[Validation of a health survey questionnaire focusing on risk factors for coronary artery disease in a group of public employees in Juiz de Fora, Minas Gerais State, Brazil].

The validity of a health survey questionnaire used in a family health promotion program to identify individuals at risk of developing coronary artery disease (CAD) was studied in a random sample of 297 State public employees in Juiz de Fora, Minas Gerais State, Brazil. The sample was stratified by sex, and subjects were 40 years of age or older. Results obtained in interviews were compared to laboratory and clinical data. We assessed the questionnaire's sensitivity, specificity, and positive and negative predictive values in relation to questions about current or previous history of diabetes, hypertension, smoking, and high cholesterol. Among individuals with at least one altered clinical or laboratory test, 81% (total sensitivity) were identified by the referred risk inquiry when reporting at least one risk factor. Sheffield and New Zealand tables were used to determine the individuals with increased risk of developing CAD and to assess if they would also be identified by the health survey questionnaire. The sensitivity for identifying this subgroup was 100%.