Psychometric properties of a brief self-report Type A questionnaire for use in primary health care

Objective - To develop a short and easily used self-report measure of Type A behaviour (Simplified Type A Questionnaire=STAQ) and to examine its capacity to predict Type A behaviour as assessed by more time-consuming measures. Design - A Videotaped Structured Interview (VSI) and a self-report measure, previously validated in Sweden, were used as comparison instruments to the STAQ. Setting - Primary health care. Patients - 206 (81 men and 125 women) aged 17-75 years attending a health centre during one year. Main outcome measures - The reliability, measured by Cronbach's alpha and split-half correlation. The specificity and sensitivity of the STAQ. The correlation between the STAQ and the two other diagnostic measures. Results - The reliability of the STAQ was 0.70 and 0.75, respectively. The specificity was 83% and the sensitivity 75%. The correlation between the STAQ and the VSI was 0.45. The correlation between the STAQ and the previously validated self-report was 0.61. Conclusions - The STAQ has good psychometric properties and, in comparison with other diagnostic measurement instruments in medical service, it has an acceptable discrimination capacity.