Figure 2.

Diagnostic concordance (Cohen κs and prevalence-adjusted and bias-adjusted κs [PABAK]) between in-person and remote assessment in patients with musculoskeletal disorders. Meta-analysis with pooled Cohen κs included studies with physiotherapists (6/32, 19%), orthopedic surgeons (23/32, 72%), and physiotherapy students (3/32, 9%) as evaluators. Meta-analysis with pooled PABAKs included studies with physiotherapists (6/30, 33%), orthopedic surgeons (21/30, 50%) and physiotherapy students (3/30, 17%) as evaluators. 95% PI: 95% prediction interval; RE: random-effect model; τ²/I²/P: test for heterogeneity [50-52,55-57,61].