[Educational intervention to improve diagnostic accuracy regarding psychological morbidity in general practice]

The aim of this pilot study was to evaluate the effectiveness of a complex educational intervention to improve the diagnostic competencies of general practitioners (GPs) regarding the detection of depression, anxiety and somatization.

Cluster-randomized controlled pilot study with six practices each in the intervention group and in the control group. Psychological morbidity was determined by patient self-report using the Patient Health Questionnaire (PHQ-D). GPs rated the extent of psychological morbidity on a numerical rating scale from 0 (no co-morbidity) to 10 (maximum) of the individual patient after the consultation, independent of the reason for encounter.

364 patients participated. There were moderate correlations between GP rating and the PHQ scales (Spearman correlation between 0.27 and 0.42). There was no significant difference between intervention and control group. Diagnostic accuracy of the GPs, as determined with areas under the curves (AUCs), ranged between 0.52 (95%KI 0.30-0.73) and 0.84 (95%KI 0.67-1.00). The AUCs showed significant heterogeneity (Cochran Q=25.0; p<0.01). The regression analysis with 'presence of psychological disorder' (in PHQ) as the dependent variable showed that longer duration of doctor-patient-relationship was negatively associated with psychological morbidity (OR 0.96; 95%KI 0.92-0.99; p=0.01). There was a significant interaction between the factors 'time of doctor-patient relationship' and 'GP rating' (ß=0.02; OR 1.02, 95%KI 1.01-1.03; p<0.001), pointing towards increasing diagnostic accuracy when patients are known for a longer time.

We found no significant effect regarding the educational intervention. The GPs' estimation regarding psychological morbidity correlated significantly with the self-rating of the patients on PHQ scales. However, there was a considerable inter-individual variation between the GPs' diagnostic accuracy. The diagnostic estimation improved with increasing duration of doctor-physician relationship.

A one-time educational intervention seems not to be sufficient to improve diagnostic competencies in the detection of psychological morbidity. The considerable variation of the diagnostic accuracy might explain why 'one-size-fits-all' educational interventions will not help improve diagnostic competencies.