Abstract
People experiencing homelessness (PEH) are at increased risk for coronavirus disease 2019 (COVID-19) infection. This study assessed COVID-19 vaccination coverage among vaccine-eligible PEH (5 years and older) stratified by demographic characteristics. PEH were less likely to complete a primary vaccination series than the Dane County population (32.0%; 95% confidence interval [CI], 30.3%–33.8% vs 82.4%; 95% CI, 82.3%–82.5%) and were less likely to have received a booster when eligible (30.8%; 95% CI, 27.8%–33.9% vs 67.2%; 95% CI, 67.1%–67.4%). Vaccination rates were lowest among young PEH and PEH of color.
Keywords: COVID-19, vaccination, people experiencing homelessness
Dane County, Wisconsin is a highly vaccinated county, with over 82% of eligible people aged 5 years and older completing an initial coronavirus disease 2019 (COVID-19) vaccination series as of 31 January 2022. However, high vaccination coverage county-wide obscures low vaccine coverage within specific populations that may be at higher risk for COVID-19 infection.
People experiencing homelessness (PEH) are at increased risk for COVID-19 infection [1–6]. The spectrum of COVID-19 illness in PEH is varied. Seroprevalence data of PEH demonstrate that many COVID-19 infections in this population are asymptomatic, which can complicate mitigation strategies in congregate shelter settings [1, 2]. Simultaneously, multiple reports show increased severity and poorer outcomes of COVID-19, including hospitalization and death [3–6]. Several factors are likely associated with vulnerability to severe COVID-19 illness among PEH, including co-occurring health conditions and living in congregate settings.
Despite widespread availability of COVID-19 vaccine in the United States, emerging evidence suggests lower vaccination coverage among PEH compared to the general population [7]. Vaccine uptake among PEH may be influenced by several factors, including less access to health care and mistrust in government and medical systems [8]. Moreover, many unvaccinated PEH may never be offered a COVID-19 vaccine [9]. Geographic transience may complicate receipt of additional doses of vaccine, which could impact PEH being up to date with COVID-19 vaccine series [10]. Given the vulnerability to COVID-19, understanding vaccination coverage in PEH is critical for targeted vaccination efforts in this specific population. The aim of this study was to estimate vaccine coverage among vaccine-eligible PEH in Dane County, Wisconsin.
METHODS
Data Sources
People Experiencing Homelessness
PEH were identified from the Homeless Management Information System (HMIS), which includes people who received housing and other services from organizations serving the homeless population in Dane County. The HMIS data included name, date of birth, and client characteristics (gender, race, ethnicity, age). Sixteen programs that provided consent to use client-level data were included in this analysis. Data were provided in 2 installments, and not all programs consented to data sharing for each installment; 6 programs only consented to data sharing for half the year. We included data for 2778 individuals aged 5 and older served by programs 1 January 2021 through 31 December 2021 to evaluate vaccine coverage in all COVID-19 vaccine-eligible age groups. Demographic information from HMIS were used to define specific populations of PEH.
COVID-19 Vaccination Records
COVID-19 vaccination records for Dane County residents were obtained from the Wisconsin Immunization Registry (WIR), Wisconsin’s centralized immunization management system. A 2015 assessment of WIR demonstrated that this system has good completeness and accuracy [11].
Population
The 2020 census bridged race population estimates for people aged 5 and older were used. These population estimates were obtained by the Wisconsin Department of Health Services Office of Health Informatics.
Variables and Statistical Analyses
Completion of a primary COVID-19 vaccination series was the primary outcome of interest, with booster coverage as a secondary outcome. Completed primary COVID-vaccine series was defined as people with 1 dose of Janssen or 2 doses of Pfizer or Moderna vaccine in WIR as of 31 January 2022. Being eligible for a booster included people aged 12 and older with a completed primary vaccine series. A booster was defined as an additional vaccine dose after a completed primary vaccine series, which was given on or after 13 August 2021, when the Centers for Disease Control and Prevention recommended a booster dose for immunocompromised people. Being up to date on vaccines was defined as being aged 5–11 years with a completed primary vaccine series, or aged 12 or older with either a completed primary vaccine series within the previous 2 months, a booster dose ≥2 months after a completed Janssen primary series, or a booster dose ≥ 5 months after a completed Pfizer or Moderna primary series [12].
To estimate vaccination and booster coverage among PEH, HMIS data were matched to the WIR using Link Plus, a probabilistic linkage tool. First name, last name, date of birth, and gender were present in both data sources and were used as matching variables. Completion of a primary series and being up to date on vaccines for PEH were calculated using the number of people aged 5 and older in HMIS for denominators. Booster coverage was calculated using the number of people eligible for a booster for denominators.
