Drug utilization according to reason for prescribing: a pharmacoepidemiologic method based on an indication hierarchy

Research output: Contribution to journalJournal articleResearchpeer-review


Purpose To develop a pharmacoepidemiologic method for drug utilization analysis according to indication, gender, and age by means of
register-based information. Statin utilization in 2005 was applied as an example.
Methods Following the recommendations for statin therapy, we constructed an indication hierarchy with eight mutually exclusive levels of
register markers of cardiovascular disease and diabetes. Danish residents, as of January 1, 1996, were followed at the individual level in
nationwide registers with respect to dispensed prescriptions of cardiovascular drugs and antidiabetics (1996–2005) along with discharge diagnoses
and surgical procedures (1977–2005). The highest current possible indication level was assigned to all cohort members. Stratified by
indication, gender, and age, statin treatment prevalence and incidence were calculated.
Results Statin treatment prevalence was highest among individuals with myocardial infarction and tended to be higher among men with
indications in the upper part of the hierarchy, but it was higher among women (especially the elderly) in the lower part of the hierarchy.
Treatment incidence rates followed roughly the same pattern. Women with no register marker or primary hypertension accounted for almost
50% of all incident female users; among men, the figure was 35%. The proportion of incident users with ischemic heart disease or myocardial
infarction increased with age.
Conclusion The proposed indication hierarchy provided new insight into prescription patterns of statins. The method can be implemented
for other drug categories and could be useful for studying trends in drug utilization, differential drug adherence, and cross-national
Original languageEnglish
JournalPharmacoepidemiology and Drug Safety
Publication statusPublished - 2011

Number of downloads are based on statistics from Google Scholar and www.ku.dk

No data available

ID: 33803745