The health impacts of the COVID-19 pandemic on adults who experience imprisonment globally: A mixed methods systematic review.

BACKGROUND: The prison setting and health status of people who experience imprisonment increase the risks of COVID-19 infection and sequelae, and other health impacts of the COVID-19 pandemic.
OBJECTIVES: To conduct a mixed methods systematic review on the impacts of the COVID-19 pandemic on the health of people who experience imprisonment. DATA SOURCES: We searched Medline, PsycINFO, Embase, the Cochrane Library, Social Sciences Abstracts, CINAHL, Applied Social Sciences Index and Abstracts, Sociological Abstracts, Sociology Database, Coronavirus Research Database, ERIC, Proquest Dissertations and Theses, Web of Science, and Scopus in October 2021. We reviewed reference lists for included studies. STUDY ELIGIBILITY CRITERIA: Original research conducted in or after December 2019 on health impacts of the COVID-19 pandemic on adults in prisons or within three months of release. STUDY APPRAISAL AND SYNTHESIS
METHODS: We used the Joanna Briggs Institute's Critical Appraisal Checklist for Qualitative Research for qualitative studies and the Joanna Briggs Institute's Critical Appraisal Checklist for Studies Reporting Prevalence Data for quantitative studies. We qualitized quantitative data and extracted qualitative data, coded data, and collated similar data into categories.
RESULTS: We identified 62 studies. People in prisons had disproportionately high rates of COVID-19 infection and COVID-19 mortality. During the pandemic, all-cause mortality worsened, access to health care and other services worsened, and there were major impacts on mental wellbeing and on relationships with family and staff. There was limited evidence regarding key primary and secondary prevention strategies. LIMITATIONS: Our search was limited to databases. As the COVID-19 pandemic is ongoing, more evidence will emerge.
CONCLUSIONS: Prisons and people who experience imprisonment should be prioritized for COVID-19 response and recovery efforts, and an explicit focus on prisons is needed for ongoing public health work including emergency preparedness. PROSPERO REGISTRATION NUMBER: 239324.

Introduction

From early in the COVID-19 pandemic, community advocates and public health experts have been sounding the alarm about the urgent public health risks of COVID-19 in prisons [1-5]. While attention has primarily focused on preventing the introduction and transmission of COVID-19 in prisons, the intersecting social and structural dynamics of transmission and public health responses suggest a much broader scope of impact.

The risk of COVID-19 introduction into prisons is high because of frequent movement of people between prisons and the community, including people being admitted to prison, people on intermittent sentences, and staff. Prison environmental conditions, such as close quarters, overcrowding, and limited individual autonomy over prevention measures increase the risks of transmission [2, 4, 6, 7]. The high prevalence of chronic health conditions among people in prison, including respiratory diseases, cardiovascular diseases, and conditions associated with immune compromise [8-11], increases the risk of serious sequelae of COVID-19 infection in this population.

The COVID-19 pandemic also affects population health status through mechanisms other than COVID-19 infection. While many countries reduced their prison population size through measures to reduce entry into custody and increase release from custody [12, 13], there may not have been commensurate increases in discharge planning or community resources to support needs, including for treatment beds and shelter beds. In addition, the transition to remote services and reduced scope and hours of services in many jurisdictions may have limited access to essential health and social services [14]. Anxiety about the pandemic, isolation, difficulties navigating public health measures, fewer employment opportunities, and reduced health and social supports may all contribute to worse mental health and increased substance use among people who experience imprisonment [14, 15]. In prisons in particular, efforts to mitigate transmission have included increased time in cells, restrictions in work and education programs, and limited visits and social interactions [13], all of which negatively impact mental health [16]. There is also evidence from some jurisdictions that the pandemic has been associated with an increasingly unstable and toxic illicit drug supply [15, 17], which may further increase the already substantially elevated risk of overdose for people who experience imprisonment [18], both while in prison and in the community post-release.

The health impacts of the COVID-19 pandemic have exacerbated existing health inequities shaped by broader patterns of marginalization and colonization for people who experience imprisonment [19]. For example, people with clinical conditions such as substance use disorders and mental illness are overrepresented in prisons [8]. In addition, certain demographic groups are overrepresented in prisons, for example, in Canada, people who are Indigenous and Black [20, 21]. This overrepresentation means that the harms of the COVID-19 pandemic experienced by people who experience imprisonment disproportionately impact specific communities.

A comprehensive review of current knowledge of the impacts of the pandemic on people who experience imprisonment is essential to inform ongoing COVID-19 prevention and response, pandemic recovery, and emergency preparedness, and to address persistent health and healthcare inequities. To address this gap, we conducted a mixed methods systematic review of evidence about the impact of the COVID-19 pandemic on the health of people who experience imprisonment.

Methods

Protocol and registration

We developed a research protocol, which we registered in PROSPERO under registration number CRD42021239324.

Search

We developed a search strategy in consultation with a research librarian. We searched Medline, PsycINFO, Embase, the Cochrane Library, Social Sciences Abstracts, CINAHL, Applied Social Sciences Index and Abstracts, Sociological Abstracts, Sociology Database, Coronavirus Research Database, ERIC, Proquest Dissertations and Theses, Web of Science, and Scopus, and the search strategy is available in S1 File or at https://www.crd.york.ac.uk/PROSPEROFILES/239324_STRATEGY_20210224.pdf. We reviewed reference lists of included studies and relevant reviews. We limited our search to studies appearing since December 2019, when the first human cases of COVID-19 were identified. We did not have any language restrictions for eligible articles, though we used only English language search terms.

We initially ran the database search on February 28th, 2021. We updated the search on October 14th, 2021.

For any identified articles that were pre-prints, we searched for a published peer-reviewed version of the article, and if a published peer-reviewed document was found (through searches in Google or Medline, or after contacting the corresponding author), we updated the extracted data and the reference based on the published peer-reviewed article. If a published peer-reviewed document was not available, we used the published pre-print.

Eligibility criteria

Studies were eligible for inclusion if they met four criteria. The first criterion was a focus on adults aged 18 and older in prisons, jails, or correctional facilities or within 3 months of release from imprisonment; we use the term “prison” in this article to describe all these carceral settings, though for each study we describe the specific setting. If studies included data for youth in detention or other populations in addition to adults in prisons, we included the study only if data were stratified for adults experiencing imprisonment. The second criterion was research that occurred in or after December 2019, in order to exclude studies of other coronaviruses. The third criterion was original quantitative or qualitative research, which could include surveillance and outbreak data. The fourth criterion was health impacts of the COVID-19 pandemic. We defined health impacts as health outcomes directly and/or indirectly attributable to the COVID-19 pandemic, including outbreaks, changes in health care and other services related to or coincident with the pandemic, COVID-19 morbidity and mortality, and changes in mental or physical health coincident with the pandemic.

We excluded studies if they focussed only on immigration detention facilities, or if they described only prison policies, which may not align with practice or experience.

Selection criteria and data extraction

Two authors independently reviewed all titles and abstracts identified in the search. We resolved discrepancies through discussion regarding whether the article was eligible for full text review based on the title and abstract. Two authors independently reviewed each full article for eligibility for inclusion. Disagreements in the decisions by the two reviewers were resolved through discussion, and the involvement of a third reviewer when necessary.

We used a data extraction form, which we modified based on the Joanna Briggs Institute (JBI) extraction tool in JBI SUMARI. We extracted study information, study characteristics, participants, methods, and outcomes. When studies presented data for correctional staff as well as people in custody, we extracted only data for people in custody. We transformed quantitative data by “qualitizing” extracted quantitative data, i.e. converting quantitative data into textual descriptions of the findings [22]. For qualitative data, we extracted themes or sub-themes with corresponding illustrations.

Synthesis

We planned a priori to use a convergent integrated approach to combine extracted data from quantitative studies and qualitative studies, which involves assembling the qualitized and qualitative data, and categorizing and pooling together these data based on similarity in meaning to produce a set of integrated findings [22]. Three authors independently coded the data, and met iteratively to review and discuss emerging codes (or groupings), and then two authors re-coded the data with the established coding framework. For most codes, we found qualitized data or qualitative data but not both types of data, so we could not pool across types of data. For some categories there were limited data, so we present the findings as a narrative synthesis [22]. When relevant, we reported data on the type of carceral setting, period of the study, and region of the study, assuming these factors may be relevant to the interpretation of the data presented, and also since epidemiological factors (such as COVID-19 variant) and social and political factors varied over time and across regions.

