How do we compare with our colleagues! Quality of general practitioner performance in consultations for non-acute abdominal complaints

Objective. To investigate what factors influence the quality of general pra ctitioner performance in consultations for non-acute abdominal complaints a nd to establish the extent to which performance quality differs between gen eral practitioners (GPs). Design. Explorative study in two parts: (i) detection of variables influenc ing quality scores of consultations; and (ii) comparison of mean quality sc ores of the consultations, selected by each GP. Setting. Sixty-two family practices across The Netherlands, Subjects. Eight-hundred and forty consultations concerning non-acute abdomi nal complaints, first encounters; 62 GPs. Method. Multilevel analysis was carried out to detect factors that influenc e quality. After correction for the effect of significant factors the mean quality scores of individual GPs were calculated and compared. Results. Eighty-eight per cent of the total variance in quality scores was located at the consultation/patient level, and 12% at the GP level. One con sultation characteristic had significant influence on quality: quality scor es were higher in consultations of longer than average duration (>15 minute s), Several patient characteristics were of significant influence. Consulta tion quality scores mere higher in consultations for patients with upper ab dominal or non-specific abdominal complaints, Quality scores were lower in consultations with female patients and with patients aged >40 years. Togeth er these characteristics explained 20% of the variance at the GP level. Non e of the GP characteristics investigated in this study appeared to have sig nificant influence on thr quality of their performance. After correction of the scores for the effect of significant factors the differences in perfor mance quality between GPs remained significant. Conclusions. Quality of performance is far more influenced by consultation and patient characteristics than by GP characteristics. After correction fo r influencing factors, the mean quality scores of GPs still differed consid erably and significantly. For many GPs the quality scores varied substantia lly between different consultations; to a large extent this variation remai ned unexplained. Consultation quality can be improved by booking more time per patient and by giving more medical/technical attention to female and ol der patients.