Annual Incidence of COVID-19 Unexpectedly Lower Among People Experiencing Homelessness

The annual incidence rate of COVID-19 was found to be lower among people experiencing homelessness than in the general populations at state and local levels, challenging expectations, according to a study recently published in JAMA Network Open. Though not representative of a national estimate, this finding indicates that opportunities exist for incorporating housing and homelessness status in infectious disease reporting to inform public health actions.

People experiencing homelessness have been found to have a higher burden of infectious diseases. However, little is known about the total number of COVID-19 cases among people experiencing homelessness due to the challenge of a lack of timely and high-quality data on COVID-19 among this population.

Researchers conducted a cross-sectional study to estimate the number of COVID-19 cases among people experiencing homelessness and compare the incidence rate among people experiencing homelessness with the general population.

Data were collected from a survey distributed by the Centers for Disease Control and Prevention to all US state, district, and territorial health departments that requested aggregated COVID-19 data among people experiencing homelessness from January 1, 2020, to September 30, 2021.

Participants included a population-based sample of all 64 US jurisdictional health departments. Among the 25 states, districts, and territories that completed the survey, 18 states (72%) and 27 localities reported COVID-19 data among people experiencing homelessness.

A total of 26,349 cases of COVID-19 among people experiencing homelessness were reported at the state level, and 20,487 cases were reported at the local level.

The annual incidence rate of COVID-19 among people experiencing homelessness at the state level was found to be 567.9 per 10,000 person-years (95% CI, 560.5-575.4 per 10,000 person-years) compared with 715.0 per 10,000 person-years (95% CI, 714.5-715.5 per 10,000 person-years) in the general population. At the local level, the incidence rate of COVID-19 among people experiencing homelessness was 799.2 per 10,000 person-years (95% CI, 765.5-834.0 per 10,000 person-years) compared with 812.5 per 10,000 person-years (95% CI, 810.7-814.3 per 10,000 person-years) in the general population.

Overall, the annual incidence rate of COVID-19 was found lower among people experiencing homelessness than in the general populations at state and local levels. These findings contradicted the expectations of researchers, considering the risks for other infectious diseases for people experiencing homelessness and the number of identified disease outbreaks in homeless shelters.

The authors suggest a few reasons for this unexpected result, including the possibility of under-ascertainment of homelessness among people with COVID-19 due to lack of data on housing status.

Although these results provide an estimate of COVID-19 incidence rates among people experiencing homelessness in multiple US jurisdictions, a national estimate and the extent of under or overestimation are unknown. The authors note that the responding jurisdictions may not be representative of those that did not collect data among people experiencing homelessness. They assert that the data should be interpreted with caution, considering the data are incomplete and may be biased toward jurisdictions that were well resourced to prevent COVID-19 among people experiencing homelessness.

Given these findings, the authors suggest that opportunities exist for incorporating housing and homelessness status in infectious disease reporting to inform public health decision-making and improve understanding of the burden of infectious diseases among disproportionately affected groups.

They encourage the integration of homeless service utilization data systems into health data systems, which may alleviate the burden on health departments in collecting housing or homelessness information during case investigations or interviews.

Additionally, they note that health departments could consider creating public-facing dashboards or regular reports with these data to allow for improved data sharing and informed decision-making at all levels of public health responses.

The study has some limitations. For example, jurisdictions that use more inclusive definitions of homelessness may have different COVID-19 incidence rates than jurisdictions that use narrower definitions of homelessness, which may contribute to incomparable and possibly skewed estimates.

The study also included a large representation of local jurisdictions from California. The study team could not control which state-level jurisdictions did or did not share the survey with their local health departments, which may have biased results at the local level.

These limitations, among others, highlight the need for more standardized data collection that uses similar definitions of homelessness across data sources among state and local levels, according to the study.

Reference

Meehan A A, Thomas I, Horter L, et al. Incidence of COVID-19 among persons experiencing homelessness in the us from January 2020 to November 2021. JAMA Netw Open. 2022;5(8):e2227248. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2795298. Published August 18, 2022. Accessed August 19, 2022.