The Effect of Housing Instability on Risky Sexual Behavior and HIV and STIs for a Sample of African American Sero-Discordant Couples

Homelessness is a major public health problem in the United States and it has largely been associated with an increase in the cost of housing [1], however, the causes of homelessness include housing conditions, welfare program administration, employment status, as well as individual predictors such as childhood exposures to physical, sexual, or substance abuse; childhood neglect; domestic violence; mental illness; sexual orientation, death of a family members and others [2-4]. Recent estimates from the National Alliance to End Homelessness suggest a national rate of homelessness of 17.7 homeless people/10,000 people in the general population [5].

Housing conditions including where, how, and with whom persons live are known as descriptive factors along with a housing instability continuum and have long been associated with physical and psychological health outcomes. Homelessness refers to not having a place to reside and can be defined as the housing deprivation in its most severe form [6], and on the other hand, unstably housed are those that may have a place to live but might be in an economically o socially insecure situation that put them at risk of not having a secure place to live. Those suffering from unstable housing can either be associated with shelters; living “doubled up”; exchanging sex for shelter, or residing in single room units or transitional housing [4]. Unsheltered living (emergency shelters or street living), renders persons the most vulnerable. Housing instability (being unstably housed) may lead to homelessness. For the purpose of this study, unstably housed/or housing instability continuum includes the homeless populations.

Evidence suggest that those who are unstably housed have riskier sexual behaviors associated with high-risk of Human Immunodeficiency Virus (HIV) such as, injection drug use, greater number of sex partners and unprotected vaginal and anal sex-at a rate of four to six times higher than that of their housed peers [7-9]. Clinical studies suggest a relation between homelessness and unstable housed people having lower odds of Viral Load (VL) non-detectability after adjusting to age, ethnicity and drug use [9]. More specifically, those experiencing unstable housing and who are sexually active report close to seventy five percent having unprotected sex, and among those persons who were HIV positive, and 57.1% reported less than 100% Antiretroviral (ARV) adherence [10,11]. Gender differences regarding high sexual risk practices exist. Unstably housed men report having earlier onset of sexual debut, greater partner concurrency, and engaging in more survival sex relative to women. Conversely, females report less condom use and more sexual transmitted infections associated with substance abuse and low social support [12,13].

Even as HIV incidence and other Sexual Transmitted Diseases (STDs) morbidity and mortality go hand in hand with housing instability it has not thoroughly acknowledged or addressed in a housing instability continuum for Sero-Discordant Couples [14]. Sero-Discordant Couples are couples with one person who is HIV-positive and one who is HIV-negative. Studies in HIV Sero-Discordant Couples provide crucial evidence on the role of antiretroviral therapy and viral load suppression to reduce the risk of HIV transmission, however they are scarce. For example, an HIV-positive partner might focus on not infecting their partner by adhering to treatment, while the HIV-negative partner may concentrate on taking care of the other person. Some evidence on the viral loads and HIV transmission risk in HIV Sero-Discordant Couples suggest that having a drug addiction and living with nonfamily members are the most significant predictors to explaining sexual risk among newly homeless or unstably housed youth [15]. Viral load on HIV transmission risk in HIV-Sero-Discordant male homosexual couples suggest, however that HIV transmission in the context of viral suppression is very low in an international sample [16], while in the US evidence suggest a strong inverse association between housing instability and viral suppression across a spectrum of unstable housing arrangements [17].

The purpose of this study is to identify how various levels of housing instability influence HIV risk in a sample of African American heterosexual Sero-Discordant Couples, after controlling for HIV risk factors and socio demographic variables across different housing status categories using secondary data.

Parent project: Data source and study population

Table 1: Project EBAN recruitment and clinical sites description.

From: [19]
*Randomized Controlled Trial (RCT) cluster of African American HIV- Sero-Discordant recruited allocated to the EBAN HIV/STD Risk Reduction or the Health Promotion (RR-HP) comparison group.
+Distribution by gender of the HIV positive diagnoses of the Sero-Discordant Couples recruited in the RR-HP.

Current study

This study specifically hypothesizes: 1) housing instability has an effect on sexual risk after controlling for HIV risk covariates; and 2) along a continuum, those with increased housing instability exhibit the most sexual behavior risk. In addition, tested for gender differences, and if additional housing members affected HIV risk behaviors. 

The sample used in the analyses was of 1063 people. Of the 1070 sample of the parent study, 7 participants did not respond to the housing questions, and therefore were excluded from the data analysis.

