UVA COVID-19 Modeling Weekly Update

Key Takeaways

  • Virginia’s average daily cases remains very high. Nevertheless cases continue to decline statewide. 29 of 35 Virginia Health Districts are in declining trajectories, and only one (Loudoun) is in a surge trajectory.
  • Variants of Concern (VoCs) continue to be identified nationally and in Virginia. Four cases of the B.1.1.7 VoC have been identified in Virginia, including 3 in Northern Virginia and 1 in Northwest Virginia.
  • Prevention and mitigation strategies may successfully contain VoCs. However, if coupled with reduced prevention effort, VoCs could lead to high transmission and extend the worst impacts of the pandemic.

Full Weekly Report.

UVA COVID-19 Model Dashboard.

UVA Biocomplexity Institute Slides.

RAND Corporation Situation & Research Update.

Eviction Moratoriums and COVID-19

By Priya Pattath, Ph.D, MPH & Michael Landen, MD, MPH

Eviction Moratoriums and COVID-19

Evictions threaten to increase the spread of COVID-19 by increasing homelessness and crowding while interfering with key public health interventions such as isolation, quarantine, social distancing and contact tracing. Historically, evictions and housing displacement have been drivers of health inequity and can be particularly harmful to individual and public health during a pandemic. Populations with higher rates of COVID-19 cases, hospitalizations and death, such as African American and Hispanic populations, are more likely to face eviction which may further exacerbate COVID-19 disparities. Specifically, 80% of persons facing eviction are people of color. Reducing evictions through eviction moratoriums helps reduce COVID-19 transmission and supports public health measures.

Evictions

Many people are at risk of eviction. When the COVID-19 pandemic began in the United States, about 20.8 million renter households (47.5% of all renter households) were already rental cost-burdened. In Virginia, 499,000 households (46.1% of renter households) are rental cost-burdened, with higher percentages for Black (53.3%) and Hispanic (55.9%) households. In 2020 women and people of color were severely affected by the COVID-19 pandemic and associated economic fallout. 51.7% of individuals that were filed against for eviction since March 15, 2020 were women.

Eviction rates are not the same throughout the country. Using data from Princeton University’s Eviction Lab and the RVA,  researchers at the Federal Reserve Bank of Richmond found that localities in Virginia, North Carolina, and South Carolina rank among cities with the most evictions in the country. In Virginia in 2016, the eviction rate was 5.1% compared to 2.8% for the U.S. Richmond had the second highest eviction rate among large cities in the United States in 2016, with an eviction rate of 11.4%, while Hampton, Newport News and Chesapeake ranked 3rd, 4th and 10th, respectively with eviction rates of 10.5%, 10.2% and 7.9%, respectively. Among mid-size cities in the United States with the highest eviction rates, Petersburg, Hopewell and Portsmouth ranked 2nd, 4th and 5th, respectively with eviction rates of 17.6%, 15.7% and 15.1%, respectively. Data from the RVA Eviction Lab, also analyzed by Richmond Fed researchers, show that state and federal policies addressing the evictions were effective early in the pandemic. Compared to 2019, filings and judgments in Virginia fell significantly between January and July of 2020.  However, their effectiveness may have waned, with more than 10,500 eviction filings and 4,500 eviction judgments issued between July and September.

Evictions and COVID-19

Evictions have been shown to increase homelessness and household crowding as people lose regular housing or move into someone else’s home. Both homelessness and household crowding have been shown to increase the risk of COVID-19. Individuals experiencing homelessness are at increased risk of severe COVID-19 due to a lack of safe housing and the high prevalence of risk factors for COVID-19 in this population. In New York City among pregnant women, neighborhood socioeconomic status and household crowding were associated with the risk of COVID-19.

Recent studies have also shown that lifting eviction moratoriums was associated with increased COVID-19 incidence and mortality in various states with effects growing over time. One study pending peer review examined whether lifting eviction moratoriums was associated with COVID-19 incidence and mortality. Among 44 states that had eviction moratoriums in place during March–September 2020, twenty-seven states (61%) lifted their moratoriums during the study period. The study found that 16 weeks after eviction moratoriums were lifted, COVID-19 incidence was twice that of states that kept their moratoriums and COVID-19 mortality was 5.4 times greater translating to a total of 433,700 excess cases and 10,700 excess deaths. The study also suggested that effects grew over time, perhaps due to displacement, crowding, and homelessness as evictions proceeded.

