Objective: To study general practitioners' (GP) clinical diagnoses of heart failure.
Design: A Clinical Judgement Analysis study.
Setting: Primary health care.
Subjects: Twenty-seven GPs from nine health centres in Stockholm County.
Intervention: Forty-five case vignettes, based on actual patients from two health centres in Stockholm, were presented to each GP. For each case vignette, the GPs judged the probability of heart failure.
Main outcome measures: The GPs' assessments of the probability of heart failure in the case vignettes. The GPs' utilization of clinical information in their judgement strategies, as measured by the regression coefficients in a multiple regression equation, with the probability assessments as dependent and the clinical criteria as independent variables.
Results: The variation between the GPs' assessments of the probability of heart failure was considerable. The judgemental strategies differed between the doctors, the most important variables for most of them being lung and heart X-rays and a history of myocardial infarction.
Conclusions: With new treatment recommendations, it has become increasingly important to identify patients with heart failure. This study demonstrates large differences in GPs' diagnoses of heart failure. An important source of this variation is the differences in how they make use of clinical information.