Significant Heterogeneity in Pooled Patient Acceptable Symptom State Threshold for Shoulder Pain: A Systematic Review and Meta-analysis.

Abstract

To identify studies reporting patient acceptable symptom state (PASS) thresholds in diverse shoulder pain populations and to investigate whether patient-reported outcome measurements measuring shoulder pain severity provide similar estimates of the PASS across groups of people experiencing shoulder pain. Web of Science, Scopus, CENTRAL, and OVID (MEDLINE) databases were searched for studies on PASS related to shoulder pain using terms for PASS, shoulder pathologies, and pain-related patient-reported outcome measurements. Risk of bias (RoB) was categorized using a modified the Quality in Prognosis Studies Tool. The meta-analysis applied random effects models, with heterogeneity assessed via I² and Cochran's Q-test. Subgroup analysis was performed, with bias evaluated using funnel plots and the Egger test. Twenty-one studies were included, involving 8.992 participants with an average age of 50.55 years. The PASS thresholds for shoulder pain varied across the interventions and follow-up periods. The pooled PASS estimate was 20.18 (95% confidence interval 16.63-23.73, prediction interval 2.98 to 37.38), with significant heterogeneity (I² = 97%). Surgical interventions had lower PASS thresholds than nonsurgical approaches. Studies with longer follow-up duration and moderate RoB had lower PASS thresholds. A significant publication bias was identified. This systematic review and meta-analysis established a pooled PASS threshold for shoulder pain, highlighting substantial heterogeneity and significant differences in PASS thresholds across RoB levels, follow-up durations, and types of intervention. Level IV, systematic review and meta-analysis of Level II-IV studies.