A population-based cohort study of mortality among users of ibuprofen in Denmark.

Using the population-based Pharmacoepidemiologic Prescription Database of North Jutland County, Denmark, we identified 113,538 persons who filled prescriptions for ibuprofen during 1989 through 1995 and determined subsequent mortality through 1996. Standardized mortality ratios [SMRs] for 25 specific causes of death were computed compared with the general population. SMRs were elevated for most causes of death, with an overall SMR of 1.21 (95% confidence interval 1.19-1.24) among persons who filled prescriptions for ibuprofen. There was a nearly threefold increase in the number of deaths from gastrointestinal bleeding within 1 year of ibuprofen prescription but no concomitant increase in hemorrhagic stroke. Elevated SMRs were seen for several cancer types, although the mortality ratios were highest within 1 year of prescription and declined with longer follow-up. For colon cancer, SMRs were below 1.0 three or more years after ibuprofen prescription. For hypertensive disease, nonhemorrhagic stroke, and diabetes, we observed slight but significant elevations of the SMRs that persisted beyond the fifth year of follow-up. Our findings indicate a slight increase in overall mortality among persons receiving ibuprofen on prescription. This excess was greatest within the first year, due partially to ibuprofen-related gastrointestinal bleeding but mostly to elevated cancer mortality among ibuprofen users. This temporal pattern is characteristic of an effect of confounding by indication, with ibuprofen being used for pain relief by patients with imminent fatal illnesses such as cancer. The slight excess mortality that persisted beyond the first few years was largely due to elevated death from hypertensive disease and diabetes, which may be explained in part by increased prescription of ibuprofen to patients with long-standing medical problems.