Risk of bias

We appraised the risk of bias in each study using the Joanna Briggs Institute’s Critical Appraisal Checklist for Qualitative Research for qualitative studies and the Joanna Briggs Institute’s Critical Appraisal Checklist for Studies Reporting Prevalence Data for quantitative studies [22]. We did not exclude studies on the basis of quality.

Results

Search results

As shown in Fig 1, we identified 2,092 unique references, 1,995 of which were assessed as irrelevant on title and abstract screen. We reviewed 97 full text studies for eligibility: 34 were irrelevant and 1 study was excluded because it had been retracted. Ultimately, 62 studies were eligible for inclusion in this review.

Fig 1

Flow diagram for studies included in a systematic review on the health impacts of the COVID-19 pandemic on people who experience imprisonment.

Characteristics of included studies

Studies were mostly conducted in the USA (n = 40), while 9 were conducted in Europe, 7 in the United Kingdom, 4 in Brazil, and 2 in Canada (Table 1). We included 54 quantitative studies and 8 qualitative studies; only 1 qualitative study was conducted in the USA. Almost half the studies (n = 28) were conducted in the first 6 months of 2020.

Table 1

Characteristics of studies included in a systematic review on the health impacts of the COVID-19 pandemic on people who experience imprisonment, N = 62.

Country	First author	Regiona	Period	Type of prison setting
Brazil	Crispim 2021	N/A	April-August 2020	prisons
	Gouvea-Reis 2021 [23]	Brasilia	April-October 2020	penitentiary
	Gouvea-Reis 2021 [24]	Brasilia	April-June 2020	penitentiary
	Ribeiro 2020	N/A	December 2019-September 2020	prisons
Canada	Blair 2021	N/A	March-May 2020	federal prisons
	McLeod 2021	British Columbia	May 2020	provincial prison
Italy	Cerrato 2021	Bologna	March-June 2020	prison
	Giuliani 2021 [25]	Lombardy	March-July 2020	prisons
	Giuliani 2021 [26]	Milan	February-April 2020	prison
	Sorge 2021	San Vittore	March-May 2020	prison
	Stufano 2021	Apulia	November 2020-January 2021	correctional facility
Spain	Marco 2021 [27]	Barcelona	March-April 2020	prison
	Marco 2021 [28]	Barcelona	Not specified	prison
Switzerland	Getaz 2021	Geneva	2016–2020	prison
United Kingdom	Coleman 2020	Not specified	2020	prison
	Gray 2021	Northern Ireland	April-December 2020	prisons
	Maycock 2021 [29]	Scotland	Not specified	prison
	Maycock 2021 [30]	Scotland	Not specified	prison
	Maycock 2021 [31]	Scotland	Not specified	prison
	Suhomlinova 2021	England, Wales	April-October 2020	prisons
	Wilburn 2021	Not specified	March-June 2020	prison
USA	Altibi 2021	Michigan	March-June 2020	state prisons
	Bandara 2020	N/A	May 2020	county jails and state prisons
	Berk 2021	Rhode Island	December 2020-February 2021	combined jail and prison
	Brinkley-Rubinstein 2021	Rhode Island	March-May 2021	combined jail and prison
	Chan 2021	New York City, New York	March-April 2020	jails
	Chin 2021 [32]	California	March-October 2020	state prisons
	Chin 2021 [33]	California	December 2020-March 2021	state prisons
	Chin 2022	California	December 2020-March 2021	state prisons
	Collica-Cox 2020	New York	March-May 2020	jail
	Dunne 2021	Idaho	July-November 2020	state correctional facilities
	Hagan 2020	N/A	April-May 2020	federal prisons, state prisons, county jails
	Hagan 2021 [34]	N/A	December 2020-April 2021	federal prisons
	Hagan 2021 [35]	Texas	July-August 2021	federal prison
	Hershow 2021	Wisconsin	August-October 2020	state prison
	Jimenez 2020	Massachusetts	April-July 2020	jails and state prisons
	Kennedy 2020	Connecticut	March-June 2020	prisons and jails
	Khairat 2021	North Carolina	June-November 2020	prisons
	KhudaBukhsh 2021	Ohio	April 2020	prison
	Lehnertz 2021	Minnesota	March-June 2020	correctional facilities
	Leibowitz 2021	Massachusetts	April 2020-January 2021	state prison
	Lemasters 2020	USA and Puerto Rico	March-July 2020	prisons
	Lewis 2021	Utah	September 2020-January 2021	correctional facility
	Maner 2021	N/A	April 2020-January 2021	prisons
	Marquez 2021 [36]	Texas	April 2019-March 2021	carceral settings
	Marquez 2021 [37]	Florida	2015–2020	prisons
	Njuguna 2020	Louisiana	May 2020	correctional and detention facility
	Nowotny 2020	USA and Puerto Rico	April-July 2020	prisons
	Pettus-Davis 2021	Midwest and Southeast states	March-November 2020	state correctional facilities
	Pocock 2020	Arizona	April 2020	correctional facility
	Puglisi 2021	Not specified	prior to May 2020	jail
	Pyrooz 2020	Oregon	April-May 2020	prison
	Saloner 2020	N/A	March-June 2020	state and federal prisons
	Stern 2021	Washington, California, Florida, Texas	September-December 2020	prisons and jails
	Toblin 2021	N/A	February-September 2020	federal prisons
	Tompkins 2021	Arkansas	April-May 2020	correctional facility
	Vest 2021	Texas	March-July 2020	prisons
	Wadhwa 2021	Chicago	May 2020	correctional facility
	Wallace 2020	N/A	January-April 2020	state prisons, federal prisons, detention facilities
	Wallace 2021	Louisiana	May-June 2020	detention centre
	Zawitz 2021	Illinois	March-April 2020	jail
Multiple countries	Montanari 2021	Europe	March-June 2020	prisons

aIf applicable and specified.

For the 54 quantitative studies (S1 Table), 29 did not satisfy the quality indicator in at least one domain, suggesting risk of bias, and 4 did not provide sufficient information on all relevant domains to be able to appraise quality for all domains of interest. For the 8 qualitative studies (S2 Table), all studies did not satisfy the quality indicator in at least one domain and 4 studies did not contain sufficient information to be able to appraise quality on all domains of interest.

Data synthesis

We grouped data into three categories with several subcategories, as shown in Table 2.

Table 2

Coding framework for systematic review on the health impacts of the COVID-19 pandemic on people who experience imprisonment.

Burden of COVID-19 infection	COVID-19 cases	Population incidence rates
		Percent positivity
		Number of cases
		Factors associated with infection
		Symptom status of cases
	COVID-19 hospitalizations	Proportion of cases hospitalized
		Factors associated with hospitalization
	COVID-19 death and mortality	Population mortality rates
		Case fatality
		Factors associated with death
COVID-19 prevention strategies	Primary prevention	Vaccination
		Hygiene
		Quarantine and isolation
	Secondary prevention	Testing
Other impacts of COVID-19 pandemic on health status	All cause-mortality
	Changes to services
	Impacts on relationships with family and staff
	Impacts on mental health

Burden of COVID-19 infection

COVID-19 cases

Population incidence rates. Five studies reported population-level incidence data for people incarcerated in US prisons. Data collected from March to July 2020 showed that 34 of the 53 prison systems had higher incidence rates than those for the general population, and for six states, the cumulative incidence rate was more than 100 cases per 1,000 higher for the prison population [38]. By June 6th, 2020, the cumulative incidence rate for people in federal and state prisons was 32.5 per 1,000, which was 5.5 times higher than the whole population rate of 5.9 per 1,000 [39]. By September 23rd, 2020, the cumulative incidence rate for people in federal prisons was 11,710.1 per 100,000, compared with 2,484.4 cases per 100,000 for adults in the general population [40]. In Massachusetts, the cumulative incidence rate in July 2020 was 2.9 times higher for people in jails and state prisons (44.3/1,000) compared with the general population [41], and in people in state prisons the incidence rate by January 11th, 2021 was 956 cases per 100,000 person-weeks, compared with 150 cases per 100,000 person-weeks for the general population [42].

