Medical care cost of Medicare/Medicaid beneficiaries with vision loss.

PURPOSE To assess the impact of vision loss on healthcare cost for patients with Medicaid and Medicare and whether these costs are adequately captured by Medicare hierarchical condition categories (HCC) risk adjustment methodology. DATA SOURCES The public use data set of the Program of All-Inclusive Care for the Elderly (PACE) for 1994-1998, and the Medicare 5% Sample datasets for 2003 and 2004. METHODS For the first analysis, up to five years of PACE data for each individual was used to calculate HCC scores (n = 3,459). For the second analysis, claim or encounter data from Medicare Fee-for-Service (FFS) and Medicare Advantage (MA) were used to estimate the cost for each beneficiary in the upcoming payment year (n = 2,108). RESULTS The increase in medical cost risk overall for visually impaired PACE participants was 10%, increasing to 13% for the non-institutionalized, community-based cohort, but PACE participants in nursing homes with vision loss did not generally result in increased costs. In the Medicare 5% sample, the HCC model under-predicts costs by about 17%. CONCLUSION Our analyses provide evidence that healthcare cost risk attributable to vision loss is not adequately captured by Medicare HCC risk adjustment methodology. We hypothesize this is due to additional morbidity and treatment patterns associated with visual impairment.