JAMAIOpen,.

Network

Original Investigation | Public Health

Extreme Heat and Suicide Watch Incidents Among Incarcerated Men
David H. Cloud, PhD, JD; Brie Williams, MD, MS; Regine Haardörfer, PhD; Lauren Brinkley-Rubinstein, PhD; Hannah L. F. Cooper, ScD

Abstract

Key Points

IMPORTANCE Extreme heat poses a distinct risk to the 2.1 million incarcerated people in the United
States, who have disparately high rates of behavioral health conditions. Suicide is a leading cause of
death among people in prisons.

Question What is the association
between exposures to extreme heat and
suicide-watch incidents in a state prison
system without air-conditioned

OBJECTIVE To examine associations of extreme heat, solitary confinement, and an indicator of

living units?
Findings This case series of 6576

suicidal behaviors among incarcerated men in a Deep South US prison system.

facility-incarceration days found that

DESIGN, SETTING, AND PARTICIPANTS This longitudinal case series panel study included adult

extreme heat was significantly

men in prisons in Louisiana, a state with one of the largest prison systems in the United States that

associated with a 30% increase in the

has been engaged in litigation due to lack of air conditioning and extreme heat. The unit of analysis

incident rate of daily suicide-watch

was prison facility-days. A facility-level data set was created by merging administrative data files,

incidents.

which included demographic characteristics, health classification, housing location and movement,
disciplinary records, and involvement in suicide-watch incidents for all incarcerated men in Louisiana
during the observation period. Individual-level variables were aggregated to facility-days to merge
in daily maximum heat index data from the US Local Climatological Data, which were linked to the zip
codes of prisons. The observation period was January 1, 2015, to December 31, 2017. Data set
construction occurred from August 2020 to September 2022, and analysis was conducted from
December 2022 to February 2023.

Meaning These findings suggest that
extreme heat may increase
vulnerabilities to situations that lead to
suicide-watch placements for
incarcerated people, bolstering calls for
heat mitigation and decarceral
interventions to assuage heat-induced
harms among incarcerated populations.

EXPOSURE The focal exposure was extreme heat days. Daily maximum heat index data were
categorized into 6 bins (<30 °F, 30-39 °F, 40-49 °F, 50-59 °F, 70-79 °F, and ⱖ80 °F) and as an
indicator for any facility-day where the maximum heat index exceeded the 90th percentile of heat
indices for total days in observation period. Conditional fixed-effects negative binomial regression
models were used to calculate incident rate ratios to test associations between extreme heat and

+ Supplemental content
Author affiliations and article information are
listed at the end of this article.

suicide watch incidents, while controlling for covariates.
MAIN OUTCOMES AND MEASURES The focal outcome was daily count of suicide watch incidents
that were recorded in a carceral system database. Covariates included daily percentages of
incarcerated persons at each prison with serious mental illness diagnosis, daily rate of solitary
confinement, and total facility population.
RESULTS The sample of 6 state-operated prisons provided 6576 facility-days for the analysis.
Results suggest a dose-responsive association between extreme heat and daily counts of suicidewatch incidents; compared with days with temperatures between 60 and 69 °F, the rate of daily
suicide incidents increased by 29% when the heat index reached the level of caution (ie, 80-89 °F)
and by 36% when reaching extreme caution (90-103 °F) (80-89 °F: incidence rate ratio [IRR], 1.29;
95% CI, 1.17-1.43; P < .001; 90-103 °F: IRR, 1.36; 95% CI, 1.15-1.61; P < .001). Compared with other
days, those with the extreme heat indicator were significantly associated with a 30% increase in the
incident rate of daily suicide-watch incidents (IRR, 1.30; 95% CI, 1.18-1.45; P < .001).
(continued)

Open Access. This is an open access article distributed under the terms of the CC-BY License.
JAMA Network Open. 2023;6(8):e2328380. doi:10.1001/jamanetworkopen.2023.28380 (Reprinted)

Downloaded from jamanetwork.com by guest on 12/27/2023

August 11, 2023

1/11

Downloaded from jamanetwork.com by guest on 12/27/2023

JAMA Network Open. 2023;6(8):e2328380. doi:10.1001/jamanetworkopen.2023.28380 (Reprinted)

incarcerated adult men in a Deep South US prison system.

