Shoulder pain: Is the outcome of manual therapy, acupuncture and electrotherapy different for people with high compared to low pain self-efficacy? An analysis of effect moderation

Introduction: High pain self-efficacy (PSE) at the initial assessment predicts a better outcome for people attending physiotherapy for musculoskeletal shoulder pain, However, mechanisms for this remain unclear.

Objectives: To investigate whether there is a statistically significant difference in outcome between participants with high compared to low PSE for the following interventions: manual therapy, acupuncture, and electrotherapy.

Methods: Participants were stratified into i) high or low baseline PSE, ii) high or low baseline Shoulder Pain and Disability Index (SPADI), iii) did or did not receive the treatment. Effect moderation of treatment effect by PSE was assessed using 95% confidence interval of the Difference of Difference which provides a test of interaction effect between PSE and treatment group at 5% significance level (p<0.05).

Results: Six-month SPADI scores were consistently lower (less pain and disability) for those who did not receive passive treatments compared to those who did. This was statistically significant in 7 of 24 models. However, PSE did not moderate the relationship between treatment and outcome: DoD (whether the effect of the intervention varies by PSE level) was statistically insignificant for all subscores; low SPADI total (DoD, -4.43; 95% CI: -11.70, 2.83), low SPADI pain (DoD, -2.10; 95% CI: -6.53, 2.32), high SPADI pain (DoD, -2.83; 95% CI: -15.48, 9.83), high SPADI disability (DoD, -4.56; 95% CI: -17.69, 8.57).

Conclusion: Our results suggest that regardless of baseline PSE, patients who receive manual therapy, acupuncture, or electrotherapy experience equal or higher pain and disability at 6-months compared to those who do not. PSE did not moderate the relationship between treatment and outcome. This study was not designed to assess treatment effectiveness, results should therefore be interpreted with caution but do indicate the need for further appropriately designed research.