People in prison have been 5.5 times more likely to get COVID-19 and have suffered a COVID-19 mortality rate 3 times higher than the general public. Now, new data from UCLA Law’s COVID-19 Behind Bars Data Project published this week in preprint1 reveals the degree to which the death rate from all causes of death — not just COVID-19 — increased in Florida state prisons during the COVID-19 pandemic.2 The findings, outlined below, give us a much fuller picture of how deadly COVID-19 has been in prisons.
More deaths, despite fewer people in prisons
The researchers collected monthly population counts for Florida state prisons from 2015 to 2020. Based on this data, the Florida state prison population decreased by about 10,000 in 2020, a reduction of approximately 11% since 2019. The Florida Department of Corrections attributes this population reduction to “fewer arrests and prosecutions, fewer individuals sentenced to incarceration, and fewer commitments received from county jails,” suggesting that while fewer people entered prison, there were no large-scale efforts to release people who were nearing the end of their sentence or those who were at high risk for death related to COVID-19. Despite the slightly smaller prison population, there were more deaths in 2020 (across all causes of death) than in any of the five years prior.
During 2020, there were 42% more “excess deaths” — from all causes — in Florida prisons
During 2020, the United States experienced 23% more deaths than expected, but the Florida prison system experienced almost double that: there were 42% more deaths than expected behind bars in Florida. Importantly, not all of these “excess deaths”3 are officially attributed to COVID-19. According to public health experts, the excess deaths in the United States may be due to undocumented COVID-19 infections, delayed access to health care, or behavioral health crises. The same is likely true in prisons, where healthcare systems frequently neglect medical complaints even during normal times, and where an increase in practices like solitary confinement during COVID-19 has exacerbated mental illnesses.
As the graph below shows, deaths spiked in Florida prisons during the pandemic- and only some were deaths from COVID-19:
Figure 1: Observed and expected mortality trends in Florida state prison population. This graph included in the article shows that deaths in Florida state prison exceeded expectations based on the past 5 years by more than 42%. Not all of those excess deaths are diagnosed COVID-19 deaths, but most are likely at least indirectly caused by the COVID-19 pandemic.
The life expectancy for people in Florida prison decreased by more than 4 years between 2019 and 2020
The authors calculate that COVID-19 reduced the life expectancy of the Florida prison population by over four years. That is far worse than the CDC estimate that the U.S. general population’s life expectancy decreased by about 1 year during the pandemic. (This mortality disparity during COVID-19 is on top of the fact that incarceration itself shortens the lives of people behind bars, as research shows that for every year of incarceration, life expectancy decreases by two years.)
Figure 2: Bootstrapped estimates of life expectancy for Florida state prison population by year. The second graph from the article shows that the life expectancy of people in Florida state prisons is the lowest it’s been in 5 years — declining by 4 years from 2019 to 2020.
Increased mortality rates in prison population
Across all age groups in Florida state prisons, the all-cause mortality rate increased from 2019 to 2020. Four age categories saw statistically significant increases: mortality rates increased 77% for people 35-44 years old, 49% for people 55-64 years old, 67% for people 65-74 years old, and 61% for those 75 or older. Although prisons are often filled with young people, older adults are making up an increasing proportion of the national prison population.
The threat of COVID-19 behind bars disproportionately impacts these older adults who are more likely to already have any number of complex medical conditions.
Conclusions
As we reckon with the consequences of states largely ignoring COVID-19 behind bars, this new data is a reminder to not limit our assessment of the harm to COVID-19 deaths alone. Since the beginning of the pandemic, we have argued that failing to depopulate prisons would not only enable massive viral outbreaks, but that it would also, indirectly, strain medical units in prison and lead to more deaths from a wide variety of illnesses. This data from Florida provides the first quantitative evidence that we and other advocates were right.
The disastrous consequences of COVID-19 behind bars could have been avoided — or at least reduced — if states acknowledged that people in prison have conditions that put them at higher risk, social distancing is impossible in prisons, and prison populations needed to be reduced significantly and quickly. Instead, states chose to leave the vast majority of incarcerated people in prison, at the mercy of a virus that thrives in crowded conditions. The consequences have extended far past the number of official diagnosed COVID-19 cases and deaths.
