Abstract
Although homelessness ranks as one of society’s most pressing and visible health equity challenges, the academic community has not actively addressed its health impacts, root causes, and potential solutions. Few schools and programs of public health even offer a basic course for students. In the COVID-19 pandemic era, academia must demonstrate urgency to address homelessness and educate learners, motivate fledgling researchers, inform policy makers, offer community-engaged and evidence-based studies, and join in the growing national debate about best approaches. At a minimum, every public health student should understand the interdisciplinary challenges of homelessness, its implications for health equity, and opportunities to address the crisis. We call for academia, particularly schools and programs of public health, to engage more fully in national partnerships to care for members of society who are most marginalized, in terms of health and behavioral health outcomes, quality of life, and connectedness.
As part of its mission, academia regularly tackles complex, seemingly intractable, health challenges—HIV/AIDS, cancer, and COVID-19, for example—to heighten societal attention, advance understanding, and offer options for evidence-based policy action. Yet, academia routinely overlooks homelessness, one of society’s most pressing health crises, for reasons that include lack of national research funding and insufficient attention to populations at increased risk. 1 Homelessness stems from a constellation of social determinants of health—such as extreme poverty, housing instability, food instability, and racism—and is complicated further by physical and mental trauma, stigma, and inadequate systems of care. This increasingly visible crisis has, throughout the COVID-19 pandemic, featured coast-to-coast media headlines about people living in shelters and on the streets, rising housing instability, heightened political battles over encampments, and the end of the national eviction moratorium, which leaves millions of people at risk of homelessness. 2
Easy solutions have defied policy makers for decades. Recently, however, renewed opportunities for national action have emerged through commitments from the US Interagency Council on Homelessness (USICH), the US Department of Housing and Urban Development (HUD), the American Rescue Plan, and nongovernmental organizations. At this critical time, we call on academia, especially schools and programs of public health, to educate students more fully about the fundamental dimensions of health and homelessness, support and mentor investigators, partner with community groups, and contribute to evidence-based approaches to care for the most marginalized members of society, in terms of health and behavioral health outcomes, connectedness, and quality of life. 3
Current Landscape
HUD defines homelessness as “lack[ing] a fixed, regular, and adequate nighttime residence.” 4 A 2018 analysis of use of shelter services during the course of a year found that annual homelessness numbers may be as high as 1.6 million people. 1 However, using a single January night (“point-in-time count”) for census tracking, HUD’s 2020 Annual Homeless Assessment Report found that approximately 580 000 people were experiencing homelessness (354 000 sheltered and 226 000 unsheltered) in the United States 4 ; the COVID-19 pandemic disrupted comprehensive tracking in the recently released 2021 HUD report. 5 Although the overall rate of homelessness between 2007 and 2020 decreased by 10% (and by nearly 50% for veterans), the prevalence of homelessness (using HUD’s point-in-time count) increased each year beginning in 2017 through 2020. 4 From 2015 to 2020, the total unsheltered homeless population (ie, individuals and families whose primary nighttime location is a public or private place not designated for, or ordinarily used as, a regular sleeping accommodation for people, eg, streets, vehicles, or parks) rose by almost 53 000 (from 173 000 to 226 000 people), comprising 39% (from 31%) of the overall homeless population. 4 For individuals in particular, in 2020, those experiencing unsheltered homelessness exceeded the sheltered population (51% vs 49%) for the first time. 4 One nationally representative sample of more than 36 000 US adults estimated lifetime and 1-year homeless prevalence of 4.2% and 1.5%, respectively.6,7 Because homelessness assessment methods differ by city and definitions vary by federal agency (eg, between HUD and the US Department of Education), future policies must harmonize efforts. 8
In 2020, almost 60% of people experiencing homelessness lived in urban areas. 4 California accounted for 28% nationally (51% of the unsheltered), and Los Angeles County alone had more than 60 000 people without homes each night. 4 The highest 2020 homelessness rates were in New York and Hawaii (47 and 46 people per 10 000 population, respectively), followed by California and Oregon (41 and 35 people per 10 000 population, respectively). 4 From 2007 to 2020, New York (45.8%), California (16.2%), Massachusetts (18.8%), and the District of Columbia (19.9%) had the largest percentage increases in homelessness. 4 The proportion of adults aged >50 years experiencing homelessness has increased as baby boomers (ie, people born between 1945 and 1964) age. 1 About half of the sheltered population is disabled. 1 Other groups disproportionately experiencing homelessness include infants and children, adolescents, immigrants, and racial and ethnic minority populations (Table).3,9-12 Notably, Black people, 12.6% of the overall US population, comprise 39.4% of those experiencing homelessness 4 ; this disproportionate percentage reflects, in part, a legacy of societal discrimination. In particular, the practice of “redlining”—a set of federal policies dating from the 1930s that used color-coded maps with red codes (hence the term) to indicate neighborhoods where Black Americans lived as risky investments—excluded those residents from government-issued mortgage programs. 1
US homelessness rates from point-in-time counts, by demographic characteristic, 2020 a
Abbreviation: —, does not apply.
