A population-based cohort study of mortality among adults prescribed paracetamol in Denmark

Objective: We examined overall and cause-specific mortality in a population-based cohort of individuals prescribed paracetamol. We evaluated rates of death due to liver and renal disease, cancer, and other major causes of death. Method: Using the population-based pharmacoepidemiology prescription database of North Jutland County, Denmark, we identified nearly 50,000 persons receiving prescriptions for paracetamol from 1989 to 1995 and evaluated mortality through 1996. Standardized mortality ratios (SMRs) for 18 specific causes of death were computed for persons prescribed paracetamol compared with persons among the general population of the county. Results: SMRs were elevated regardless of cause of death; with a nearly doubled overall death rate among persons prescribed paracetamol. The mortality ratios were highest within 1 year of paracetamol prescription and tended to decline with increasing follow-up, especially for cancers. The magnitudes of the excess mortality risks for liver and renal diseases were not materially higher than for other causes of death. Conclusion: Paracetamol is a first-line analgesic for patients with a wide variety of chronic or serious illnesses and therefore is associated with increased rates of nearly all causes of death. The findings are an example of confounding by indication. Paracetamol may be a drug most prone to this form of bias in pharmacoepidemiologic studies. Future observational studies reporting adverse effects of paracetamol must thoroughly evaluate confounding by indication before making causal inferences. (C) 2003 Elsevier Inc. All rights reserved.