2006 HSR&D National Meeting Abstract
3093 — Effect of Gender and Race on Receipt of Influenza and Pneumococcal Immunizations in VA
Bean-Mayberry BA (CHERP, VA Pittsburgh)
Brucker N (CHERP, VA Pittsburgh)
Bayliss N (CHERP, VA Pittsburgh)
Xu X (CHERP, VA Pittsburgh)
Mor M (CHERP, VA Pittsburgh)
Crick E (CHERP, VA Pittsburgh)
Yano E (Center for the Study of Healthcare Provider Behavior, Greater Los Angeles Healthcare System)
Fine M (CHERP, VA Pittsburgh)
The VA uniformly promotes and reviews clinical preventive care for all veterans. However, little data are available to inform us about the equity of preventive care delivery. The objective of this study was to assess the performance of key immunization measures by gender and race in a national sample of veterans.
We examined the rates of influenza and pneumococcal immunization by gender and race in a cross-sectional sample of veterans selected by VA External Peer Review Program from 2001-2003 and linked with the National Patient Care Database (N=91,570). We performed univariate comparisons by gender for each demographic and clinical factor. We used multiple logistic regression to measure the association between gender and race (white, black, other, and unknown) with each immunization, adjusting for patient confounders such as age, marital status, VA eligibility status, clinical conditions recommended for immunization, geographic region (e.g., VISN), and clustering for repeated sampling of unique veterans across fiscal years.
Women were younger, unmarried, less frequently identified with clinical conditions for immunization, yet often sampled across fiscal years. Among persons >65, gender (OR 0.8, 95%CI 0.7, 0.8), black race (OR 0.7, CI 0.6, 0.8), and unknown race (OR 0.7, CI 0.7, 0.8) were significantly associated with a lower odds of pneumonia immunization, while age and marital status had a positive association. For influenza, black (OR 0.7, CI 0.6, 0.8) and unknown race (OR 0.9, CI 0.9, 1.0) were associated with a lower odds of immunization; yet age and marital status remained positive. Among younger veterans, gender was not associated with pneumonia immunization, but black (OR 0.8, CI 0.7, 0.9) and unknown race (OR 0.7, CI 0.6, 0.7) remained significantly associated with lower immunization. For influenza, gender displayed a significantly positive association (OR 1.6, CI 1.4, 1.8), yet age remained negative (OR 0.9, CI 0.9, 0.9) and race had no effect.
Our findings indicate that disparities exist in preventive care. Effects of gender vary with age and immunization, while effects of race appear consistently associated with a decreased odds of either immunization except in younger adults receiving influenza.
Gender and race may display complex relationships in VA that require novel and thoughtful considerations of differential veteran experiences. Our data warrant intensive study to understand these relationships and develop appropriate interventions to ensure equity.