Of 1063 participants, 605 (57%) were stably housed and 458 (43%) were unstably housed. Of 458 unstably housed participants, 104 (23%) lived with a family member, 189 (41%) lived with his/her sexual partner, and 165 (36%) lived in a rooming house, single room, group living arrangement, welfare-type living, or were unsheltered (“other living arrangement”). Baseline characteristics between participants in varying housing groups were dissimilar and, subsequently, statistically significant in more women (57%) were stably housed than men (43%), (x₂=49.15, p=0.000); the greatest within-group disparity among women’s and men’s housing groups were those who were stably housed (women: 65%, men 49%) and those living with a sexual partner (women: 10%, men: 26%).

As shown in table 2, bivariate analysis showed there were statistically significant differences in sexual risk behaviors based upon housing status. Specifically, dichotomized and grouped housing variables show there is significant at p < 0.05 difference between the expected and observed result for HIV and STI status. After controlling for covariates, adjusted odds ratios for primary outcomes revealed persons living with family members had a 48% decrease in likelihood of being HIV+ when compared to stably housed participants, yet they were 2.4 times more likely to be HIV+ than those who were living with a partner. Individuals living with a partner had a 78% decrease in likelihood of HIV+ status compared to stably housed participants. Persons living in group or unsheltered accommodations had a 44% decrease in likelihood of being HIV+ when compared to stably housed, yet they were 2.6 times more likely to be HIV+ than those who were living with a partner and 1.08 times more likely to be HIV+ than those who were living with family members. Similarly, persons living with family members had a 15% decrease in likelihood of being STI+ when compared to stably housed participants, though this result did not reach statistical significance?

Table 2: Describes the study sample, and distribution in dependent and independent variables.

Numbers rounded 
~Missing data on the different categories of unstable housing
^bivariate analysis of dichotomous variables t-test
*Statistically significant p < 0.005

Individuals living with a partner were nearly 3 times more likely to be STI+ compared to stably housed participants (AOR=2.687, p=0.007) table 3. Participants living with partners were 1.28 times more likely to be STI+ compared to persons in group or unsheltered living arrangements and 2.7 times more likely to have an STI compared to those living with family members. Persons living in group or unsheltered accommodations were 2 times more likely to be positive for an STI compared to stably housed participants and were 2 times more likely to have an STI compared to persons living with family members.

p<0.01

The results of this study revealed that housing instability has a statistically significant association with HIV status and sexual risk behavior. For each sexual risk outcome, housing instability (solely or when combined with an interaction term) was statistically significant as an exposure attributing to additional risk behaviors. These findings are consistent with epidemiological data showing an association between HIV status outcomes with housing instability. This study’s sample had a sixty eight percent of HIV- seropositives owned or rented their own homes. Equally, HIV-positive persons within this analysis were more likely to be older; being in a relationship with his or her partner for a longer time and were more likely to have known their status for nearly 1.5 years longer than unstably housed persons. Conversely, unstably housed groups were more at risk for positive STI status, partner concurrency, and number of unprotected sex acts. This merely confirms, along with previous literature, that after controlling for HIV risk covariates, housing instability hails as a statistically significant exposure variable for HIV and sexual risk behaviors.

The bivariate and multivariate analysis suggests that housing instability led to a significant decrease in HIV+ serostatus. This result is contrary to previous literature which documents housing instability should have a positive association with HIV status and, thus, risk for HIV-positive serostatus should increase as housing instability increases. These results may be explained as the study sample had more HIV+ women who were stable housed, while their male partners reported being unstably housed and were HIV-. Gender differences regarding high-risk sexual practices and factors that contribute to such risk are clearly demonstrated in the literature. Unstably housed men report having earlier onset of sexual debut, greater partner concurrency, and engaging in more survival sex relative to women. Conversely, females report less condom use and more STIs associated with substance abuse and low social support [20,21].

Our findings suggest that housing instability has a continuum that increases or decreases sexual risk. We proposed that housing instability and subsequent risk outcomes occur along a continuum. As persons live in situations that seem more vulnerable, their level of risk would increase. This work purported housing stability and risk along this continuum (from least risk to most risk): 1) stably housed, 2) living with family members, 3) living with partner, and 4) living in group or unsheltered accommodations. Our findings support that a continuum exists; however, it fails to establish groups that place participants at most or least risk consistently across outcomes. A potential limitation is that different behavioral, social, and structural factors impact behaviors-for instance, factors that place a person at risk for partner concurrency may be different than those that lead to unprotected sex.

Overall, our findings suggest that a better understanding is needed of what are the different permutations and granularity of unstable housing and how it relates with HIV risk. The over simplification of living arrangements may mislead findings on the risk factors for HIV and may misrepresent a vulnerable group in great need of public health and health services.

