Low agreement was found between pharmacy data and physician reported use of hospital-administered antenatal corticosteroids
2007 (English)In: Journal of Clinical Epidemiology, ISSN 0895-4356, E-ISSN 1878-5921, Vol. 60, no 5, 512-517 p.Article in journal (Refereed) Published
Objective: To evaluate physicians' recall accuracy on starting year of routine use of antenatal corticosteroids (ACS) to women facing imminent preterm delivery. Study Design and Setting: Starting year of routine treatment with ACS was evaluated by a questionnaire mailed to all 52 Swedish maternity wards. The information was compared with that obtained from a telephone interview with physicians involved in the introduction of routine ACS and with pharmacy data. From pharmacy data, routine use of ACS was defined as >= 20 mg betamethasone or dexamethasone purchased per maternity ward, year, and preterm delivery. Results: 24 hospitals with >= 10 preterm deliveries/year, having started ACS 1976-1997 and with information from questionnaire, interview, and pharmacy data were included in the analyses. There was fair agreement (kappa = 0.38, P < 0.001) on starting year between questionnaires and telephone interviews with physicians and slight agreement (kappa = 0.06, P = 0.098) between questionnaires and pharmacy data. Three hospitals had complete agreement on starting year according to questionnaire, telephone interview, and pharmacy data. Conclusion: Agreement between information obtained from questionnaires, telephone interviews with physicians, and pharmacy data on use of ACS was low. Physician reported past drug use needs to be validated and pharmacy data can be useful for such purpose.
Place, publisher, year, edition, pages
2007. Vol. 60, no 5, 512-517 p.
pharmacoepidemiology, drug use, questionnaire, interview, corticosteriods, comparative study
Medical and Health Sciences
IdentifiersURN: urn:nbn:se:uu:diva-150858DOI: 10.1016/j.jclinepi.2006.09.008ISI: 000245868300010PubMedID: 17419962OAI: oai:DiVA.org:uu-150858DiVA: diva2:409011