re-checked

Protocol

A one-stage random-effects dose–response meta-analysis with restricted cubic splines over 69 randomised trials that each compared an interval arm against a continuous moderate arm for at least 4 weeks (mean 10 ± 6 weeks). 2387 participants (HIIT 1199, MICT 1188), 55 ± 36% men, healthy or with elevated cardiometabolic risk (pre-hypertension, overweight/obesity, metabolic syndrome, pre-diabetes).

"HIIT" pools near-maximal intervals (bouts typically >80% V̇O₂max or >85% HRmax, most often 80–95% HRmax) AND all-out sprints of ≤30 s. MICT was continuous work at 65–80% HRmax. Session dose = minutes of the main set, warm-up and cool-down excluded, recoveries included. Mean sessions: 25.3 ± 11.6 min HIIT vs 43.2 ± 11.8 min MICT; weekly 74 vs 130 min.

Key Findings

Interval training raised V̇O₂max more than continuous training: d = 0.38 (95% CI 0.27–0.49), a mean difference of 1.20 mL·kg⁻¹·min⁻¹. Within-group effects d = 0.83 (HIIT) and d = 0.48 (MICT).
HIIT dose–response is NON-LINEAR. 50% of the maximal effect at 5.8 min/session, 80% at 11.4 min/session (95% CI 9.5–40.2), maximal observed effect at 23.2 min, with limited additional gain beyond ~23 min. Maximal modelled gain 4.31 mL·kg⁻¹·min⁻¹ (so 80% = 3.45).
Counted as hard work only, 80% of the gain came from ~5 min of work per session (5.2, CI 4.5–7.5) and ~15 min of work per week (14.7, CI 12.8–19.0), plateauing near 10 min of work per session. Recovery made up a median 50% of a HIIT session (range 17–90%); work-to-rest ranged from 1:9 to 5:1.
MICT dose–response is LINEAR across the doses studied: 80% of its maximal observed effect at 52.4 min/session (CI 30.4–55.8), maximal at 63 min. "Because no plateau was reached, further gains at higher doses cannot be excluded."
Time ratios MICT:HIIT — 6.2-fold at ED50, 4.6-fold at ED80, 2.7-fold at the maximal effect.
Neither baseline V̇O₂max, age, nor the prescribed MICT intensity moderated the effect; the curve did not differ between healthy and cardiometabolic-risk groups.
Blood pressure, lipids, glucose markers and body composition improved comparably in both arms.

Read the direction carefully

Two things in this paper are easy to carry too far. First, the population: "the included samples were predominantly sedentary to a low fitness level, leaving under-represented the well-trained end of the spectrum — where a reduced trainability ('ceiling') might be expected".

The 11-minute figure is how fast an untrained cardiovascular system responds, not a statement that a trained cyclist needs only 11 minutes. Second, the shape of the two curves says different things: HIIT saturates, MICT did not.

The honest reading is not "intervals beat volume" but "intervals buy the first gains cheaply and then stop paying, while easy volume keeps paying for as long as anyone measured".

The confidence interval on the headline number is wide — 9.5 to 40.2 minutes — and the authors add that these are "population-level averages and should not be read as optimal individualised prescriptions".

Practical Application

For a beginner or a returning athlete with little time, a short interval session is a complete session, not a compromise: after a warm-up, 10–12 minutes of intervals including recoveries — for example 5 × 1 min hard / 1 min easy — captures most of the available V̇O₂max gain. Two or three of these a week is about 15 minutes of hard work, which is where the weekly curve flattened.
Do not lengthen a beginner's interval set past ~20–25 minutes in search of more V̇O₂max. The curve is flat there; the extra minutes cost recovery and adherence and buy nothing measurable. Add an easy session instead.
Easy continuous minutes did not plateau — so when a new athlete has more time, spend it on longer easy rides, not on longer interval sets. This is the same split SYNTH-26 arrives at from the other direction: intensity distribution is a function of hours.
Do not transfer the 11-minute number to a trained or competitive athlete. For them the base's own VO₂max entries apply (time above 90% V̇O₂max, 30/15s, 4–5 × 4–5 min); this paper contains almost no such people and says a ceiling is expected.
Say "11 minutes of the main set, about 5 of them hard" when quoting it. "11 minutes of HIIT" is heard as 11 minutes of hard work, which is double what was measured.
Cardiometabolic health markers improved equally with both — for a health-motivated beginner the choice between the two is preference and adherence, not efficacy.

Limitations

Sedentary-to-low-fitness samples; no stratification by training status was possible. Generalisation to Tier 2+ athletes is not supported by the data.
"HIIT" pools near-maximal intervals with all-out sprint protocols, so the curve is an average over very different sessions with work-to-rest from 1:9 to 5:1.
The ED80 confidence interval for HIIT is wide (9.5–40.2 min); the point estimate is more precise-sounding than the data.
MICT's linear fit holds only inside the observed range (roughly up to an hour); nothing here says what happens at two or three hours.
Interventions averaged 10 weeks — this is the early, fast part of adaptation. Most V̇O₂max values were not verified with a plateau or supramaximal bout.
Outcome is V̇O₂max and cardiometabolic markers; no endurance performance, threshold or durability outcome was analysed.
How this entry was checked and changed
Content correction 2026-09-27
Source note (truth pass): the numbers in this entry were taken from the paper's full text when it was ingested; many of them are not in the abstract. They were not re-checked against the full text on 2026-09-27.