Footnotes
A preprint reports new medical research that has yet to be peer-reviewed and so should not be used to guide clinical medical practice. ↩
The data in this study is limited to state prisons in Florida. However, because the age and sex distribution of Florida’s prison population is typical of other state prison systems, we expect that studies of other states’ prisons would likely show similar results. ↩
Excess deaths are the number of deaths from all causes that were greater than the expected number of deaths. ↩
Low rates of vaccine uptake among correctional staff make it clear that withholding the vaccine from people who are locked up -- or offering it only to a small fraction of the prison population -- is senseless.
Correctional staff in most states have been eligible for COVID-19 vaccination for months, prioritized ahead of many other groups because of the key role staff play in introducing the virus into prisons and jails and then bringing it back out to surrounding communities. Against the recommendations of medical experts, many states chose to vaccinate correctional staff before incarcerated people, often claiming that staff would serve as a barrier against the virus entering prisons and infecting people who are locked up. Now it’s becoming clearer than ever that this policy choice was a gigantic mistake: New data suggests that most prison staff have refused to be vaccinated, leaving vast numbers of incarcerated people — who have been denied the choice to protect themselves — at unnecessary risk.
We compiled data from the UCLA Law COVID-19 Behind Bars Data Project, The Marshall Project/AP, and other sources,1 and calculated the current rate of staff immunizations in 36 states and the Bureau of Prisons. We found that across these jurisdictions, the median vaccination rate — i.e. the percentage of staff who had received at least one COVID-19 vaccine dose — was only 48%. The numbers are even more disturbing in states like Michigan and Alabama, where just over 10% of staff have gotten at least one dose of a COVID-19 vaccine.
Figure 1.Data compiled from the UCLA Law COVID-19 Behind Bars Data Project, The Marshall Project, and several state-specific data sources (see footnote 1). See the appendix to this article for a table with details about all 37 prison systems for which we gathered data.
This data confirms what we’ve learned anecdotally over the past few months through local news reporting. For example:
In Colorado, vaccine uptake among correctional staff has been so poor that the state is now offering staffers $500 each to get the vaccine.
The Marshall Projectreported in mid-March that “In Massachusetts, more than half the people employed by the Department of Correction declined to be immunized. A statewide survey in California showed that half of all correctional employees will wait to be vaccinated. In Rhode Island, 30% of prison staff have refused the vaccine, a higher rate than the incarcerated, according to the state’s Department of Corrections. And in Iowa, early polling among employees showed a little more than half the staff said they’d get vaccinated.”
These low rates of vaccine uptake among correctional staff make it clear that withholding the vaccine from people who are locked up — or offering it only to a small fraction of the prison population — is senseless. No policymaker in any state should assume there is a firewall of vaccinated staffers protecting incarcerated people from the coronavirus.
Especially as the U.S. experiences a potentially disastrous “fourth surge” of the pandemic, it remains urgently necessary to:
Offer the vaccine to all incarcerated people — now. As we’ve discussed before, incarcerated people are much more likely to contract and die from the coronavirus, because outbreaks behind bars are common and a disproportionate number of incarcerated people have chronic medical problems that make the virus more deadly. (In many of the states we researched, officials and journalists have noted that incarcerated populations have had much higher uptake rates than staff.)
Depopulate prisons and jails. The coronavirus thrives in dense environments, so releasing people is still the best way to stop outbreaks behind bars — and as long as staff and incarcerated people aren’t vaccinated, outbreaks are certain to continue. States should be considering the most medically vulnerable incarcerated people first, and not excluding people automatically based on whether they committed a violent crime (we’ve written at length about the perils of leaving behind whole categories of incarcerated people). Unfortunately, prison releases have been very sparse so far.
