a. Burden of disease
i. Seroprevalence, transmission rates
In the United States, the Center of Disease Control (CDC) has been at the forefront of the COVID-19 pandemic, providing informational tools to the public and guidelines for federal government entities. From late March to April 2020, the CDC responded to reports of several infections in five homeless shelters across Boston, Massachusetts (n = 1); San Francisco, California (n = 1); and Seattle, Washington (n = 3).6 The study tested a total of 1,192 residents and 313 staff members using reverse transcriptase polymerase chain reaction to determine the infection prevalence.6 The percent of residents and staff members who tested positive following the initial reports of positive cases were: 17% and 17% in Seattle, 36% and 30% in Boston, and 66% and 16% in San Francisco, respectively. In the Boston shelter, specifically, only 11.5% of those tested experienced symptoms of the disease.7 Furthermore, while there had been no reported cases of COVID-19 at two shelters in Atlanta, Georgia during this time, testing was performed from April 8–9, 2020 to rule out the presence of asymptomatic individuals; the combined infection rate at these shelters was 4% among residents and 2% among staff members.6 A second, more comprehensive study with similar objectives and methods tested homeless individuals and shelter workers from April 7 - May 6, 2020 at 24 shelters and 9 unsheltered outreach events in Atlanta.8 This study found a SARS-CoV-2 infection rate of 2.1% among sheltered PEH, 0.5% among unsheltered PEH, and 1.3% among shelter staff. Moreover, a subset of those tested were screened for symptoms, determining that 76.2% of those who tested positive for COVID-19 experienced no symptoms during the previous week.
These studies made attempts at understanding the risk of COVID-19 infection among PEH as compared to the general population. Yoon et al. cited a door-to-door household survey conducted in Fulton and DeKalb counties in Georgia from April 28-May 3, 2020, which found an estimated 2.5% seroprevalence of SARS-CoV-2 antibodies.8 Also, data collected from the CDC and state and local public health agencies demonstrated a community incidence of 14.4% in Boston and 5.7% in San Francisco at the time that the aforementioned CDC studies were conducted.6 These results reveal significantly higher infection rates among PEH. However, several limitations—which include missing or uncontrolled data and varying sensitivity of RT-PCR tests—exist, requiring more information regarding infection rates and testing protocols in the general population to better determine the presence of varying COVID-19 risk.
Similar study objectives and results have been found outside of the US. Do Couto et al. conducted serum testing at a shelter in Sao Paulo, Brazil that makes up 19.6% of the total homeless population in the city.9 The seroprevalence of SARS-CoV-2 IgM and IgG antibodies in homeless individuals was 2.5% and 54.7%, respectively. No statistical difference was found between this group and shelter workers, who had seropositivity rates of 5.7% and 47.1%. Again, information provided on seropositivity in the general population was limited as it only cited the crude seroprevalence of 17.1% observed one month later in blood donors from Sao Paulo.9 A study by Loubiere et al. gathered SARS-CoV-2 prevalence data on the homeless population in Marseille, France from 48 different locations, including streets, slums, squats, emergency or transitional shelters, and drop-in centers. A 5.6% seropositivity rate among homeless individuals was found along with a 6% rate within the subset who reported experiencing no common COVID-19 symptoms.10 Importantly, the authors also cited a national seroprevalence survey that determined 3.6% seropositivity throughout Marseille, suggesting a higher incidence of COVID-19 among the homeless population. Though, it is unclear whether the participant samples of these two surveys were mutually exclusive.
ii. Risk factors
A recent report found that PEH had a case fatality rate 1.3 times that of the general population.11 To address concerns of heightened risk to COVID-19 among the homeless population, researchers described potential contributing factors to such vulnerability. Insecure living conditions have been cited as a major contributor to SARS-CoV-2 exposure, as common strategies to lower transmission rates, such as social distancing and sanitation, are difficult to adhere to in crowded shelters and public areas.9,12,13 In France, it was determined that 56.9% of PEH who tested positive for the virus had spent more than one month in emergency shelters while only 29.5% of those who tested negative did so.10 Furthermore, Sandoval-Olascoaga et al. determined that in states that lifted their eviction moratorium before the CDC issued their national moratorium, individuals had a higher hazard ratio (up to 1.82, with P < 0.001) of a COVID diagnosis.14 Additionally, the risk of contracting other transmittable diseases that have been associated with worse clinical onset and outcomes of COVID-19, such as HIV/AIDS, is also heightened by insecure shelter.9 Furthermore, PEH face greater barriers to healthcare and overall public health information and education, leading to several comorbidities. In the study conducted by Yoon et al., 32.0% of the participants reported having cardiovascular disease, 14.7% reported having chronic lung disease, and 10% reported having diabetes. Smoking was also reported by 48.8% of sheltered participants and 63.0% of unsheltered participants.8 In migrant populations, such comorbidities have been shown to increase the likelihood of contracting COVID-19.13
Another commonly cited risk factor for COVID-19 is poor mental health. A recent study analyzing electronic health records from 61 million adult patients in the US found that a diagnosis of mental health disorder increased risk of COVID-19 infection, specifically in regard to depression (AOR = 7.64, p < 0.001) and schizophrenia (AOR = 7.34, p < 0.001).15 Hospitalization and death rates were also higher in populations diagnosed with COVID-19 and a mental health disorder; patients had hospitalization rates of 27.4% and death rates of 8.5%, compared to 18.6% and 4.7% in patients with no mental disorder (p < 0.001). This is particularly concerning for PEH as they are especially vulnerable to psychiatric disorders. Data collected from homeless shelters in France found 30% of residents had symptoms of moderate to severe depression (n = 527).16 Furthermore, a study conducted by Beghi et al. analyzed emergency room admissions at four sites in Northern Italy from March 9-May 3, 2020 to investigate the correlation between the pandemic and mental health. While a significant decrease in psychiatric referrals was observed for patients who lived in a residential facility, a significant increase was observed for homeless patients.17 The heightened mental health disorders among PEH can be partially explained by the onset and exacerbation of such disorders by environmental factors associated with inadequate housing quality, such as noise and inadequate sleep.18 Thus, it is revealed that PEH experience several risk factors that may increase their risk for contracting and having poor outcomes for SARS-CoV-2 infection.
