[ Dose·Response ]
Not yetQuality score: 40/100

Effectiveness of novel facial stretching with structured exercise versus conventional exercise for Bell's palsy: a single-blinded randomized clinical trial.

Scientific reports · 2026

Statistically significant does not mean clinically proven: this 24-patient Bell's palsy trial needs replication with real numbers before it changes practice.

neurologyfacial-palsyexercise-rehabilitationrandomised-controlled-trialphysical-therapyrehabilitation

What they found

The experimental group (contralateral facial stretching plus structured exercise) showed greater improvement in SFGS score than conventional exercise alone (p = 0.002), but the abstract reports no mean change values, effect size or confidence intervals.

Study design

Parallel-group single-blinded randomised controlled trial, evidence level 2b, n=24 (12 per arm), acute Bell's palsy patients at a single hospital in Mangalore, India, with outcomes at baseline, day 10 and day 20.

Methodology & bias

Parallel-group single-blinded RCT, 24 acute Bell's palsy patients (12 per arm) at one Indian hospital, randomised to contralateral facial stretching plus structured exercise versus conventional exercise for 3 weeks, assessed on SFGS at baseline, day 10 and day 20. Key risk-of-bias issues: no stated power calculation so the trial is almost certainly underpowered for a condition with variable natural recovery; "single-blinded" leaves assessor blinding status ambiguous, which matters because SFGS scoring has a subjective component; no reported allocation concealment method; no intention-to-treat statement or accounting for dropouts/losses to follow-up; and only an abstract-level p-value with no effect size, confidence interval or mean change reported, which prevents separating statistical from clinical significance. The 20-day endpoint is also too short to capture Bell's palsy's natural history, which commonly continues improving over months, so between-group differences at day 20 could reflect differing recovery trajectories rather than a durable treatment effect.

The appraisal

A p-value of 0.002 only tells you the groups differ; with no reported mean SFGS change, effect size or 95% CI, you cannot judge whether that difference is clinically meaningful to a patient looking at their own face in a mirror. With 12 per arm, no stated power calculation, unclear assessor blinding, and only 20 days of follow-up, this is a hypothesis-generating pilot, not practice-changing evidence, and the statistical significance should not be read as proof of a durable or substantial clinical gain.

The gap

No effect size, confidence interval or mean SFGS change is reported, so a statistically significant p-value (0.002) cannot be interpreted as clinically meaningful, and the tiny unpowered sample with a 20-day endpoint cannot establish a durable treatment effect.

Landmark context

Bell's palsy rehabilitation conventionally centres on affected-side neuromuscular re-education, mirror biofeedback and symmetry exercises; this trial's novel addition of contralateral (unaffected-side) stretching is a plausible extension of that framework but this single small RCT cannot establish it as a new standard.

What to do Monday

Do not add contralateral facial stretching to standard Bell's palsy care on the strength of this trial alone; continue affected-side neuromuscular re-education as the evidence-based core and treat this technique as an unproven adjunct to monitor, not adopt.

Apply this in practice

Physiotherapists: Bell's palsy still needs ascertainment before any exercise programme starts, confirm acute onset, rule out central cause (forehead sparing suggests UMN lesion, needs urgent medical/stroke referral), and grade baseline facial function with a validated tool such as the Sunnybrook Facial Grading System or House-Brackmann scale, ideally photographed for comparison. If you choose to trial contralateral (unaffected-side) stretching alongside standard affected-side neuromuscular re-education, do so as an adjunct, not a replacement, and re-score with the same tool at day 10 and day 20 to track within-patient change, because this trial gives you no transportable effect size to predict how much improvement to expect. S&C coaches: Bell's palsy in an athlete is a medical presentation, not a training issue, your role is recognising red flags (sudden unilateral facial droop, inability to close the eye, ask about recent viral illness) and referring promptly to physio/medical rather than attempting any facial exercise intervention yourself; general training can continue as tolerated but corneal protection (eye care, lubricating drops) takes priority given exposure risk from incomplete eyelid closure. Both professions should treat this paper as a signal to watch rather than a protocol to adopt, since the absence of effect size and confidence intervals means you cannot yet counsel a patient on expected magnitude or timeline of recovery from adding contralateral stretching.

Read the paper ↗

Appraise any paper on demand

This is one critical appraisal from Dose-Response. Join and appraise any sports-medicine, rehab or performance paper yourself — plus the five that matter, every week.

Join Dose-Response
Shared from a Dose-Response member’s library. Educational critical appraisal, not medical advice. doseresponse.org