State-Level Healthcare Spending and Health Outcomes: An Application of Spatial Durbin Panel Approach to Mortality

Posted: 19 Apr 2015 Last revised: 12 Sep 2015

See all articles by Srimoyee Bose

Srimoyee Bose

Georgia State University; West Virginia University, Davis College of Agriculture, Natural Resources & Design, Students

Tesfa Gebremedhin

West Virginia University

Usha Sambamoorthi

West Virginia University

Date Written: April 16, 2015

Abstract

There exists a complex relationship between state-level healthcare spending and health outcomes. As seen at the national level, healthcare expenditures are not associated with a corresponding improvement in health outcomes. This may or may not be the case at the state-level. Furthermore, the health outcomes of residents in a state are influenced by the residents’ demographic profiles, socio-economic factors, environmental factors, and healthcare needs. The ever-increasing healthcare spending without a corresponding improvement in health outcomes of individuals in the U.S. requires a closer examination of state-level polices and characteristics. As state governments are a vital driver of healthcare implementation and as healthcare policy responses in containing healthcare expenses and outcomes vary among states based on the underlying state-level factors, it is critical to examine state-level variations in healthcare outcomes.

The aim of this study is to assess the determinants of state-level mortality rates using a spatial Durbin fixed effect model and evaluate the association of mortality rate with healthcare/hospital expenditures using panel data from 2000 through 2009. The study revealed the presence of a significant positive spatial dependence of mortality rate among neighboring states. While population composition (percentage of African-Americans, percentage of male population and percentage of individuals over 65 years of age) and employment rate significantly increased the mortality rate of a state; percentage of Hispanic population, number of active physicians, percentage of married population, proportion of population aged below 5 years and percentage of population with a college degree (bachelors or higher) reduced the mortality rates of a state. Additionally, higher rates of Hispanic population, percentage of population below age 5, Gini coefficient (income inequality) and total number of hospitals per 1000 people of an individual state increased the mortality rates of the neighboring states and higher education level of the state decreased the mortality rate of the neighboring states. Finally the study reported that there is no impact of increasing health/hospital spending on the health outcomes (mortality rate).

Results of the study identified the importance of the role of social-determinants as well as up-stream factors such as income, social interaction and education in improving health outcomes (mortality rate). Hence, focusing on the economic, social and population factors of the state is necessary to reduce the mortality rate without further burden of increasing health spending on the states.

Keywords: Mortality rate, Spatial Durbin Fixed Effect, State-level, income inequality, married, education

JEL Classification: C21, C23, I10, H75, I180, J16, I13, I14

Suggested Citation

Bose, Srimoyee and Gebremedhin, Tesfa and Sambamoorthi, Usha, State-Level Healthcare Spending and Health Outcomes: An Application of Spatial Durbin Panel Approach to Mortality (April 16, 2015). Available at SSRN: https://ssrn.com/abstract=2595435 or http://dx.doi.org/10.2139/ssrn.2595435

Srimoyee Bose (Contact Author)

Georgia State University ( email )

35 Broad Street
Atlanta, GA 30303-3083
United States

West Virginia University, Davis College of Agriculture, Natural Resources & Design, Students ( email )

Morgantown, VA
United States

Tesfa Gebremedhin

West Virginia University ( email )

PO Box 6025
Morgantown, WV 26506
United States

Usha Sambamoorthi

West Virginia University

PO Box 6025
Morgantown, WV 26506
United States

Here is the Coronavirus
related research on SSRN

Paper statistics

Abstract Views
368
PlumX Metrics