Date of Award

Spring 5-17-2018

Degree Type


Degree Name

PhD Health Sciences


Health and Medical Sciences


Ning J. Zhang, Ph.D.

Committee Member

Terrence F. Cahill, Ed.D.

Committee Member

Genevieve Pinto Zipp, Ed.D.


Access, utilization, limited English proficiency (LEP), self-rated health status, LEP Health Outcomes Assessment and Decision tool, immigrant, cultural competence, structural equation modeling (SEM), path analysis, health disparity.


Background and Study Purpose: Findings from previous studies suggest that, in a health care delivery context, individuals with limited English proficiency (LEP) are adversely impacted by lack of patient-provider language concordance. Yet, the concept of LEP has been mostly studied in the context of cultural competence and language has been generally considered a demographic or cultural characteristic. There is a growing body of research concerning LEP and health status; however, it is limited. This study sought to evaluate the effects of LEP on access, utilization, and self-rated health status (SRHS) among LEP respondents to a large health interview survey by comparing LEPs to two groups: English only (EO) and English and another language (E+OL).

Methods: The study design was retrospective, cross-sectional, and observational. Quantitative statistical analyses were required. Secondary data from the 2013-2014 California Health Interview Survey was used. N = 40,240 non-institutionalized Californian adults. The predictor was levels of English proficiency. EO was a reference group. The outcomes were access, utilization, and SRHS. Covariates were age, sex, race, income and education.

Results. Logistic regressions showed that compared to the E+OLs, LEPs had: (1) Lower odds ratio on all observed variables measuring access with statistical significance for some variables and others no statistical significance. (2) Lower odds ratio on all observed variables measuring utilization with statistical significance. Further, correlations among the all measurement variables were positive and effect sizes ranged from low to medium. Finally, results from a path analysis for LEPs showed a recursive inverse effect on access (p < .05, B = -0.27, 95% CI [-0.36, -0.18]), utilization (p < .05, B = -.80, 95% CI [-0.97, 0-.62]), and SRHS (p < .05, B = -.88, 95% CI [-1.04, -0.73]). In addition, there was a predictive effect of access on SRHS and access had a mediating effect related to LEP on SRHS (p = 0.003, 95% CI [0.01, 0.06]) and a predictive effect of utilization on SRHS and utilization had a mediating effect related to LEP on SRHS (p < .05, 95% CI [0.03, 0.06]). Further analysis showed that, when levels of English proficiency was not allowed a direct path to SRHS and access and utilization had respective direct paths to SRHS, path loadings were equal across EOs, E+OLs, and LEPs and were statistically significant across groups (access: p < .05; utilization: p < .05). These results suggest that levels of English proficiency contribute to the disparities observed among LEPs.

Conclusion: There are disparities in access, utilization, and SRHS among individuals with limited English proficiency. Those disparities can be reduced through decreasing barriers to access and utilization. Based on findings from this study, the LEP Health Outcomes Assessment and Decision model was developed and is being proposed for used in studying perceived health outcomes in LEPs.