REF-1382 · Morrissey et al., 2021 · British Journal of Sports Medicine · Paper · PubMed · Free full text
What helps heel pain on the first steps in the morning?
For 4-6 weeks: taping, plantar-fascia stretching and advice on shoes and load; if not better, shockwave, then custom orthoses. No runner subgroup; no tested return-to-run plan.
Main caveat: Few powered trials. Of 51 high-quality trials only 9 could establish efficacy.
Morrissey and colleagues built a best-practice guide for plantar heel pain by deciding WHAT to include from a systematic review restricted to high-quality RCTs (PEDro ≥8/10, low risk of bias on a PEDro subset) and HOW and WHEN to apply it from interviews with 14 international expert clinician-researchers and a survey of 40 patients.
Of 362 trials appraised, 51 qualified (4351 participants; median sample 75); only 9 were powered (38 per group, for a 19-point change in first-step pain on a visual analogue scale) to establish efficacy, for 10 interventions. Evidence was graded strong, moderate or limited, positive or neutral, by the number of powered high-quality trials.
The resulting core approach, used together for about 4-6 weeks, is low-dye taping (moderate evidence, first-step pain at 1 week, SMD 0.47) and plantar-fascia stretching (moderate evidence of a large short-term effect against radial shockwave, SMD 1.21, mainly in acute heel pain) — both labelled "DO" — plus individualised education labelled "DECIDE": load management, pain education with pain-monitoring, attention to long-term conditions and body mass, and footwear that is supportive, comfortable, has a rearfoot-to-forefoot drop and is socially acceptable.
Education "was not tested for efficacy but was recommended by experts and appreciated by patients".
If the core approach is not working, shockwave comes next (focused: strong evidence for short-term first-step pain, OR 1.89; radial: moderate), then custom orthoses (strong evidence for short-term pain vs sham, SMD 0.41, from conflicting trials), then experimental options or a return to the core.
Prefabricated orthoses (as used in the trials), calf stretching, magnetised insoles and wheatgrass cream showed no effect; injections were inadequately tested; and the review found no high-quality trial of progressive strengthening, with the experts divided.
Our reading: for the heel row of SYNTH-73 ("heel raises are worth asking about", from one trial of 48 non-runners, REF-920) this guide puts taping, plantar-fascia stretching and education first and leaves strength training outside the proven package — not refuted, untested at the guide's quality bar.
Protocol
Review: Ovid MEDLINE, EMBASE, CINAHL, SPORTDiscus, Cochrane CENTRAL and Web of Science to October 2019, three trial registries, reference lists and citations; no language restriction.RCTs of any intervention in people over 16 with clinically diagnosed plantar heel pain (pain under the heel, most noticeable on weight-bearing after rest and worse after prolonged weight-bearing), follow-up at least 2 weeks; fractures, tumours, infection and rheumatological, neural, vascular or dermatological causes excluded.
Quality filter: PEDro ≥8/10 retained; then a risk-of-bias score from six PEDro items, under 5 of 6 excluded; the 10 trials behind proof of efficacy rechecked with RoB 2 (48 of 50 domains low risk, none high).
Efficacy rules: primary proof = superior to sham or placebo; secondary proof = superior or equivalent to a treatment of proven efficacy; adequately powered = 38 per group. Evidence levels: strong (several high-quality trials in a meta-analysis), moderate (one powered high-quality trial), limited (one underpowered high-quality trial), each positive or neutral.SMD 0.20-0.49 small, 0.50-0.79 medium, ≥0.80 large. Short term 1 week to 3 months, medium >3 to 6 months, long >6 months.
Trials included: 51 (n = 4351; median 75, IQR 62; mean symptom duration 13 months, range 0.8-68.3; 49% women; mean follow-up 20 weeks, range 2-104).Interventions: shockwave 14, foot orthoses 8, corticosteroid injection 11, other injections 8, manual therapy and exercise 2, low-level laser 2, pulsed radiofrequency 2, single trials of stretching, taping, taping with iontophoresis, dry needling, night splints, footwear, flip-flops, magnetised insoles, radiation, electrolysis and wheatgrass cream.