explored the association between extreme heat, solitary confinement, and suicidal behaviors among

extreme heat on suicidality and its psychological antecedents among people in prison. Therefore, we

extreme isolation and deprivation.27 However, no studies have examined the potential influence of

solitary confinement, can increase vulnerabilities to self-injury and suicide due to exposure to

overrepresented in carceral settings.23-26 Living conditions in jails and prisons, including exposures to

People with serious mental illnesses are at increased risk of self-harm and suicide and are also

use; mood disorders; schizophrenia and delusional disorders; and nonsuicidal self-harm.15-22

escalations in community hospitalization rates for behavioral health symptoms, including substance

mental health issues.12-14 Additionally, studies have linked heat waves and rising temperatures to

trigger or exacerbate feelings of lethargy, irritability, and sadness, especially for people with existing

health symptoms by altering the body’s ability to thermoregulate and regulate emotions.12 This may

temperatures and incidence of suicide.9-11 At a biophysical level, heat stress may worsen mental

the climate crisis, recent studies have reported positive associations between higher ambient

Research has long demonstrated that suicides tend to increase in hotter seasons. In the wake of

violence per year.”8

violent incidents, and that “unmitigated exposure to heat generates an additional 44 cases of intense

in Mississippi found that “intensely hot days” were associated with a 20% increased risk of severe

deaths may be attributable to extreme heat in the state’s non–air-conditioned prisons. Another study

associated with a 15.1% increased all-cause mortality risk and estimated that approximately 13% of

among incarcerated persons. A study in Texas prisons2 found that an extreme heat day was

2 prior studies that we are aware of have explored associations between extreme heat and health

related stress than those in general population units of prisons or in community settings.1,4,7 Yet, only

susceptible to the hazards of extreme heat because they are less able to avoid or mitigate heat-

As recounted in litigation and commentaries, people in solitary confinement are especially

a public health and human rights crisis.5,6

approximately 22 hours per day, which is defined by international bodies as torture and recognized as

extreme isolation (known as solitary confinement), generally defined as being confined in a cell for

physiological and psychological stress of heat exposures.1,4 The flip side of extreme overcrowding is

into poorly ventilated dormitories or small cells (single or double-bunked), which can intensify the

Overcrowding is rampant in the US carceral system, with hundreds or thousands of people cramped

circulation, which create conditions for indoor temperatures that exceed those outdoors.1,4

as stone, metal, and concrete, that retain heat and have small or closed windows that impede air

constructed to endure rising temperatures. Carceral structures are mostly built with materials, such

have disparately high rates of behavioral health conditions.1-3 Few jails and prisons have been

Extreme heat poses a distinct risk to the 2.1 million incarcerated people in the United States, who

Introduction

JAMA Network Open. 2023;6(8):e2328380. doi:10.1001/jamanetworkopen.2023.28380

carceral contexts.

for legal interventions and advocacy seeking to abate heat-induced morbidity and mortality in

linkages between climatological events and health outcomes in prisons, and may have implications

indicator of suicidality among an incarcerated sample, contribute to an emerging literature exploring

August 11, 2023

2/11

Extreme Heat and Suicide Watch Incidents Among Incarcerated Men

CONCLUSIONS AND RELEVANCE Findings suggest an association between extreme heat and an

Abstract (continued)

JAMA Network Open | Public Health

c..G

Downloaded from jamanetwork.com by guest on 12/27/2023

JAMA Network Open. 2023;6(8):e2328380. doi:10.1001/jamanetworkopen.2023.28380 (Reprinted)

index, dew point temperature, relative humidity, degree days (heating and cooling), and daily

resource that tracks hourly, daily, and monthly maximum, minimum, and average temperature, heat

Heat-related data were downloaded from the US Local Climatological Data (LCD), a publicly available

Focal Independent Variable: Extreme Heat Days

facility) of each suicide-watch incident for each day between January 1, 2015, and December 31, 2017.

were obtained at the individual level, which allowed us to determine the date and location (ie,

potential suicide risk and notifies a supervisor, and a person is placed under observation. These data

placed on suicide watch. A suicide watch occurs when a staff member believes that a person is a

facility level. These incidents were recorded in CAJUN by correctional staff any time a person was

The focal dependent variable was daily count of newly initiated suicide-watch incidents at the prison-

Focal Dependent Variable

the subsequent sections.

