Objective: Most studies of potentially inappropriate medications (PIMs) among older adults have focused on prevalence rather than incidence. This study determined one-year incidence of PIMs use among Indiana Medicaid beneficiaries 65 years old or older who resided in nursing homes and examined associations between incident PIM use and hospitalizations and mortality. Methods: A retrospective analysis was conducted using Indiana Medicaid claims and enrollment files. Individuals were included in the sample if they were 65 years old or older, received Medicaid covered nursing home services from October 2002 through 12 months after starting a PIM in 2003 or until death in 2003, and were prescribed at least one new medication in 2003. Individuals who received any PIM in the three months prior to January 2003 were excluded. The 2003 Beers criteria were used to identify PIMs. Associations between incident PIM use, hospitalization and mortality were assessed using logistic regression models that controlled for age, gender, race, marital status, Charlson comorbidity scores, number of medications prescribed in 2003, and nursing home location. Selection bias was examined using seemingly unrelated bivariate probit models. STATA Intercooled for Windows was used for all statistical analyses. Results: The study sample consisted of 7594 individuals. One-year incidence PIM use was 42.1%. Rhos, correlations of error terms from equations predicting hospitalizations and mortality, were not significant indicating no selection bias. Incident PIM users were more likely to be hospitalized (odds ratio {OR} = 1.27, 95% C.I. 1.10 - 1.47) and more likely to die (OR= 1.45, 95% C.I. 1.31 - 1.61) in 12 months after controlling for demographic and clinical characteristics. Conclusion: Incidence of PIM prescribing was high among elderly Indiana Medicaid beneficiaries residing in nursing homes. Individuals who began use of a PIM in 2003 were at a higher risk of hospitalization and at higher risk of dying.