Studies in Canada, Brazil, and Italy similarly showed high incidence rates for people in prisons. The cumulative incidence for the federal prison population in Canada was 10 times higher in Quebec, 2 times higher in Ontario, 6 times higher in British Columbia compared with the general population [43]. In Brazil, a survey of state penitentiaries identified 23,054 confirmed cases of COVID-19 by September, 2020, for a cumulative incidence rate of 30.9 per 1,000, compared with 19.7 per 1,000 for the total population [44].

Percent positivity. Three studies reported on cases identified and percent positivity based on enhanced testing initiatives. In 16 US facilities that conducted mass testing between April and May 2020, the incidence ranged from 0% to 86.8%, with a median of 29.3%, and mass testing increased total known cases from 642 before mass testing to 8,239 after testing [45]. In an Italian correctional facility for people with chronic diseases, between November 2020 and January 2021, 2 people tested positive of 426 people tested in an initial mass testing campaign (0.5%), and 0 people tested positive of 480 in a second mass testing campaign from December 2020 to January 2021 (0%) [46]. In state correctional facilities with work release programs in Idaho, USA, the percent positivity in mass testing between July and November 2020 ranged from 1–92% [47].

Sixteen studies reported the percentage of people who tested positive in the context of outbreaks [23, 26–28, 35, 48–58], as shown in Fig 2, with a wide range within and across countries from 2.5% to 90.6%, but with very high proportions of people testing positive in most outbreaks.

Fig 2

Percent testing positive in outbreaks in studies included in a systematic review on the health impacts of the COVID-19 pandemic on people who experience imprisonment, by country.

Eight studies presented test positivity data outside of enhanced testing initiatives or outbreaks. In New York City, USA jails, 58% people tested positive of 978 people tested from March to April 2020 [59]. In prisons and jails in Connecticut between March and June 2020, 12% (n = 1,240) of people tested positive of the 10,304 people tested for COVID-19 [60]. A study of 53 US prison systems from March to July 2020 found that percent positivity varied widely from 0% to 42%, and in most states, percent positivity was higher for those in prison compared with the general population [38]. In 18 prisons in Lombardy, Italy, 102 cases were identified in 5,777 people tested during the first COVID-19 wave between March and July 2020, for an incidence of 2.7% [25]. A Massachusetts, USA study found 5% test positivity for 8,455 tests conducted in people in state prisons and 14% test positivity in 1,843 tests conducted in county jails [41]. In California, USA, state prisons, test positivity was 19.2% by October 2020 [32]. Between March and November 2020, 9.9% of 317 people across 38 correctional facilities in the Midwest and Southeast USA had had confirmed COVID-19 [61]. In a Rhode Island, USA, correctional facility from March to May 2021, the COVID-19 incidence was 1.3% (20/1,539) for people who were vaccinated [62].

Number of cases. Eight studies reported on the number of cases without providing denominators. A US study of surveillance data from 32 state and territorial health department jurisdictions found that there were 4,893 COVID-19 cases among incarcerated or detained people between January 21st and April 21st, 2020 [63]. In a prison outbreak in northern Italy between March and June 2020, 34 cases were identified [64]. Between March and July 2020, there were 11,799 confirmed COVID-19 cases among incarcerated people in the Texas Department of Criminal Justice [65]. In a penitentiary outbreak in Brasília, Brazil from April to June 2020, 859 cases were identified [24]. In Brazil, between April and August 2020, a total of 18,767 COVID-19 cases were identified among incarcerated people [66]. A European study reported 26 confirmed COVID-19 cases in Belgium between March and June 2020, 119 cases in France by mid-May 2020, and 315 cases in Italy between March 9 and May 2020 [67]. In state correctional facilities with work release programs in Idaho, USA, there were 382 cases identified between July and November 2020 [47]. In a UK prison outbreak in 2020, there were 88 possible, probable, or confirmed cases [68].

Factors associated with infection. Regarding factors associated with COVID-19 infection, a study in a correctional facility in Arkansas, USA, in April and May 2020 found that pre-existing chronic lung disease was associated with infection, but other pre-existing medical conditions (hypertension, diabetes, cardiovascular disease, chronic kidney disease, chronic liver disease) were not [49]. In a prison in Barcelona, Spain, age, history of diabetes, and HIV infection, respectively, were not associated with infection [27]. In an outbreak in a Texas prison in July and August 2021, vaccination status and diabetes were associated with infection [35].

Symptom status of cases. Several studies described the proportion of cases that were symptomatic and the proportion of cases with specific symptoms. The proportion of cases identified as symptomatic was 89.8% (510/568) in New York City, USA jails from March to April 2020 [59], 76.3% (479/628) in an outbreak in an Illinois, USA jail in March and April 2020 [50], 4% (3/71) in a Louisiana, USA correctional facility in outbreak in April to May 2020 [52], 18.8% in a men’s correctional facility in Arkansas, USA in April and May 2020 [49], 43% (48/111) in a Louisiana, USA detention facility outbreak in May and June 2020 [53], 37% (12/19) in a correctional facility in Illinois, USA in May 2020 [56], and 53.5% (68/127) for people aged 60 and older in an outbreak in a penitentiary in Brazil from April to October 2020 [23].

In a men’s correctional facility in Arkansas, the most common symptoms identified for cases were headache, runny nose, chills, and cough, however, each of these symptoms was reported by less than or equal to 6% of cases [49]. In the correctional facility in Illinois, USA, the most commonly reported symptoms for symptomatic cases were loss of taste or smell (47%), headache (32%), and chills (26%) [56]. In the Louisiana jail, most commonly reported symptoms for cases were headache (32%), loss of taste or smell (31%), and nasal congestion (26%) [53]. In the outbreak penitentiary in Brazil, the most common symptoms in cases were headache (34.9%), followed by cough (30.2%), fever (28.9%), ageusia/anosmia (19.7%), dyspnea (16.7%), myalgia (10.5%), sore throat (8.0%), nasal congestion (5.7%), and diarrhea (4.0%) [24].

COVID-19 hospitalizations

Proportion of cases hospitalized. Several studies presented data on the percentage of cases hospitalized, ranging from 0.0% to 10.0% [24, 27, 32, 35, 47, 53–55, 57, 59, 60, 63, 68], as shown in Fig 3.

Fig 3

Percent of COVID-19 cases hospitalized for people who experience imprisonment in studies included in a systematic review on the health impacts of the COVID-19 pandemic on people who experience imprisonment, by month of start of study period.

Some studies also provided data on the percentage of cases admitted to ICU: 0.3% of 13,636 cases in state prisons in California, USA, between March and October 2020 [32], 1.6% (20/1,240) in Connecticut, USA state prisons from March to June 2020 [60], 1.4% (8/568) in New York City, USA jails in March and April 2020 [59], and 0.0% (0/46) in a penitentiary in Barcelona, Spain [27].

A Michigan, USA study of people hospitalized with COVID-19 found that a higher proportion of those who were incarcerated were admitted to ICU (29.6%) compared with the proportion of people who were not incarcerated who were admitted to ICU (18.7%), and the median time from self-reported onset of symptoms to hospital admission was longer for people who were incarcerated (6 days) compared with those who were not incarcerated (5 days) [69].

Factors associated with hospitalization. In two studies, older age was associated with hospitalization and ICU admission in cases [32, 60]. In prisons and jails in Connecticut, heart disease was the strongest predictor of hospitalization for those who tested positive for COVID-19, and heart disease, age, and autoimmune disease were each associated with ICU admission [60]. In an outbreak in Texas, USA from July to August 2021, the proportion of cases hospitalized was 8% for those unvaccinated (3/39) and 1% of those fully vaccinated (1/129) [35].

