Abstract
BACKGROUND AND PURPOSE: Although the incidence of hip fracture during the past 50 years has increased, a break in this trend has been reported in the last decade. Whether this change is attributable to changes in bone mineral density (BMD) or whether it varies between urban and rural regions is unknown. METHODS: We evaluated changes in annual hip fracture incidence in women aged > or = 50 years in one urban population (n = 51,757) and one rural population (n = 26,446) from 1987 to 2002. We also examined secular differences in BMD (mg/cm(2)), evaluated by single-photon absorptiometry at the distal radius, prevalence of osteoporosis, and several other risk factors for hip fracture in one population-based sample of urban women and one sample of rural women aged 50-80 years at two time points: 1988/89 (n = 257 and n = 180, respectively) and 1998/99 (n = 171 and n = 118, respectively). RESULTS: No statistically significant changes were evident in annual age-adjusted hip fracture incidence per 10(4) when analyzing all women (-0.01 per year (95% CI: -0.37, 0.35)), rural women (-0.38 per year (-1.05, 0.28)), or urban women (0.19 per year (-0.28, 0.67)). BMD (expressed as T-score) was similar in 1988/99 and 1998/99 when analyzing all women (-0.09 (-0.26, 0.09)), urban women (-0.04 (-0.27, 0.19)), or rural women (-0.15 (-0.42, 0.13)) women. INTERPRETATION: Since no changes in age-adjusted hip fracture incidence and no differences in BMD were found during the study period, changes evident in the other risk factors for hip fracture that we investigated (such as gait velocity and balance) are either of minor importance or are counteracted by changes in other risk factors.