The risk of hip fracture among noninstitutionalized older adults

The risk of hip fracture among the 7,527 respondents to the Longitudinal Study on Aging (LSOA) is prospectively modeled using logistic regression techniques. Based on existing studies, a seven-stage hierarchical model serially introduces ecological, demographic, and social factors as well as general health status before considering symptoms and diseases conducive to hip fracture, falling history and body mass, and previous hip fracture. Interaction terms involving age and White women are then introduced to explore novel hypotheses. Of the LSOA respondents, 368 (4.9%) experienced hip fracture between 1984 and 1991. Significant risks of hip fracture were associated with age, female gender, White race, being hospitalized (for any cause) in the year prior to baseline, having fallen at least once in the year prior to baseline, and leaner body mass. The risk associated with increasing age diminishes over the life course. Similarly, the protective effect of body mass diminishes over the life course. Finally, previous ecological findings are clarified by identifying an elevated risk for White women living in the rural South. HIP fractures occur frequently among older adults, withsome studies indicating that by their ninth decade, 17 percent of men and 32 percent of women will have suffered a hip fracture (Gallagher et al., 1980). Virtually all hip fractures require hospitalization and result in considerable health care expenditures (Fitzgerald et al., 1987). Sequelae from hi