OBJECTIVE: To evaluate the quality of medical diagnosis in the sick leave certificate. DESIGN: Retrospective, observational study. SETTING: Castelldefels Primary Care Centre, Barcelona, Spain. MEASUREMENTS AND INTERVENTIONS: 205 sick leave episodes of patients reported as non-occupational diseases and injuries were analyzed. The diagnosis of the sick leave certificate was checked against the clinical report carried out by the same physician who made out the sick leave certificate. To assess the quality of medical diagnosis, sensitivity, positive predictive value and the percentage of positive agreement were calculated. The variability of these indicators was also assessed according to the duration of the sick leave episode. RESULTS: Simple agreement between the diagnoses of the sick leave certificate and the clinical report was 92%. The sensitivity of the sick leave certificate in relation to the clinical report by diagnostic group ranged from 75% to 97%; positive predictive value, from 81% to 100%; and the percentage of positive agreement, from 86% to 97%. Short-term sick leave episodes (< or =15 days) had lower values than long-term sick leave (>15 days). CONCLUSIONS: Our findings confirm that the quality of diagnosis on the sick leave certificate is high. The availability of more medical information about the reasons for the sick leave episode does not modify the quality of the diagnosis substantially.
OBJECTIVE: To evaluate the quality of medical diagnosis in the sick leave certificate. DESIGN: Retrospective, observational study. SETTING: Castelldefels Primary Care Centre, Barcelona, Spain. MEASUREMENTS AND INTERVENTIONS: 205 sick leave episodes of patients reported as non-occupational diseases and injuries were analyzed. The diagnosis of the sick leave certificate was checked against the clinical report carried out by the same physician who made out the sick leave certificate. To assess the quality of medical diagnosis, sensitivity, positive predictive value and the percentage of positive agreement were calculated. The variability of these indicators was also assessed according to the duration of the sick leave episode. RESULTS: Simple agreement between the diagnoses of the sick leave certificate and the clinical report was 92%. The sensitivity of the sick leave certificate in relation to the clinical report by diagnostic group ranged from 75% to 97%; positive predictive value, from 81% to 100%; and the percentage of positive agreement, from 86% to 97%. Short-term sick leave episodes (< or =15 days) had lower values than long-term sick leave (>15 days). CONCLUSIONS: Our findings confirm that the quality of diagnosis on the sick leave certificate is high. The availability of more medical information about the reasons for the sick leave episode does not modify the quality of the diagnosis substantially.