Louisiana by aggregating individual-level data to facility-level indicators described in more detail in

Vera Institute of Justice. We created a panel of daily data for each of the 6 state-operated prisons in

December 2017. Access to this data set was obtained through a data-sharing agreement with the

and solitary confinement exposures for all persons in Louisiana prisons between January 2015 and

demographic characteristics, sentencing, mental health, disciplinary records, housing assignments,

information on all persons sentenced to imprisonment in Louisiana. These files included

Unified Network (CAJUN), which is the LCDR’s administrative database that tracks individual-level

Segregation Initiative (SAS-I). These data included raw data files from the Criminal and Justice

produce a report on solitary confinement practices in Louisiana for the Safe Alternatives to

These data were sourced from the Vera Institute of Justice (Vera) and were previously used to

Creating the Analytic Data Set

75% of the days within the observation period (January 1, 2015, to December 31, 2017).

considerations. Our sample was limited to people incarcerated in 1 of the 6 state-operated prisons for

aggregating individual-level variables to the facility-days based on theoretical and practical

January 1, 2015, and December 31, 2017. We used several steps to create a facility-level data set by

The sample for this study included adult men who were in LDCR custody (6 facilities) between

Sampling

health-related harms of extreme heat and lack of air conditioning.

by 2050.28 Moreover, LDCR has been embattled in litigation over both solitary confinement and the

more frequent, longer, and more severe heat waves, with an average of nearly 115 danger days a year

state averages 35 days a year when heat exceeds dangerous levels and is projected to experience

Louisiana has one of the largest and most densely populated prison systems in the United States. The

Setting and Study Sample

Institutional Review Board, and data are protected by a federal certificate of confidentiality.

approved for this secondary data analysis. This study was approved by the Emory University

consent waiver and Health Insurance Portability and Accountability Act waiver of authorization was

days, where facility refers to the 6 prisons in the sample, as described in later sections. A complete

suicide-watch incidents across 6 Louisiana prison facilities. The unit of analysis was prison-facility

models to assess associations between daily exposures to extreme heat and daily incident rate of

(LDCR) (January 1, 2015, to December 31, 2017) and used fixed-effects negative binomial regression

maximum heat index and daily data from the Louisiana Department of Corrections and Rehabilitation

August 11, 2023

3/11

Extreme Heat and Suicide Watch Incidents Among Incarcerated Men

This longitudinal case series panel study merged climatological data with measures of the daily

Methods

JAMA Network Open | Public Health

c..G

Downloaded from jamanetwork.com by guest on 12/27/2023

JAMA Network Open. 2023;6(8):e2328380. doi:10.1001/jamanetworkopen.2023.28380 (Reprinted)

the accuracy of our data merging procedures by comparing results of our descriptive analysis with

incidents, extreme heat, solitary confinement, and mental health severity over time. We confirmed

First, we conducted descriptive analysis to explore facility-level variations in suicide-watch

2017 and were not influenced by the presence or absence of other variables in our data.

missing completely because Vera did not have solitary confinement data for the last 6 months of

solitary confinement for approximately 15% of the 6576 facility-incarceration days. These data were

went from January 1, 2015, to July 3, 2017. Therefore, we had missing data on facility-level rates of

The Vera Institute of Justice data set from which solitary confinement exposures were obtained only

There were very minimal missing data in our data set, and only for the solitary confinement variable.

