Housing Status and HIV Risk Behaviors Among Homeless

Housing Status and HIV Risk Behaviors Among Homeless and Housed Persons with HIV in the United States Daniel P. Kidder, Ph. D, 1 Richard J. Wolitski, Ph. D, 2 Sherri L. Pals, Ph. D, 2 & Michael L. Campsmith, DDS, MPH 2 U. S. Centers for Disease Control and Prevention National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention 1 Global AIDS Program 2 Division of HIV/AIDS Prevention XVII International AIDS Conference, Mexico City 6 August 2008 The findings and conclusions in this presentation have not been formally disseminated by the Centers for Disease Control and Prevention and should not be construed to represent any agency determination or policy.

Background: Homelessness in the U. S. § Determining numbers of homeless persons in U. S. is difficult – Often episodic: People cycle in and out – Chronically homeless persons: Difficult to locate § Social and economic factors associated with HIV in the U. S. are also associated with homelessness – Male, black, substance use, mental health, lack of education, history of physical & sexual abuse § Homeless persons: More chronic diseases and physical health problems than general population.

Background § Homeless persons are more likely than stably housed people to engage in HIV risk behaviors – Injection drug use – Risky sexual practices – Exchanging sex for money, drugs, or a place to stay § HIV/AIDS diagnosis rates are 3 to 9 times higher among homeless persons compared with those in stable housing in U. S. § Yet few studies conducted among homeless people living with HIV/AIDS – Most studies have been single site, small scale

Study Goals § Examine differences between homeless and housed people living with HIV/AIDS on: – HIV sexual transmission risk behaviors – Substance use (drug and alcohol) risk behaviors § Examine independent effects of homelessness on sex and substance use risk, controlling for potentially confounding factors

Methods U. S. CDC’s Supplement to HIV/AIDS Surveillance (SHAS) project § Behavioral survey of adults § All participants recently (past 2 years) diagnosed with HIV or AIDS in U. S. § Extends information routinely collected in AIDS surveillance § Cross-sectional § Multi-site

Participating SHAS Sites WA MN Detroit Chicago Denver Hartford, CT New Haven, CT Jersey City, NJ Paterson, NJ Philadelphia DE Baltimore, MD KS Los Angeles County, CA AZ NM Atlanta Austin/Travis County Houston Edisto Health District, SC Charleston County, SC Richland County (Columbia), SC Jacksonville Tampa Population-Based Facility-Based Miami

Methods (continued) § Questionnaire modules: • • • Sociodemographics Drug use Sexual behavior HIV testing and medications Health and social services use § Data collected from 1990 – 2003 § Current study includes data from May 2000 – December 2003 – n = 8075

Housing Status § At time of interview § Homeless – Living in shelter or on streets – n = 310 (4%) § Housed – Living in house, apartment, medical care facility, or correctional institution – n = 7765 (96%)

Sociodemographic Characteristics • Bivariate analyses Homeless Housed Age (average) 40 40 NS Gender (% male) 78 72 <0. 05 Race (%) p NS Black 60 55 Hispanic 20 19 White 18 22 Other 3 4 < High School Education (%) 46 31 <0. 001 Unemployed (%) 90 63 <0. 001 Annual household income <$10, 000 (%) 87 49 <0. 001 5 11 <0. 001 Marital status (% married)

Sexual Behavior • Bivariate analyses Homeless Housed p 63 69 <0. 05 >9, ever 84 73 <0. 001 >1, past 12 months 42 31 <0. 001 Ever 46 21 <0. 001 Past 12 months 15 4 <0. 001 16 10 <0. 05 Sexually Active, past 12 months (%) Sex Partners (%) Sex Exchange for Money/Drugs (%) Unprotected Sex at Last Sex with an Unknown Serostatus Partner (%)

Substance Use • Bivariate analyses Homeless Housed p Possible Alcohol Abuse (%) 55 35 <0. 001 Drug Use, past 12 months (%) 59 34 <0. 001 Injection Drug Use (IDU), ever (%) 37 19 <0. 001 46 31 <0. 001 Injection Drug Use, past 12 months (%) (among those who had ever injected)

Multivariate Logistic Regression Analyses § Dependent variables: – Sexual risk behaviors • Sexually active, # sex partners, sex exchange, unprotected sex with unknown status partner – Substance use • Alcohol abuse, drug use, injection drug use § Controlling for sociodemographic, HIV risk group factors often associated with homelessness – Age, gender, race/ethnicity, marital status, education, income, employment – Men who have sex with men (MSM), IDU, MSM/IDU, high risk heterosexual

Sexual Behavior Regression Analyses • Reference Group: Housed Participants Adjusted OR 95% CI 0. 97 0. 75, 1. 26 >9, ever 1. 59 1. 11, 2. 27 >1, past 12 months 1. 67 1. 29, 2. 15 Ever 2. 48 1. 89, 3. 25 Past 12 months 2. 91 1. 99, 4. 25 1. 99 1. 29, 3. 07 Sexually Active, past 12 months Sex Partners Sex Exchange (money/drugs) Unprotected Sex at Last Sex with an Unknown Serostatus Partner

Substance Use Regression Analyses • Reference Group: Housed Participants Adj. OR 95% CI Possible Alcohol Abuse 1. 63 1. 26, 2. 10 Drug Use, past 12 months 2. 31 1. 77, 3. 00 Ever 1. 90 1. 46, 2. 47 Past 12 months 2. 75 1. 87, 4. 06 Injection Drug Use

Summary of Results Compared to housed respondents, homeless had greater odds of having… § § § more sex partners exchanged sex for money or drugs unprotected sex with unknown status partner possible alcohol abuse used drugs injected drugs …even after controlling for possible confounding factors

Limitations § SHAS not representative of all people living with HIV/AIDS in the U. S. § Some sites only interviewed people with AIDS § Not a study of homeless people § Self-report data § Cross-sectional

Conclusions § Little research has been reported on homeless people living with HIV/AIDS in the U. S. § Results indicate importance of screening for housing status and behavioral interventions for PLWHA § Homeless PLWHA should be a special priority for substance abuse and HIV prevention programs