OBJECTIVE: Severe disabilities are common among older people who have impairments in a range of physiologic systems. It is not known, however, whether the presence of multiple impairments, or coimpairments, is associated with increased risk of developing new disability. The aim of this study was to determine the combined effects of two impairments, decreased knee-extension strength and poor standing balance, on the risk of developing severe walking disability among older, moderately-to-severely disabled women who did not have severe walking disability at baseline. DESIGN: The Women's Health and Aging Study is a 3-year prospective study with 6 semi-annual follow-up data-collection rounds following the baseline. SETTING: At baseline, knee-extension strength and standing balance tests took place in the participants' homes. PARTICIPANTS: 758 women who were not severely walking disabled at baseline. MEASUREMENTS: Severe walking disability was defined as customary walking speed of >0.4 meters/second and inability to walk one quarter of a mile, or being unable to walk. RESULTS: Over the course of the study, 173 women became severely disabled in walking. The cumulative incidence of severe walking disability from the first to the sixth follow-up was: 7.8%, 12.0%, 15.1% 19.5% 21.2%, and 22.8%. In Cox proportional hazards models, both strength and balance were significant predictors of new walking disability. In the best balance category, the rates of developing severe walking disability expressed per 100 person years were 3.1, 6.1, and 5.3 in the highest- to lowest-strength tertiles. In the middle balance category, the rates were 9.6, 13.2, and 14.7, and in the poorest balance category 21.6, 12.7, and 37.1, correspondingly. The relative risk (RR) of onset of severe walking disability adjusted for age, height, weight, and race was more than five times greater in the group with poorest balance and strength (RR 5.12, 95% confidence limit [95% CI] 2.68-9.80) compared with the group with best balance and strength (the reference group). Among those who had poorest balance and best strength, the RR of severe walking disability was 3.08 (95% CI 1.33-7.14). Among those with best balance and poorest strength, the RR was 0.97 (95% CI 0.49-1.93), as compared with the reference group. CONCLUSION: The presence of coimpairments is a powerful predictor of new, severe walking disability, an underlying cause of dependence in older people. Substantial reduction in the risk of walking disability could be achieved even if interventions were successful in correcting only one of the impairments because a deficit in only one physiologic system may be compensated for by good capacity in another system.