Statistical Analysis

observation period.

using housing files to count the unique individuals in each prison for every day in the

given day divided by Louisiana’s total imprisoned population. This control variable was calculated

Daily facility population was defined as the total number of people incarcerated at each prison on a

Daily Facility Population

CAJUN: administrative segregation, extended lockdown, closed-cell restriction, or death row.

showed they were assigned to 1 of the following types of units, designated by location codes in

that day. A person was counted as being in solitary confinement on a particular day if housing records

in solitary confinement each day and then divided it by the total population at each prison-facility

We aggregated individual housing files to calculate the total number of incarcerated persons residing

Daily Rate of Solitary Confinement

and a history of substance dependency.

the Diagnostic and Statistical Manual of Mental Disorders [Fourth Edition]) diagnosis other than SMI

functionality measures for at least 6 months. People on level 4 typically have an axis I (according to

months. People classified as level 3 have an SMI diagnosis but have been stable on medication and

2 typically were diagnosed with an SMI and a pattern of functional instability within the past 6

intensive clinical care, designated housing units, and ongoing management. Those classified as level

People classified as level 1 are assessed as having the most severe level of impairment and requiring

is a time-varying measure, and the date and result of most recent classification is recorded in CAJUN.

and severity of incarcerated persons’ mental health status during the course of their incarceration. It

and relative levels of impairment at a facility level.33 LDCR uses a level system to classify the acuity

persons at each prison classified as levels 1 to 3 as a proxy of serious mental illness (SMI) diagnosis

We aggregated individual-level files to create an indicator of daily percentages of incarcerated

Daily Rate of Serious Mental Illness

in the observation period, based on guidance of prior studies’ definition of extreme heat.2

facility-day where the maximum heat index exceeded the 90th percentile of heat indices for all days

the reference category was 60 to 69 °F.8,31,32 Second, we created a dichotomous indicator for any

40-49 °F, 50-59 °F, 70-79 °F, and ⱖ80 °F) based on the distribution of this variable. For modeling,

extreme heat. First, we categorized the daily maximum heat index into 6 bins (<30 °F, 30-39 °F,

effects of extreme heat on health.30 Guided by extant literature,2,8 we created 2 indicators of

relative humidity is combined with the air temperature”29 and is more frequently used to assess the

temperature, heat index is a measure of “what the temperature feels like to the human body when

index recorded by the weather station linked to the zip code of each of the 6 prisons. Distinct from

Weather Observing System stations. For this study, we used LCD data on the daily maximum heat

as well as observations collected every 20 minutes from approximately 1400 US Automated

August 11, 2023

4/11

Extreme Heat and Suicide Watch Incidents Among Incarcerated Men

precipitation. These data are collected from 950 US Automated Surface Observing System stations

JAMA Network Open | Public Health

c..G

1000

2000

3000

4000

5000

0

5

10

Suicide watch incidents, No.

15

Downloaded from jamanetwork.com by guest on 12/27/2023

JAMA Network Open. 2023;6(8):e2328380. doi:10.1001/jamanetworkopen.2023.28380 (Reprinted)

Facility days, No.

Figure 1. Distribution of Daily Counts of Recorded Suicide Watch Incidents

prisons (Table 1). Across all 6 prisons, the mean (SD) average percentage of people in solitary

mean (SD) daily maximum heat index across all prisons was 84.5 (17.1) °F with little variation between

which had 152 days (13.9%) in the extreme heat range for the observation period (Figure 2). The

percentile maximum daily heat index) with the exception of Raymond Laborde Correctional Center,

the observation period that met the definition of an extreme heat day (ie, exceeded the 90th

Each facility experienced similar percentage of days (mean [SD] days, 108 [0.29] days; 9.8%) during

Across all prisons, the mean (SD) daily number of suicide-watch events was 0.79 (1.56) (Figure 1).

average daily population and demographic composition of each facility are reported in Table 1.

The sample included 6 state-operated prisons in Louisiana, with a total of 6576 facility-days. The

Results

was conducted from December 2022 to February 2023. Statistical significance was set at P < .05.

