The impact of functional disability trends on health care expenditures of the growing US elderly population has not been well studied. OBJECTIVE: To examine the association between transitions in functional disability status and medical expenditures among elderly persons. METHODS: Data used were the 1998-1999 Medicare Current Beneficiary Survey (MCBS), a nationally representative survey of the Medicare population. The study sample included community-dwelling beneficiaries aged 65 years who were continuously enrolled in fee-for-service Medicare during 1998 and 1999. Functional disability status was measured in each fall as dependence in Activities of Daily Living (ADLs). Functional disability transitions (1998/1999) were coded into 6 categories based on stability, decline and improvements in ADL dependence. To capture expenditures as closely associated with ADL measures, we calculated Medicare (Part A&B) expenditures in the last quarter of each year. We examined changes in expenditures by the six transitional categories. OLS-regression analyses were conducted to control for sociodemographic, geographic, health status, and health insurance characteristics. RESULTS: The sample of 4452 beneficiaries (weighted N=14.3 million) had a mean age of 76.2±6.9 years; a large proportion were women (58.3%) and white (88.6%). While 60% of the sample maintained stable independence, about 7% improved to independence and 10.9% had an emergence of dependence in 1999. Beneficiaries with stable independence had a minimal change in total expenditures (+$178) between the last quarters of 1998 and 1999. The highest increase in expenditures (+$2886) was seen among beneficiaries who increased in dependence. Emergence of dependence among independent beneficiaries was also associated with an increase of $1318 while improvement to independence among dependent beneficiaries was related to savings of $315. Similar trends were also observed across inpatient, outpatient and physician expenditures. The impact of increase and emergence of dependence on medical expenditures continued to hold in the regressions. CONCLUSION: Significantly higher costs are associated with onset or increase in functional disability among the aged. Interventions to prevent or improve functional disability can be potentially cost-effective.