The analysis of data was replicated with methodology from Evans et al. 3 Risk ratios (RRs) are the ratio of subpopulation-specific homelessness rates to non–subpopulation-specific homelessness rates. An RR of 1 indicates no difference in the probability of homelessness between the subpopulation of interest and the general population. An RR of 2 signifies that the subpopulation of interest has double the chance of homelessness as the general population. A sample calculation of RR for people aged <18 years experiencing homelessness shows the following:
Population breakdown from the US Census Bureau; includes US states and Washington, DC; both Hispanic and non-Hispanic members are included in each category. 9
From the Substance Abuse and Mental Health Services Administration. 10
From the Centers for Disease Control and Prevention. 11
Providing Care
Major health conditions both cause and exacerbate homelessness. One study documented substantially higher mortality risk in the unsheltered homeless population of nearly 3-fold versus the sheltered homeless population and nearly 10-fold versus the general population. 13 Major causes of mortality include drug overdoses, cancer, and chronic heart and lung diseases. 13 One major meta-analysis estimated that 76% of people experiencing homelessness have behavioral health conditions, including severe mental health disorders (eg, schizophrenia spectrum disorders, mood disorders) 14 and alcohol, tobacco, and other substance use disorders. Furthermore, people experiencing homelessness are more likely to have other major health conditions, such as HIV and hepatitis C infections, and are more likely to be exposed to weather extremes and physical violence. 8 The escalating national opioid crisis has also engulfed homeless populations. A complex constellation of physical and emotional trauma, including adverse childhood experiences, poverty, debt, mental health disorders, and incarceration history, substantially escalates the risk of homelessness.1,6,7
Health care for people experiencing homelessness begins by establishing trust between them and health teams, especially since the homeless population often faces stigma, disjointed lives, and pressing needs for social connection and support. Care coordination, led by broad health teams of physicians, nurses, case workers, education specialists, and recovery coaches, among others, optimally spans shelter clinics, outreach sites, hospitals, emergency medical services, and jails. Quality care requires not only explicit attention to health but also to housing, food, transportation, legal services, and employment needs. 15
Roughly 300 Health Care for the Homeless (HCH) programs (a subset of federally qualified health centers overseen by the US Department of Health and Human Services) coordinate health services for people experiencing homelessness. 16 The National Health Care for the Homeless Council offers technical assistance and advocacy support to HCH programs. 16 Although academic affiliations among HCH programs vary widely, the Boston Health Care for the Homeless Program (BHCHP), for example, collaborates with 2 area medical schools (at Harvard and Boston University) to allow joint staff appointments and provision of care at clinics at their major teaching hospitals. Moreover, BHCHP programs include respite care, mobile health units, supportive home health care by peer specialists with lived homelessness experience, oral health care, and street outreach that includes behavioral health services and tracking patients with electronic medical records. Such work occurs in an era in which the Affordable Care Act has helped Medicaid expansion states reduce uninsurance rates for HCH patients from 51% (2013) to 22% (2020) versus 63% (2020) in nonexpansion states. 16
During the COVID-19 pandemic, HCH programs have, in collaboration with city and state leaders, conducted major efforts to prevent homelessness, provide COVID-19 vaccines to people experiencing homelessness, expand telehealth for behavioral health and substance use counseling, and coordinate efforts to address financial and social needs. For example, in 2020, BHCHP’s citywide COVID-19 care model provided services such as symptom screening at shelters, expedited testing at pop-up sites, isolation and management for those infected and quarantine for those exposed, and contact investigation and tracing. 17 Early results of the model in spring 2020 noted that 33.1% (429/1297) of homeless adults who underwent polymerase chain reaction testing for SARS-CoV-2 had a positive test result 17 ; broader national evaluations can inform future outbreak protection efforts.