Descriptive statistics were calculated to examine vaccine coverage among specific populations of PEH compared to the general Dane County population with 95% confidence intervals (CIs) calculated using the Wilson method. Vaccination coverage was stratified by age group, gender, ethnicity, race, and homeless services program type. Programs were classified as providing services to unsheltered PEH (3 programs), sheltered PEH (8 programs), and PEH housed in permanent or transitional housing (5 programs). For county vaccination coverage, we used Dane County WIR records to identify people with a completed primary vaccine series and a booster dose. Receipt of primary COVID-19 vaccination series by vaccine manufacturer was also examined for PEH and Dane County.
χ2 tests were conducted to assess whether vaccination rates differed for categories within demographic groups for PEH, with P values <.05 indicating statistical significance. Statistical testing was not performed on vaccination coverage by program type, as some PEH participated in multiple programs within different categories. For comparisons to Dane County, estimates with nonoverlapping CIs were considered statistically different. All analyses for PEH were conducted using SAS (version 9.4; SAS Institute) and Dane County CIs were estimated using OpenEpi (version 3.01).
This project was evaluated using the University of Wisconsin-Madison Health Sciences Institution Review Board’s Self-Certification Tool and was deemed to be quality improvement and exempt from Institution Review Board review.
RESULTS
PEH were less likely to complete a primary vaccination series than the Dane County population overall (32.0%; 95% CI, 30.3%–33.8% vs 82.4%; 95% CI, 82.3%–82.5%) and for all demographic groups (Table 1). Among those with a completed primary vaccine series, PEH were less likely than the Dane County population to have received the recommended booster dose (30.8%; 95% CI, 27.8%–33.9% vs 67.2%; 95% CI, 67.1%–67.4%). This disparity was observed across all demographic groups (Table 2). Among eligible individuals, only 16% of PEH were up to date on their COVID-19 vaccines compared to 61% of the Dane County population.
Table 1.
Completed COVID-19 Primary Vaccination Series Among People Experiencing Homelessness Compared to All Residents in Dane County, Wisconsin
| Dane People Experiencing Homelessness | All Dane Residents | ||||||
|---|---|---|---|---|---|---|---|
| Characteristic | No. PEH | No. Completed Vaccine Series | % Completed Vaccine Series (95% CI) | P Valuea | No. Residents | No. Completed Vaccine Series | % Completed Vaccine Series (95% CI) |
| Overall | 2778 | 889 | 32.0 (30.3–33.8) | 523 060 | 430 929 | 82.4 (82.3–82.5) | |
| Age, yb | |||||||
| 5–11 | 182 | 9 | 4.9 (2.6–9.1) | <.001 | 43 400 | 24 794 | 57.1 (56.7–57.6) |
| 12–17 | 104 | 17 | 16.3 (10.5–24.6) | 37 339 | 31 007 | 83.0 (82.7–83.4) | |
| 18–24 | 196 | 33 | 16.8 (12.3–22.7) | 73 703 | 48 110 | 65.3 (64.9–65.6) | |
| 25–34 | 581 | 130 | 22.4 (19.7–25.9) | 87 018 | 72 484 | 83.3 (83.1–83.5) | |
| 35–44 | 581 | 174 | 29.9 (26.4–33.8) | 75 534 | 66 634 | 88.2 (88.0–88.5) | |
| 45–54 | 466 | 198 | 42.5 (38.1–47.0) | 61 874 | 54 913 | 88.7 (88.5–89.0) | |
| 55–64 | 485 | 224 | 46.2 (41.8–50.6) | 63 090 | 56 913 | 90.2 (90.0–90.4) | |
| 65+ | 178 | 104 | 58.4 (51.1–65.4) | 81 102 | 76 071 | 93.8 (93.6–94.0) | |
| Genderc | .15 | ||||||
| Female | 1031 | 312 | 30.3 (27.5–33.1) | 263 351 | 222 903 | 84.6 (84.5–84.8) | |
| Male | 1720 | 567 | 33.0 (30.8–35.2) | 259 709 | 207 288 | 79.8 (79.7–80.0) | |
| Ethnicityd | .045 | ||||||
| Hispanic/Latino | 209 | 54 | 25.8 (20.4–32.2) | 33 618 | 26 410 | 78.6 (78.1–79.0) | |
| Not Hispanic/Latino | 2446 | 801 | 32.8 (30.9–34.6) | 489 442 | 379 052 | 77.4 (77.3–77.6) | |
| Racee | <.001 | ||||||
| White | 1257 | 468 | 37.2 (34.6–39.9) | 451 456 | 343 128 | 76.0 (75.9–76.1) | |
| Black or African American | 1256 | 352 | 28.0 (25.6–30.6) | 33 382 | 17 722 | 53.1 (52.6–53.6) | |
| Multiracial | 132 | 31 | 23.5 (17.1–31.4) | NA | NA | NA | |
| Other race | 59 | 21 | 35.6 (24.6–48.3) | NA | NA | NA | |
| Program type | NA | ||||||
| Unsheltered | 255 | 79 | 31.0 (25.6–36.9) | NA | NA | NA | |
| Sheltered | 2418 | 739 | 30.6 (28.8–32.4) | NA | NA | NA | |
| Permanent/transitional housing | 278 | 150 | 54.0 (48.1–59.7) | NA | NA | NA | |
Abbreviations: CI, confidence interval; PEH, people experiencing homelessness; NA, not available.
