What actually prevents running injuries — and what doesn’t

Running injuries are common — how common depends entirely on how you define an injury, and the honest range across studies is wide. An enormous amount has been written about preventing it, most of it is not supported, and some of it has been directly tested and failed. The genuinely useful part of this literature is not what works — it is the long list of things that demonstrably do not, which you can stop worrying about and stop paying for.

Start with the number that matters

Pooling the studies that measured injuries per hour of running:

• Novice runners: 17.8 injuries per 1,000 hours (95% CI 16.7 to 19.1)
• Recreational runners: 7.7 per 1,000 hours (95% CI 6.9 to 8.7)

Beginners are injured at roughly 2.3 times the rate of established runners. The largest pooled analysis reports an injury incidence of 40.2% (from 10,941 runners) and a prevalence of 44.6% (from 22,823) — considerably higher than the “one in four” figure that circulates. That is the problem worth solving, and the club’s newest members are the people most exposed.

What has actually been disproven

This is the most valuable section on the page, because every item here is something you can stop doing, stop buying, or stop feeling anxious about.

Choosing shoes by your arch or foot type. This is the best-evidenced negative in the whole field. Across 7,203 participants in three trials, prescribing shoes on foot posture gave a rate ratio of 1.03 (95% CI 0.94 to 1.13) — with moderate certainty, the highest grade anywhere in this literature. Those three trials were themselves large military trials, reporting 0.97 in both men and women — so treat this as one body of evidence, not two. The Cochrane conclusion is blunt: “There is no evidence that prescribing footwear based on foot type reduces running-related lower-limb injuries in adults.”

Correcting “overpronation”. A study tracked 927 novice runners over 326,803 km. Runners with pronated feet did not get injured more. A secondary analysis even found them injured slightly less (the study’s primary, BMI-adjusted comparison was null at p = 0.09, so read this as no evidence of harm rather than as protective) — incidence rate difference −0.37 per 1,000 km (95% CI −0.03 to −0.70, p = 0.03). In a separate trial, motion-control shoes prescribed for pronation produced more pain than neutral shoes, in both neutral and pronated feet.

Stretching before you run. A randomised trial of 1,538 army recruits found a hazard ratio of 0.95 (95% CI 0.77 to 1.18). Cochrane pooled six trials and 5,130 people: RR 0.85 (0.65 to 1.12). A separate meta-analysis: 0.963. Stretch if you enjoy it. It will not stop you getting hurt.

The 10% rule. It has been tested exactly once, properly, in 532 novice runners randomised to a 13-week programme built on it or a standard 8-week one. Injury incidence: 20.8% versus 20.3%, p = 0.90. The paper’s title says it plainly: “No effect of a graded training program on the number of running-related injuries in novice runners.” We could not find any source establishing where the 10% figure came from in the first place.

The acute:chronic workload ratio — the “sweet spot” your watch may show you. Researchers replaced the real training-load data with randomly generated numbers and reproduced the same apparent effect. Neither the ratio nor acute load beat a model containing no information at all. The methodologists’ verdict: “There is no evidence supporting the use of ACWR in training-load-management systems… it adds noise and creates statistical artifacts.” It was never studied in distance runners.

Screening yourself on biomechanics. A review tested more than 100 biomechanical and musculoskeletal measures across 3,404 runners in prospective studies. Two reached significance, both trivially. The conclusion: “recommendations for injury prevention in non-elite runners cannot be made based on biomechanical and musculoskeletal measurements alone.”

A two-year study of 300 runners with injuries diagnosed by an orthopaedic surgeon put it even more strongly: “contrary to several long-held beliefs, flexibility, arch height, quadriceps angle, rearfoot motion, lower extremity strength, weekly mileage, footwear, and previous injury are not significant etiologic factors across all overuse running injuries.”

The claim this article used to make

We previously told you that strength training cuts running injuries to about a third. That figure is real — RR 0.315 (95% CI 0.207 to 0.480) — but it comes from trials of footballers, handball players and military recruits doing supervised neuromuscular programmes. It was not measured in runners.

When someone searched the entire literature for randomised trials of exercise-based prevention in endurance runners, they found nine trials covering 1,904 people. Pooled:

• Injury risk: RR ≈ 0.81 (95% CI 0.63 to 1.05), p = 0.110 — not significant
• Injury rate: IRR ≈ 0.86 (95% CI 0.64 to 1.16), p = 0.329 — not significant, and graded high certainty

Their conclusion: “Exercise-based interventions do not appear to reduce the risk and rate of running-related injuries.” Seven of the nine trials were at high risk of bias.

The individual runner trials tell the same story. The largest — 720 first-time marathoners randomised to a strength programme — found RR 0.97 (95% CI 0.57 to 1.63). A trial of 129 novice runners across three arms found nothing, and the stretching control group had the numerically lowest injury rate. The one positive result was a foot-strengthening programme in 118 runners, OR 0.40 (95% CI 0.15 to 0.98) — a secondary analysis that barely cleared significance.

We were wrong to state it the way we did, and this is the correction.