b. Interventions and Outcomes
Current methods for limiting the spread of COVID-19 involve early screening, diagnosis, isolation, and treatment.19 These measures are also suggested to help prevent the spread of disease among homeless populations despite differences in living situations that may change what the most ideal prevention practices are. According to Badiaga et al., these measures from the CDC should be amended to be easily implemented in shelters and on streets.20 In this section, we will highlight some of these measures and their potential effectiveness and limitations in preventing transmission of SARS-CoV-2 infection.
i. Isolation
Isolation and quarantine spaces are crucial especially for PEH as they may be unable to effectively practice distancing measures.21,22 In April 2020, PEH and service staff in 24 homeless shelters and nine outreach events in Atlanta, Georgia were tested for SARS-CoV-2.8 Seroprevalence in sheltered PEH was 2.14% (36/1684), a rate 4x that of prevalence in unsheltered PEH (0.48% or 3/628) and almost double that of staff (1.28% or 7/548). According to Sisson, because of a lack of permanent accommodations, PEH often cannot be isolated or quarantined.23 With this in mind, homeless shelters may currently lack the resources to allow for adequate physical distancing, leading to increasing infection rates. Thus, off-site isolation sites seem to be promising in preventing transmission.
COVID-19 Recuperation Units (CRUs) served as a novel method to improve hospital patient flow during the peak of the pandemic by providing space for isolation and quarantine for PEH. In a CRU that operated adjacent to Boston Medical Center, PEH accounted for 84% of the population census. In addition, there was a 28% reduction in hospitalizations after the CRU had opened.24 CRUs were hence highly effective as it allowed hospital beds to be used for patients who required acute care and provided a safe location for PEH to isolate.
Isolation hotels staffed with nurses and healthcare workers under the supervision of a physician were also employed to act as an isolation area for PEH. An isolation hotel in San Francisco, California set up from March and May of 2020 provided isolated housing for 1009 hotel guests of which 501 (49.7%) were PEH.25 Furthermore, 346 of 549 patients (63%) were transferred to the hotel from the county hospital with only 13 of those patients (4%) being readmitted back into the hospital due to worsening conditions.25 Thus, referrals by the emergency department and inpatient units to such isolation hotels are promising methods to control transmission rates while alleviating the need for hospitalization.
ii. Vaccination
Perhaps the most significant step in reducing health disparities of COVID-19 is increasing access to vaccinations and boosters. Full vaccination against SARS-CoV-2 has been shown to be 85–95% effective.19 Furthermore, fully vaccinated individuals make up only 3% of total COVID-19 case hospitalizations. Vaccination also results in decreased transmission rates, likely due to a decreased viral load.19 Therefore, homeless populations should be considered priority groups for vaccination given their vulnerability to infection.
Unsurprisingly, there is huge variation in vaccine distribution plans internationally and even domestically. In the United Kingdom, homeless populations were grouped together with individuals aged 16 to 65 years with health problems.26 However, many individual cities offered mass vaccination earlier to vulnerable groups. On the other hand, a vaccine shortage delayed rollout in Canada, resulting in many PEH waiting to receive their first dose as recently as September 2021. A recent report by the CDC studied vaccination coverage and social vulnerability.27 Social vulnerability was defined using several different indicators, with four main underlying themes of socioeconomic status, household composition and disability, racial and ethnic minority status and language, and housing type and transportation. Overall, higher vaccination coverage was observed in counties classified as low social vulnerability compared to counties classified as high social vulnerability (15.8% vs 13.9%). When addressing PEH specifically, surveys conducted across nine shelters in King County, Wyoming found that only 54.3% of residents (n = 361) were either accepting of the vaccine or already vaccinated.28 Accordingly, there are major lapses in vaccinations for PEH in this population, contributing to a greater risk of COVID-19 cases.