Barriers include challenges in care continuity—as reflected by high levels of emergency department use—and the profound lack of robust mental health services for the homeless population.7,18 Access to care is also limited by lack of health insurance and transportation challenges, while mistrust of the health care system leads many to shun medical treatments and health care settings altogether. 7 Costs are high; one BHCHP study found health care spending ($18 764 per person per year) for their Medicaid clients to be 2.5 times more than for a comparison nonhomeless Massachusetts Medicaid population. 18
Reconnecting People to Housing
To rehouse individuals and families, HUD employs a network of continuums of care, which are local planning bodies that coordinate homelessness services in a geographic area. 4 Emergency housing services involve temporary shelters (including slightly fewer than 400 000 beds nationally) and permanent housing (roughly 546 000 beds nationally) 4 ; in the latter category, 68% of beds are for permanent supportive housing, which also includes voluntary supportive services for people with chronic illnesses, disabilities, mental health issues, or substance use disorders. 4 Since 2010, the US government has endorsed a “Housing First” approach that provides permanent supportive housing without prerequisites, such as sobriety or mandatory participation in supportive services.8,19 A 2018 National Academy of Medicine report found that, although permanent supportive housing initially improved housing and possibly health care utilization (eg, emergency department visits), neither health outcomes nor costs improved in follow-up studies that rarely exceeded 24 months. 8 Further evaluations must document long-term status, especially since one 14-year follow-up study of chronically unsheltered individuals in a Housing First program in Boston reported high mortality rates, frequent moves to avoid eviction, and poor housing retention. 20
Legal challenges regularly mark the policy landscape concerning homelessness. Various courts have upheld rights to temporary emergency shelters, although the legal “right to shelter” exists only in New York City (for individuals) and in Massachusetts and the District of Columbia (for families). 21 Sacramento is considering the nation’s first “right to housing” law, which would create a city obligation to house all people experiencing homelessness. 22 In 2018, the US Court of Appeals for the Ninth Circuit held that cities cannot enforce anticamping ordinances for people sleeping outside when shelters are unavailable, as Boise and Los Angeles had attempted to do. 22
Rising home prices and rents in the United States have exacerbated the challenges of housing affordability and availability. In 2020, 46% (20 million) of renter households paid more than 30% of income for rent, with 24% of households paying more than half their income. 23 For every 100 “extremely low-income households” (defined as <30% of area median income) in 2020, only 36 units of affordable housing were available. 24 The COVID-19 pandemic has further burdened lower-income renters, who are disproportionately people from racial and ethnic minority groups. 24
Federal rental assistance attempts to close the supply-and-demand gap through (1) project-based programs in which public housing agencies contract with private owners to rent units to low-income families, (2) tenant-based programs to help families rent privately owned housing, and (3) public housing that is owned and managed by HUD or a local housing authority. 25 The Section 8 project-based rental assistance program today serves about 1.2 million households (>2 million people) through multiyear rental assistance agreements between private owners and HUD. 26 The Housing Choice voucher program, a tenant-based program, provides 2.3 million households (5 million people) with vouchers to help pay rent for privately owned, market-rate units. 27 Federal public housing developments, often located in low-income, racially and ethnically segregated neighborhoods, operate 958 000 units (serving 1.8 million people), a number that has fallen since the mid-1990s as a result of federal limits on adding units and public-to-private housing conversions. 28 All 3 programs limit housing costs to 30% of income. 25 Notably, however, housing assistance is not an entitlement, and, in 2019, only about one-fourth of eligible households received housing assistance through HUD programs. 29