An important complication

Before you conclude that nothing works, there is a genuine problem with all those null trials. Researchers went back and checked how many participants actually followed the training programmes they were assigned:

• One 839-person trial: 0 participants fully compliant
• Another with 612: 0
• The 532-person 10%-rule trial: 8

So these trials tested prescribing a schedule, not following one. That is a real reason to hold the null results a little more loosely — and equally a reason to doubt that any prescribed progression rule is something real runners can actually implement.

What the field itself says

A Cochrane review of 25 trials and 30,252 participants reached this verdict: “the evidence base for the effectiveness of interventions to reduce soft-tissue injury after intensive running is very weak, with few trials at low risk of bias.” Of those 25 trials, 19 were in military recruits and only three were in ordinary runners.

There is a decent explanation for why prediction keeps failing. A proposed framework argues that injury depends on the load reaching a specific structure — this Achilles, that kneecap — relative to what that structure can currently take. Whole-body numbers like weekly kilometres or a workload ratio cannot capture that, because the same run loads different tissues completely differently in different people.

So what does hold up?

Honestly, very little — and pretending otherwise is how the bad advice above got established. Here is what we can defend.

Proven, though not in runners: a randomised, double-blind, placebo-controlled trial of 5,201 female navy recruits found that calcium (2,000 mg) plus vitamin D (800 IU) daily reduced stress fractures by 20% (5.3% vs 6.6%, p = 0.0026). That is the single most robust positive prevention result adjacent to this field. Worth knowing given how common vitamin D deficiency is in India.

Strength training, for a different reason. The injury case is weak, but the performance case is solid: heavy or explosive strength work improves running economy, with a large effect in trained runners and moderate-certainty evidence across roughly 650 athletes. Note the qualifier — submaximal loads and isometric work produced no improvement. Lift because it makes you a more efficient runner, not because it will keep you healthy.

Energy availability, especially for women. This is the best-evidenced risk factor in the whole report. In a prospective study of 259 exercising girls and women, combining high training volume with low bone density gave OR 5.1 (95% CI 2.2 to 12.1) for bone stress injury; a separate combination in the same study — 12 or more hours a week, a leanness sport and dietary restraint, without low bone density in the model — carried OR 8.7 (2.7 to 28.3). In female runners specifically, each one-point increase in a cumulative risk score raised trabecular bone stress injury risk by 26% (p = 0.0007).

Plausible but unproven: increasing your cadence by 5 to 10% reliably reduces knee loading — but only two of 37 studies had any injury outcome, and both were treating existing pain rather than preventing anything. It also moves load off the kneecap and onto the ankle and Achilles. We previously put a number on that shift; the figures we used compared a tendon force against a joint force and did not mean what we said they meant, so we have removed them. The direction is supported; the size is not well established, and the direct forefoot-versus-rearfoot comparison of Achilles force has not come out significant. Reasonable if your knee hurts; think twice if your Achilles does.

Gait retraining with real-time feedback produced a striking result — HR 0.38 (95% CI 0.25 to 0.59) in 320 novice runners over a year, from a two-week intervention. That is a very large effect for this field, from one centre, never replicated. Interesting; not yet advice.

And comfort as your shoe-selection criterion. There is no prevention trial behind it. It is the best available heuristic mainly because every alternative has been falsified.

Sleep, and how weak that evidence actually is

You will read that poor sleep causes injuries. The evidence is thinner than its reputation: the most-cited study surveyed 112 adolescents about their sleep and checked injury records retrospectively, finding those sleeping under 8 hours were 1.7 times more likely to be injured (95% CI 1.0 to 3.0) — a lower bound of exactly 1.0. In that same study, school grade was a stronger predictor than sleep. A meta-analysis of seven studies gives OR 1.58 (1.05 to 2.37).

There is no trial in which improving sleep reduced injuries, and none of this evidence is in distance runners. Sleep well for the reasons in our sleep guide — the recovery and mood evidence is better than the injury evidence.

Which injuries you are actually likely to get

Two different rankings, and the distinction matters more than it sounds.

By incidence — what you are most likely to newly develop: Achilles tendinopathy 10.3%, medial tibial stress syndrome (shin splints) 9.4%, patellofemoral pain 6.3%, plantar fasciitis 6.1%.

By prevalence — what runners are most likely to be carrying at any moment: patellofemoral pain 16.7%, medial tibial stress syndrome 9.1%, plantar fasciitis 7.9%, iliotibial band syndrome 7.9%.

You will often read that runner’s knee is the most common running injury. That is true of prevalence and false of incidence, where it ranks third. The difference is that patellofemoral pain lasts a long time, so it accumulates.

That longevity is worth knowing about. Among novice runners who got injured, median time to recovery was 71 days, and half still could not run 500 metres twice without pain after ten weeks.

The gap nobody can fill for you

Every number on this page comes from Europe, North America, Australia or Hong Kong. We could not find a single study of running injury rates in Indian or South Asian recreational runners. Whether Danish novice-runner injury rates apply to Gurugram — different surfaces, different heat, different footwear market, different baseline activity — is untested.