Experts: n = 14 (Australia 6, Denmark 3, UK 2, USA 2, Canada 1; 7 physiotherapists, 6 podiatrists including 2 podiatric surgeons, 1 rheumatologist), at least 5 years' experience and active in research; about 9 patients with heel pain a month and 51 publications on average; interviewed to thematic saturation, Framework analysis.
Patients: n = 40 online survey responses on interim results; 95% of 266 treatment components they mentioned matched the core approach.
Synthesis rule (predetermined): inclusion of interventions from the review; their application, combination and timing from the interviews and survey.
Funding and interests: the Private Physiotherapy Education Fund (UK), which took no part in design, conduct or reporting; DM was part-funded by the NIHR/HEE Senior Clinical Lecturer scheme (NIHR CAT SCL-2013-04-003); CJB is supported by an MRFF TRIP Fellowship (APP1150439). Competing interests: "Three authors (MC, BV, MSR) each conducted a randomised controlled trial that was evaluated in this review."
Key Results
Efficacy and strength of evidence (Table 1; between-group SMD or OR, 95% CI; positive favours the intervention; levels in the guide's own terms):
| Intervention (comparator) | Outcome, time | Effect | Evidence level | Place in the guide |
|---|---|---|---|---|
| Low-dye taping + sham ultrasound (sham ultrasound) | first-step pain, 1 week | SMD 0.47 (0.05 to 0.88) | moderate positive (one trial) | core, "DO" |
| pain; function | 0.30 (−0.11 to 0.71); −0.05 (−0.46 to 0.36) | moderate neutral | ||
| Plantar-fascia stretching (radial shockwave), mainly acute heel pain | first-step pain, short / medium / long | SMD 1.21 (0.78 to 1.63) / 0.64 (0.24 to 1.04) / −0.04 (−0.43 to 0.35) | moderate positive / moderate positive / moderate neutral (secondary proof, one trial) | core, "DO" |
| Education: load, pain (with pain-monitoring), comorbidity and body mass, footwear | — | not tested as an intervention | recommended by experts, appreciated by patients | core, "DECIDE" |
| Focused shockwave | first-step pain, short | OR 1.89 (1.18 to 3.04) | strong positive | step 2 |
| pain, short; function, short | SMD 0.36 (0.11 to 0.61); 0.36 (0.10 to 0.61) | moderate positive | ||
| first-step pain, medium / long | SMD 1.31 (0.61 to 2.01) / 1.67 (0.88 to 2.45) | limited positive (one underpowered trial) | ||
| Radial shockwave | first-step pain, short / long | OR 1.66 (1.00 to 2.76) / 1.78 (1.07 to 2.96) | moderate positive | step 2 |
| pain, short (pooled) | SMD 1.64 (−1.06 to 4.33) | "strong positive", from the two trials' own between-group statistics | ||
| Custom orthoses (sham) | pain, short / medium / long | SMD 0.41 (0.07 to 0.74) / 0.55 (0.09 to 1.02) / 0.04 (−0.37 to 0.45) | strong positive (trials conflicting) / limited positive / moderate neutral | step 3 |
| function, short | −0.21 (−0.48 to 0.06) | moderate neutral | ||
| Prefabricated orthoses | pain, short; function, short | −0.25 (−0.59 to 0.09); −0.06 (−0.40 to 0.28) | moderate neutral | not one prescription for all |
| Calf stretching | first-step pain; pain; function, short | −0.39 (−0.80 to 0.03); 0.00 (−0.40 to 0.41); −0.24 (−0.65 to 0.17) | moderate neutral | not in the guide |
| Dry needling | pain; first-step pain, short | −0.33 (−0.76 to 0.10); −0.42 (−0.85 to 0.02) | moderate neutral (the trial's own covariance analysis found a small effect) | low-priority adjunct |
| Magnetised insoles; wheatgrass cream | pain, short | 0.00 (−0.39 to 0.39); not given | moderate neutral | not in the guide |
| Progressive strengthening / resistance exercise | — | no high-quality trial found | experts divided | not in the guide |
| Corticosteroid or platelet-rich plasma injection | — | inadequately tested; no placebo- or sham-controlled RCT | — | may be considered, preferably in an RCT or with structured evaluation |
Core approach content (expert reasoning): load management — break up long periods of static loading such as standing, reduce injurious compressive and stretch loading from activities such as running in the more active, avoid problematically rapid training changes in athletic populations; pain education — what pain means for the tissue, a good prognosis but slow resolution, "Techniques such as pain-monitoring were strongly recommended" (no thresholds in the main text); comorbidities (e.g. type 2 diabetes) and body mass addressed sensitively; footwear — supportive, comfortable, with a rearfoot-to-forefoot drop, socially acceptable, no barefoot walking or flat unsupportive shoes until symptoms have entirely resolved; one expert: "I don't think minimalist (footwear) is made for everybody".
