An issue that arises when selecting therapy is whether patient or clinician preferences for treatment moderate the effect of treatment.
To evaluate this question we conducted a secondary analysis of the results of a randomized controlled trial of exercise treatment of chronic whiplash. Immediately prior to randomization, treatment preference ratings were collected from each patient and from the physiotherapist who assessed each patient. Patients were then randomized to receive advice alone or exercise and advice with the primary treatment outcomes of pain and function measured immediately after conclusion of treatment. Treatment effect modification was evaluated with a treatment group by preference interaction term in the regression analyses.
The analysis revealed that neither patient nor therapist treatment preferences moderated the effect of exercise treatment for chronic whiplash. The interaction effect of treatment group by patient preference was 0.1 (−0.3 to 0.5, p=0.68) on the 0–10 pain intensity scale and −0.1 (−0.5 to 0.3, p=0.64) on the 0–10 function scale. The interaction effect of treatment group by therapist preference was 0.0 (−0.3 to 0.4, p=0.786) on the 0–10 pain intensity scale and −0.2 (−0.4 to 0.1, p=0.296) on the 0–10 function scale.
Our findings do not provide evidence that patient or therapist treatment preferences moderate the effect of exercise treatment for chronic whiplash.