Rural/Urban differences in Inpatient Health Services Utilization in New England

Session number: 1139

Abstract title: Rural/Urban differences in Inpatient Health Services Utilization in New England

Author(s):
William Weeks, MD, MBA - Veterans Rural Health Initiative Senior Scholar, VA Quality Scholars Program VISN 1 Patient Safety Center of Inquiry White River Junction VAMC
Rebecca P Lamkin, MA - CHQUER, Boston, MA
Steven M Wright, PhD - OPQ, Washington DC CHQUER, Boston, MA

Objectives: Recent reports suggest that the general population that lives in rural settings have less access to health services than their urban counterparts. We wanted to examine whether these trends existed in the veteran population, across multiple systems of care.

Methods: We used the merged VHA/Medicare database to examine inpatient service utilization across systems of care for veterans who were enrolled in VHA in VISN 1 from 1997 through 1999. We identified six age strata (<35, 35-44, 45-54, 55-64, 65-74, and 75+) and classified veterans into urban, suburban, or rural categories using their zip code of residence and the Department of Agriculture’s Rural/Urban Commuting Area codes. We used the Wilcoxon signed ranks test and regression analysis to compare the mean number of inpatient stays for rural to urban veterans.

Results: Over the three years, the proportion of enrolled veterans who lived in a rural setting increased from 18.3% to 19.8 % for all veterans, 17.9% to 20.2% for veterans over 65, and 21.3% to 26.2% for veterans who concurrently used Medicare. Veterans who lived in rural and suburban settings had fewer VA inpatient stays (for urban, suburban, and rural: 0.50, 0.37, and 0.37 in 1997; 0.29, 0.23, and 0.23 in 1998; and 0.40, 0.27, and 0.27 in 1999, p<0.0001 for each year). Veterans who lived in rural and suburban settings had fewer hospitalizations funded through Medicare than their urban counterparts (for urban, suburban, and rural: 0.46, 0.36, and 0.36 in 1997, p=0.2; 0.40, 0.35, and 0.33 in 1998, p<0.0001; and 0.61, 0.49, and 0.42 in 1999, p<0.0001). Regression analysis showed that, for each age group, increasingly rural designation was associated with fewer VHA inpatient stays and bed days of care (p<.01 for each year).

Conclusions: In New England, veterans who live in rural settings constitute a growing proportion of the enrolled veteran population. In comparison to their urban counterparts, they use fewer inpatient services, both through VHA and through Medicare.

Impact statement: The long travel distances required when inpatient services are regionalized may implicitly restrict access to care for rural veterans. VHA should consider different delivery models to improve access for rural veterans.