Date of Award

January 2013

Document Type

Open Access Thesis

Degree Name

Master of Public Health (MPH)


School of Public Health

First Advisor

Debbie L. Humphries

Second Advisor

Mayur M. Desai


Background: A `test-and-treat' strategy for HIV prevention in the United States is a method that was proposed in an effort to curb and reduce HIV transmission. However, the magnitude of the limitations observed at the various stages in the spectrum of engagement in care can threaten the `test-and-treat' HIV prevention initiative because successful retention remains problematic. Based on the United States' retention statistics, approximately one fourth to one half of those who were diagnosed with HIV were out-of-care. This figure suggests we need more information about the best ways to retain HIV-infected individuals in care.

Objective: In order to determine the best methods to retain HIV positive patients in care over time, primary predictors driving or influencing HIV positive patients' retention status must be assessed to better focus intervention strategies that would promote retention in care for those at high risk of falling out of care. This investigation sought to query into how one may employ the CAREWare database, used by Ryan White funded agencies, to study predictors of engagement in care at the Yale-New Haven Hospital Nathan Smith Clinic, which is the largest Ryan White funded clinic in Connecticut. The identification of key barriers influencing retention in care will allow health care providers and public health officials to devise a multidisciplinary team model that would specifically cater to the needs of HIV/AIDS patients at the Clinic, so that interventions may be aimed at optimizing the longitudinal care initiative and improving clinical health outcomes.

Method & Materials: A cross-sectional study was conducted to identify key predictors associated with the retention status of HIV/AIDS patients who attended the Nathan Smith Clinic between October 1, 2011, and October 1, 2012. The timeframe of one year was chosen because the Health Resources and Services Administration (HRSA) definition of retention is, over a given year, an HIV/AIDS patient must have at minimum two primary care visits, with each visit being at least 3 months apart. Clients who failed to meet the HRSA retention definition were designated as out-of-care. The community CAREWare database and Logician (GE Centricity), the Yale-New Haven Hospital's electronic medical records system, were employed to abstract patient information. The Pearson X2 and independent samples t-tests were used to assess the unadjusted associations between the patients' baseline characteristics and retention status. Logistic regression analyses and a backward elimination method of selecting variables were utilized to discover potential contextual factors influencing retention in care.

Results: From October 1, 2011 to October 1, 2012, 696 patients obtained outpatient HIV/AIDS primary care services at the Nathan Smith Clinic and were eligible for the study. Of this total, 134 (19.3%) HIV/AIDS clients were determined to be out-of-care under the HRSA definition for retention in care. The model selection technique yielded three statistically significant predictors of retention that best represented the data. Males were observed to be 1.81 times as likely to fall out-of-care compared to females [adjusted OR 95% CI (1.17, 2.82)]. Patients with no insurance [adjusted OR: 3.30, p=0.004], with private insurance [adjusted OR: 1.86, p=0.040] or with Medicaid [adjusted OR: 2.47, p=0.002] were at an increased risk of falling out-of-care compared to individuals with Medicare. Lastly, HIV/AIDS clients who had received none of the three HIV/AIDS support services (i.e. medical case management, mental health or substance abuse) had a 3.29 increased odds of falling out-of-care compared to individuals who had any one of the three support services [adjusted OR 95% CI (2.05, 5.26)].

Conclusions: These findings suggested that barriers affecting engagement in care at the Nathan Smith Clinic involved gender and insurance differences as well as engagement in HIV/AIDS support services. In order to optimize the longitudinal care initiative and improve clinical health outcomes, an HIV/AIDS multidisciplinary team model should be devised to target these predictors of retention.


This is an Open Access Thesis.

Open Access

This Article is Open Access