COVID-19 deaths and mortality

Population mortality rates. Surveillance and health administrative data show high COVID-19 mortality rates, i.e. deaths from COVID-19 per population (not per case), and large numbers of COVID-19 deaths in prisons. By July 2020, 32 out of 50 US state departments of corrections had reported at least one COVID-19 related death [70], and for those states, 10 reported little to no difference in mortality between the prison and general populations, 6 reported a substantially higher mortality rate in the general population, and 3 reported a slightly higher mortality rate in the prison population [70]. In Texas prisons, 120 people died from COVID-19 by July 15th, 2020, for a COVID-19 crude mortality rate (CMR) of 79.4 per 100,000, compared with 11.8 per 100,000 for the general population [70]. Data from Brazil between December 2019 and September 2020 showed the COVID-19 CMR was 14.5 per 100,000 for people in prison compared with 60.4 per 100,000 for the whole population [44], with substantial variation in COVID-19 mortality rates for people in prison across states.

Four US studies compared COVID-19 mortality rates for people in prison compared with people in the general population after controlling for age and sex [36, 39, 40, 70], as shown in Table 3.

Table 3

Estimates of COVID-19 mortality for people in prison relative to the general population, controlling for age and sex, in studies included in a systematic review on the health impacts of the COVID-19 pandemic on people who experience imprisonment.

Indicator	Location	Period and demographic group		Point estimate
COVID-19 standardized mortality ratio	US prisons [70]	April 25, 2020 week		1.2
		May 2, 2020 week		1.6
		May 9, 2020 week		2.0
		May 16, 2020 week		2.1
		May 23, 2020 week		2.2
		May 30, 2020 week		2.3
		June 6, 2020 week		2.4
		June 13, 2020 week		2.4
		June 20, 2020 week		2.5
		June 27, 2020 week		2.6
		July 4, 2020 week		2.7
		July 11, 2020 week		2.8
	US federal prisons [40]	February to September 2020	All	2.6
			Males	2.5
			Females	4.6
	Texas carceral settings [36]	April 2020-March 2021	Black	25.4
			Hispanic	33.2
			White	16.4
COVID-19 relative mortality rate	US federal and state prisons [39]	March-June 2020		3.0

aData for the same prisons and time period may have been included in multiple studies.

Case fatality. Several US studies reported the proportion of cases that died: 0.5% (3/568) in New York City jails from March to April 2020 [59], 1.1% (7/628) in a jail in Illinois from March to April 2020 [50], 0.6% (7/1,240) in prisons and jails in Connecticut from March to June 2020 [60], ranging from 0% to 0.9% based on surveillance data for prison systems from March to July 2020 [38], 0.5% in state prisons in California from March to October 2020 [32], 0% of 382 cases in Idaho, USA between July and November 2020 [47], and 0.7% in a Utah outbreak from September 2020 to January 2021 [55]. A European study reported 0 deaths of 26 cases (0%) in Belgium between March and June 2020, 1 death of 119 cases (0.8%) in France by mid-May 2020, and 4 deaths in 315 cases (1.3%) in Italy between March 9th and May 4th, 2020 [67]. In Quebec, Ontario and British Columbia, Canada, the case fatality ratio in federal prisons was higher than that of the general population between March and May 2020: 0.5% compared with 0.3% [43].

In Michigan, USA, in-hospital all-cause mortality was higher in incarcerated individuals with COVID-19 (29.6%) compared with people with COVID-19 in the general population (20.1%); after controlling for age, sex, obesity, and comorbidities, the adjusted odds ratios for those incarcerated compared to those not incarcerated was 2.3 for in-hospital mortality and 2.0 for 30-day mortality [69].

Factors associated with death. Two additional studies presented data on factors associated with COVID-19 mortality. Data for people in prisons and jails in Connecticut, USA found that older age was a risk factor for death, whereas chronic conditions, other demographic characteristics, and facility type were not [60]. A separate California, USA study identified comorbidities, BMI over 40 and age 65 years or older as mortality risk factors [32].

COVID-19 prevention strategies

Primary prevention

Vaccination intentions and rates. One study explored vaccination intentions. In a study of 5,110 people in correctional facilities in Washington, California, Florida and Texas, USA, from September to December, 2020, 44.9% said they would receive a vaccination, 9.7% said they would hesitate, and 45.4% said they would refuse [71]. There was similar willingness to vaccinate reported among men and women, greater willingness for people who were older, and lower willingness for Black people.

Three studies reported vaccination rates. In a correctional facility in Rhode Island, USA, vaccination was offered for 6 weeks between December 2020 and February 2021, and 76.4% of incarcerated individuals received the vaccine; 90.9% of the first group offered the vaccine (age >65 years old, immunocompromised, or age >55 with comorbidities) received the vaccine, and by 4 months later, 77.7% of the incarcerated population was fully vaccinated [72]. There were no significant adverse events reported after vaccination [72]. In California, USA, by March 2021, 49% of people that met inclusion criteria in prisons had received at least one dose of the vaccine, and 22% had received 2 doses [33]. By April 2021, of the total population incarcerated in US federal prisons, 44.8% had received at least one vaccine dose, 29.9% had been fully vaccinated, 69.8% had been offered vaccination, and 64.2% of those offered vaccination had accepted [34]. For 2,514 people who had signed a vaccine declination form and were offered vaccination a second time, 1,415 accepted on the second offer [34].

In three studies in the USA, factors associated with higher vaccination were a history of smoking and Black, non-Hispanic race/ethnicity in Texas [35]; age, female sex, non-Hispanic White compared with non-Hispanic Black or Asian race/ethnicity, being born outside of the USA or with an unknown country of birth, number of medical conditions associated with severe COVID-19 illness, as well as institution type for people in federal prisons [34]; and older age, medical vulnerability, and Hispanic or White race or ethnic group compared with Black race or ethnic group in people in California state prisons [33].

Hygiene. In an Italian prison, people reported an inability to adhere to public health guidance between March and May 2020 [73].

“Here, in prison, it’s not easy to maintain social distances, to have the right masks and to sanitize everything."

Quarantine and isolation. Quarantine data for 9 US states show a wide range in the maximum quarantine rates from April 2020 to January 2021, from 36.3 per 1,000 people (n = 970) in Indiana to 843 per 1,000 people (n = 40,827) in Ohio [74].

A study of people at the time of release from a correctional facility in Arizona, USA in April and May 2020 found that release of cases to a medical recovery site for isolation substantially decreased the number of secondary infections in the community; from 7 released cases leading to 6 secondary cases associated with 4 hospitalizations to 12 released cases leading to 0 secondary infections after implementation [75].

Secondary prevention

Testing. Eight studies reported on COVID-19 testing rates. In federal prisons, state prisons, and jails across the USA from April to May 2020, 16,392 incarcerated people were offered testing, representing a range of 2.3% to 99.6% across facilities (median 54.9%) [45]. Testing rates similarly varied widely in a study of 53 US prison systems from March to July 2020, ranging from 6 tests administered per 1,000 incarcerated people in Hawaii to 1,530 tests administered per 1,000 incarcerated residents in Minnesota; in most states, testing rates were higher for people in prison compared with the general population [38]. A study of people in US federal prisons also found higher testing rates in prisons compared with the general population; as of September 2020, 50.3% of people in US federal prisons had been tested, compared to 32.5% of the U.S. population (assuming 1 test per person) [40]. In contrast, from March to April 2020, 64% of federal prisons in Canada (32/50) had lower testing rates than the general population [43]. This was also the case for a jail in Westchester County, New York, USA, in which over 20% of people were tested from March to May 2020, which was lower than New York State’s testing rates of 25% in the general population [76]. In New York City, USA jails, 15% (978/6,311) of incarcerated residents were tested from March to April 2020 [59]. A Massachusetts, USA study found that by July 2020, 8,455 tests had been administered in state prisons with 7,735 people in custody, while 1,843 tests were administered in county jails with 7,252 people in custody [41]. In California, USA, state prisons, 81.8% of 96,440 people had been tested for COVID-19 by October 2020 [32]. A separate study in California state prisons identified that testing rates varied by vaccination status and were lower for people who were unvaccinated: in January 2021, the testing rate was 933 tests per 10,000 person-days for unvaccinated people, 1,167 tests per 10,000 person-days for partially vaccinated people, and 2,018 per 10,000 person-days for those fully vaccinated [77].