As the new data shows, it’s simply not true that “offering” the vaccine to correctional officers amounts to protecting incarcerated people or the public from the rapid spread of the virus in correctional facilities. What states must do is make the vaccine truly accessible to both corrections staff and people who are locked up, and immediately begin increasing prison releases through commutations, good time credits, and expansions of parole. As long as states ignore and neglect incarcerated people, there will be no end in sight to the pandemic in prisons and jails.
Update April 23, 2021: The percentage of North Carolina Dept. of Public Safety staff who had received at least one dose of a vaccine as of April 20, 2021 was erroneously reported in an earlier version of this article as 85%. That calculation was based on the number of correctional staff (7,774) as reported by The Marshall Project/AP, but the Dept. of Public Safety clarified that the staff vaccination counts include all DPS employees, not just correctional staff. All related calculations (e.g. the total reported in the appendix) have been updated as well to reflect the corrected data.
Update June 21, 2021: The percentage of Virginia Dept. of Corrections (DOC) staff who had received at least one dose of a vaccine as of April 20, 2021 was originally reported as 72%. That calculation was based on the number of correctional staff (8,895) as reported by The Marshall Project/AP, but the DOC clarified in an email that our denominator (the number of correctional staff) may not be correct, although they did not provide the correct number of staff. They did, however, report that as of June 4, 2021, only 57% of staff had received at least one dose of a vaccine. Without the details of the number of staff and number receiving a first dose, we are unable to update the other calculations in this analysis (e.g. the total reported in the appendix).
Footnotes
Source notes: In addition to the UCLA and The Marshall Project/AP data sets, we sought staff vaccination data from state Department of Corrections websites, news articles, and in one case, the Covid Prison Project’s media-sourced data set. Our vaccination rate calculations are based on total staff numbers, most of which come from The Marshall Project/AP data set; other sources are noted in the appendix table. The types of employees included in the total staff counts vary by state, and those details were not always clear in the data set. Data from UCLA, The Marshall Project/AP, and state Department of Corrections websites were accessed on April 20, 2021.
It’s important to note that states do not report vaccination data consistently, so we made every effort to avoid double-counting staff and overestimating vaccination rates. Specifically, we typically defined staff receiving “at least one dose” of a vaccine as those who were reported as “partially” vaccinated, or having “initiated” vaccination or “received first dose.” This is because many states record vaccinated staff members twice – once when a two-dose vaccine schedule is started and once when it’s completed; those receiving the one-shot Johnson & Johnson vaccine may be included in both categories as well (as a “first dose” and as “completed”). In states where the available data suggested a different definition, we have noted those differences in “notes/clarifications” in the appendix table. ↩
Appendix
Prison system
Number of staff who have received at least one dose
Total number of staff
Percentage of staff who have received at least one dose
Source for staff vaccination counts
Source for total staff count
Notes/Clarifications
Alabama
824
6,259
13%
The Marshall Project/AP
The Marshall Project/AP
Arkansas
1,421
4,045
35%
The Marshall Project/AP
The Marshall Project/AP
California
27,758
46,000
60%
UCLA Law Covid-19 Behind Bars Data Project
The Marshall Project/AP
Colorado
2,972
6,267
47%
UCLA Law Covid-19 Behind Bars Data Project
The Marshall Project/AP
Connecticut
2,697
6,170
44%
UCLA Law Covid-19 Behind Bars Data Project
The Marshall Project/AP
Delaware
1,268
2,530
50.1%
UCLA Law Covid-19 Behind Bars Data Project
The Marshall Project/AP
Idaho
567
1,999
28%
The Marshall Project/AP
The Marshall Project/AP
In addition to the 224 staff vaccinated at the department, an additional 343 self-disclosed they received both doses from outside providers.
We used the staff number from the DOC because the number of vaccinated employees was described as those “who work in Louisiana’s state prisons” (not all DOC employees).
Included in our calculation of the number of staff who received at least one dose are 1,091 who received the J&J vaccine, 743 who received a first dose from the MDOC, 485 who “completed external [outside of the MDOC] vaccination process,” and 123 who “started external vaccination process.” Because the number that “started” an external vaccination process is much smaller than the number that have completed it, we assumed that the 123 who “started” were not also included in the “completed” group, as is the case in other data sets.
