Diagnostic Accuracy and Clinical Decision‐Making Among Greek Physiotherapists in Direct Access Practice: A Cross‐Sectional Clinical Vignette Study

Background Direct‐access physiotherapy requires physiotherapists to make accurate clinical decisions whilst ensuring patient safety through appropriate triage and referral. Evidence regarding these competencies among Greek physiotherapists is lacking. Objective To evaluate the diagnostic accuracy and clinical management decision‐making abilities of Greek physiotherapists using clinical vignettes representative of direct‐access practice. Methods A cross‐sectional online survey was conducted among 253 licenced physiotherapists in Greece. Participants completed 12 clinical vignettes classified according to the original instrument as musculoskeletal, non‐critical medical and critical medical presentations. Diagnostic and management responses were scored dichotomously and summarised as accuracy scores and percentages. Descriptive statistics were used to summarise performance, a paired‐samples t ‐test was used to examine within‐participant differences between diagnostic and management scores and repeated‐measures ANOVA was used to compare diagnostic accuracy across the three clinical categories. Spearman correlation and general linear model analyses explored associations of diagnostic and management accuracy with clinical experience and screening behaviours, whereas group differences according to specialisation were examined separately. Results Mean diagnostic accuracy was 5.57 ± 2.49/12, whereas responses classified as correct were more frequent for management decisions (8.51 ± 1.45/12; p < 0.001). However, the diagnostic and management components differed in their response and scoring structures, limiting direct interpretation of this comparison. Diagnostic accuracy differed significantly across clinical categories ( p < 0.001), with the highest accuracy for musculoskeletal presentations (56.60 ± 27.01%), followed by non‐critical medical (50.89 ± 25.52%) and critical medical presentations (23.32 ± 25.30%). Diagnostic accuracy ranged from 13.8% to 96.8% across vignettes. Several clinical scenarios showed low diagnostic accuracy but comparatively higher management accuracy, suggesting that an acceptable management decision may sometimes be selected despite diagnostic uncertainty. No statistically significant associations were identified in the categorical‐factor general linear models, and no statistically significant differences in diagnostic or management accuracy were observed across physiotherapy specialisation groups. Conclusions Greek physiotherapists demonstrated variable diagnostic accuracy across clinical presentations and more frequently selected responses classified as appropriate for management; however, this difference should be interpreted cautiously because of the differing response and scoring structures of the two components. Diagnostic accuracy varied substantially across clinical categories and was particularly low for critical medical presentations. The vignette‐level findings suggest that appropriate management decisions may sometimes occur despite diagnostic uncertainty. Targeted education in differential diagnosis and red‐flag screening may improve clinical preparedness for first‐contact practice.