χ2 tests conducted to determine difference in vaccine coverage within demographic groups.
Missing age: PEH = 5 (0.2%); Dane County residents n = 3 (<0.1%).
Missing gender: PEH n = 8 (0.3%); Dane County residents n = 902 (0.2%). Nonbinary gender is not included.
Missing ethnicity: PEH n = 123 (4.4%); Dane County residents n = 30 427 (5.8%).
Missing race: PEH n = 74 (2.7%); Dane County residents n = 22 843 (4.4%).
Table 2.
Receipt of COVID-19 Booster Dose Among People Experiencing Homelessness Compared to All Residents in Dane County, Wisconsin
| Dane People Experiencing Homelessness | All Dane Residents | ||||||
|---|---|---|---|---|---|---|---|
| Characteristic | No. Booster Eligible | No. Received Booster | % Received Booster (95% CI) | P Valuea | No. Booster Eligible | No. Received Booster | % Received Booster (95% CI) |
| Overall | 880 | 271 | 30.8 (27.8–33.9) | 406 132 | 273 055 | 67.2 (67.1–67.4) | |
| Age, yb | <.001 | ||||||
| 12–17 | 17 | 2 | 11.8 (3.3–34.3) | 31 007 | 13 603 | 43.9 (43.3–44.4) | |
| 18–24 | 33 | 4 | 12.1 (4.8–27.3) | 48 110 | 22 698 | 47.2 (46.7–47.6) | |
| 25–34 | 130 | 25 | 19.2 (13.4–26.9) | 72 484 | 42 641 | 58.8 (58.5–59.2) | |
| 35–44 | 174 | 46 | 26.4 (20.4–33.5) | 66 634 | 45 421 | 68.2 (67.8–68.5) | |
| 45–54 | 198 | 60 | 30.3 (24.3–37.0) | 54 913 | 39 276 | 71.5 (71.2–71.9) | |
| 55–64 | 224 | 89 | 39.7 (33.6–46.3) | 56 913 | 44 175 | 77.6 (77.3–78.0) | |
| 65+ | 104 | 45 | 43.3 (34.2–52.9) | 76 071 | 65 241 | 85.8 (85.5–86.0) | |
| Genderc | .03 | ||||||
| Female | 308 | 109 | 35.4 (30.3–40.9) | 210 741 | 147 636 | 70.1 (69.9–70.3) | |
| Male | 562 | 158 | 28.1 (24.6–32.0) | 194 682 | 125 304 | 64.4 (64.2–64.6) | |
| Ethnicityd | .12 | ||||||
| Hispanic/Latino | 53 | 11 | 20.8 (12.0–33.5) | 24 631 | 11 112 | 45.1 (44.5–45.7) | |
| Not Hispanic/Latino | 793 | 248 | 31.2 (28.1–34.6) | 358 016 | 249 626 | 69.7 (69.6–69.9) | |
| Racee | .1 | ||||||
| White | 466 | 161 | 34.6 (30.4–40.0) | 324 168 | 231 247 | 71.3 (71.2–71.5) | |
| Black or African American | 347 | 92 | 26.5 (22.1–31.4) | 16 854 | 7017 | 41.6 (40.9–42.4) | |
| Multiracial | 30 | 10 | 33.3 (19.2–51.2) | NA | NA | NA | |
| Other race | 21 | 7 | 33.3 (17.2–54.6) | NA | NA | NA | |
| Program type | NA | ||||||
| Unsheltered | 79 | 15 | 19.0 (11.9–30.0) | NA | NA | NA | |
| Sheltered | 731 | 207 | 28.3 (25.2–31.7) | NA | NA | NA | |
| Permanent/transitional housing | 149 | 67 | 45 (37.2–53.0) | NA | NA | NA | |
Abbreviations: CI, confidence interval; PEH, people experiencing homelessness; NA, not available.