While the source for the COVID Prison Project data is unavailable, its data seems to be corroborated by an April 6 MDOC Employee newsletter, which states, “Thousands of … team members have been vaccinated against the COVID-19 virus.”
Specifically, the NDOC update reports “1,230 – first dose, 822- second dose.” Because it is unclear whether those who received second doses are also counted among those who have received a first dose, as is true in other data sets, we used the first dose counts to avoid double-counting.
N.C. Dept. of Public Safety, via email (4/23/21). This includes all DPS employees, not just correctional officers.
We included both “partially” and “fully” vaccinated staff because the number of “fully” vaccinated staff was much greater than the number “partially” vaccinated, suggesting that unlike other data sources, the “fully” vaccinated staff are not double-counted in the NCDOC’s “partially” vaccinated staff counts.
Ohio
7,057
12,192
58%
The Marshall Project/AP
The Marshall Project/AP
Pennsylvania
3,094
15,073
21%
UCLA Law Covid-19 Behind Bars Data Project
The Marshall Project/AP
Rhode Island
927
1,339
69%
The Marshall Project/AP
The Marshall Project/AP
Tennessee
3,247
5,179
63%
UCLA Law Covid-19 Behind Bars Data Project
The Marshall Project/AP
Texas
11,893
36,073
33%
The Marshall Project/AP
The Marshall Project/AP
Vermont
467
1,001
47%
The Marshall Project/AP
The Marshall Project/AP (does not include health care workers, who are contractors).
Virginia
6,416
8,895
72%
UCLA Law Covid-19 Behind Bars Data Project
The Marshall Project/AP
See update of June 21, 2021 in the text above. The VA DOC reported that as of June 4, 2021, only 57% of staff had received at least one dose of a vaccine; further details were not provided.
According to a Patch.com article (3/16/21), correctional staff were only eligible for vaccination starting March 17, which was much later than many other states.
Specifically, the DCR reports 1,914 first doses and 1,774 second doses administered to 3,687 employees (including contract staff). We used the count for first doses to avoid double-counting those who have received second doses, because it was unclear in the data whether these are mutually exclusive groups.
The Marshall Project/AP report a much smaller DOC staff number (4,640), but it varied so dramatically from the WDOC number that we decided to use the count from the dashboard.
We are excited to welcome Naila Awan, who will serve as the first-ever Director of Advocacy at the Prison Policy Initiative. Naila is a civil and human rights lawyer with years of experience collaborating with, supporting, and representing Black- and Brown-led grassroots organizations in policy reform and litigation efforts. Prior to joining Prison Policy Initiative, Naila worked for multiple civil rights organizations and served on the legislative staff for Congresswoman Tammy Baldwin.
Most recently, Naila served as Senior Counsel at Dēmos, where her work centered on combating voter suppression and expanding access to the ballot for traditionally marginalized communities. In this role, she led a cross-functional project to end the disenfranchisement people experience when then come into contact with the criminal legal system, testified before Congress, and served as counsel in A. Philip Randolph Institute v. Husted, a U.S. Supreme Court case challenging Ohio’s voter purge practices, and Mays v. LaRose, a class action seeking to expand access to the ballot for voters detained in jail. She also co-authored Enfranchisement for All: The Case for Ending Penal Disenfranchisement in Our Democracy and How to End De Facto Disenfranchisement in the Criminal Justice System.
Naila holds a L.L.M in International Studies from the New York University School of Law, a J.D from the Ohio State University’s Moritz College of Law, and a B.A from Miami University of Ohio.
We are excited to welcome our new Communications Director Mike Wessler. Mike has more than a decade of experience helping campaigns, political parties, nonprofit organizations and elected officials accomplish their goals through strategic communication. Mike has done communications work at the Massachusetts Office of the State Auditor and the Office of the Montana Governor, as well at the Montana Department of Labor and the Montana Democratic Party. Mike has a Bachelor’s degree in Political Science from Florida State University.
