Learn physiotherapy · BPT-205 Exercise Therapy
Range of Motion and Stretching
Stretching is among the most prescribed things physiotherapists do. On high-certainty evidence it changes joint mobility by about one to two degrees, and it does not prevent injury. It still has uses — just not the two it is usually prescribed for.
In one line. Stretching reliably increases the range you can measure in the short term. What it does not do, on the best available evidence, is produce clinically important lasting change in joint mobility or reduce injury risk.
Both of those are high-certainty negative findings against near-universal practice, which is exactly the kind of thing worth knowing before you build a career of prescribing it.
The contracture finding is GRADE high, and negative
A Cochrane review found that in neurological conditions stretch produced a pooled mean difference of 2 degrees (95% CI 0 to 3), a 2% relative change; in non-neurological conditions 1 degree (95% CI 0 to 2), a 1% change. The GRADE level of evidence was high for both meta-analyses. The authors' conclusion is unambiguous: stretch does not have clinically important effects on joint mobility. [1] This is not a call for more research. It is a settled answer.
What the evidence shows
| Question | Finding | Source and quality |
|---|---|---|
| Does stretch treat or prevent contracture? | Not to any clinically important degree. 2 degrees (95% CI 0 to 3) in neurological conditions; 1 degree (0 to 2) in non-neurological, SMD 0.2 (0 to 0.3) | 18 studies and 549 participants, and 18 studies and 865 participants respectively. GRADE high for both [1] |
| Does stretching prevent injury? | No. Risk ratio 0.963 (95% CI 0.846 to 1.095) — the interval comfortably includes no effect | 25 trials, 26,610 participants, 3,464 injuries. Publication bias and heterogeneity formally tested [2] |
| What does prevent injury? | Strength and proprioceptive work. Proprioception training RR 0.550 (0.347 to 0.869); multiple combined exposures RR 0.655 (0.520 to 0.826) | Same analysis [2]. In hamstrings specifically, eccentric training reduced injury incidence by 56.8% to 70.0% across 108 included studies [3] |
| Does stretching increase measured flexibility? | Yes, and more than the alternatives. Static stretching gave a mean difference of 10.89 degrees (95% CI 8.92 to 12.86) against 9.73 degrees for proprioceptive neuromuscular facilitation | Hamstring review, 108 studies from 2,602 identified [3] |
| Does stretching one muscle affect others? | Yes. Unilateral passive static stretching produced moderate increases in passive range at non-stretched joints (SMD 0.86), larger beyond 240 seconds (SMD 1.24) than under 120 seconds (SMD 0.72) | Sex, trained state, intensity and duration did not moderate the overall result [4] |
| Does prolonged stretching reduce force? | Slightly, and inconsistently. Force deficits of -6.7% in the stretched muscle (d = -0.35) and -4.0% contralaterally (d = 0.22) | Only six studies; three measures showed deficits and four showed trivial change. Treat as a signal, not a rule [5] |
Where students get this wrong
1. Prescribing stretch to prevent or treat contracture
This is the commonest use and the one with the clearest negative evidence: 1 to 2 degrees, GRADE high. [1] A degree or two is not nothing to measure, and it is nothing to a patient. If contracture management is the goal, the honest position is that passive stretch alone is not an effective treatment, and saying so is more useful than prescribing it because it is expected.
2. Including stretching in warm-ups to prevent injury
Across 25 trials and 26,610 participants, stretching showed a risk ratio of 0.963 with an interval spanning no effect. [2] In the same analysis, proprioceptive training nearly halved injury risk and combined programmes cut it by a third. For hamstrings, eccentric training reduced injuries by 57% to 70%. [3] The warm-up minutes spent stretching have a much better available use.
3. Concluding that stretching is therefore useless
It is not. It produces the largest measured flexibility gains of the stretching methods compared — 10.89 degrees for static stretching. [3] It increases range at joints you did not stretch. [4] Patients often report it feels good. The error is not doing it; the error is claiming it does something the evidence says it does not.
4. Assuming the effect is local
Stretching one muscle group produced moderate increases in passive range at non-stretched joints (SMD 0.86), [4] and small force deficits in the contralateral limb. [5] Whatever mechanism produces short-term range change is at least partly systemic rather than tissue-specific — which sits awkwardly with the tissue-lengthening explanation most students are taught.
5. Stretching immediately before something requiring force
Prolonged static stretching was associated with small force deficits in the stretched muscle and, less consistently, the opposite limb. [5] The evidence base is six studies and the effects were inconsistent, so this is a reason for sensible sequencing — save prolonged static stretching for after, not before, a strength or power task — rather than for alarm.
6. Reading a range gain as a tissue change
An increase in measured range can reflect changed stretch tolerance rather than changed tissue length. The non-local findings [4] and the failure of stretch to alter contracture over time [1] both point that way. When you record improved range after stretching, you have documented what the patient allowed, not necessarily what the tissue became.