Along with inadequate supply of vaccines, vaccine hesitancy may have an important role in the low vaccination rates among PEH. A pilot mobile survey in Los Angeles studying vaccine hesitancy found that 48% (n = 90) of PEH were vaccine hesitant.29 Interviews with PEH in Oakland and San Francisco revealed similar results and cited mistrust in the government, doubts about the safety of the vaccine, and desire for trial data as reasons for hesitancy.30
iii. Public Health Education
According to Lewer and colleagues, COVID-19 does not impart equal risks to all constituents of society, therefore, the form of education used to communicate prevention and risk factors also must differ.31 PEH are highly vulnerable to life-threatening infectious disease worldwide due to illiteracy and barriers to proper public health education. The homeless community is oftentimes unable to follow the instructions set forth by the CDC as they have poor access to media outlining relevant guidance. This in turn may contribute to the high positivity rates for the homeless population during the pandemic.
Power et al. stated that a framework of health education is imperative for primary and secondary infectious disease prevention for PEH residing in shelters.32 Health education is crucial in mitigating the homeless community's vulnerability by various contagious disease outbreaks, predominantly COVID-19. Proper guidelines and recommendations should be communicated to the PEH. The most significant challenge towards addressing implications is the communication barrier. Under-education among the homeless population is the primary cause of the communication gap between public health education professionals and the homeless community. They should be addressed with necessary interim guidelines and recommendations from CDC in simple language.33
To mitigate the prevalence of COVID-19 in homeless communities, safety interventions and good hygienic practices are considered the most effective tools. This involves taking adequate preventive measures such as proper handwashing, using sanitizers, wearing masks, maintaining social distancing, and getting vaccinated.34 However, an illiterate homeless population has no direct access to healthcare facilities; therefore, all the possible means should be utilized to provide them with the necessary personal protective equipment and directions for application. With rapid viral spread in the homeless community globally, governments should be directed to developing different planning guides regarding pandemics that cater to PEH specific needs. Governments should implement valuable strategies for community programs to educate and spread awareness to PEH.35
iv. Access to Supplies
Lack of safe housing, drop-in facilities, and poor shelter increases the risks of PEH contracting COVID 19.18 The spread of COVID 19 in these settings may be rapid, and the detection of the illness is not feasible because of the limited access to social services and health care.36 Lima et al. argue that there is a high risk of severe COVID-19 on homeless individuals because of the increased ubiquity of medical comorbidities, e.g., heart disease, liver disease, and respiratory conditions arising from high smoking rates in homeless populations.18 Furthermore, the number of individuals aged 65 and above among those in homeless populations exacerbates the risk of flourishing severe COVID-19.7
Regular hand washing with soap and running water or using an alcohol-based sanitizer helps reduce the rate at which COVID-19 is spread.19 In homeless settings, though, access to water, soap, or sanitizer is difficult. Therefore, this lack of access to sanitary supplies is a factor that increases the risk of homeless individuals contracting the infection. This, therefore, is a significant challenge that draws back the proposed strategies for reducing the prevalence of the disease. There is a need for the government and other organizations to consider supplying homeless environments with sanitation facilities in line with the fight against the pandemic. Moreover, most homeless individuals perceive hand hygiene as a piece of common knowledge, meaning that the homeless are likely not to put extra effort into ensuring soap and running water exist within their settings. Therefore, the provision of supplies for handwashing should instead be implemented, in addition to providing educational services on why and how to ensure hand hygiene.18
Furthermore, the World Health Organization (WHO) emphasizes using a face mask to cover one’s nose and mouth to avoid spreading the infection. Access to affordable masks by homeless individuals is problematic; thus, this heightens the risks of them contracting the disease. Therefore, the priority should be supplying affordable masks to the homeless. Additionally, no significant achievement will be realized from training the homeless individuals on how to use the mask, which is helpful for the containment of COVID-19 infection, without the provision of such masks. The concerned organizations should thus, consider providing the masks to homeless individuals rather than plainly educating them on how to use masks they have no access to.33
Because PEH constitute a group of more vulnerable individuals to COVID-19, it is uneconomical to prioritize them for testing SARS-CoV-2.36 To prevent the rapid spread of SARS CoV-2, meal programs, shelter, and other organizations supporting PEH need to create conditions that gather for the recommended strategies to reduce the risk of these individuals contracting the infection. On this view, MacKenzie et al. argue that it requires more funding, commitment, and additional human resource that makes it infeasible.37
COVID-19 and its accompanying transmission control measures pose particular challenges and increased risks of harm for individuals experiencing homelessness. Such actions that can be put in place to increase the safety of these homeless individuals include making arrangements for temporary housing, creating isolation sites for the homeless individuals that test positive for the infection, and enlarging shelter spaces. The different needs for the other homeless individuals, though, should be factored in when implementing these strategies to ensure the functioning and equity-focused interventions. COVID-19 pandemic thus has highlighted the significance of housing as a social determinant of health and raised a question on whether the current strategies of addressing the issue of homelessness are effective.