Costs and health effects of osteoporotic fractures

Abstract

The objective of this study is to estimate the short- and long-term direct health care costs of hip, spine, and forearm fractures occurring during the remaining lifetimes of white post-menopausal women and to provide a prototype for estimating avoided costs of fractures in future cost-effectiveness analyses. A Markov model that uses population-based data and Monte Carlo simulations to estimate lifetime fracture risk, fracture-related functional impairment, and fracture-related costs for cohorts of individuals while controlling for competing causes of functional impairment and, hence, long-term costs is presented. Five cohorts of 10,000 women, each of a different age group, were simulated to obtain estimates of remaining lifetime fractures, functional impairment, and costs. Additional simulations tested the impact of discharge rates to and continued residence in nursing homes. Acute fracture care costs, nursing home annual cost, cost of community-based long-term care, and discount rates were varied in sensitivity analyses. Cohort costs were applied to the 1990 population distribution of U.S. white women to project future costs of women currently aged 45 years of age and older. Among hip, spine, and forearm fractures, hip fractures were determined to account for 36-50% of remaining lifetime fractures depending on age group of the cohort. Similarly, hip fractures were determined to account for 67-79% of fracture-related dependent functioning, 87-100% of fracture-related nursing home placement, and 87-96% of short-term fracture costs. Among white U.S. women aged 45 or older, an estimated 5.2 million hip, spine, and forearm fractures; 2 million person-years of fracture-related functional impairment; and $45.2 billion total direct medical costs can be expected in the next 10 years. Women aged 65-84 were estimated to experience the largest number of fractures, person-years of fracture-related impaired function, and fracture care costs in the next 10 years. Estimated lifetime cost was particularly sensitive to assumptions about fracture-related nursing home utilization rates. The future health and economic impact of established osteoporosis is expected to be substantial. Because we have included only three fracture sites, our estimates are likely to be conservative. Osteoporosis interventions that can reduce the need for fracture-related extended nursing home care and that are effective among women aged 65-84 are likely to be particularly cost-effective.