A recent simulation study pending peer review demonstrated that, by forcing households to merge, evictions can alter the shape of COVID-19 epidemic curves in U.S. cities. The study modeled the effect of evictions on SARS-CoV-2 epidemics, simulating viral transmission within and among households in a theoretical metropolitan area. By applying the model to Philadelphia using locally specific parameters, the researchers showed that infections particularly increase in heterogeneous cities where both evictions and contacts occur more frequently in poorer neighborhoods.

Evictions may interfere with key public health measures for reducing COVID-19. Evictions may interrupt COVID-19 quarantine and isolation. Because of an acute change of address, both case investigation and contact tracing may be harder to complete.  Evictions may also lead to lower access to COVID-19 testing and access to care by driving families to poorer, under resourced neighborhoods and medically underserved geographic areas.

 Interventions

For these reasons, the Centers for Disease Control and Prevention issued an order to temporarily halt residential evictions nationwide effective September 4, 2021 to prevent the spread of COVID-19 by facilitating self-isolation, supporting stay-at-home and social distancing, and thus reducing the risk of overcrowded living environments. The order has been renewed to continue until January 31, 2021.

The initial financial stimulus from the Coronavirus Aid, Relief, and Economic Security (CARES) Act included both direct payments and enhanced unemployment benefits. Additional COVID financial relief through additional direct payments, extended unemployment benefits and rental assistance was provided at the end of December 2020 through another COVID relief package.

During the pandemic, many states have issued additional eviction bans and provided financial assistance programs. The District of Columbia offers the most comprehensive assistance by suspending all eviction proceedings. Virginia has a rent and mortgage relief program to assist households facing evictions due to COVID-19. Administered by the Virginia Department of Housing and Community Development, the Virginia Rent and Mortgage Relief Program may provide one-time payments, with an opportunity for renewal, to persons with financial need for past due rent or mortgage payments beginning April 2020.

Conclusions

Evidence strongly supports the use of eviction moratoriums and other eviction prevention efforts to reduce COVID-19 and its complications. Evictions are more common among those groups that already are experiencing a greater burden from COVID-19. Evictions increase homelessness and household crowding. Homelessness and household crowding are associated with an increased risk of COVID-19. And evictions interfere with key public health measures such as isolation, quarantine, social distancing and contact tracing. Therefore, eviction moratoriums and related policies should be effective at reducing COVID-19 cases until transmission has declined substantially, which may not occur until the summer of 2021.

References

Benfer EA, Vlahov D, Long M et al. (in press). Eviction, Health Inequity, and the Spread of COVID-19: Housing Policy as a Primary Pandemic Mitigation Strategy. Journal of Urban Health (in press).

CDC (2020). Federal Register Notice: Temporary Halt in Residential Evictions to Prevent the Further Spread of COVID-19. https://www.cdc.gov/coronavirus/2019-ncov/downloads/eviction-moratoria-order-faqs.pdf

Dolkart P & Norris S (2020). Is a Wave of Evictions Coming? Regional Matters. Federal Reserve Bank of Richmond. 

Emeruwa UN, Ona S, Shaman JL et al. (2020). Association between Built Environment, Neighborhood Socioeconomic Status, and SARS-CoV-2 Infection among Pregnant Women in New York City. JAMA. 2020;324(4):390-392.

Eviction Lab (2020). Princeton University. 

Greenberg D, Gershenson C & Desmond M (2016). Discrimination in Evictions: Empirical Evidence and Legal Challenges. Harvard Civil Rights-Civil Liberties Law Review. 2016;51(1):115-158.

Harvard Joint Center for Housing Studies (2020). America’s Rental Housing 2020 Report. 

Leifheit KM, Linton SL, Raifman J et al. (2020). Expiring Eviction Moratoriums and COVID-19 Incidence and Mortality November 30, 2020.

Lemmerman E, Louis R & Hepburn P (2020).  Preliminary Analysis: Who is being filed against during the pandemic? Eviction Lab. Princeton University.

Perri M, Dosani N & Hwang SW (2020). COVID-19 and People experiencing Homelessness: Challenges and Mitigation Strategies. 2020;192:E716-E719.

Sheen J, Nande A, Walters EL et al. (2020). The effect of eviction moratoriums on the transmission of SARS-CoV-2. medRxiv. Published online November 1, 2020:2020.10.27.20220897.