90th percentile of heat-index from January 1, 2015, through December 31, 2017. The data analysis

as reference group) and a second model using the binary indicator for facility-days exceeding the

between extreme heat and suicide watch incidents: one using the heat-index-bin indicator (60-69 °F

XTNBREG commands in Stata version 17 (StataCorp). We ran 2 models to assess the association

period. Conditional fixed-effects negative binomial regression models were performed using

considering the ubiquitous absence of air conditioning in the sampled prisons during the observation

shape the incidence rates of suicide watches within and between prisons, especially when

watch incidents. Extreme heat is an exposure that is independent of facility-level exposures that may

models to account for potential seasonality effects that may influence the incident rate of suicide-

variations in unobserved factors over time and facility. We also added a fixed effect for month to our

fixed effects for day and facility in the model to account for clustering and potential influence of

heat and suicide-watch incidents, while controlling for aforementioned covariates.35,36 We included

conditional fixed-effects negative binomial regression models to test associations between extreme

not normally distributed (Figure 1) and overdispersed (variance exceeded the mean value), we used

Since our focal outcome was a count (ie, daily frequencies of suicide-watch incidents) that was

comparable results.

including all factors in the model, running 500 iterations.35 As discussed later, each approach had

confinement at each facility for roughly 6 months). We also performed multiple imputation by

confinement data. First, we ran the models with the data as missing (ie, excluded daily rate of solitary

For robustness, we compared results from 3 different approaches to handling missing solitary

factor prior to running full models.

August 11, 2023

5/11

Figure 1 shows the distribution of the daily counts of
suicide watch incidents recorded in the administrative
records of the Louisiana Department of Corrections
and Rehabilitation between January 2015 and
December 2017.

Extreme Heat and Suicide Watch Incidents Among Incarcerated Men

those in other sources.33,34 Next, we conducted bivariate analysis of each theoretically relevant

JAMA Network Open | Public Health

c..G

JAMA Network Open | Public Health

Extreme Heat and Suicide Watch Incidents Among Incarcerated Men

confinement was 21.2% (11.6), although it varied by facility and ranged from 6.9% to 28.9%. The
estimated daily mean (SD) population with an SMI and classified as having more severe functional
impairment (ie, level 1-3 mental health status) varied by facility as follows: Louisiana State
Penitentiary, 10.2% (1.1); Elayn Hunt Correctional Center, 17.4% (1.1); Raymond Laborde Correctional
Center, 14.5% (1.6); Rayburn Correctional Center, 10.3% (1.1); David Wade Correctional Center, 9.9%
(1.0); and Dixon Correctional Center, 5.8% (0.9). In bivariate analysis, all putative factors reached
statistical significance and were included in the final models (Table 2).
Results of the multivariable model suggested a strong and dose-responsive association
between extreme heat and the daily incident rate of suicide watches across analyses (Table 3).
Within the hotter daily heat index bins (ie, those exceeding 60-69 °F), the incidence rate of daily
suicide incidents increased by 29% when the heat index reached the level of caution (ie, 80-89 °F)
and by 36% when reaching extreme caution (90-103 °F) (80-89 °F: incidence rate ratio [IRR], 1.29;
95% CI, 1.17-1.43; P < .001; 90-103 °F: IRR, 1.36; 95% CI, 1.15-1.61; P < .001). Moreover, while the cooler
heat-index bins (ie, those below the reference group of 60-69 °F) were not statistically significant,
the model indicates that cooler days may decrease the incidence rate of suicide-watch incidents.
Results of model 2 corroborated model 1 and suggested that compared with other days, those falling
into the extreme heat category were significantly associated with a 30% increase in the incident rate
of daily suicide-watch incidents (IRR, 1.30; 95% CI, 1.18-1.45; P < .001). In other words, extreme heat
days contained 30% more suicide-watch incidents than days below that range.

Table 1. Descriptive Statistics for Facility-Level Variables for 6576 Facility-Days, January 1, 2015, to July 3, 2017
Prison facility
Characteristic

LSP

EHCC

DWCC

DCC

RLCC

RCC

Daily facility population, mean (SD), No.

4200 (117.6)

1330 (23.0)

854 (14.5)

1259 (16.2)

1273 (17.4)

940 (6.7)

Total count of suicide watch incidents, No.

1561

3236

109

46

29

195

Daily count of all suicide watch incidents, mean (SD),
No.

