VHA's Rural category is very large and broadly dispersed; policy makers should supplement analyses of Rural veterans' health care needs with more detailed breakdowns. Most of VHA's Highly Rural enrollees live in the western United States where distances to care are great and alternative delivery systems may be needed.

This article describes a statistical modeling study designed to improve targets of need for registered nurse (RN) workforce. The model is place-based and incorporates the concepts of clinical need and regional service utilization. A cross-sectional study was conducted in Nebraska (1993-1999), and the unit of study was the county (N = 66). A mixed-model approach was used, and five predictor variables (% age 20-44,% age 45-64,% age 65+,% White non-Hispanic, and area) were significantly (p < .001) associated with service demand. Coefficient estimates were applied to various population projection scenarios, and the model’s algorithm converted service demand into number of RNs needed to compare numbers of RNs employed with projected need. The implications for RN workforce policy and funding decisions—at both federal and state levels—are significant. Further research with a larger, multistate database will be conducted to refine the model and demonstrate generalizability.

The results are discussed in regard to how a place-based approach can advance the study of rural health needs. By focusing on the needs of the people residing in a defined area, as determined from the aggregate characteristics of the population, a model is generated that can be used to predict special circumstances confronting any service provider. The public policy implications of the findings are also considered. Special payment policies could be written on the basis of place instead of provider characteristics, and grant programs providing technical assistance could be targeted to places of greatest need.

In order better to inform policymakers about financing uncompensated hospital care through appropriate allocation of resources among Nebraska communities, this study used seven years (1996-2002) of county-level data from multiple sources to examine the relationship between population economic factors and the hospital inpatient care use by uninsured patients. The generalized estimating equation (GEE) regression analysis showed that, at the county level, the population uninsurance rate and other economic factors (e.g., per capita income, the percentage of population receiving welfare) are statistically significant predictors of average hospital self-pay inpatient charge per resident. Residents in the three western regions of the state also incurred statistically higher per-resident hospital self-pay inpatient charges than did their counterparts in the three eastern regions. State policymakers in Nebraska can use our study results to allocate resources on the basis of community economic characteristics and geographic location, to help reduce the financial burden of caring for the uninsured to safety net hospitals. The study may have a wide application to other states examining the same policy issues. The model used in this study can be easily created for other states, as the required data are readily available.

No abstract