OBJECTIVE: To define a typology of primary care (PC) practices based on a mixed inductive/deductive approach that uses a large number of variables describing organizational and demographic characteristics of practices and a priori hierarchical structuring of the data. DESIGN: Secondary analysis of the Swiss part of the QUALICOPC study using a multiple factor analysis approach incorporating 74 variables hierarchically structured and including information on infrastructures, clinical care, workforces, accessibility and geographic location of PC practices. SETTING: Switzerland. PARTICIPANTS: Two hundred randomly selected PC practices. MAIN OUTCOME MEASURES: Typology of PC practices based on axes identified through the multiple factorial approach. RESULTS: The factorial analysis extracted two uncorrelated axes summarizing 17% of the global variance. The first axis is mainly associated with two dimensions related to the comprehensiveness of services, namely 'clinical care provided' (Pearson's r = 0.73) and 'available infrastructures' (r = 0.78). The second axis is mainly associated with the workforce in the practice such as the number of general practitioners or other health workers (r = 0.69). Swiss PC practices were mapped using these two axes. CONCLUSIONS: This innovative approach allows defining a global typology of PC practices. Based upon Swiss data, two axes were identified to globally describe PC organization: comprehensiveness of services and workforces development. This exploratory study demonstrates a promising way, first to characterize globally one or several PC models that emerge from complex features, second to compare more accurately PC organization between countries and finally to assess how these models might be associated with patients' outcomes.
OBJECTIVE: To define a typology of primary care (PC) practices based on a mixed inductive/deductive approach that uses a large number of variables describing organizational and demographic characteristics of practices and a priori hierarchical structuring of the data. DESIGN: Secondary analysis of the Swiss part of the QUALICOPC study using a multiple factor analysis approach incorporating 74 variables hierarchically structured and including information on infrastructures, clinical care, workforces, accessibility and geographic location of PC practices. SETTING: Switzerland. PARTICIPANTS: Two hundred randomly selected PC practices. MAIN OUTCOME MEASURES: Typology of PC practices based on axes identified through the multiple factorial approach. RESULTS: The factorial analysis extracted two uncorrelated axes summarizing 17% of the global variance. The first axis is mainly associated with two dimensions related to the comprehensiveness of services, namely 'clinical care provided' (Pearson's r = 0.73) and 'available infrastructures' (r = 0.78). The second axis is mainly associated with the workforce in the practice such as the number of general practitioners or other health workers (r = 0.69). Swiss PC practices were mapped using these two axes. CONCLUSIONS: This innovative approach allows defining a global typology of PC practices. Based upon Swiss data, two axes were identified to globally describe PC organization: comprehensiveness of services and workforces development. This exploratory study demonstrates a promising way, first to characterize globally one or several PC models that emerge from complex features, second to compare more accurately PC organization between countries and finally to assess how these models might be associated with patients' outcomes.