Although the most successful federal program, the Low-Income Housing Tax Credit program, has produced more than 3.3 million units of affordable housing since 1987, 30 units may be converted to market-rate housing when affordability restrictions expire. In 2021 and early 2022, new housing developments were challenged by high and rising building-related costs, 23 a trend likely to continue given recent record-high inflation. While some local jurisdictions have addressed local zoning codes that favor single-family homes to limit density, more policy changes are needed to increase the supply of affordable and multi-unit housing 23 (eg, removing barriers to shared housing [multiple families in the same household], permitting “accessory dwelling units” and “tiny homes,” and increasing the share of affordable housing in new developments).
Recent Opportunities for National Action
HUD’s 2021 “House America” initiative, supported by USICH, aims to reconnect 100 000 people to housing and add 20 000 affordable housing units. 31 The 2021 American Rescue Plan included $21.5 billion for emergency rental assistance (adding to $25 billion previously allocated under the Consolidated Appropriations Act of 2021) for landlords and tenants, including eviction diversion programs; $5 billion for emergency Housing Choice vouchers; $5 billion for housing and services; and $80 million to support COVID-19 testing and mitigation measures.32,33 USICH plans to release an updated national strategic plan (to be released in 2022). 34 Meanwhile, since Los Angeles passed Proposition HHH in 2016, a $1.2 billion bond bill to build 10 000 supportive housing units, its slow implementation has led to a recent lawsuit that will result in up to $3 billion more for affordable housing. 35 Despite these substantial developments (including the passage of the bipartisan infrastructure bill in 2021 and establishment of the National Housing Trust Fund in 2008, which provides federal revenue to build, rehabilitate, and preserve housing for low-income people), 36 every sector of society needs to demonstrate greater involvement. As an example, the private sector now features some commitments, such as technology companies pledging hundreds of millions of dollars to mitigate Silicon Valley’s affordable housing crisis 37 and hospital and medical systems (“anchor institutions”) addressing housing and other social determinants of health as part of delivery of care. 38
Integrating and Coordinating National Efforts
National groups pushing for more coordinated, multisectoral approaches include the National League of Cities, the National Alliance to End Homelessness, the National Low Income Housing Coalition, the Corporation for Supportive Housing, and ChangeLab Solutions. Community Solutions’ Built for Zero campaign complements national efforts by supporting a network of 98 communities to integrate local programs and agencies, facilitate community-level measurements, invest in high-impact housing, and match housing solutions to individuals. 39 Although not a Built for Zero city, Houston, Texas, has also taken a similar highly coordinated approach and cut the number of people deemed homeless by 63% since 2011, as reported by local officials. 40
A 2010 USICH plan to address homelessness among veterans as a national priority and substantial federal funding and programs helped drop rates of homelessness in half from 2009 to 2020. 4 The US Department of Veterans Affairs (VA) National Center on Homelessness Among Veterans promotes research and recovery-oriented care; clinicians at VA hospitals regularly track patient housing status, coordinate immediate shelter provision if necessary, and offer transitional and permanent housing support. Programs also include HUD-VA Supportive Housing, which provides rental assistance housing vouchers for permanent supportive housing, and the Supportive Services for Veteran Families Program, which offers light case management and temporary rent and utility relief. A report from 2022 suggested that, thus far, 82 communities—including 3 states—have effectively ended homelessness among veterans by coordinating housing, health care, and services. 41
Building Academic Communities to Address Homelessness
To join in these national efforts, higher education, especially schools and programs of public health, can convene diverse disciplines—including health, medicine, law, policy, and social work, among others—to form academic communities to address homelessness. Education can include not only formal courses but also newsletters, seminars, debates, and public lectures to reach broader audiences. Fellowship and mentorship programs can create a critical pipeline of investigators and public health practitioners, bolster the workforce, improve training, and aim to destigmatize care.
