HIIT vs. SIT vs. RST for VO2max: What the Rankings Mean

A new network meta-analysis reportedly puts repeated sprint training (RST) ahead of high-intensity interval training (HIIT) and sprint interval training (SIT) for improving VO2max in competitive athletes. That sounds like a clean winner. It is not—at least not yet.

The 2025 review by Yang et al. is the broadest comparative analysis available: 51 studies, four training nodes, and a formal ranking framework that compares RST, HIIT, SIT, and continuous training (CT). But the accessible publication text does not provide the actual effect sizes, confidence intervals, P-scores, heterogeneity statistics, or dose-response results. Without those numbers, nobody can tell whether RST's apparent edge over HIIT is meaningful, tiny, or too imprecise to guide a training overhaul.

For athletes deciding what to do next week, the defensible conclusion is narrower: interval training improves aerobic capacity, HIIT has substantial support across several athlete groups, and the claim that RST is the best option requires a closer look at the full results—especially because a direct randomized trial in well-trained men favored aerobic HIIT over sprint intervals.

What the network meta-analysis tried to answer

Pairwise trials can compare one program against another. A network meta-analysis can estimate comparisons that individual trials did not test directly, provided the underlying trials are sufficiently comparable. Yang and colleagues pre-registered their review (PROSPERO CRD42023435021) and used a frequentist network meta-analysis to compare interval-training formats against each other and against CT.

The analysis used Hedges' g, a standardized effect-size measure, and P-scores, which rank treatments by the probability that each outperforms the alternatives. The reported hierarchy is RST first, HIIT second, SIT third, with CT behind the interval methods. That ranking is a useful hypothesis, not a prescription carved into a treadmill console.

The authors also used a three-level meta-analytic structure, allowing them to account for multiple effect sizes from a single study rather than treating each as fully independent. Their meta-regression controlled for baseline VO2max and examined athlete training tier. They compared linear, polynomial, and restricted cubic-spline dose-response models using AICc, a model-selection criterion that penalizes needless complexity.

Those are sensible methods. The interpretive problem is simpler: the available text does not report the results those methods produced. Specific claims about ideal work-bout duration, work-to-rest ratio, recovery limits, or minimum training duration cannot be verified without the paper's Results section. Readers should treat any exact protocol numbers attributed to this review as provisional.

A ranking cannot replace direct comparison

The most relevant counterweight comes from Hov et al.'s 2023 randomized trial in well-trained men. It directly compared aerobic high-intensity intervals with sprint intervals and found that aerobic HIIT produced the greater VO2max improvement.

That result does not automatically disprove an RST-first network ranking. RST and the sprint-interval protocol used by Hov et al. may not be identical. The studies may also differ in sport context, participant characteristics, and program design. Still, direct randomized evidence deserves particular weight when it addresses the same practical choice: should a well-trained athlete use aerobic intervals or sprint-based intervals to raise VO2max?

Network estimates depend on transitivity—the assumption that trials connected through a shared comparator can reasonably inform one another. That assumption gets harder to defend when 51 studies span sports, sexes, and athlete tiers. If RST versus HIIT rests heavily on indirect pathways, differences in participant fitness or training design can masquerade as a treatment advantage.

HIIT has the deeper supporting record

The evidence for HIIT improving VO2max in trained team-sport populations is consistent, even though those studies do not settle the RST-versus-HIIT contest. Yuan et al.'s 2024 meta-analysis found that HIIT improved VO2max, repeated-sprint ability, and speed in male team-sport athletes. Clemente et al.'s meta-analysis likewise found improvements in aerobic capacity and physical fitness among male soccer players across randomized and non-randomized studies.

Those findings establish that HIIT works in these populations. They do not show that HIIT beats RST or SIT, because neither review included those modalities as direct comparative arms. It is a common evidence mistake to confuse "effective" with "best." Training research has enough of those already.

Sport-specific comparisons also suggest that the label on the workout is not the entire intervention. Song et al. compared technique-specific high-intensity intervals with short sprint intervals in taekwondo athletes. Zhao et al. randomized 24 male collegiate volleyball players to six weeks of interval running or interval drop jumps: both programs used 15 seconds of work alternated with 15 seconds of passive recovery for 11 minutes and both improved VO2max. Movement demands can change the adaptation profile even when the session looks similar on a stopwatch.

That is one reason modality-level rankings should not outrank sport demands, injury history, access to equipment, or the training already filling the rest of the week.

Elite athletes should expect smaller gains

VO2max is harder to move once an athlete has spent years moving it. Two 2025 meta-analyses by Wiesinger et al. (Part I) and Wiesinger et al. (Part II) found that HIIT can improve VO2max and other performance measures in elite athletes, but reported smaller average responses than in less-trained groups and limited evidence in the highest training tiers.

This ceiling effect changes the practical stakes. A protocol that produces a worthwhile average response in a Tier 2 or Tier 3 athlete may produce a small, variable change in a Tier 5 athlete. In that setting, recovery, total weekly load, competition schedule, and periodization may determine more than whether the hard session belongs in the HIIT, SIT, or RST drawer.

It also limits confidence in a single pooled optimum. A meta-regression can identify patterns across studies, but it cannot prove that one work-to-rest ratio will optimize VO2max for every sport or performance level. Nonlinear curves are especially vulnerable when few studies test the extremes of work duration or recovery.

Do athletes need individualized interval prescriptions?

Probably not as a default requirement. In a randomized parallel study of 46 male youth soccer players, Zhu et al. assigned players to individualized HIIT based on locomotor profiles, standardized long-interval HIIT, RST, or a regular-training control group. The intervention lasted six weeks. The individualized and non-individualized approaches produced similar aerobic, anaerobic, and neuromuscular adaptations.

That does not mean every athlete should receive the same session. It means a well-structured standardized program can work without a specialized profiling scheme. Coaches still need to adjust workload for fitness, fatigue, injury risk, and the athlete's sport. Individualization remains useful when it solves a real programming problem, rather than when it merely gives a spreadsheet more to do.

What to do with the RST-first claim

RST may prove to be the leading interval format for athlete VO2max, but the current ranking should remain conditional until readers can inspect its numerical results: the Hedges' g estimates, confidence intervals, P-scores, heterogeneity, direct-versus-indirect evidence balance, and certainty ratings. A P-score ranks probabilities; it does not tell an athlete how many milliliters of oxygen uptake they will gain or whether the difference changes race, match, or season outcomes.

For now, choose an interval format that fits the sport and the athlete, then execute it consistently enough to create an aerobic stimulus without wrecking the next key session. HIIT has broad supporting evidence in team-sport athletes, aerobic HIIT beat sprint intervals in one direct randomized trial of well-trained men, and RST's reported network advantage remains plausible but unquantified in the available results. The sensible response is not to crown a winner early. It is to demand the numbers.