1.5 (1.7)

3.1 (2.3)

1.0 (0.4)

0.7 (0.3)

0.6 (0.2)

0.2 (0.5)

Facility-days with a suicide watch incident, %

66.2

89.5

8.9

4.0

2.5

15.9

Daily maximum heat index, mean (SD), °F

84.3 (17.7)

84.8 (16.3)

82.7 (18.3)

84.6 (16.5)

85.7 (17.7)

84.8 (16.3)

Facility-days at extreme heat, %

9.0

9.4

8.3

8.9

13.9

8.9

Daily percentage of population in solitary confinement, 25.8 (5.2)
mean (SD), %

30.2%(6.4)

35.8 (7.0)

7.9 (3.1)

10.0 (4.4)

17.8 (5.3)

Daily percentage of population with serious mental
illness, mean (SD), %

17.4 (1.1)

9.9 (1.0)

5.8 (0.9)

14.5 (1.6)

10.3 (1.1)

10.2 (1.1)

Abbreviations: DCC, Dixon Correctional Center; DWCC, David Wade Correctional Center; EHCC, Elayn Hunt Correctional Center; LSP, Louisiana State Penitentiary; RCC, Rayburn
Correctional Center; RLCC, Raymond Laborde Correctional Center.

Figure 2. Estimated Maximum Daily Heat Index for Total Sample, 2015 to 2017

Daily maximum heat index, ° F

120

100

80

60

40

20

40

60

80

100

Daily mean dry-bulb temperature, ° F

JAMA Network Open. 2023;6(8):e2328380. doi:10.1001/jamanetworkopen.2023.28380 (Reprinted)

Downloaded from jamanetwork.com by guest on 12/27/2023

Figure 2 displays the distribution of the maximum heat
index for each facility-day among the prisons in the
total sample. These are plotted by heat index bins.

August 11, 2023

6/11

0.83 (0.78-0.87)
0.86 (0.84-0.89)
0.69 (0.66-0.71)
1 [Reference]
1.07 (1.05-1.12)
1.43 (1.39-1.46)
1.33 (1.26-1.40)
1.37 (1.33-1.40)
1.01 (1.00-1.02)
1.04 (1.03-1.04)
0.99 (0.99-0.99)

40-49 °F
50-59 °F
60-69 °F
70-79 °F
80-89 °F
90-103 °F
Extreme heat day
Daily percentage of population in solitary confinement
Daily percentage of population with serious mental impairment
Daily facility population

1.29 (1.17-1.43)
1.36 (1.15-1.61)

80-89 °F
90-103 °F

1.01 (0.92-1.02)
1.06 (1.10-1.80)
1.39 (0.87-8.12)

Percentage of population in solitary confinement
Percentage of population with serious mental illness
Total daily facility population

.71

<.001

.06

<.001

.83

<.001

.02

<.001

<.001

.14

NA

.21

.17

.05

.39

P value

Downloaded from jamanetwork.com by guest on 12/27/2023

JAMA Network Open. 2023;6(8):e2328380. doi:10.1001/jamanetworkopen.2023.28380 (Reprinted)

1.30 (1.18-1.45)

Extreme heat day

Model 2

1.23 (0.200-7.535)

1.08 (0.98-1.19)

70-79 °F

Daily facility population

1 [Reference]

60-69 °F

1.06 (1.04-1.07)

0.92 (0.81-1.05)

50-59 °F

Percentage of population with serious mental illness

0.89 (0.77-1.05)

40-49 °F

1.01 (1.00-1.02)

0.77 (0.59-1.00)

Percentage of population in solitary confinement

0.72 (0.34-1.52)

30-39 °F

IRR (95% CI)

21-29 °F

Heat index bins (reference group 60-69 °F)

Model 1

Variable

Table 3. Significant Factors for Daily Suicide Watch Incidents in 6 Louisiana State Operated Prisons,
2015 to 2017a

0.24 (0.11-0.52)

30-39 °F

Incident rate ratio (95% CI)a

21-29 °F

Heat index bins

Variable

Table 2. Bivariate Associations With Suicide Watch Incidents Using Conditional Negative Binomial Regression
With Fixed Effects

exposures for incarcerated populations are well-founded: the incident rate of suicide watches was

incarcerated adult men in Louisiana. According to our findings, concerns about the perils of heat

