Back to Annual Meeting

Back to Annual Meeting

APHA Scientific Session and Event Listing

Recorded presentation

Aaron J. Siegler, MHS¹, Ralph DiClemente, PhD², Gina Wingood, MPH ScD², Lara DePadilla, MS², and Delia Lang, MPH PhD². (1) Rollins School of Public Health and Center for AIDS Research, Emory University, 1518 Clifton Rd NE, Atlanta, GA 30322, 404-982-0853, asiegle@emory.edu, (2) Rollins School of Public Health and Center for AIDS Research, Emory Univeristy, 1518 Clifton Rd NE, Atlanta, GA 30322

Background: Risk behaviors associated with HIV are well researched, but within HIV seropositive (HIV+) women less is known. We assessed whether distinct risk patterns could be identified among 265 HIV+ women.

Methods: Women were recruited from two southern states as HIV prevention trial participants. Using a two-step cluster analysis, participants were assigned into four clusters based on alcohol use, marijuana use, other illicit drug use, unprotected sex with primary partner, unprotected sex with non-primary partner, commercial sex, and medication compliance.

Results: The first cluster is a small (n=10), high-risk group that contains every participant engaged in commercial sex (n=6). It has significantly more illicit drug use and non-significantly higher levels of every other risk category except unprotected sex with primary partners, which is significantly lower. The second cluster (n=34) is comprised of participants with significantly higher levels of marijuana use, unprotected sex with non-primary partners, and medication non-compliance. The third cluster (n=36) is characterized by alcohol use and unprotected sex with primary partners. The fourth cluster (n=185) has significantly lower rates of every risk category.

Conclusions: The analysis suggests that high-risk groups exist within the population of HIV+ women that need tailored interventions. A potential explanation for the discrepancy in rates of protection among primary and non-primary partners in the first cluster is that higher-risk behaviors outside the primary partnership necessitate protective behaviors within the primary partnership. This analysis indicates that protective patterns of HIV+ women may not be accurately captured by dichotomous outcomes, but instead necessitate deeper investigation.

Learning Objectives:

• Define the statistical technique of cluster analysis.
• Articulate several important risk behaviors of HIV seropositive women.
• Identify key risk groups of HIV seropsitive women.

Presenting author's disclosure statement:

Any relevant financial relationships? No

Recorded presentation