Individual research investigations are under way, but stronger national coordination can amplify impact. For example, increasing academic attention to implementation science and using community-engaged research to more fully involve HCH programs, nongovernmental organizations, and people with lived homelessness experience can help build better systems of housing and care. Studies can further investigate the impacts of racism and the social determinants of health that drive homelessness (eg, housing affordability, income inequality); improve national data systems; rethink emergency shelter systems, which are rarely integrated into surrounding communities; explore better policies to narrow the gap between supply and demand for affordable housing; address costs; and support long-term evaluations of Housing First programs. Studies can also identify better models of care for chronic health conditions and advance national efforts to build more robust behavioral health systems to address mental illness and substance use disorders. Variability among states in the adoption of Medicaid expansion has resulted in natural experiments on how different health insurance coverage and eligibility have affected patient outcomes. Legal scholars can study both the potential and the limits of “right to housing” standards. New policy studies can explore the future of financing, how private businesses can innovatively contribute to housing solutions, and how to braid public funding involving HUD, the US Department of Health and Human Services, and Medicaid. Research can also identify best practices in VA homelessness efforts, outcomes after expiration of the eviction moratorium, and the impact of the American Rescue Plan and other monetary resources on achieving House America and USICH goals.
Unfortunately, obstacles have limited academia’s involvement to date, with the most notable being the near absence of dedicated national research funding. The federal research infrastructure has traditionally targeted diseases rather than “vulnerable populations,” that is, those without access to health care or quality care and who experience suboptimal health outcomes, often as a result of social injustices related to race and ethnicity, poverty, gender, sexual orientation, age, first language, or physical or mental health conditions. For example, the National Institutes of Health, with 27 institutes related to specific diseases, has no dedicated funding for homelessness research; research on housing efforts is also suboptimal. The relative lack of senior academic faculty with substantial interdisciplinary expertise in homelessness hinders the development of a pipeline of young investigators routinely seen in other areas of public health. Even basic education is lacking, leaving some schools and institutions relatively unaware of issues just outside their walls. In an informal web-based search of curricula from leading schools of public health, medicine, and social work, we identified only a handful of dedicated courses addressing health, housing, and homelessness.
Exceptions include the Johns Hopkins Bloomberg School of Public Health, which has offered a dedicated course on homelessness since the 1980s, and the University of Southern California’s course for physician assistants. Princeton University hosts an Eviction Lab. The newly established Benioff Homelessness and Housing Initiative at the University of California, San Francisco, provides policy-oriented research about homelessness and inequitable patterns of eviction. Recently, the Harvard T.H. Chan School of Public Health launched an Initiative on Health and Homelessness, a pilot program in close collaboration with BHCHP.
Call to Action
The long-standing homelessness crisis, more visible than ever, needs urgent attention and commitment from all sectors of society. Now is the time for academia to engage more fully to educate students, motivate researchers, and better prepare the next generation of health professionals to care most for “those whom society counted least and put last.” 42 Doing so can help the nation move a step closer to health equity and heighten attention to a dire humanitarian crisis that can no longer be overlooked.
Footnotes
Acknowledgements
The authors are grateful to Karen Emmons, PhD; William N. Evans, PhD; Aayush Gupta, MPH; Bob Heckart, JD; Rajarshi Mukherjee, PhD; Natalie Slopen, ScD; Penny Sun, MPH; Nancy Turnbull, MBA; Kirk Vanda, MBA; and Barbara Wei, MPH, for their contributions to this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
