OBJECTIVES: To examine how local sales levels of antidepressive agents (ADs) correlated with GPs' conceptions of depressive disorders and of factors that may influence their work with depressed patients. DESIGN: A postal questionnaire survey to GPs requesting their conceptions of depression and their opinions of additional factors that may influence their work with depressed patients. GPs' conceptions and opinions were compared with local sales rates of ADs. SETTINGS: Three selected groups of Swedish municipalities: those with the highest, the average, and the lowest sales rates of ADs. SUBJECTS: All 535 GPs who worked in the selected municipalities. MAIN OUTCOME MEASURES: Spearman rank correlations for responses to the questionnaire with the sales levels of ADs. RESULTS: High sales levels correlated positively with a high evaluation of ADs' effectiveness in depression and panic disorders and were inversely correlated with the degree of appreciation of psychotherapy-based treatments. High sales levels were also associated with a high evaluation of GPs' own clinical and private experience, with a positive appreciation of the work with depressed patients and with a high level of participation in the pharmaceutical companies' activities. The demonstrated statistical correlations were not particularly strong and included less than half of the items. CONCLUSIONS: This ecological study confirms a number of statistical associations between sales levels of ADs and GPs' prevailing conceptions of factors related to depression. However, their explanatory value of the geographical sale variation appears limited. To further clarify this variation, studies employing information on individual GPs' conceptions and prescribing are required.
OBJECTIVES: To examine how local sales levels of antidepressive agents (ADs) correlated with GPs' conceptions of depressive disorders and of factors that may influence their work with depressedpatients. DESIGN: A postal questionnaire survey to GPs requesting their conceptions of depression and their opinions of additional factors that may influence their work with depressedpatients. GPs' conceptions and opinions were compared with local sales rates of ADs. SETTINGS: Three selected groups of Swedish municipalities: those with the highest, the average, and the lowest sales rates of ADs. SUBJECTS: All 535 GPs who worked in the selected municipalities. MAIN OUTCOME MEASURES: Spearman rank correlations for responses to the questionnaire with the sales levels of ADs. RESULTS: High sales levels correlated positively with a high evaluation of ADs' effectiveness in depression and panic disorders and were inversely correlated with the degree of appreciation of psychotherapy-based treatments. High sales levels were also associated with a high evaluation of GPs' own clinical and private experience, with a positive appreciation of the work with depressedpatients and with a high level of participation in the pharmaceutical companies' activities. The demonstrated statistical correlations were not particularly strong and included less than half of the items. CONCLUSIONS: This ecological study confirms a number of statistical associations between sales levels of ADs and GPs' prevailing conceptions of factors related to depression. However, their explanatory value of the geographical sale variation appears limited. To further clarify this variation, studies employing information on individual GPs' conceptions and prescribing are required.