χ2 tests conducted to determine difference in booster coverage within demographic groups.
Age: Dane County residents n = 3 (<0.1%).
Missing gender: PEH n = 3 (0.3%); Dane County residents n = 738 (0.2%). Nonbinary gender is not included.
Missing ethnicity: PEH n = 34 (3.9%); Dane County residents n = 25 468 (6.3%).
Missing race: PEH n = 16 (1.8%); Dane County residents n = 18 385 (4.5%).
Among PEH, completion of a primary vaccine series was lowest among those aged 5–11 years (4.9%; 95% CI, 2.6%–9.1%), 12–17 years (16.3%; 95% CI, 10.5%–24.6%), 18–24 years (16.8%; 95% CI, 12.3%–22.7%), 25–34 (22.4%; 95% CI, 19.7%–25.9%), those who were multiracial (23.5%; 95% CI, 17.1%–31.4%) or Hispanic (25.8%; 95% CI, 20.4%–32.2%). For program type, both completion of primary series and receipt of a booster dose were highest among people receiving permanent or transitional housing services (primary series, 54.0%; 95% CI, 48.1%–59.7%; booster, 45.0%; 95% CI, 37.2%–53.0%).
About 1 in 5 PEH (22.5%) received the single-dose Janssen vaccine, 36.8% received the Moderna vaccine, and 40.7% received the Pfizer vaccine. In contrast, 10.1% of the Dane County population received the Janssen vaccine, 31.2% received Moderna vaccine, and 58.7% received Pfizer vaccine.
DISCUSSION
PEH in Dane County, Wisconsin have low COVID-19 vaccination rates across all age groups despite living in a highly vaccinated county. These findings are consistent with a recent analysis that found lower vaccination coverage among adult PEH in 6 urban public health jurisdictions [7]. Children were included in our analysis to assess overall vaccine coverage in all vaccine-eligible PEH utilizing services. Rates of COVID-19 vaccination in children experiencing homelessness (CEH) aged 5 to 11 years have not been reported previously. While CEH had the shortest time of vaccine eligibility (approximately 3 months prior to analysis), the large disparity between CEH and their counterparts in the general population further highlight the challenges and delays in reaching all PEH with vaccine efforts. Disparities in vaccination coverage among PEH of color were observed, which is concerning due to the increased risk of severe outcomes from COVID-19 among racial and ethnic minorities [13].
Low vaccination and booster coverage among PEH in Dane County indicates that this population remains vulnerable to COVID-19 infection, particularly during periods of high COVID-19 incidence. Mobile vaccination services have been identified as an approach to improve equitable vaccine access to PEH and increase vaccination coverage in this population [14]. While several mobile vaccination teams have offered on-site vaccination clinics at several shelters and community organizations serving PEH in Dane County, COVID-19 vaccination rates remain low, indicating barriers to vaccine uptake beyond access. As public health interventions focus on increasing COVID-19 vaccine coverage among PEH, efforts to mitigate the impact of COVID-19 including continuing shelter-specific infection control strategies such as mask use and physical distancing, and maintaining local isolation/quarantine sites, are critical to reducing COVID-19 transmission in this population [15].
This analysis has several limitations. The people included as PEH are a subset of the entire population of PEH in Dane County, as not all programs gave consent to use data, while others only consented to provide data for a partial year. Some PEH may not access services, and are not represented in HMIS, and others were not included because the programs in which they were served did not consent to providing client-level information for this analysis. Consequently, our findings may not be generalizable to all PEH locally and outside of our jurisdiction. Vaccination records for PEH may have been missed during matching due to inconsistencies in name or date of birth. In addition, vaccination records were limited to people with a Dane County residence, which may be problematic for populations such as PEH who may change their geographic location. PEH vaccinated outside of Dane County and those who provided an out-of-county residential address are not represented in the vaccination records used for this analysis. This highlights the need for better data sharing between local and state health departments monitoring health outcomes, particularly for special populations that may be represented in multiple jurisdictions, as well as the need for rigorous data sharing agreements to safeguard the privacy of individuals and prevent harm among populations such as PEH. This is particularly relevant for local health departments without epidemiological capacity to conduct analyses that require specialized methods such as probabilistic matching. Furthermore, this analysis does not explore potential drivers of the disparity in vaccination coverage among the general population compared to PEH, such as employer vaccination policies. Additional research to identify successful vaccination policies and strategies to reduce the disparity is needed. Despite these limitations, these findings highlight an urgent need for better understanding of barriers to vaccine uptake to improve vaccination coverage among all PEH as well as targeted interventions to improve vaccine equity among specific populations within PEH.