Streeter RA, Snyder JE, Kepley H et al. (2020). The geographic alignment of primary care Health Professional Shortage Areas with markers for social determinants of health. PLOS ONE 15(4): e0231443.

 

Holiday Gatherings Increase Risk of Spreading COVID-19

By Carter C. Price

Virginia is averaging more than 3,500 confirmed cases of COVID-19 per day (more than 40 cases per 100,000) over the last few weeks. This case rate is the highest we have seen since the first case was announced in Virginia in March. Further, it is more than four times the levels we saw during this summer’s surge, as can be seen in Figure 1. Hospitalizations have also been increasing and trail behind the case level by about two or three weeks. Just as cases increased rapidly after the Thanksgiving weekend, holiday gatherings in the coming weeks could push levels even higher. It has never been more important to limit the size of gatherings this holiday season, not only to protect those gathering but to keep hospital beds open for all who might need them.

 

Figure 1: Seven-day average rate for confirmed cases and hospitalizations
Source: https://covidtracking.com/data/state/virginia

 

Figure 2 shows the average daily cases per 100,000 by county from December 8 to 15. Most counties in the Southwest and many counties in the Northwest and Central regions have daily cases rates well above 60 per 100,000. However, just about every county in Virginia has high rates of COVID-19, over 20 per 100,000.

 

 

Figure 2: Average Daily Case Count per 100,000 for December 8th to 15th
Source: VDH

 

When case rates are this high, the risk from even small gatherings of people outside of one’s household increases substantially. As seen in Table 1, we used regional COVID-19 prevalence to estimate the risk that at least one person will show up to a holiday gathering infected with COVID-19 depending on the number of attendees. Some gatherings will face higher risk of being exposed to the virus than these estimates because of the presence of people with occupations or behaviors that raise their risk of being infected by COVID-19. Other gatherings, such as those involving a couple of families that have been social distancing, may face lower risks of exposure.

 

Table 1: Estimated Likelihood a Gathering has at Least One Person Infected with COVID-19 by County and Group Size
Note: * Current limit on social gathering and **Previous limit on social gathering
Source: Author’s calculations using a binomial distribution where attendees are independently sampled with probability of infections based on the 14-day infectious period and assuming one-fourth of infections are confirmed by testing.

 

In the Southwest, more than half of events with 20 people would be expected to have at least one attendee with COVID-19. The odds are only slightly lower for most of the rest of the Commonwealth. Even small gatherings come with risks. For gatherings with just ten people, the current limit on social gathering size, we expect about one in five in Central and Eastern Virginia to include at least one person with COVID-19; that increases to about one in four in Northern and Northwest Virginia. In Southwest Virginia, we expect about one in three gatherings of ten people will include at least one person with COVID-19. Given that case rates are expected to continue to grow, the risks will be even higher by the end of this week.

Governor Northam’s Executive Order Seventy-Two lowered the limit on social gatherings to 10 people from 25. If followed, this lower limit will cut the risk of spread from a gathering of the maximum allowed size nearly in half across the Commonwealth, but even at this lower level, the likelihood that an attendee at a gathering has COVID-19 is still sizable, ranging from one-in-five to one-in-three. That is why it is important to continue to follow the guidance from VDH and the CDC regarding safe gatherings.

Just as cases increased after Thanksgiving weekend in much of Virginia, Christmas, New Years, and other holiday gatherings could lead to increased COVID-19 spread. If that happens, we’ll see cases rise rapidly during the first two weeks of January, hospitalizations will rise shortly after that, and by the end of January the number of deaths per day from COVID-19 will be substantially higher than today. January and February could be quite grim if what follows this holiday season is similar to the spread of disease after Thanksgiving weekend.

UVA COVID-19 Model Weekly Updates

Key Takeaways

  • Models are designed to project what could happen based on current trends but do not forecast what will happen. Behavioral responses drive changes in current trends.
  • The reproduction rate has increased in all districts, and is above 1.0 statewide and in all but one region.
  • Weekly incidence in Virginia (12/100K) is increasing but is still below the also increasing national average (19.6/100K).
  • 5 Health Districts entered surge trajectories and the number with declining trajectories dropped significantly.
  • While early, national and global trends suggest concerns about winter case growth and schools reopening may be founded.

Full Weekly Report.

UVA COVID-19 Model Dashboard.

UVA Biocomplexity Institute Slides.

RAND Corporation Situation & Research Update.