[Accuracy of a self-administered questionnaire on the use of antibiotics].
BACKGROUND: To document the ambulatory use of drugs and the indications for that use, investigators often rely on self-administered questionnaires of questionable accuracy. The present study assessed the accuracy of a French Canadian self-administered questionnaire with regard to documenting current antibiotic drug use. METHODS: The information independently obtained from physicians and pharmacists was compared with the information reported by 340 patients. Patients were asked to participate in the study by their pharmacist at the time that their prescribed antibiotic was dispensed. The proportion of agreement between the data sources was calculated with regard to antibiotic regimen characteristics and indications for use and the nature of the treated infection. RESULTS: Self-reported information demonstrated a high level of agreement with data provided by physicians (k = 0.87) and pharmacists (k = 0.94), with regard to antibiotic names. Regarding the nature of the treated infection, the agreement between self-reported information and data obtained from physicians was substantial (k = 0.63). A total of 242 patients completed the questionnaire twice at two-week intervals. Test-retest reliability was high regarding both the antibiotic name (k = 0.72) and the nature of the treated infection (k = 0.86). CONCLUSIONS: The self-administered questionnaire assessed in this study can reliably and accurately document the name of antibiotics used and the nature of the treated infections. Further work is needed to improve the accuracy of the questionnaire with regard to other components of antibiotics use.