Service member and veteran mental distress rates and military-civilian residential segregation in South Carolina counties

ABSTRACT

We examine service member/veteran (SMV) mental distress rates in South Carolina counties and military-civilian residential segregation (MCRS). In phase one of our analysis, we utilized small area estimation (SAE) via a generalized-linear mixed model (GLMM) to calculate the probability of mental distress rates among SMVs, using individual-level data from the 2017 Behavioral Risk Factor Surveillance System (BRFSS). We applied these probabilities to demographic population counts (i.e. age by race by sex by military status) prepared at the county level by the U.S. Bureau of the Census in order to develop rates of SMV mental distress for each county. In phase two, we used these calculated mental distress rates and block-group-level 2013–2017 five-year American Community Survey (ACS) data to calculate MCRS for counties and to assess the relationship between SMV mental distress rates and MCRS. Phase one results showed that the average predicted mental distress rate among SMVs was 9.33 percent, although we found geographic variation across counties. Phase two results showed that the average mental distress rate was higher in counties with high compared to low MCRS (9.58 vs. 9.22) (Cohen’s d = 0.62). Social connection opportunities for SMVs and civilians are needed where high MCRS occurs.

KEYWORDS:

• Veterans
• mental health
• housing
• segregation
• civilians

Disclosure statement

No potential conflict of interest was reported by the authors.

Human Subjects Approval Statement

Because this study involved secondary data analysis of a publicly available dataset, no additional institutional review board examination was required.

Notes

1. We were unable to include ethnicity because the U.S. Bureau of the Census did not allow us to obtain tabulations for age by race by sex by ethnicity by SMV status due to privacy concerns.

2. We utilized the five-year population estimates from the U.S. Bureau of the Census because these estimates have a smaller margin of error than the single-year estimates for small geographies.

3. See the equations in Harris (2017).

4. We were unable to obtain household income by SMV status for each county in South Carolina from the U.S. Bureau of the Census. Therefore, we utilized per capita income in our analysis.

Additional information

Notes on contributors

Justin T. McDaniel

Justin T. McDaniel, PhD, is an Assistant Professor of Public Health at Southern Illinois University. His research interests include using GIS to illuminate health disparities among service members and veterans.

Alyssa B. Mayer

Alyssa B. Mayer, MPH, PhD, is an Assistant Professor of Health Promotion in the Department of Nursing and Health Professions at the University of South Carolina, Beaufort.

Robert J. McDermott

Robert J. McDermott, PhD, is a Professor of Public Health in the Department of Public Health and Recreation Professions at Southern Illinois University.

David L. Albright

David L. Albright, PhD, is an Associate Professor of Social Work and the Hill Crest Foundation Endowed Chair in Mental Health in the School of Social Work at The University of Alabama.

Hee Yun Lee

Hee Yun Lee, PhD, is a Professor of Social Work and the Endowed Academic Chair on Social Work and Health in the School of Social Work at The University of Alabama.

Eva Harara

Eva Harara, MA, is a doctoral student in the Department of Public Health and Recreation Professions at Southern Illinois University.