Contributor Information
Crystal Gibson, Public Health Madison and Dane County, Madison, Wisconsin, USA.
Casey Schumann, Public Health Madison and Dane County, Madison, Wisconsin, USA.
Kimberly Neuschel, Public Health Madison and Dane County, Madison, Wisconsin, USA.
Joseph A McBride, Departments of Medicine and Pediatrics, Division of Infectious Diseases, University of Wisconsin-Madison School of Medicine and Public Health, Madison, Wisconsin, USA.
Notes
Financial support. No financial support was received for this work.
Supplement sponsorship. This article appears as part of the supplement “Homelessness and Infectious Diseases: Understanding the Gaps and Defining a Public Health Approach,” sponsored by the Centers for Disease Control and Prevention.
References
- 1. Roederer T, Mollo B, Vincent C, et al. Seroprevalence and risk factors of exposure to COVID-19 in homeless people in Paris, France: a cross-sectional study. Lancet Public Health 2021; 6:e202–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Baggett TP, Gaeta JM. COVID-19 and homelessness: when crises intersect. Lancet Public Health 2021; 6:e193– 4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Montgomery MP, Hong K, Clarke KEN, et al. Hospitalizations for COVID-19 among US people experiencing incarceration or homelessness. JAMA Netw Open 2022; 5:e2143407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Cawley C, Kanzaria HK, Zevin B, et al. Mortality among people experiencing homelessness in San Francisco during the COVID-19 pandemic. JAMA Netw Open 2022; 5:e221870. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Ghinai I, Davies ES, Mayer S, et al. Risk factors for severe acute respiratory syndrome coronavirus 2 infection in homeless shelters in Chicago, Illinois—March–May, 2020. Open Forum Infect Dis 2020; 7:ofaa477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Cha S, Henry A, Montgomery MP, et al. Morbidity and mortality among adults experiencing homelessness hospitalized with COVID-19. J Infect Dis 2021; 224:425–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Montgomery MP, Meehan AA, Cooper A, et al. Notes from the field: COVID-19 vaccination coverage among persons experiencing homelessness six U.S. jurisdictions, December 2020–August 2021. MMWR Morb Mortal Wkly Rep 2021; 70:1676–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Meehan AA, Yeh M, Gardner A, et al. COVID-19 vaccine acceptability among clients and staff of homeless shelters in Detroit, Michigan, February 2021. Health Promot Pract 2022; 23:35–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Tucker JS, D’Amico EJ, Pedersen ER, Garvey R, Rodriguez A, Klein DJ. COVID-19 vaccination rates and attitudes among young adults with recent experiences of homelessness. J Adolesc Health 2022; 70:504–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Ozawa S, Yemeke TT, Evans DR, Pallas SE, Wallace AS, Lee BY. Defining hard-to-reach populations for vaccination. Vaccine 2019; 37:5525–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Koepke R, Petit AB, Ayele RA, et al. Completeness and accuracy of the Wisconsin immunization registry. J Public Health Manag Pract 2015; 21:273–81. [DOI] [PubMed] [Google Scholar]
- 12. Centers for Disease Control . Stay up to date with your COVID-19 vaccines. https://www.cdc.gov/coronavirus/2019-ncov/vaccines/stay-up-to-date.html. Accessed 23 March 2022.
- 13. Romano SD, Blackstock AJ, Taylor EV, et al. Trends in racial and ethnic disparities in COVID-19 hospitalizations, by region—United States, March–December 2020. MMWR Morb Mortal Wkly Rep 2021; 70:560–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Rogers JH, Cox SN, Hughes JP, et al. Trends in COVID-19 vaccination intent and factors associated with deliberation and reluctance among adult homeless shelter residents and staff, 1 November 2020 to 28 February 2021—King County, Washington. Vaccine 2022; 40:122–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Centers for Disease Control and Prevention . Interim guidance for homeless service providers to plan and respond to coronavirus disease 2019 (COVID-19). https://www.cdc.gov/coronavirus/2019-ncov/community/homeless-shelters/infection-control-inventory-planning-tool.pdf. Accessed 4 June 2021.