Other impacts of the COVID-19 pandemic on health status

All-cause mortality

Three US studies examined changes in mortality during the COVID-19 pandemic. The Departments of Corrections for Delaware, Michigan, New Jersey, and Ohio reported COVID-19 deaths in the first 6.5 months of 2020 in excess of 50% of deaths from all causes for the most recent year for which mortality data were available, which was 2016 [70]. The CMR was significantly higher in the Florida state prison population in 2020 compared with 2019, and life expectancy was significantly lower in 2020 than in any of the other year over the study period (2015–2019), with a life expectancy drop of 4.1 years between 2019 and 2020 for this population [37]. The all-cause mortality rate for incarcerated individuals in Texas increased by 85% from April 1st, 2019 to March 31st, 2021, with a 126% increase in the Black population, a 107% increase in the Hispanic population, and a 52% increase in the White population [36].

Changes to services

Seven studies described changes to services and programs. Five studies identified reduced access to health services during the pandemic for people who experience imprisonment, with one participant in a study of incarcerated males in Scotland describing health services as “at a minimum” [29]. For example, a study of transgender women and non-binary individuals in men’s prisons in the UK found that during the pandemic, access to health care worsened as staff cared for COVID-19 cases [78].

“[N]o dentist, no opticians, no diabetic clinic, no asthma clinic, and only essential doctor appointments[.]”

In a men’s prison in Oregon, USA, 71% of incarcerated individuals reported that COVID-19 prevented them from receiving programming, which included the suspension of dialectical behavioural therapy [79]. Similarly, a study of 16 carceral systems offering opioid agonist treatment (OAT) programs across the USA found that COVID-19 resulted in the downsizing of operations: of the 16 systems surveyed, 10 reduced the scale of their OAT programs, 7 changed their medication dispensation process, which included limiting the frequency of assessments, and 1 discontinued follow-up appointments upon release [80]. A study of blog posts from people in prison in San Vittore, Italy identified a suspension of substance use disorder treatment: “This epidemic suspends the opportunity we had to treat ourselves” [73]. A study of correctional facilities in Ireland for men and women highlighted a lack of treatment for substance withdrawal [81].

"Was coming off alcohol in custody. No treatment for alcohol withdrawal Started me on librium in police station. [I] didn’t get them in here. [I] didn’t receive any support from healthcare—withdrawal off alcohol."

One study reported a change in health care delivery during the pandemic. A telemedicine program was implemented in North Carolina, USA prisons for specialty care, and 94.0% of patients (453/482) reported a positive overall telemedicine experience [82].

Three studies described a loss of services outside of health care. In an Oregon, USA prison, Getting Out by Going In, a leadership program, Step-Up, a skills training program, and General Education Development (GED) classes were all suspended [79]. The suspension of education programs was also identified in studies in two UK prisons [30, 78].

Impacts on relationships with family and staff

Four studies described the impact of the pandemic on incarcerated individuals’ relationships with their families. In a men’s prison in Scotland, one study participant expressed his concern about how the lockdown impacted his relationship with his family [31].

“Personally, am incredibly worried about progression so I can be back with my family supporting them at such hard times.”

Incarcerated men in Scotland noted that suspension of visitation had deteriorated relationships with family. One person stated that the lack of visitation resulted in “a breakdown in [his] family life” [30], and another said that the lockdown was “tearing families apart” [29]. One participant highlighted the use of phone calls as a means to connect with family during the suspension of visitation [79].

“They cut off visiting for obvious reasons. It’s cut off til further notice, but that’s why Telmate [a private company providing inmate phone calls] has been giving us free calls so we can still stay connected with our families because family is a big part of DOC and keeping sane.”

Participants in a study of men’s and women’s correctional facilities in Ireland similarly cited the value of phone calls and virtual visits with family for mental health [81].

Phone calls really helped to take my mind off thing[s], I spoke with my mum and felt much better, I got to sleep that night.

Even phone access could be problematic, however; one participant in the study in Scotland described his reluctance to use the phone due to concerns about lack of phone sanitation and minimal access to hand sanitizer [31].

Four studies commented on the effects of COVID-19 on relationships between people in custody and staff. One study noted that the pandemic strained relationships between incarcerated individuals and staff, with study participants describing heightened tensions and an increase in conflict [31]. Frustrations associated with the pandemic were expressed as “scuffles with prisoners and staff” and “unnecessary violence” [31]. One participant described the feeling of being a burden to staff, while another felt that staff were more detached from their roles since the start of the COVID-19 pandemic, for example with staff not responding to cell buzzers when individuals required assistance [31]:

"Too many officers sit at the desk. No social distancing that equals to what’s forced on prisoners. Buzzers are being pressed and on more than one occasion it’s taken 30 minutes plus to answer, one was 90 minutes until more people started kicking doors on various landings."

A study in UK prisons found that relationships between incarcerated individuals and staff had suffered partly because incarcerated individuals blamed correctional staff for COVID-19 restrictions which they perceived as “harsh” and “an additional punishment” [78].

In contrast, three studies found that during the pandemic, staff were increasingly supportive of incarcerated individuals. One participant in Northern Ireland stated that “[isolation] was absolutely terrible, but the staff were lovely. Really helpful was staff” [81]. In a study in Italy, a participant expressed appreciation for the ongoing work of prison workers [73].

"Every day our counsellors show up on time and we continue treatment, even if it is reduced. We always see them smiling at us, trying to minimize the problem, giving us hope and bringing us news. They are always present to listen to our problems…"

Similarly, in a study from British Columbia, Canada, peer health mentors supporting people leaving prison reported more compassion from correctional staff towards incarcerated individuals [83].

“They’re not about punishment right now, and they’re really about care and understanding and, you know, really worried about their health.”

Impacts on mental health

Several studies explored the impact of the pandemic on mental health. A person in a UK prison described a “worsening situation, feelings of isolation, and the increased weight of time in prison” [31]. People incarcerated in an Italian prison from March to May 2020 described worry, psychological pain, and fear [73]. Isolation emerged as a common theme across studies: “I stay in my cell 24 hours a day, except for those minutes dedicated to phone calls with my loved ones” [73], with study participants describing the impacts of isolation on mental health, exacerbating “feelings of boredom, frustration and stress” as well as suffering [29, 30, 73].

“There has been a rise in mental health issues during the lockdown due to the amount of time spent in isolation. There has even been a suicide in [name of prison].” 67

Many incarcerated individuals described how isolation exacerbated depression, anxiety, thoughts of self-harm, and suicidal ideation [78, 81].

“My depression is coming back. It was really hard not speaking to anyone either on landing or family. I had suicidal thoughts and self-harmed but didn’t tell anyone.”

“Struggling, have self harm on a number of occasions in secret […] its lonely and depressing especially as I have no one outside prison supporting me anyhow, so now completely isolated so increases my suicide and self-harm thoughts”

In another study, a participant expressed feeling grateful about being in solitary confinement, as a strategy to prevent transmission of COVID-19 [79].

“I definitely think, I mean, this is the one time in my life I’ve been grateful to be in isolation.”

Another common theme was anxiety surrounding infection risk, with study participants particularly concerned about infection transmission from staff, given that they had limited contact with the outside community otherwise [31].

“Let’s hope it doesn’t get into prison, god help us if it does!!"

In an Oregon, USA prison, study participants had varying levels of concern about infection risk, with 74.2% of participants reporting that they were “either not at all or somewhat worried” that their institution would become infected with COVID-19, while 25.8% were “either pretty or extremely worried” [79]. Many participants indicated that they were not worried about a COVID-19 outbreak in their facility because they had no control over the disease’s spread [79].

Three studies identified participants’ desire for greater transparency from the prison system regarding the pandemic. In one study, participants described anxiety surrounding a lack of communication from prison officials [31]. Participants reported that they were not provided information about the outside community, while the information they were able to access was outdated [31]. In another study, participants described feeling that newsletters from prison officials were not adequate, as the information within was sparse and not tailored to their specific unit [79]. A study in a Northern Ireland prison found that 71% of respondents felt they had not received the appropriate information, which further added to anxiety [81].

“There was an expectation that we should just know things, nothing was communicated to me. Mental [h]ealth has been up and down, feel a panic attack coming on.”

A study of peer health mentors supporting people leaving prison found that COVID-19 had also intensified anxiety around release from custody [83]. Peer health mentors reported that incarcerated individuals were reaching out for increased emotional support [83].

[They’re] grateful that we’re still here and we’re still answering our phone. And a lot of them just reach out just to talk. ‘Cause there’s so much unknown out here. And they know everything’s shutdown and they’re, like, ‘what’s going to happen to me?’

