Talk to the Veterans Crisis Line now
U.S. flag
An official website of the United States government

Health Services Research & Development

Veterans Crisis Line Badge
Go to the ORD website
Go to the QUERI website

IIR 10-144 – HSR&D Study

New | Current | Completed | DRA | DRE | Portfolios/Projects | Centers | Career Development Projects

IIR 10-144
Racial and Ethnic Disparities in Satisfaction with VA Care
Susan L. Zickmund PhD
VA Pittsburgh Healthcare System University Drive Division, Pittsburgh, PA
Pittsburgh, PA
Funding Period: November 2011 - February 2015

The VA is committed to equal treatment for all Veterans; however, some data suggest that racial and ethnic minorities are less satisfied with their VA healthcare than are Whites. We examined racial and ethnic differences in satisfaction with VA care in a racially-diverse patient sample.

Specific Aim 1. To examine the associations between patient race/ethnicity and satisfaction with overall, outpatient, inpatient, and 12 domains of VA health care based on a quantitative survey.

Specific Aim 2. To identify themes of satisfaction and dissatisfaction with overall, outpatient, inpatient, and 12 domains of VA health care in a diverse sample of Blacks, Hispanics, and Whites based on a qualitative interview.

Specific Aim 3: To use mixed methods to examine differences in themes of satisfaction and dissatisfaction with overall, outpatient, inpatient, and 12 domains of VA health care among Blacks, Hispanics, and Whites.

We identified a sampling frame of potentially eligible Veterans who received outpatient care within FY12 and FY13 at each of the 25 predominately minority-serving VA medical centers (VAMCs). The sampling frame was stratified by race/ethnicity and gender at each of the 25 sites. The target sample size was 1350, i.e., 9 Veterans per stratum per site.

Recruitment was conducted at all study sites simultaneously in 18 waves. During each wave, an initial mailing was sent to a stratified random sample of potential participants from the sampling frame and who were then called to confirm eligibility. Contact information for enrollees was sent to a University center which conducted the interviews. Our script asked Veterans to rate their satisfaction with their VA care on 15 domains important in the patient satisfaction literature and followed up with open-ended questions per domain.
Baseline demographic information, health literacy, comorbidity, and health status was ascertained by self report. Frequencies and percentages are reported for categorical data, and means and standard deviations are reported for continuous variables. Satisfaction was quantified based on a 5-category Likert scale. Preliminary analyses indicated sparse codes in the latter 3 categories, which we collapsed into one category: less than satisfied (LS).
We used mixed-effects multinomial regression to compare the proportions of participants who were somewhat satisfied (SS) or LS to those who were very satisfied (VS, the reference level). We modeled the gender-specific associations between race/ethnicity and satisfaction with the 15 domains of satisfaction. These primary domain-specific models included fixed effects for gender, age, race/ethnicity, and the gender-by-race/ethnicity interaction. A random effect for VAMCs accounts for clustering of Veterans within facilities. We assessed the overall gender-specific differences between race/ethnicity groups based on joint 0.05 level tests for both non-reference levels and constructed gender-specific linear contrasts of satisfaction for Black vs. White and Hispanic vs. White Veterans.

We also qualitatively assessed the audio-recorded qualitative interviews. We listened to over 200 interviews and developed an initial codebook using an iterative approach based on the editing style developed by Crabtree and Miller. We used 150 additional interviews to refine the codebook. The final in-depth codebook contained codes for satisfaction, dissatisfaction, and suggestions for VA improvements. Six qualitative coders listened to the interviews and coded each domain separately. All codes were transcribed verbatim and entered into the qualitative database.

Between June 2013 and November 2014, 8,010 Veterans were mailed invitations, 2,501 expressed interest in participating, 1,928 were screened as eligible, 1,387 consented, and 1,219 completed interviews, including 421 Whites, 388 Blacks, 394 Hispanics, and 16 of other race/ethnicity (excluded from the final analysis). Whites in the sample were on average older than Blacks and Hispanics (p=0.0004 and p<0.0001). They were more likely to be married than were Blacks (p<0.0001) and were less likely to receive all of their care at the VA than Blacks and Hispanics (p=0.0004 and p=0.0004).

Specific aim 1: 47% of patients described themselves as VS with their overall VA care. The 3 domains with the highest VS rating were: cost (78%), clinic physical facility (75%), and pharmacy services (74%). The 3 domains with the lowest VS rating were: interpersonal communication (58%), pain management (54%), and access (46%). We found significant non-adjusted rates by race in terms of being VS in the domains of respect (p<0.01) and outpatient care (p=0.02). Based on the estimated Age-Adjusted RRs from multinomial logistic regression, Black males and female Hispanics were more satisfied with access than were White males and White females, as they were less likely to be LS (RR=0.47 and 0.48, respectively, relative to satisfied). Hispanic females were less likely than were White females to be SS with Pharmacy (RR=0.49 relative to satisfied). With respect to Cost, Hispanic males were more likely than were White males to be SS (RR=2.85), i.e., Hispanic males were relatively less likely to be VS.

Specific aim 2: For the qualitative analysis, 521 interviews have been coded thus far, with 158 being independently double coded and adjudicated. We focus here on overall VA care where participants shared their unprompted views on any area of care. We examined the frequency of both satisfaction and dissatisfaction codes used to capture participants' views. In the overall domain 524 satisfaction codes and 325 dissatisfaction codes were identified. Of the unique satisfaction codes, 27% of the total focused on interpersonal communication, 27% on quality of care, and 26% on access. Of the unique dissatisfaction codes, 47% of the total focused on access, 21% on quality of care, and 15% on interpersonal communication.

This large and detailed study used stratified sampling to identify the rating of satisfaction with VA care in different subgroups of Veterans. We identified few significant racial and ethnic differences, even prior to adjusting for relevant covariates, although concerns related to the cost of VA care emerged for Hispanic males. Our qualitative results reinforce concerns frequently voiced related to VA access. Future results will provide specific information about whether racial and ethnic differences in satisfaction with VA care emerge in the qualitative coding.


Journal Articles

  1. Hausmann LRM, Jones AL, McInnes SE, Zickmund SL. Identifying healthcare experiences associated with perceptions of racial/ethnic discrimination among veterans with pain: A cross-sectional mixed methods survey. PLoS ONE. 2020 Sep 3; 15(9):e0237650.
  2. Zickmund SL, Burkitt KH, Gao S, Stone RA, Jones AL, Hausmann LRM, Switzer GE, Borrero S, Rodriguez KL, Fine MJ. Racial, Ethnic, and Gender Equity in Veteran Satisfaction with Health Care in the Veterans Affairs Health Care System. Journal of general internal medicine. 2018 Mar 1; 33(3):305-331.
  3. Hausmann LRM, Canamucio A, Gao S, Jones AL, Keddem S, Long JA, Werner R. Racial and Ethnic Minority Concentration in Veterans Affairs Facilities and Delivery of Patient-Centered Primary Care. Population health management. 2017 Jun 1; 20(3):189-198.
Conference Presentations

  1. Zickmund SL, Borrero S, Burkitt KH, Hausmann L, Switzer GE, Stone RA, Gao S, Rodriguez K, Fine MJ. Gender differences in satisfaction with VA health care. Paper presented at: VA HSR&D / QUERI National Meeting; 2015 Jul 10; Philadelphia, PA.

DRA: Health Systems
DRE: none
Keywords: none
MeSH Terms: none

Questions about the HSR&D website? Email the Web Team.

Any health information on this website is strictly for informational purposes and is not intended as medical advice. It should not be used to diagnose or treat any condition.