Question | Discussion | References | CME Credit

Case 5: Discussion

Overview of Homelessness in the HIV-Infected Population

At present, more than 13 million individuals in the United States are homeless or have been homeless[1]. Behaviors most often reported by homeless adults that increase the risk of HIV infection include injection drug use and exchange of sex for money or drugs. As might be expected, the homeless and marginally housed have a significantly higher HIV prevalence than in the general United States population. Specifically, one study from San Francisco noted an HIV seroprevalence of 10.5% in homeless and marginally-housed individuals, with most HIV infections involving men who have sex with men, injection-drug users, or those who exchange sex for money or drugs (Figure 1). In a separate 3-year study conducted in Philadelphia in 1992-1994, investigators found that users of homeless shelter had a 9-fold increased risk of developing AIDS over a 3-year period when compared with the general Philadelphia population[3]. Homelessness is also associated with increased mortality in HIV-infected persons[4].

Barriers to Routine Medical Care

For HIV-infected homeless individuals, basic needs, such as food, clothing, and shelter, often supersede considerations regarding routine HIV medical care. As described in Maslow's hierarchy of needs, food, shelter, and safety needs must be met before individuals can consider other issues that affect their lives. Even when basic needs regarding food and shelter are met for homeless HIV-infected individuals, other factors, such as mental health disorders, drug and alcohol addiction, and lack of insurance, can serve as additional barriers that prevent them from engaging in care for their HIV disease. Moreover, since homelessness is nearly always associated with poverty, the lack of access to free medical care can serve as a further barrier to HIV therapy. Taken together, multiple factors exist that have resulted in HIV-infected homeless individuals more frequently seeking medical attention for acute illness than for routine HIV care. For example, one study reported that HIV-infected persons with a concomitant substance abuse disorder utilized emergency departments, urgent care, and inpatient services significantly more than housed HIV-infected individuals[5].

Housing Resources for Homeless HIV-Infected Persons

The Housing and Urban Development (HUD) office established the Office of HIV/AIDS Housing, with one of its major functions being management of the Housing Opportunities for Persons with AIDS (HOPWA) program. Currently, HOPWA is the only federal program dedicated to housing issues for HIV-infected persons[6]. Specific state-by-state HOPWA information is available at the HIV/AIDS HOPWA website ( In addition, other HUD programs may also provide valuable housing assistance, particularly for persons disabled by AIDS[6]. Providers and case managers should be aware that CARE funds can be used for housing needs of HIV-infected persons; details of the funding is contained in the HIV/AIDS Bureau policy 99-02: Use of Ryan White Care Act Funds for Housing Referral Services and Short-Term or Emergency Housing Needs[8].

Primary Care for the Homeless HIV-Infected Person

Establishing a stable primary care relationship with homeless HIV-infected persons is essential for providing optimal medical care. Community health centers, outreach programs, and case managers play an essential role in establishing and maintaining care for these individuals. Sustained engagement in primary care is often challenging due to issues such as ongoing substance abuse or mental illness[9], and it is essential to address these issues. In addition, clinicians should make every effort possible to involve case management as an integral component of primary care. Case managers or social workers can coordinate primary care and assist with referrals to housing, mental health, and substance abuse services[10]. In addition, clinics that work with the homeless need access to short-term emergency housing and multidisciplinary services to address the complex medical and social problems of the HIV-infected homeless population. In addressing the medical aspects of primary care for homeless HIV-infected persons, special emphasis should be placed on diseases that occur with increased frequency in this patient population. Accordingly, primary care services should include:

  1. Screenings for tuberculosis with purified protein derivative (PPD) skin testing every 6-12 months.
  2. Immunization to prevent pneumococcal pneumonia
  3. Annual influenza immunization
  4. Screening for hepatitis B and C infection among injection drug users.

As with all individuals with HIV, healthcare personnel should provide risk reduction counseling and education regarding HIV and other sexually transmitted diseases. Risk reduction interventions with homeless individuals have resulted in decreased high-risk sexual behavior[11].

Initiating Antiretroviral Therapy in the Homeless

The medical indications for initiating antiretroviral therapy in homeless HIV-infected persons are the same as in those HIV-infected persons who are not homeless[12,13]. For homeless HIV-infected persons, however, social factors often affect the feasibility and timing of initiating antiretroviral therapy. Prior to initiating antiretroviral therapy, medical providers should address the homeless person's basic needs, ongoing drug or alcohol problems, and mental health disorders[14]. In one report from San Francisco, only about 30% of homeless HIV-infected persons received combination antiretroviral therapy, as compared with 88% of HIV-infected men who have sex with men[15]. Multiple barriers may prevent homeless persons from receiving antiretroviral therapy, including provider reluctance, lack of a safe place to store medications, and concomitant alcohol, drug, and mental health problems[14]. Early protease inhibitor-based antiretroviral therapy regimens were notoriously complex and required a high number of pills, multiple daily doses, and timing around meals. These factors posed significant barriers for homeless HIV-infected patients[16]. Now that multiple recommended options exist for low-pill burden regimens taken once or twice daily[12,13], medical providers should attempt to use a convenient and compact regimen for homeless HIV-infected persons. In general, initial antiretroviral therapy for homeless HIV-infected individuals should consist of preferred components recommended in the DHHS antiretroviral therapy guidelines[12,13]. In addition, when choosing an initial regimen, medical providers must also consider mental health issues, life circumstances, and the potential interactions between antiretroviral medications and illicit substances.

