Patterns of Risk Behaviors and Utilization of Care among Homeless and Housed HIV-Positive Patients Hospitalized in the Southeastern United States
1Elyse LaFond, Christine Bell MPH, Carlos del Rio MD
1Emory University Rollins School of Public Health, Atlanta, GA



Introduction

Homelessness and HIV

Homelessness and HIV-infection are important problems that commonly affect the same populations. In addition, homelessness is a risk factor for HIV, and HIV is a risk factor for homelessness.
The literature suggests that, when compared to HIV-positive housed persons, HIV positive-homeless persons exhibit:

Lower utilization of care
More likely to have never had an HIV care provider
More likely to not be taking HIV medication
More hospital visits
Poorer mental health

Increased risk behaviors
More sex partners
Greater unprotected sex with a partners of unknown serostatus
Greater sex in exchange for drugs or money
More likely to have used and/or injected drugs

Accelerated disease progression
Progression to AIDS approximately 9 times faster than general population


Methods and Materials

Recruitment
Bedside interviews were conducted with 2,656 stable, HIV-infected patients at Grady Memorial Hospital and Jackson Memorial Hospital from June 2006-July 2009. 349 (13%) patients were enrolled as of July 2009.

Screener Questionnaire to Determine Eligibility
Anonymous, verbal consent, conducted at bedside, data collected on handheld device → $5 compensation

Baseline Questionnaire
Randomized trial began August, 2006 → $35 compensation

Baseline interview: 1.5 hours, focusing on demographics, physical health, mental health, sexual history, drug use history, housing, and HIV attitudes and beliefs

Data Analysis of Baseline Interviews
Univariate analyses were used to describe demographics

Bivariate analyses were used to assess variables associated with homelessness


Results

Table 1. Demographic characteristics of study participants (n = 349; values are presented as percentages with frequencies in parentheses).



Table 2. Significant variables associated with currently being homeless by self report (bivariate analyses).






Conclusions and Future Studies

HIV-positive homeless persons are more likely to:
Never have been to HIV care
Engage in risk behaviors such as crack use and multiple sex partners
Exhibit depressive symptoms

These findings suggest that homeless persons are at increased risk for transmitting HIV and fueling the HIV/AIDS epidemic



Recommendations

Outreach intervention programs are needed to specifically target HIV-positive homeless populations in order to:

Link to primary HIV care and drug treatment
Reduce hospitalizations
Reduce high-risk behaviors Prevent transmission

Such programs should be tailored to the unique circumstances surrounding homelessness, including an increased emphasis on linkage to care through case management, reduction of risk behaviors, and improving mental health status


Resources

Funding: NIH/NIDA (RO1 DA017612), NIH/NIAID (P30 AI050409), and NIDA (R21 DA016133-01A1). This material is based upon work supported by the Howard Hughes Medical Institute under Grant No. 52005873 and by a fellowship award by the Concerned Parents for AIDS Research to Elyse LaFond.