SEARCH

SEARCH BY CITATION

Keywords:

  • case–control study;
  • corticosteroids;
  • epidemiology;
  • fracture;
  • risk

Abstract.

Vestergaard P, Olsen ML, Paaske Johnsen S, Rejnmark L, Toft Sørensen H, Mosekilde L (Aarhus University Hospital, Denmark; and Aarhus and Aalborg University Hospitals; Aarhus, Denmark). Corticosteroid use and risk of hip fracture: a population-based case–control study in Denmark. J Intern Med 2003; 254: 486–493.

Background. Corticosteroids (CS) are used in a wide range of conditions but have several possible adverse effects including an increased risk of osteoporotic fractures.

Objective. To examine the association between cumulative CS dose and risk of hip fracture.

Design. Population-based case–control design.

Subjects and methods. A total of 6660 subjects with hip fracture and 33 272 age-matched population controls were identified using the County Hospital Discharge Registry in North Jutland County, Denmark and the Danish Central Personal Registry, respectively. Data on redeemed prescriptions for CS within the last 5 years before the index date were retrieved from a population-based prescription database, and recalculated to prednisolone equivalents. Cases and controls were categorized according to cumulative CS dose: (i) no use; (ii) <130 mg (e.g. equivalent to 30 mg of prednisolone for 4 days given for an acute exacerbation of asthma); (iii) 130–499 mg (e.g. equivalent to a short course of prednisolone of 450 mg for acute asthma); (iv) 500–1499 mg (e.g. equivalent to 7.5 mg prednisolone daily for 6 months or 800 μg day−1 of inhaled budesonide for 1 year); and (v) ≥1500 mg (e.g. equivalent to >4.1 mg day−1 for 1 year, a long-term high dose). Data were analysed using conditional logistic regression adjusted for potential confounders including gender, redeemed prescriptions for hormone replacement therapy, antiosteoporotic, anxiolytic, antipsychotic and antidepressant drugs.

Results. Compared with never users, an increased risk of hip fracture was found for CS users, with increasing cumulative doses of any type of CS use during the preceding 5 years [adjusted odds ratio (OR) = 0.96, 95% confidence interval (CI) = 0.89–1.04] for <130 mg prednisolone; OR = 1.17 (CI = 1.01–1.35) for 130–499 mg; OR = 1.36 (CI = 1.19–1.56) for 500–1499 mg; and OR = 1.65 (CI = 1.43–1.92) for ≥1500 mg. An increased risk was also found when the study population was stratified according to gender, age and type of CS (systemic or topical).

Conclusions. Even a limited daily dose of CS (more than an average dose of approximately 71 μg prednisolone per day) was associated with an increased risk of hip fracture.