Antiretroviral Therapy Adherence in the Homeless

Prior studies in homeless individuals with latent tuberculosis infection have shown poor adherence to therapy without interventions to improve adherence. Conversely, excellent adherence was achieved when medications could be conveniently obtained and simple incentives were used[17]. Antiretroviral adherence in homeless persons is a complex issue and not easy to predict. In one study, a substantial proportion of homeless and marginally-housed individuals had good adherence with protease-inhibitor based regimens (Figure 2). In a subsequent study, however, investigators reported that one-third of homeless HIV-infected individuals discontinued antiretroviral therapy within a one-year period of starting therapy[19]. Depression, injection-drug use, and African-American ethnicity were identified as predictors for discontinuing treatment. In this same study, the remaining individuals who continued on medications had adherence levels similar to other clinical populations of HIV-infected persons[19]. The authors emphasized the importance of addressing depression and drug-use issues before or during antiretroviral therapy in this patient population. A separate study involving HIV-infected injection-drug users found depression, but not housing status, was associated with adherence, with 63% of homeless or marginally housed patients reporting 100% adherence[20]. Multidisciplinary approaches involving case management, pharmacy, and nutrition focusing on adherence interventions are likely to improve these outcomes. Factors associated with improved adherence to antiretroviral therapy in homeless individuals have included case management[10] and incentive programs that provide cash or food[15].

Specific Considerations for Development of Resistance

Among homeless individuals taking unboosted protease-inhibitor based regimens, adherence strongly correlated with viral load measurement, with even a 10% decrease in adherence associated with a doubling of HIV RNA levels[18]. Accordingly, the implications of non-adherence are possible virologic failure and the development of resistant virus[21]. The different classes of antiretroviral medications have different thresholds for the development of resistance, with resistance to non-nucleoside reverse transcriptase inhibitor therapy occurring most often at low to moderate levels of adherence, resistance to single protease inhibitor therapy occurring most often at moderate to high levels of adherence, and resistance to ritonavir (Norvir)-boosted protease inhibitor therapy most likely occurring at middle ranges of adherence (Figure 3). When considering what regimen to use for homeless persons when initiating therapy, providers face a dilemma in that the most simple efavirenz (Sustiva)-based regimens also have a very low genetic barrier to resistance if adherence is problematic. Prescribing a simple, once-a-day regimen to an unstably housed person with HIV may not mitigate the need for sustained adherence support.

Directly Observed Therapy in the Homeless

Directly observed therapy for homeless individuals with active tuberculosis led to improved adherence and control of tuberculosis in some endemic areas[24]. Unfortunately, unlike therapy for tuberculosis, treatment for HIV does not have a finite duration of 6 months or less, but requires lifetime therapy. Although investigators have conducted multiple studies using directly observed antiretroviral therapy for HIV-infected persons in marginalized populations, limited data exist regarding directly observed antiretroviral therapy specifically for HIV-infected homeless persons. One study evaluated a directly observed antiretroviral therapy program in HIV-infected drug users, 35% of whom were homeless; a third of individuals in this study missed medical appointments and approximately 50% had been evaluated in an emergency department for an acute complaint. The study coordinators administered medications through a mobile clinic, resulting in significantly higher adherence in the directly-observed supervised therapy group than in the group that received unsupervised medication.


Clinical care programs with HIV-infected homeless persons must have access to the tools needed to support the extended needs of the homeless, such as short and long-term housing, multi-disciplinary case management, mental health programs, and drug treatment programs. Homeless HIV-infected persons are faced with additional barriers to obtain the long term health benefits of comprehensive HIV medical care. Although no single intervention works to erase these barriers, improved outcomes can be achieved with programs that address the specific needs unique to homeless patients.

[Back to Case 5 Question | See Case 5 References]

CME Credit | Back to Top

  • The following link will open in a new window.
    Figure 1. Relative Risk of HIV Infection Among Homeless and Marginally Housed Adults in San Francisco, by Risk Group

    Abbreviations: MSM = men who have sex with men; IDU = injection drug use

    This analysis of HIV risk factors is based on data from 2508 adults in shelters, meal programs, and low-cost hostels in San Francisco. These data were collected during a 21-month period that started in April 1996. Data from Robertson MJ, Clark RA, Charlebois ED, Tulsky J, Long HL, Bangsberg DR, Moss AR. HIV seroprevalence among homeless and marginally housed adults in San Francisco. Am J Public Health. 2004; 94:1207-17.

    Figure 1
  • The following link will open in a new window.
    Figure 2. Median Adherence to Protease-Inhibitor Based Regimen in an Indigent Population

    Abbreviations: AEMD = adjusted electronic monitored doses; PDDT = percentage days doses taken

    These data are based on the mean value of three periodic adherence assessments during a 6-10 week time frame. The investigators defined the percentage of days doses taken as the percentage of days that the subject opened the electronic medication monitor at least once during the entire study period. All patients were taking a protease-inhibitor based regimen.

    Data from Bangsberg DR, Hecht FM, Charlebois ED, et al. Adherence to protease inhibitors, HIV-1 viral load, and development of drug resistance in an indigent population. AIDS 2000;14:357-66.

    Figure 2
  • The following link will open in a new window.
    Figure 3. Correlation of Medication Adherence and Risk of Development of Antiretroviral Resistance

    Reproduced from Bangsberg DR, Moss AR, Deeks SG. Paradoxes of adherence and drug resistance to HIV antiretroviral therapy. J Antimicrob Chemother. 2004;53:696-9. Reproduced with permission from Oxford University Press and the British Society for Antimicrobial Chemotherapy.

    Figure 3