Finally, a study of reported self-harm events in a people in a Swiss prison from 2016 to 2020 found that comparing 2016–2019 to 2020, there was a 57% increase in severe suicide attempts from 4.4/100 people in detention/year to 7.0/100 people in detention/year, and a 57% increase in other self-harm events from 7.9/100 people in detention/year to 12.5/100 people in detention/year [84].

Discussion

This systematic review identified 62 studies on the health impacts of the COVID-19 pandemic on people who experience imprisonment. Key findings are that incarcerated populations have experienced disproportionately high rates of COVID-19 infection, COVID-19 mortality, and all cause-mortality. The COVID-19 pandemic was associated with worse access to many health and non-health services, increased isolation and feelings of powerlessness, and negative impacts on relationships with family and on mental wellbeing, as well as varied impacts on relationships with staff. There is a lack of data on the health impacts of the pandemic beyond the risk and short-term outcomes of COVID-19 infection itself, and in particular, a lack of data on the experiences of people who experience imprisonment globally and a lack of quantitative data for people outside of the USA.

Regarding limitations of individual studies, most studies were at risk of bias in one or more domains examined. Common limitations were the lack of data on the population included and on the setting, the lack of comparator population data, the lack of stratification or adjustment for important risk factors (such as age) for outcomes such as incidence, hospitalization, and death, the lack of denominator data to be able to understand rates in addition to counts, and a lack of clarity regarding methods used to identify outcomes of interest.

There are two main limitations to this review. First, we limited our search to published peer-reviewed studies and published pre-print studies. Since published peer-reviewed and published pre-print studies may have a unique influence on public policy compared with surveillance data alone, including these published studies is of value in and of itself, but that notwithstanding, future work could include searches of additional grey literature to enhance the data identified. Second, we updated our search in October 2021, which was over 18 months after the World Health Organization declared SARS-CoV-2 a pandemic on March 13th, 2020. We appreciate that in the context of a global emergency such as this pandemic, additional scholarly work will emerge in the coming months and years, and that there is often a substantial lag from the time of the completion of research to publication. This review serves to summarize the literature to date, which is important to inform ongoing pandemic response, recovery planning and implementation, and future emergency preparedness.

The variation in findings across studies for certain important COVID-19 outcomes is striking. For example, percent positivity in outbreaks varied between 2.5% and 90.6%. This variation could reflect true differences, for example differences in prison structures and policies, prevention measures, and population characteristics and behaviours, but could also reflect how the study was conducted, for example over what follow up period data were collected, who was considered at risk of infection and included in the denominator, and who accessed testing. For studies comparing COVID-19 mortality between people in prison and the general population, the magnitude of the difference in risk varied substantially across studies, and in one study even the direction of association was different. These differences may be due to differences in the populations compared, for example the study that found that people in prison had lower risk of COVID-19 mortality compared with the general population did not control for age [44], or may be due to circumstances at specific points in time in prison settings and in the community, which could be attributable to prevention strategies, epidemiology, or chance.

While this systematic review was designed to be descriptive rather than explanatory, additional work would be valuable to quantitatively summarize aspects of the burden of disease of COVID-19, including evidence for risk factors and risk mechanisms. Further, research is needed to elucidate the impacts of specific COVID-19 response strategies for this population and setting, recognizing the unique opportunity to compare findings across person, place, and time, though this work would notably require considerable data harmonization efforts. Understanding which prevention strategies have reduced the burden of COVID-19 and other harms during the COVID-19 response is necessary to inform effective emergency planning and response, as well as to support ongoing health promotion and health care.

Given the major burden of the COVID-19 pandemic on incarcerated populations, prisons and people who experience imprisonment should be prioritized in ongoing COVID-19 response and recovery efforts [85]. This could involve implementing evidence-based strategies now to mitigate the risks of adverse health impacts of the pandemic, for example interventions to promote mental health and treat mental illness. As part of a broader public health agenda to address the harms of the carceral system [19], future emergency prevention and preparedness efforts should explicitly consider this setting and population. Additional work is needed to understand the impacts of the pandemic on the health status of this population, in both the short and long-term, particularly regarding the experiences of people in prisons including their health care access and quality.

Risk of bias for quantitative studies included in a systematic review on the health impacts of the COVID-19 pandemic on people who experience imprisonment.

(DOCX)

Click here for additional data file.

Risk of bias for qualitative studies included in a systematic review on the health impacts of the COVID-19 pandemic on people who experience imprisonment.

(DOCX)

Click here for additional data file.

Database search strategy.

(DOCX)

Click here for additional data file.

PRISMA checklist.

(DOC)

Click here for additional data file.

23 Mar 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Seth Blumberg

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that this manuscript is a systematic review or meta-analysis; our author guidelines therefore require that you use PRISMA guidance to help improve reporting quality of this type of study. Please upload copies of the completed PRISMA checklist as Supporting Information with a file name “PRISMA checklist”.

Additional Editor Comments:

Both reviewers and I found the manuscript’s focus on people who experience imprisonment to be compelling. I agree with the reviewers that the mixed-methods approach for conducting the systemic review was thoughtfully planned and well-articulated. Meanwhile, there is also general consensus that the impact of the systemic review gets watered down in the results and discussion as there was not enough interpretation of the findings. The reviewers provide excellent feedback for improving the article in this regard. I look forward to your review in which the reviewer feedback is incorporated.

Also, a few small thoughts as I read the article:

- I found the wording of lines 44-46 of the abstract awkward to interpret.

- Comparison of test positivity is a bit challenging because it depends on who gets test, and when they are test, etc. I think it would help to mention all these nuances so that the wide variability in percentages can be put into place

- Similarly, can you provide contextualization for why covid mortality results may be higher in general population for some studies and prison population for others?

- Vaccine hesitancy vs acceptance evolved as the pandemic progressed. Can that explain any of the variance in vaccination rates?

- The first paragraph of the discussion has several claims that are a bit challenging to abstract from the results. However, with improved presentation of the results, I think the link will be more evident

- The material in the second paragraph of the discussion can probably be incorporated into the results section. (i.e., items #15/#22 of the PRISMA checklist)

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: N/A

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Thank you for your work! This paper is methodologically sound, well written and framed, the work is thorough, and the studies used were clearly stated. More detailed feedback has been included within the attached review.

Reviewer #2: 1) The authors conducted a well-documented, far-reaching review of articles linked to people with imprisonment in COVID-19. The article provides high-quality, easily accessed information that will be useful to researchers working on COVID-19 in the carceral setting. I appreciated the detailed review of data extraction methods for both the quantitative and qualitative approaches, and the explanation of their move towards a narrative synthesis. I also thought that the Table format detailing the thematic areas as a review was a helpful orientation of the succeeding sections, and that the discussion gave a helpful synthesis of a large amount of information collected. Finally, I certainly agree with the authors that it helps to convey the importance of continued research on health in the carceral setting, and does a good job identifying gaps.

There are several areas where I would like clarification:

- In the introduction, a number of broad statements are made about "prisons" and various response efforts and consequences with relatively few citations. Given that the systematic review encompasses a global population, this section would benefit from further specificity both in describing where the data comes from, and the broader implications. For example: "While many jurisdictions reduced their prison population size through early releases, pardons, diversions, and release on bail or parole [12], there may not have been commensurate increases in discharge planning or community resources to support needs, including for treatment beds and shelter beds. In addition, the transition to remote services and reduced scope and hours of services may have limited access to essential health and social services [13]." How does this extend to the global prison population, given that these citations are each from Canada and from Germany?

- When distinguishing between "published" and "pre-print" literature, are the authors intended to refer to peer-reviewed literature in journals as "published," or all published literature regardless of peer review? Please make this clear in a revision.

- "If studies included data for youth in 131 detention or other subpopulations in addition to adults in prisons" on line 131; who are other subpopulations?

- For the case-fatality, could you be clear why you report some as percentages as some as fractions? Could also put this in methods.

On the whole, I think it would be helpful to clearly characterize the study as a global review of adults experiencing imprisonment. I also think for certain sections, and particularly test positivity, a long section packed with numbers that are difficult to contrast when reading, a graphic would more effectively compare and present very interesting results (and could simultaneously show country, # of people tested, percent positivity, and time point). Related to this, I think further use of comparison and contrast throughout the reporting sections would also strengthen the readability of the article.

