Effectiveness of stratified blended physiotherapy versus usual physiotherapy for neck and shoulder pain: a cluster-randomized trial

BACKGROUND: Neck and shoulder pain are common in primary care. A Stratified Blended Physiotherapy Approach (SBPA) combines prognostic risk stratification with blended care delivery, but its effectiveness compared with usual physiotherapy has not been established.

OBJECTIVES: To evaluate the effect of SBPA on combined pain and disability over 9 months, compared to usual physiotherapy in people with neck and/or shoulder pain.

METHODS: A multi-center, pragmatic, 2‑arm, parallel cluster‑randomized controlled trial was conducted in Dutch primary care. Ninety‑three physiotherapy practices were randomized (1:1) to SBPA or usual physiotherapy. In total, 139 people were recruited (SBPA = 74, usual care = 65) from 24 practices, and 136 completed the first questionnaire. Recruitment was affected by the COVID‑19 pandemic, resulting in a smaller sample than required (n = 238). SBPA combined prognostic risk stratification using the Keele Subgrouping for Targeted Treatment Tool (Keele STarT MSK Tool) with suitability for blended care assessed using the Dutch Blended Physiotherapy Checklist. People suitable for blended care received app‑based e‑Exercise modules; others received an equivalent paper‑based workbook. Usual physiotherapy followed guideline‑based care. The primary outcome was combined pain and disability, measured using the Neck Pain and Disability Scale (NPAD) or the Shoulder Pain and Disability Index SPADI (0-100). Secondary outcomes included: pain intensity, using the Numeric rating scale (NRS 0-10), Item Short Form Health Survey (SF‑36), Brief Illness Perception Questionnaire (B‑IPQ), Patient Activation Measure (PAM‑13), physical activity (Actigraph accelerometer), the Exercise Adherence Rating Scale (EARS), Global perceived effect (GPE), satisfaction and treatment content, and the number of sessions. Mixed‑effects models were used.

RESULTS: The SBPA arm decreased by 20.2 points compared with 21.6 points in the usual physiotherapy arm. Mixed-effects models showed no statistically significant differences between arms over time (P = 0.70). Adjusted mean difference was -0.91 points (95% CI, -7.39 to 5.56).

CONCLUSION: No statistically significant or clinically relevant differences were observed over 9 months. The trial was underpowered and subject to selection bias, limiting certainty regarding the effectiveness of SBPA in this context. Further research should focus on the economic implications of SBPA.

TRIAL REGISTRATION: Netherlands Trial Register, NL-OMON52723, registered 27 December 2019, first enrolment 30 September 2020.