What we would actually tell a member

1. Stop paying for gait analysis and prescribed shoes. That is the clearest negative finding in the field. Buy shoes that feel comfortable.
2. Stop stretching to prevent injury. It does not, in three separate lines of evidence.
3. Do not trust your watch’s training-load score. The metric behind it reproduces itself from random numbers.
4. Build up gradually anyway — not because a progression rule is proven, but because the downside is a median 71 days out.
5. Do strength training — for running economy, which is well evidenced, and accepting that the injury benefit is plausible rather than demonstrated. Make it heavy or explosive; light circuits did nothing in the trials.
6. Eat enough. If you are training seriously and restricting food, that is the best-evidenced risk factor here — particularly for women, and particularly for bone.
7. Get pain that lasts more than two weeks looked at rather than searching for a mechanical cause yourself. The screening measures do not predict anything, and a clinician who can examine you will do better than a checklist.

The honest summary: we cannot reliably predict who gets injured, and no intervention has been shown in a replicated, low-bias trial in runners to change it. Anyone selling you certainty about this is selling you something.

References

• Wu H, et al. Do exercise-based prevention programs reduce injury in endurance runners? Sports Med, 2024;54(5):1249–67. doi:10.1007/s40279-024-01993-7
• Lauersen JB, Bertelsen DM, Andersen LB. The effectiveness of exercise interventions to prevent sports injuries. Br J Sports Med, 2014;48(11):871–7. doi:10.1136/bjsports-2013-092538
• Toresdahl BG, et al. A randomized study of a strength training program to prevent injuries in runners of the New York City Marathon. Sports Health, 2020;12(1):74–9. doi:10.1177/1941738119877180
• Buist I, et al. No effect of a graded training program on the number of running-related injuries in novice runners. Am J Sports Med, 2008;36(1):33–9. doi:10.1177/0363546507307505
• Relph N, et al. Running shoes for preventing lower limb running injuries in adults. Cochrane Database Syst Rev, 2022;8:CD013368. doi:10.1002/14651858.CD013368.pub2
• Knapik JJ, et al. Injury-reduction effectiveness of prescribing running shoes on the basis of foot arch height. J Orthop Sports Phys Ther, 2014;44(10):805–12. doi:10.2519/jospt.2014.5342
• Nielsen RO, et al. Foot pronation is not associated with increased injury risk in novice runners wearing a neutral shoe. Br J Sports Med, 2014;48(6):440–7. doi:10.1136/bjsports-2013-092202
• Pope RP, et al. A randomized trial of preexercise stretching for prevention of lower-limb injury. Med Sci Sports Exerc, 2000;32(2):271–7.
• Impellizzeri FM, et al. What role do chronic workloads play in the acute to chronic workload ratio? Sports Med, 2021;51(3):581–92. doi:10.1007/s40279-020-01378-6
• Peterson B, et al. Biomechanical and musculoskeletal measurements as risk factors for running-related injury in non-elite runners. Sports Med Open, 2022;8(1):38. doi:10.1186/s40798-022-00416-z
• Messier SP, et al. A 2-year prospective cohort study of overuse running injuries (TRAILS). Am J Sports Med, 2018;46(9):2211–21. doi:10.1177/0363546518773755
• Nielsen RO, et al. Randomised controlled trials in sports injury research: authors — please report the compliance with the intervention. Br J Sports Med, 2020;54(1):51–7. doi:10.1136/bjsports-2019-100858
• Yeung SS, Yeung EW, Gillespie LD. Interventions for preventing lower limb soft-tissue running injuries. Cochrane Database Syst Rev, 2011;7:CD001256. doi:10.1002/14651858.CD001256.pub2
• Videbæk S, et al. Incidence of running-related injuries per 1000 h of running in different types of runners. Sports Med, 2015;45(7):1017–26. doi:10.1007/s40279-015-0333-8
• Kakouris N, Yener N, Fong DTP. A systematic review of running-related musculoskeletal injuries in runners. J Sport Health Sci, 2021;10(5):513–22. doi:10.1016/j.jshs.2021.04.001
• Lappe J, et al. Calcium and vitamin D supplementation decreases incidence of stress fractures in female Navy recruits. J Bone Miner Res, 2008;23(5):741–9. doi:10.1359/jbmr.080102
• Barrack MT, et al. Higher incidence of bone stress injuries with increasing female athlete triad-related risk factors. Am J Sports Med, 2014;42(4):949–58. doi:10.1177/0363546513520295
• Kulmala JP, et al. Forefoot strikers exhibit lower running-induced knee loading than rearfoot strikers. Med Sci Sports Exerc, 2013;45(12):2306–13. doi:10.1249/MSS.0b013e31829efcf7
• Chan ZYS, et al. Gait retraining for the reduction of injury occurrence in novice distance runners. Am J Sports Med, 2018;46(2):388–95. doi:10.1177/0363546517736277
• Llanos-Lagos C, et al. Effect of strength training programs in middle- and long-distance runners’ economy at different running speeds. Sports Med, 2024;54(4):895–932. doi:10.1007/s40279-023-01978-y