Finally, I do have a concern with the exclusion of studies that "described only properties of 143 the virus in prisons, such as reproduction ratio"; the reproduction ratio is the product of contact rate, i.e. human behavior and interventions, susceptibility and other relevant factors, and not inherent to the virus itself. That said, I would agree with the decision to exclude studies focused exclusively on mathematical modelling and not reporting original data.

2) Not applicable, no statistical analysis

3) Data referenced is generated from the published and peer-print literature, and to my understanding all articles are cited in the bibliography and supplementary tables.

4) On the whole, the article is well-written and clearly conveys the ideas. There are a handful of grammatical errors; specific ones identified include lack of capitalization of "Google" on line 122, "in" instead of "on" on line 133, missing a "the" before "most" on line 307.

There are also a couple of where language is unclear: "The high prevalence of health conditions"; which conditions? In the section on test positivity, I would ask that please restrict the use of percentages to percent positive (e.g. "30.5% in a men’s prison in Arkansas in April and May 247 2020, in which 99.2% of 1,661 people in prison were tested") as it becomes very easy to get lost in this section.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Reviewer feedback PONE-D-22-05150.docx

Click here for additional data file.

11 Apr 2022

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

We have reviewed these documents and we think that our manuscript meets all style requirements.

2. We note that this manuscript is a systematic review or meta-analysis; our author guidelines therefore require that you use PRISMA guidance to help improve reporting quality of this type of study. Please upload copies of the completed PRISMA checklist as Supporting Information with a file name “PRISMA checklist”.

We have renamed the file as requested.

Additional Editor Comments:

Both reviewers and I found the manuscript’s focus on people who experience imprisonment to be compelling. I agree with the reviewers that the mixed-methods approach for conducting the systemic review was thoughtfully planned and well-articulated. Meanwhile, there is also general consensus that the impact of the systemic review gets watered down in the results and discussion as there was not enough interpretation of the findings. The reviewers provide excellent feedback for improving the article in this regard. I look forward to your review in which the reviewer feedback is incorporated.

Thank you very much. We have revised the article based on the helpful feedback from reviewers and we think the manuscript is much improved.

Also, a few small thoughts as I read the article:

-I found the wording of lines 44-46 of the abstract awkward to interpret.

We appreciate this comment and we have revised the language.

-Comparison of test positivity is a bit challenging because it depends on who gets test, and when they are test, etc. I think it would help to mention all these nuances so that the wide variability in percentages can be put into place.

We have provided additional detail in the Results and added information regarding this important issue in the fourth paragraph of the Discussion.

-Similarly, can you provide contextualization for why covid mortality results may be higher in general population for some studies and prison population for others?

We have revised relevant content in the Methods section to present data separately when controlling for age and sex, and we discussed this specific finding in the fourth paragraph of the Discussion.

-Vaccine hesitancy vs acceptance evolved as the pandemic progressed. Can that explain any of the variance in vaccination rates?

We think this is unlikely, given that the periods for the three studies that reported vaccination rates were similar: December 2020-February 2021, until March 2021, and until April 2021.

We have added a comment in the Synthesis section of the Methods to note that we reported on study period as well as other factors consistently, to support interpretation of data that may vary over time.

-The first paragraph of the discussion has several claims that are a bit challenging to abstract from the results. However, with improved presentation of the results, I think the link will be more evident.

We think that the revisions have made the findings more clear.

-The material in the second paragraph of the discussion can probably be incorporated into the results section. (i.e., items #15/#22 of the PRISMA checklist)

As we did not conduct a meta-analysis, we do not have information relevant to PRISMA Items 15 and 22, i.e. assessing for publication bias or selective reporting bias (as per https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1000100).

In the second paragraph of the Discussion, we have summarized the quality of studies at a very high level (i.e. to summarize what was presented in more detail in the Results section including relevant tables), and elaborated on common limitations, which we hope will inform interpretation as well as future work.

Reviewers’ comments:

Reviewer #2:

1) The authors conducted a well-documented, far-reaching review of articles linked to people with imprisonment in COVID-19. The article provides high-quality, easily accessed information that will be useful to researchers working on COVID-19 in the carceral setting. I appreciated the detailed review of data extraction methods for both the quantitative and qualitative approaches, and the explanation of their move towards a narrative synthesis. I also thought that the Table format detailing the thematic areas as a review was a helpful orientation of the succeeding sections, and that the discussion gave a helpful synthesis of a large amount of information collected. Finally, I certainly agree with the authors that it helps to convey the importance of continued research on health in the carceral setting, and does a good job identifying gaps.

We appreciate these comments.

There are several areas where I would like clarification:

-In the introduction, a number of broad statements are made about "prisons" and various response efforts and consequences with relatively few citations. Given that the systematic review encompasses a global population, this section would benefit from further specificity both in describing where the data comes from, and the broader implications. For example: "While many jurisdictions reduced their prison population size through early releases, pardons, diversions, and release on bail or parole [12], there may not have been commensurate increases in discharge planning or community resources to support needs, including for treatment beds and shelter beds. In addition, the transition to remote services and reduced scope and hours of services may have limited access to essential health and social services [13]." How does this extend to the global prison population, given that these citations are each from Canada and from Germany?

We have revised the content in the third paragraph of the Introduction to clarify the text and provide additional references.

-When distinguishing between "published" and "pre-print" literature, are the authors intended to refer to peer-reviewed literature in journals as "published," or all published literature regardless of peer review? Please make this clear in a revision.

We have clarified this content in the Search section of the Methods, as well as in the third paragraph of the Discussion.

-"If studies included data for youth in 131 detention or other subpopulations in addition to adults in prisons" on line 131; who are other subpopulations?

We have revised the language to say “populations” rather than “subpopulations.” We specified this criterion since we anticipated (based on prior reviews we have conducted) that we might identify studies that presented data for people in prisons together with people in the community, rather than stratified data for adults in prison.

-For the case-fatality, could you be clear why you report some as percentages as some as fractions? Could also put this in methods.

We have revised the presentation of the data to consistently report case fatality ratios as percentages.

On the whole, I think it would be helpful to clearly characterize the study as a global review of adults experiencing imprisonment.

Thank you for this suggestion. We have revised content in the title from “people who experience imprisonment” to “adults who experience imprisonment globally.”

I also think for certain sections, and particularly test positivity, a long section packed with numbers that are difficult to contrast when reading, a graphic would more effectively compare and present very interesting results (and could simultaneously show country, # of people tested, percent positivity, and time point). Related to this, I think further use of comparison and contrast throughout the reporting sections would also strengthen the readability of the article.

We appreciate this suggestion, and we have added figures to summarize content in two sections of the Results (Percent positivity, as you suggested, and Proportion of cases hospitalized) and a table in the Population mortality rates section. We have also added content in the Discussion regarding the comparability of findings across studies.

Finally, I do have a concern with the exclusion of studies that "described only properties of 143 the virus in prisons, such as reproduction ratio"; the reproduction ratio is the product of contact rate, i.e. human behavior and interventions, susceptibility and other relevant factors, and not inherent to the virus itself. That said, I would agree with the decision to exclude studies focused exclusively on mathematical modelling and not reporting original data.

On reflection, we agree that it makes sense to include studies that report on reproduction ratio for people in prison using original data. We have revised the methods section, and reviewed papers for eligibility that reported on reproduction ratio. We ended up including one additional study: Puglisi et al., though we did not include the data on reproduction ratio since it included staff in prisons and our explicit focus is on people who experience imprisonment, not staff.

2) Not applicable, no statistical analysis

This is correct.

3) Data referenced is generated from the published and peer-print literature, and to my understanding all articles are cited in the bibliography and supplementary tables.

This is correct.

4) On the whole, the article is well-written and clearly conveys the ideas. There are a handful of grammatical errors; specific ones identified include lack of capitalization of "Google" on line 122, "in" instead of "on" on line 133, missing a "the" before "most" on line 307.

We have revised the text, and we apologize for these errors.

There are also a couple of where language is unclear: "The high prevalence of health conditions"; which conditions?

