Contact Us


Frequently Asked Questions

ETD Help

Policies and Procedures

Copyright and Patents

Access Restrictions

Search ETDs:
Advanced Search
Browse by:
Browse ProQuest
Search ProQuest

Laney Graduate School

Rollins School of Public Health

Candler School of Theology

Emory College

Emory Libraries

New ETD website is now LIVE and located here:

Quality of Care and Healthcare utilization among HIV infected individuals

Josephs, Joshua Saul (2015)
Dissertation (309 pages)
Committee Chair / Thesis Adviser: Sullivan, Patrick S
Committee Members: Del Rio, Carlos ; Gebo, Kelly (Johns Hopkins); Skarbinski, Jacek (CDC); Brent, Johnson (University of Rochester);
Research Fields: Health Sciences, Public Health
Keywords: quality of care; utilization; HIV; emergency department; structural equation model
Program: Laney Graduate School, Epidemiology
Permanent url:


HIV infection remains a common cause of healthcare utilization and an important driver of healthcare costs in the United States today. An estimated 1.1 million individuals are currently living with the disease. Although healthcare utilization among HIV-infected individuals has been studied extensively, there have been no national probability estimates of the frequency of healthcare utilization since the Healthcare Services and Utilization Study in 1998. We aimed to analyze two common measures of healthcare utilization, use of the emergency department and hospital admissions. We sought to estimate the frequency of these measures as well as explore new methods for modeling healthcare utilization. In addition to the standard technique of logistic regression, we used structural equation modeling (SEM). SEM has not been used to evaluate healthcare utilization; in particular, we examined the Gelberg-Andersen-Aday model, a commonly invoked, but infrequently analyzed, model of healthcare utilization. We also explored the predictive validity of the logistic regression model compared to the SEM. Finally, prompted by the National AIDS Strategy, which has measuring quality of care as one of its goals, we assessed five different composite quality measures along with their variance properties. We explored emergency department and hospital utilization using data from the Medical Monitoring Project (MMP). The MMP is a CDC-funded surveillance system of HIV-infected individuals in care in the United States. A smaller percentage of participants, 10.8% and 7.4%, respectively made visits to the emergency department or hospital, in 2009 than in prior studies. Using logistic regression we found that socio-demographic disparities and clinical variables such as CD4 count and viral load remain associated with healthcare utilization. Structural equation modeling generally found that the associations proposed by the Gelberg-Andersen-Aday model were supported by the data. Comparisons of the logistic regression and SEM found that the logistic model produced better specificity, while the SEM provided greater sensitivity. Using data from the HIV Research Network, a national longitudinal study, we found that the type of scoring system used produced radically differing scores, which ranged from 20%-80% depending on the score type. Scores increased uniformly over the study period. We also found that regardless of distribution used the variances of the quality metrics were similar. In conclusion, we found that socio-demographic disparities and clinical variables remain important risk factors for emergency department and hospital utilization. Quality of care composite measurements differed considerably. The MMP should continue to monitor changes in disparities over time and research should be conducted on the effect on mortality of reporting composite quality of care measures to providers.

Table of Contents

Table of Contents -- Chapter 1. Introduction to the Overall Project -- 1.0 Introduction to HIV and HIV care in the United States -- 1.1 A history of HIV treatment and the state of HIV care in the United States -- today -- 1.2 Theoretical models of healthcare access and utilization -- A. Figure 1.1: The Andersen model of healthcare utilization -- B. Figure 1.2: A directed acyclic graph of the measured variables in the -- Medical Monitoring Project and their effect on healthcare utilization -- C. Figure 1.3: Structural equation modeling diagram of the relationship -- between the measured variables, the latent factors, and healthcare -- utilization -- 1.3 Introduction to Structural Equation Modeling -- A. Figure 1.4: Directed acyclic graph of the maternal age, birth order, and Down Syndrome example -- 1.4 Issues and considerations in the measurement of quality of care -- A. Figure 1.5: Recommended quality of care measures by Horberg -- 1.5 Structure of the dissertation -- Chapter 2. Emergency Department Utilization Literature Review -- 2.1 Introduction to emergency department utilization -- 2.2 Literature review of emergency department utilization -- 2.2.1 Methods of literature review -- 2.2.2 Search strategy -- A. Figure 2.1 Flowchart of ED utilization literature searches 2.2.3 Literature review of factors associated with ED utilization -- 2.3 Conclusions -- Chapter 3. Hospital Utilization -- 3.1 Introduction to hospital utilization -- 3.2 Literature review of hospital utilization -- 3.2.1 Methods of literature review -- 3.2.2 Search strategy -- A. Figure 3.1 Flowchart of hospital utilization literature searches 3.2.3 Literature review of factors associated with hospital utilization -- 3.3 Conclusions -- Chapter 4. Quality of Care Literature Review -- 4.1 Review of quality of care -- Chapter 5. Emergency Department Utilization -- 5.1 Aims of the emergency department analysis -- 5.2 Main emergency department paper -- 5.3 Biases and limitations of the emergency department analysis -- Chapter 6 Structural Equation Modeling: Assessment of the Gelberg-Andersen-Aday Model of Healthcare Utilization -- 6.1 Aims of the SEM paper -- 6.2 Main SEM paper -- Table 6.1-6.3 Characteristics of individuals included in the study, coefficients and odds ratios for ED and hospital admissions. -- Figures 6.1-6.4 SEM diagrams for ED and hospital utilization, and receiver operating curves for ED and hospital utilization -- 6.3 Additional biases and limitations -- Chapter 7. Quality of Care Analysis -- 7.1 Aims of the quality of care analysis -- 7.2 Main quality of care analysis paper -- Table 7.1-7.4 Indicators included in the study, patient demographic data, percent of indicators completed, composite measures -- Figure 7.1 Graph of composite measures over time -- Table 7.5 Range of composite measures by site -- Table 7.6 Quality of care correlates -- 7.3 Biases and Limitations of the Quality of Care analysis -- Chapter 8 Conclusions and Future Directions for Research -- 8.1 Review of overall results -- 8.2 Limitations -- 8.3 Dissertation in context, innovation, and significance -- 8.4 Future directions for research -- 8.5 Final conclusions -- -- -- -- -- --


application/pdf Dissertation 309 pages (2.9 MB) [Access copy of Dissertation]
Supplemental Files
application/msword frontmatter4-8-15.doc (43.5 KB) [Front matter for Dissertation]
Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.