HIIT, Yoga, and Depression in College Students: What Wins?

HIIT ranked first for depression in college students. That does not make it the winner.

A 2023 network meta-analysis by Li et al. asked a useful question that most depression trials skip: if college students want non-drug treatment, which option looks best? Across 24 studies and 1,458 participants, the authors compared music therapy, massage, acupuncture, and several exercise types, then ranked them with a network meta-analysis. High-intensity interval training, or HIIT, came out on top with a SUCRA score of 96%, followed by yoga at 94.9% and dance at 78.3%.

If you only read that sentence, you would think the case is closed. It is not. The entire HIIT node in this analysis included just 14 experimental participants. That is the smallest evidence base for any modality in the paper. A ranking built on 14 people can wobble if one study overestimates the effect, if one group differs at baseline, or if one outcome measure behaves oddly. So the headline is real, but it is not sturdy.

The paper matters anyway, because it tackles a population that often gets stuck between two bad options: symptoms serious enough to hurt school and daily life, but evidence too thin to guide campus-scale treatment. Li et al. focused specifically on college students with depression rather than lumping them into a generic adult sample. In China, where the paper notes that the Ministry of Education has required depression screening in student physical exams, that question has direct policy value. If schools are going to recommend or provide non-pharmacological options, they need more than vibes and a yoga flyer.

What the study actually did

Li et al. pooled 24 controlled studies: 17 randomized controlled trials and 7 non-randomized controlled trials. That mix alone should slow anyone down. Randomization reduces allocation bias. Non-randomized designs do not. Putting both into one network can make the estimates look more precise than the underlying evidence deserves.

The strength of the method is the method itself. Network meta-analysis lets researchers compare treatments that have not faced each other directly. If one study tests HIIT against a control group and another tests yoga against a control group, the network can estimate HIIT versus yoga indirectly. For sparse evidence, that is often the best available tool. It is also a tool that can produce very tidy-looking rankings from messy inputs. The SUCRA score is a probability-based ranking metric, not a medal ceremony.

That matters here because the gap between HIIT and yoga was tiny: 96% versus 94.9%. With a well-powered evidence base, a small gap can still be meaningful. With 14 HIIT participants, it mostly tells you the two interventions landed near the top of a fragile network. It does not tell you HIIT beats yoga in any practical sense for depressed college students.

Why exercise still looks credible

The weak point is the ranking, not the broader idea that exercise can help depression. On that question, the evidence is much better. A 2024 systematic review and meta-analysis by Tao et al. found that HIIT reduces depressive symptoms across populations. A 2024 meta-analysis of randomized trials by Gaia et al. found that HIIT reduced depressive and anxiety symptoms in healthy individuals and argued that moderate-to-vigorous intensity exercise may have larger antidepressant effects than lower-intensity exercise. A 2022 systematic review and meta-analysis by Martland et al. concluded that HIIT is safe and feasible across the general population and people with physical illnesses.

Those papers do not prove that HIIT is the best choice for a depressed 20-year-old student. They do show that the basic antidepressant signal for exercise, including higher-intensity exercise, is not a fluke from one campus study.

The mechanistic case points in the same direction, with the usual caution that biomarkers are not the same thing as feeling better. A 2025 network and dose-response meta-analysis by Zhu et al. found that exercise raises brain-derived neurotrophic factor, or BDNF, in patients with depression, and that higher-intensity exercise produced larger BDNF increases than moderate-intensity approaches. Gaia et al. also discussed intensity-related effects on monoamine pathways. That makes HIIT biologically plausible. It does not settle the college-student question, because those mechanistic data come from depressed patients more broadly, not specifically from 18-to-25-year-old students under campus conditions.

Why the ranking probably overstates certainty

The cleanest challenge to Li et al. comes from the structure of their own evidence. First, the HIIT result rests on very sparse data. Second, the analysis combined randomized and non-randomized studies without reporting a design-stratified sensitivity analysis. Third, the paper did not use GRADE-NMA or CINeMA to rate confidence in the network estimates. That omission matters because network meta-analysis can generate a full ordering even when the supporting comparisons are mostly indirect and low-confidence.

There is also a demographic gap. About 63% of the sample was female, and the study did not report sex-stratified analyses. Depression prevalence and treatment response can differ by sex, so the lack of subgroup analysis limits how far you can generalize the rankings.

Safety is another place where readers should keep their feet on the ground. Martland et al. support HIIT as broadly safe and feasible, but that is not the same as proving safety for unsupervised high-intensity exercise in students with severe or treatment-resistant depression. The campus version of HIIT may involve low supervision, variable adherence, and a very human tendency to quit hard things during finals week.

What contradicting evidence tells us

A 2025 network meta-analysis by Wang et al. looked at physical activity interventions for depressive symptoms in menopausal women. Different population, same basic analytical idea. Their result was flatter: multiple exercise modalities showed comparable antidepressant effects, with no consistent superiority for one type. That does not refute Li et al. directly, because menopausal women are not college students. It does show what often happens when a treatment network has better evidence density across its nodes: rankings compress.

That is the more believable pattern for exercise and depression right now. Exercise helps. Moderate-to-vigorous intensity may matter. But once you compare specific modalities against each other, especially in narrow populations, the hierarchy gets much less certain than the marketing copy would like.

The current evidence supports exercise for depression in college students far more strongly than it supports HIIT as the single best exercise format.

What campuses and students can take from this now

If you are making decisions today, the practical message is modest but useful. Colleges should treat exercise as a serious part of depression care, not as a cheerful afterthought. Students who dislike HIIT should not read this paper and conclude that yoga, dance, or other active options are second-rate. The difference between doing a sustainable exercise program and doing none is almost certainly larger than the paper's 1.1-point SUCRA gap between HIIT and yoga.

That does not mean all exercise prescriptions are interchangeable. Intensity probably changes biology, and adherence probably changes outcomes even more. A hard protocol that people abandon is less impressive than a slightly less intense one they actually finish. Li et al. did not answer that tradeoff. The published evidence still needs head-to-head randomized trials that compare HIIT, yoga, dance, and other modalities in depressed college students, with enough participants to estimate real differences and enough follow-up to track dropout, adherence, and symptom change under real-world conditions.

The verdict

Li et al. asked the right question and used the right broad strategy for a thin evidence base. That is the good part. The bad part is that the evidence under the most eye-catching result is paper-thin. HIIT ranked first, but on 14 participants, with no formal confidence grading, in a network that mixed randomized and non-randomized studies.

So the honest conclusion is narrower than the ranking suggests. Exercise is one of the better-supported non-pharmacological tools for depression, and higher-intensity exercise has a plausible biological and clinical case behind it, supported by Tao et al., Gaia et al., Zhu et al., and Martland et al.. What the literature does not yet show is that HIIT clearly beats yoga, dance, or other active formats for college student depression. For now, the directional science is credible. The ranking is not ready to boss everyone around.