• Presentation

Impact of Influenza Vaccination on the Cost and Use of Medical Services Among a Nationally Representative Sample of Elderly Americans

Citation

Gilman, B., & Bonito, A. J. (2005, December). Impact of Influenza Vaccination on the Cost and Use of Medical Services Among a Nationally Representative Sample of Elderly Americans. Presented at American Public Health Association Annual Meeting, Philadelphia, PA.

Abstract

Influenza and pneumonia (P&I) represent the 5th leading cause of death and account for over $12 billion in health care services among the elderly in the United States. However, most prevention effectiveness studies are based on non-representative samples of elderly beneficiaries in managed care, rely on administrative records to identify the immunized subgroup, use a narrow set of health status controls, and include only inpatient services. This study corrects these methodological shortcomings by using survey and claims data for a nationally representative sample of Medicare beneficiaries in fee-for-service over a four-year period. The study uses a four-part expenditure model to compare service use and costs between Consumer Assessment of Health Plan Survey (CAHPS) respondents who reported receiving an influenza immunization and those who reported not receiving a flu shot during four consecutive flu seasons (1999-2002). The outcome variables include: (1) the probability of having an inpatient hospitalization, (2) the probability of having an outpatient visit only, (3) the cost of inpatient services conditional on being hospitalized, (4) the cost of outpatient services conditional on using outpatient services only, and (5) total costs. Outcomes were defined over two sets of conditions, based on principal diagnoses, likely to be affected by influenza vaccination: P&I only and all acute and chronic respiratory conditions (ACRCs). Preliminary results suggest that beneficiaries who received an influenza vaccination are less likely to be hospitalized and to use services in an ambulatory setting for both P&I-related illnesses and ACRCs than those who did not receive an immunization. Among those who used inpatient services, the vaccinated group incurred fewer P&I- and ACRC-related costs than the non-vaccinated group. The impact of flu vaccination on outpatient costs was less clear. Average total costs among the vaccinated subgroup were lower for both P&I-related conditions and ACRCs. The magnitude of the savings varied depending on the duration and intensity of the flu activity and the match between vaccine and virus. The study offers further evidence that increasing the proportion of the elderly population who get vaccinated against influenza should result in lower net medical expenditures.