Controversy and Debate on Meta-epidemiology. Paper 1: Treatment effect sizes vary in randomized trials depending on the type of outcome measure.

ABSTRACT

OBJECTIVE: To compare estimated treatment effects of physical therapy (PT) between patient-reported outcome measures (PROMs) and outcomes measured in other ways. STUDY DESIGN AND SETTING: We selected randomized trials of PT with both a PROM and a non-PROM included in Cochrane systematic reviews (CSRs). Two reviewers independently extracted data and risk-of-bias assessments. Our primary outcome was the ratio of odds ratios (RORs), used to quantify how effect varies between PROMs and non-PROMs; an ROR > 1 indicates larger effect when assessed by using PROMs. We used REML-methods to estimate associations of trial characteristics with effects and between-trial heterogeneity. RESULTS: From 90 relevant CSRs, 205 PT trials were included. The summary ROR across all the comparisons was not statistically significant (ROR, 0.88 [95% CI 0.70-1.12]; P = 0.30); however, the heterogeneity was substantial (I2 = 88.1%). When stratifying non-PROMs further into clearly objective non-PROMs (e.g., biomarkers) and other non-PROMs (e.g., aerobic capacity), the PROMs appeared more favorable than did clearly objective non-PROMs (ROR, 1.92 [95% CI 0.99-3.72]; P = 0.05). CONCLUSION: Estimated treatment effects based on PROMs are generally comparable with treatment effects measured in other ways. However, in our study, PROMs indicate a more favorable treatment effect compared with treatment effects based on clearly objective outcomes.