Due to the cross-sectional study design, it is difficult to estimate causality and to generalize the results across African American generally or to sero discordant heterosexual couples. Equally, co-variation of cause and effect could not be well established as the time when unstable housing began was unknown in relation to HIV onset for the seropositive individuals. Future research would do well to relate the health behavior and biological HIV-related health outcomes with qualitative research to understand the continuum of housing instability, social support, and where and with whom a person lives.

1. Glynn C, Fox E (2017) Dynamics of homelessness in urban America. Cornell University, Ithaca, New York.
2. Lee BA, Tyler KA, Wright JD (2010) The New Homelessness Revisited. Annual Review of Sociology 36: 501-521.
3. Bonnefoy X (2007) Inadequate housing and health: an overview. Int. J. of Environment and Pollution 30: 411-429.
4. Sermons MW, Witte P (2011) State of Homelessness in America. National Alliance to End Homelessness, Washington, DC.
5. Feldman BJ, Calogero CG, Elsayed KS, Abbasi OZ, Enyart J, et al. (2017) Prevalence of Homelessness in the Emergency Department Setting. West J Emerg Med 18: 366-372.
6. Crowley S (2003) The Affordable Housing Crisis: Residential Mobility of Poor Families and School Mobility of Poor Children. The Journal of Negro Education 72: 22-38.
7. Kipke MD, Weiss G, Wong CF (2007) Residential status as a risk factor for drug use and HIV risk among young men who have sex with men. AIDS Behav 11: 56-69.
8. Marshall BD, Kerr T, Shoveller JA, Patterson TL, Buxton JA, et al. (2009) Homelessness and unstable housing associated with an increased risk of HIV and STI transmission among street-involved youth. Health & place 15: 753-760.
9. Loh J, Kennedy MC, Wood E, Kerr T, Marshall B, et al. (2016) Longer duration of homelessness is associated with a lower likelihood of non-detectable plasma HIV-1 RNA viral load among people who use illicit drugs in a Canadian setting. AIDS Care 28: 1448-1454.
10. Wenzel SL, Henwood B, Harris T, Winetrobe H, Rhoades H (2017) Provider perceptions on HIV risk and prevention services within permanent supportive housing. AIDS Care 29: 1331-1335.
11. Wenzel SL, Rhoades H, Harris T, Winetrobe H, Rice E, et al. (2017) Risk behavior and access to HIV/AIDS prevention services in a community sample of homeless persons entering permanent supportive housing. AIDS Care 29: 570-574.
12. Flentje A, Leon A, Carrico A, Zheng D, Dilley J (2016) Mental and Physical Health among Homeless Sexual and Gender Minorities in a Major Urban US City. J Urban Health 93: 997-1009.
13. Harris T, Rice E, Rhoades H, Winetrobe H, Wenzel S (2017) Gender Differences in the Path From Sexual Victimization to HIV Risk Behavior Among Homeless Youth. J Child Sex Abus 26: 334-351.
14. Wolitski RJ, Kidder DP, Pals SL, Royal S, Aidala A, et al. (2010) Randomized trial of the effects of housing assistance on the health and risk behaviors of homeless and unstably housed people living with HIV. AIDS Behav 14: 493-503.
15. Solorio MR, Rosenthal D, Milburn NG, Weiss RE, Batterham PJ, et al. (2008) Predictors of Sexual Risk Behaviors Among Newly Homeless Youth: A Longitudinal Study. J Adolesc Health 42: 401-409.
16. Van de Ven P, Mao L, Fogarty A, Rawstorne P, Crawford J, et al. (2005) Undetectable viral load is associated with sexual risk taking in HIV serodiscordant gay couples in Sydney. AIDS 19: 179-184.
17. Clemenzi-Allen A, Geng E, Christopoulos K, Hammer H, Buchbinder S, et al. (2018) Degree of Housing Instability Shows Independent "Dose-Response" With Virologic Suppression Rates Among People Living With Human Immunodeficiency Virus. Open Forum Infect Dis 5: 035.
18. NIMH Multislice HIV/STD Prevention Trial for African American Couples Group (2008)Project Eban: An HIV/STD Intervention for African American Couples. J Acquir Immune Defic Syndr 49: 15-27.
19. El-Bassel N, Jemmott JB, Landis JR, Pequegnat W, Wingood GM, et al. (2010) National Institute of Mental Health Multisite EBAN HIV/STD Prevention Intervention for African American HIV Serodiscordant Couples: a cluster randomized trial. Arch Intern Med 170: 1594-1601.
20. Wachter K, Thompson SJ, Bender K, Ferguson K (2015) Predictors of multiple arrests among homeless young adults: Gender differences. Children and Youth Services Review 49: 32-38.
21. Tevendale HD, Lightfoot M, Slocum SL (2008) Individual and environmental protective factors for risky sexual behavior among homeless youth: an exploration of gender differences. AIDS Behav 13:154-164.