We have revised this text in the second paragraph of the Introduction.

In the section on test positivity, I would ask that please restrict the use of percentages to percent positive (e.g. "30.5% in a men’s prison in Arkansas in April and May 247 2020, in which 99.2% of 1,661 people in prison were tested") as it becomes very easy to get lost in this section.

As noted above, we have substantially revised this section, and the data are now presented as percentages consistently, including in Figure 2.

Reviewer #1:

Summary

Thank you for taking the time to conduct a thoughtful and careful review of the current published/pre-printed literature on COVID-19’s impact on those who have been incarcerated; this is crucial and important work! The methodological approach was transparent and sound, and I applaud the authors for their diligent work. Because the purpose of this article is a review, I believe that there are substantial revisions that are necessary for this to sufficiently serve that purpose to readers. It’s clear that there was a lot of data extraction done, and this is fantastic, but there was minimal synthesis, contextualization, and organization of these findings, which is crucial for providing a thorough and manageable review of literature. While suggested revisions (particularly for the data synthesis and discussion sections) are considerable, much of the content is already in the paper, but could use some reorganizing, reframing, or a deeper synthesis. As you aptly highlighted, there are more studies that will be coming out and more work in this area that will need to be done, and therefore, providing more clear recommendations based on the large breadth of information reviewed (taking into account the sampled studies’ limitations,) would be a wonderful addition to help drive home the value of all this great work. More detailed and specific comments are provided below – thank you for allowing me the opportunity to read your paper, and I hope these comments prove useful to you!

We appreciate these comments and have addressed specific comments as detailed subsequently.

Introduction – Comments

•This section was compelling, concise, well-framed! Below are just a few comments on some areas to further strengthen this section:

o I think some clarification around the risk of COVID-19 being introduced into prisons being high could be useful here. The criminal justice system (including courts and prisons) altered its practices considerably throughout the pandemic, and so acknowledging that even in the presence of substantial changes and policy shifts (e.g., prisons massively changing how/when people moved, implementing lockdowns, etc.), it was still impossible to effectively halt the introduction of COVID-19 into prisons.

We appreciate this comment, but we think this type of statement would be out of scope for this review and isn’t directly supported by what we found.

o Clarifying point about addressing increases in drug supply is advised, (e.g., that this is specific to individuals upon release.) If you want to address use in prison more explicitly, suggesting the potential for an increase in relapses (or new use) would be warranted and follows from the statement about worsened mental health for those who have been incarcerated during the pandemic.

We have added content to this section in the third paragraph of the Introduction to clarify that changes in the illicit drug supply may impact people both while in prison and in the community post-release.

Eligibility Criteria – Comments

•This section was thorough and clear, and your methodological approach was sound! Your clarity and transparency in how use language is especially appreciated! I want to especially thank you for bringing up the point about policies not necessarily aligning with how prisons are run in practice! Below are just a few comments on some areas to further strengthen this section:

o Just highlighting a really small typo (nothing major): “The first criterion was a focus adults…”

We have fixed this mistake.

Characteristics of Included Studies – Comments

•Table 1 is incredibly helpful and a great addition to this paper! Below are just a few comments on some areas to further strengthen this section:

o While this is only a very minor point, if the table was created in such a way that allows you to easily resort it, that might be very helpful for readers to understand the distribution of studies in your overall sample more easily. I think either resorting based on either time or study location could be particularly helpful.

We have revised the formatting of the table so the studies are now presented by country.

Risk of bias – Comments

•Thank you for your transparency in addressing the quality of the studies included! Elaborating on the domains where quality indicators were not met would be helpful (e.g., do they all tend to be different, or are there patterns amongst which indicators are not met?) Knowing this information would be helpful for informing the interpretation of results.

We have discussed common limitations in the second paragraph of the Discussion.

Data Synthesis – Burden of COVID infection – Comments

•This section is very thorough! However, I found myself getting very bogged down by a lot of numbers and felt that synthesis, organization, and interpretation of findings was lacking. I also found myself having trouble with interpretation of these figures given the information about study quality (i.e., might the bias be impacting reported figures, and if so, do we expect these numbers to be under/over-estimates? Or maybe not representative?) Below are some specific comments on potential routes to help strengthen this section:

o There are a lot of different numbers for incidence rates/percent positivity presented in this section, which is excellent, but I found myself getting a bit bogged down in the details. Graphs here I think could be particularly helpful and given the fact that the number of figures/tables are not limited, this is strongly advised for helping readers be able to make sense of all of these figures. Then, the written sections can focus more heavily on synthesis. For example, instead of listing something like “Study A found Incident Rate A in Time A and Location A,” starting with something like, “The range in cumulative incidence rates ranged from x to y, with the lowest being in (Time B/Location C, etc.)”

We appreciate this suggestion. As noted above, we have shifted data from the text into two figures and a table, and in several places we have added content to summarize ranges. We think that these revisions have substantially improved the paper.

o For the figures, even if it’s just a bar graph, ordered by date (when available) with location-specific details included (and patterns/colors associated with location potentially,) with both prison-specific and overall population-specific side-by-side bars. Then the text can focus on describing how much higher these figures were in-prison, compared to respective populations, over time/space. That or a table could be helpful!

As above, we have made changes as suggested.

o As mentioned above, quality of studies needs to be explicitly woven into interpretation so that we can more accurately and appropriately interpret the figures reported from these studies. I do not think this necessarily means calling out the studies where quality indicators were not met.

We have included content on study size and specified study period, location, and facility type (when applicable) to support interpretation of the internal and external validity of the studies.

Discussion – Comments

•I appreciated the transparency about how early this study is, as this was important to acknowledge, and is helpful! In this section, I was really hoping for a lot more in terms of synthesis, recommendations, and limitations. Below are a few more specific suggestions that may help support this section:

o Limitations: I was hoping to see more discussion of what seemed to be some geographic-specific differences in the types of studies conducted. For example, the U.S. seemed to focus a lot more on the epidemiological data (e.g., incidence rates, percent positivity, etc.), while data outside of the U.S. – namely Europe – seemed to be the primary source of qualitative data. This is crucial, and likely reflects considerable differences in penological culture.

We have added a comment in the Characteristics of included studies section that only 1 qualitative study was conducted in the USA. We also added a comment in the first paragraph of the Discussion to specify that there is a lack of data on the experiences of people globally and a lack of quantitative data for people outside of the USA.

o I would love to see more on recommendations! You (rightly!) suggest individuals who are incarcerated need to be prioritized in handling COVID-19 and other public health emergencies. How do we accomplish this? What future studies need to be conducted, based off of what you found? Did it seem that some places were doing better than others in terms of reducing the burden of COVID-19 on its incarcerated populations? Even just highlighting this is important – more work can be done in future studies that could illuminate why this might have been the case. Especially in regards to mental health of residents in prisons – what can be done to support these individuals while still keeping their risk of infection as low as possible (e.g., increasing access to tablets/phone calls, etc.)?

We have added further content in the Discussion, e.g. regarding the need for research to quantitatively summarize the COVID-19 burden of disease and the effectiveness of specific interventions. Based on the nature of the synthesis we conducted as well as the lack of specific data and the lack of comparable data on many outcomes, we don’t think we have sufficient information to be able to comment on whether some places did better than others. We have acknowledged in the final paragraph of the Discussion the importance of implementing interventions now that could mitigate the adverse health impacts of the COVID-19 pandemic in this population, and specifically noted the need to implement strategies to promote mental health and treat mental illness.

1 May 2022

4 May 2022

See attached Response to Reviewers.

Submitted filename: Response to reviewers_May 4 2022.docx

Click here for additional data file.

10 May 2022

The health impacts of the COVID-19 pandemic on adults who experience imprisonment globally: A mixed methods systematic review

PONE-D-22-05150R2

Dear Dr. Kouyoumdjian,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Seth Blumberg

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Thank you for your thoughtful incorporation of reviewer feedback. Congratulations on a job well done!

Reviewers' comments:

N/A

12 May 2022

PONE-D-22-05150R2

The health impacts of the COVID-19 pandemic on adults who experience imprisonment globally: A mixed methods systematic review

Dear Dr. Kouyoumdjian:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Seth Blumberg

Academic Editor

PLOS ONE