What the evidence supports — and what it does not
Supported
- Stretching to increase measured flexibility. Largest gains of the methods compared. [3]
- Strength and proprioceptive training for injury prevention. [2]
- Eccentric training for hamstring injury. 57% to 70% reduction. [3]
- Longer stretch durations where range is the goal. [4]
Not supported
- Stretch for treating or preventing contracture. 1 to 2 degrees, GRADE high. [1]
- Stretching to prevent injury. RR 0.963. [2]
- Treating range gains as evidence of tissue lengthening. [1][4]
- Prolonged static stretching immediately before force tasks. [5]
How certain is this?
Evidence grade: High for the contracture finding, moderate for the rest.
The contracture conclusion is as certain as rehabilitation evidence gets: a Cochrane review, 36 studies across two populations, roughly 1,400 participants, and the authors themselves rate the evidence high under GRADE. [1] The injury-prevention analysis is large (26,610 participants, 3,464 injuries) and formally tested for publication bias and heterogeneity. [2]
The supporting findings are weaker. The non-local force work rests on six studies with inconsistent results, [5] and the non-local range finding, while a formal meta-analysis, addresses a phenomenon whose mechanism is unresolved. [4]
Sourcing note. This is a curriculum resource written against sources that can be opened and checked. It reproduces no figures or tables from any textbook.
Common questions
Should I stop prescribing stretches?
Not necessarily — but change what you tell the patient it is for. The evidence does not support it for contracture [1] or injury prevention. [2] It does support it for increasing measured flexibility, [3] and patients often value how it feels. Prescribe it for what it does.
What should I do about contracture instead?
This page cannot give you a validated alternative, because the review that closed the stretch question did not open another. [1] What it does establish is that continuing to prescribe passive stretch and expecting joint mobility to change is not supported. Discuss positioning, activity and function with the team rather than defaulting to a stretching programme.
How long should a stretch be held?
If range is the goal, longer appears better: beyond 240 seconds total produced large increases (SMD 1.24) against moderate increases under 120 seconds (SMD 0.72). [4] Note these are totals across a session, and note that this is range measured shortly after — not lasting joint mobility, which is the finding in [1].
What should go in a warm-up then?
On the injury-prevention data, strength and proprioceptive work: proprioception training RR 0.550 and combined programmes RR 0.655, against RR 0.963 for stretching. [2] For hamstrings specifically, eccentric training is the intervention with the reduction of 57% to 70%. [3] See also exercise dose for how much of it to prescribe.
Why does my range improve if the tissue does not lengthen?
Most likely because your tolerance to stretch changes rather than the tissue. Two findings point that way: stretching one muscle increases range at joints you did not stretch, [4] and sustained stretch programmes do not change joint mobility over time. [1] The mechanism is not settled, and you should present it as unsettled rather than repeating the tissue-lengthening story as fact.
References
- Harvey LA, Katalinic OM, Herbert RD, et al. Stretch for the treatment and prevention of contracture: an abridged republication of a Cochrane Systematic Review. Journal of Physiotherapy. 2017 Apr;63(2):67–75. doi:10.1016/j.jphys.2017.02.014 PMID 28433236 Cochrane systematic review (abridged republication)
- Lauersen JB, Bertelsen DM, Andersen LB. The effectiveness of exercise interventions to prevent sports injuries: a systematic review and meta-analysis of randomised controlled trials. British Journal of Sports Medicine. 2014 Jun;48(11):871–7. doi:10.1136/bjsports-2013-092538 PMID 24100287 Systematic review and meta-analysis
- Rudisill SS, Varady NH, Kucharik MP, et al. Evidence-Based Hamstring Injury Prevention and Risk Factor Management: A Systematic Review and Meta-analysis of Randomized Controlled Trials. The American Journal of Sports Medicine. 2023 Jun;51(7):1927–1942. doi:10.1177/03635465221083998 PMID 35384731 Systematic review and meta-analysis
- Behm DG, Alizadeh S, Anvar SH, et al. Non-local Acute Passive Stretching Effects on Range of Motion in Healthy Adults: A Systematic Review with Meta-analysis. Sports Medicine. 2021 May;51(5):945–959. doi:10.1007/s40279-020-01422-5 PMID 33459990 Systematic review and meta-analysis
- Behm DG, Alizadeh S, Drury B, et al. Non-local acute stretching effects on strength performance in healthy young adults. European Journal of Applied Physiology. 2021 Jun;121(6):1517–1529. doi:10.1007/s00421-021-04657-w PMID 33715049 Systematic review
About this resource
How to use this
Written to be learned from, not memorised.
This page reports a high-certainty negative finding against a near-universal practice, and then says what stretching is actually good for. Faculty may use this page in teaching with attribution. It carries its review date and its next review date, so you can see at a glance whether it is current before you put it in front of a cohort.