Timing (from the qualitative components): all core components together for about 4-6 weeks before adjuncts; the flow chart adds shockwave at 4 weeks if the core is not working and orthoses at 12 weeks if improvement is still suboptimal; adjuncts when the core is only partly successful or taking more than 6 weeks; then experimental approaches, or a return to the core with attention to accuracy and adherence.
Strength training: "our stringent systematic review could not identify evidence in favour of more comprehensive exercise approaches", and the experts were divided on whether resistance exercise helps; trials in development may settle it.
Diagnosis (experts): clinical, from history (first-step pain, pain after inactivity, localised to the medial-inferior heel) and palpation; ultrasound to exclude tears or lumps, MRI for resistant cases or bone oedema. Expert subgroups: high body mass with long standing at work, and "the lean runner maybe doing too much too soon".
Returning to running: no return-to-running criteria, timeline or pain limit; running appears only in the load advice above.
Mechanisms
Measured: first-step pain, overall pain and function in the trials; no mechanism was measured.
Authors' interpretation: taping supports the plantar fascia; some experts tape first and, if symptoms fall, try to replicate the effect with shoes or orthoses; orthoses unload tissue under the heel where rest is not feasible; a small heel drop takes tension off the calf and therefore the fascia (expert quote).
Authors' interpretation: prefabricated orthoses failed in the trials perhaps because each trial gave one device to everyone without the usual education-and-modification process; a range of prefabricated devices fitted individually may work and needs testing.
Inference: the stretching evidence is one trial against radial shockwave in mainly acute heel pain, the taping evidence one trial at 1 week; "core" means best available and patient-acceptable, not proven over months.
Practical Application
What helps heel pain on the first steps in the morning? For 4-6 weeks: taping, plantar-fascia stretching and advice on shoes and load; if not better, shockwave, then custom orthoses. No runner subgroup; no tested return-to-run plan.
Graded by evidence vs by people, in the guide's own terms: trial-graded — taping (moderate, first-step pain at 1 week), plantar-fascia stretching (moderate, short and medium term), focused and radial shockwave (strong/moderate), custom orthoses (strong for short-term pain, trials conflicting); untested and recommended by experts with patient support — the education package (load management, pain education and pain-monitoring, footwear, body mass and comorbidities) and the whole sequence and its 4-6-week and 12-week timings.
Where heel raises and strength work sit: outside the guide. The review found no high-quality trial of progressive strengthening and the experts were split; that is untested, not shown useless.For the heel row of SYNTH-73 ("heel raises are worth asking about", from REF-920: 48 patients, not runners, better only at 3 months): name taping, plantar-fascia stretching and education first, and heel raises as an option a clinician may add.With 48 patients in two arms, that trial falls below this guide's 38-per-group power rule (our arithmetic; the main text does not list the 51 trials).
Calf stretching is not plantar-fascia stretching: in one powered trial calf stretching showed no effect on first-step pain, pain or function in the short term; the stretch with evidence targets the fascia.
Shoes, for a runner with heel pain (experts' advice, untested): comfortable, cushioned, with a small heel-to-toe drop; avoid barefoot walking and flat unsupportive shoes until symptoms have fully gone; no single prefabricated insert for everyone.
Load, for a runner (experts' advice, untested): reduce the running that loads the heel and avoid rapid training changes; break up long standing; monitor pain during and after activity (the guide gives no threshold). Not from this paper; the base's own practice: the pain ceiling the assistant gives is ≤3/10, and outside the Achilles that rule is transfer (SYNTH-73).
