Objectives: The precise relationship between practice management (structure) and the doctor's actual performance (process) in general practice is tenuous. Analysis of their mutual relationship may yield insight into the way they contribute to outcome and into corresponding assessment procedures.
Method: In a cross-sectional study, consultations of 93 GPs were videotaped in their own practice and assessed by peer-observers on medical performance and on communication with patients, followed by a practice visit by a non-physician observer using a validated Visitation Instrument to assess Practice management and organization (VIP). Pearson correlations (observed and disattenuated for unreliability of the instruments) between scores on 22 practice management dimensions and scores of 16 selected cases on medical performance and communication were calculated. The predictive value of specific practice management aspects for actual performance was determined by multiple regression analysis, with performance scores as dependent variables and scores on the 22 management dimensions and GPs' professional characteristics as independent variables.
Results: Nine practice management dimensions correlated significantly with medical performance and so did five dimensions with actual communication. Overall, most associations were weak. Combined with demographic variables (age for medical performance and working single-handedly for communication), 26% of variance in medical performance scores could be explained by only three practice management dimensions. One practice dimension (delegation of medical tasks to the practice assistant) explained 11% of variance in communication with patients. Organization of quality assessment activities explained most of the variation in medical performance.
Conclusions: Practice management (structure) and actual performance (process) seem to be largely autonomous constructs. Quality improvement and assessment activities should emphasize that practice management is different from actual performance. Structure and process may contribute to patient outcome independently of each other.