Background: It is not known whether the recently described break in the trend in hip fracture incidence in many settings applies in both women and men, depends on changes in bone mineral density (BMD) or changes in other risk factors, or whether it is apparent in both urban and rural settings.
Methods: We evaluated changes in annual hip fracture incidence from 1987 to 2002 in Swedish men aged ≥60 years in one urban (n = 25,491) and one rural population (n = 16,432) and also secular differences in BMD, measured by single-photon absorptiometry at the distal radius and multiple other risk factors for hip fracture in a population-based sub-sample of the urban and the rural men aged 60-80 years in 1988/89 (n = 202 vs. 121) and in 1998/99 (n = 79 vs. 69).
Results: No statistically significant changes in the annual age-adjusted hip fracture incidence per 10,000 were apparent from 1987 to 2002 in urban (0.38 per year, 95% CI -0.12 to 0.88) or rural men (-0.05 per year, 95% CI -0.63 to 0.53). BMD was similar in 1988/89 and 1998/99 when examining both urban (-19.6 mg/cm(2), 95% CI -42.6 to 3.5) and rural (-23.0 mg/cm(2), 95% CI -52.1 to 6.1) men.
Conclusions: Since no secular change in age-adjusted hip fracture incidence was found during the study period, a levelling off in hip fracture incidence is present also in Swedish men. Because BMD on a group level was similar in 1988/89 and 1998/99, changes in other risk factors ought to be either of minor importance or counteracted by changes in different risk factors.