Not from this paper; the base's own practice: the assistant uses a clinician's diagnosis as given and never names the condition; not improving within about two weeks of sensible changes, or recurring, means a clinician (SYNTH-73). The guide's own escalation (shockwave at about 4-6 weeks) is a clinician's step.
What it is good for as a source: what to try first for diagnosed plantar heel pain and in what order, with the evidence grade of each piece and which pieces rest on experts and patients. What it does not show: effects in runners, any long-term benefit of taping or stretching, whether strengthening helps, or how and when to return to running.
Limitations
Few powered trials. Of 51 high-quality trials only 9 could establish efficacy; taping and plantar-fascia stretching each rest on one trial, and "strong" for radial shockwave pain was assigned from the trials' own statistics although the pooled SMD crossed zero (−1.06 to 4.33); "strong" for custom orthoses came from conflicting trials.
Mostly short term. 96% of trials measured the short term, 27% the medium term and 13% the long term.
The core is partly untested. Education was never tested on its own or added to another treatment in a trial; the timings came from the qualitative work; the progression to orthoses after shockwave is "extrapolated from expert interviews and systematic review findings, rather than from trials including failed previous treatment".
Who. Over-16s with heel pain, 49% women; the authors describe a condition of sedentary middle-aged and older adults; no runner subgroup.
No GRADE. The adapted van Tulder levels count trials and power; they do not rate imprecision or indirectness as GRADE does.
Reporting inconsistencies. The abstract reports 7 themes and 38 subthemes from the interviews, the Results 6 themes and 30 subthemes; the abstract lists focused shockwave's medium- and long-term first-step results next to the short-term one without saying that Table 1 grades them limited (one underpowered trial); the flow chart starts shockwave at 4 weeks while the text gives 4-6 weeks or more than 6 weeks.
Funding and interests: the Private Physiotherapy Education Fund (no role in the research) and NIHR/HEE and MRFF fellowships; three authors each ran a trial evaluated in the review. No device maker, federation or data platform is named; shockwave devices are discussed as a resource commissioners may fund.
How this entry was checked and changed
Why this level
a systematic review of quality-filtered RCTs is the rubric's A design; held at A- because only 9 RCTs (10 interventions) met the 38-per-group power rule for proof of efficacy and most graded findings rest on a single trial, 96% of trials reported short-term outcomes but only 13% long-term, certainty was not rated with GRADE, and the education, footwear, load-management and timing parts of the core approach were not tested and come from experts and patients; three authors ran trials that the review evaluated. Read from the full text (Europe PMC, CC BY 4.0)
Full text (checked 2026-10-28)
Fifty-one eligible trials enrolled 4351 participants, with 9 RCTs suitable to determine proof of efficacy for 10 interventions; forty people with PHP completed the online survey and 14 experts were interviewed (full text)
Taping (SMD: 0.47, 95% CI 0.05 to 0.88) and plantar fascia stretching (SMD: 1.21, 95% CI 0.78 to 1.63) for first step pain in the short term (full text)
Focused shockwave for first step pain in the short-term (OR: 1.89, 95% CI 1.18 to 3.04); custom foot orthoses for general pain in the short term (SMD: 0.41, 95% CI 0.07 to 0.74) and medium term (SMD: 0.55, 95% CI 0.09 to 1.02) (full text)
All recommended core approach components should be used simultaneously for approximately 4–6 weeks before consideration of adjunctive interventions such as ESWT or orthoses (full text)
Education was not tested for efficacy but was recommended by experts and appreciated by patients (full text)
Our stringent systematic review could not identify evidence in favour of more comprehensive exercise approaches; there is an absence of high-quality trials of progressive strengthening while the expert interviews showed very divergent views (full text)
Most studies had a small sample size (median=75, IQR=62); the mean duration of symptoms was 13 months and 49% were female participants (full text)
Ninety-six per cent of studies evaluated outcomes in the short term, but only 27% in the medium term and 13% in the long term (full text)
Calf stretching, first step pain: moderate neutral, −0.39 (−0.80 to 0.03) (full text)
Three authors (MC, BV, MSR) each conducted a randomised controlled trial that was evaluated in this review (full text)