We studied the associations of extreme heat, solitary confinement, and suicidal behaviors among

Discussion

increases by 6.0%, a finding that is corroborated in model 2 (Table 3).

health need (ie, level 1-3 mental health status), the incident rate of daily suicide-watch incidents

SD increase in the daily percentage of incarcerated people classified as having a higher level of mental

increase in the incident rate of daily suicide-watch incidents. These models also suggest that for each

increase in the percentage of people held in solitary confinement was associated with a 1.0%

need were significantly associated with suicide-watch incidents. First, both models indicate that 1-SD

All variables reached statistical significance at
P < .05.

a

August 11, 2023

7/11

Final model only included variables whose parameter
estimates were significant at the P < .05 level. Model
included fixed effects for day and prison facility.
Extreme heat day refers to any facility-day that
exceeded the 90th percentile of the daily heat index
for the total observation period.

Abbreviations: IRR, incidence rate ratio; NA, not
applicable.

a

Extreme Heat and Suicide Watch Incidents Among Incarcerated Men

Both models suggest facility-level indicators of solitary confinement and level of mental health

JAMA Network Open | Public Health

c..G

Downloaded from jamanetwork.com by guest on 12/27/2023

JAMA Network Open. 2023;6(8):e2328380. doi:10.1001/jamanetworkopen.2023.28380 (Reprinted)

carceral settings.

expanding the observation period are important goals for subsequent studies on heat and health in

number of facilities, enhancing measurement instrumentation, including additional data sets, and

variations in exposures and covariates on the focal association, where not possible. Increasing the

techniques, which are better suited for accounting for the influence of between- and within-facility

died by suicide. Additionally, due to the small number of prisons in our sample, hierarchical modeling

have reliable data on race and ethnicity and were unable to determine whether anyone in the facility

and temperature control capacities are needed.27 Similarly, in our facility level analysis, we did not

incarcerated women; therefore, studies that better account for gender and variations in geography

limited the sample to adult men. Nationally, self-injury and suicide are highly prevalent among

the only prison designated for women in Louisiana was evacuated during the study period, which

because the sample only included incarcerated adult men: incarcerated women were displaced after

infrastructure, as documented in recent litigation.7,44 The generalizability of this study is limited

likely that the heat index inside prison cells exceeded the outdoor measure due to the physical

other tier in 1 prison, none of the living spaces in the 6 prisons was air-conditioned.33 As a result, it is

exposures within and between facilities. However, aside from several cells in a death row unit, and 1

recorded outdoor exposures and therefore could not account for variations in indoor heat index

confinement vs dormitories and other types of units. Additionally, our measurement of heat index

differences in how extreme heat may influence suicidality vulnerabilities in spaces used for solitary

magnitudes of suicidality and self-injury in these prisons and did not permit accounting for

There are several limitations in this study. Drawing on administrative data likely underestimated

Limitations

environmental justice and public health imperatives.41-43

policy makers should heed to calls to pursue decarceration and anticarceral solutions as

Additionally, due to the excess harms prison conditions can impose in the context of excess heat,

climate change (eg, the Southeast) should allocate funds to update buildings accordingly.7

and jails do not have air conditioning. Carceral systems in settings that expect to be most impacted by

change is needed. Our results may help amplify the need for systematic changes. Most US prisons

attuned to the psychological and behavioral effects of extreme heat. However, more systemic

typically intended to avert heat exhaustion or heat stroke and dehydration and have been less

are vulnerable and providing access to fans, ice, and cold showers.7 However, these policies are

In response to extreme heat, prisons often implement strategies such as flagging people who

such studies and visualize hot spots for emergent environmental injustices.40

insights into the promises of merging climatological, carceral, and other sources of data to carry out

preventing and redressing associated harms. Our methods, alongside previous studies, provide

scientists to forge partnerships to advance science and formulate intersectoral interventions for

spaces as form of “thermal (in)equity”1 and urged geographers, epidemiologists, and environmental

the 80 °F range. Echoing others, Colucci et al conceptualized the hazards of extreme heat in carceral

heat with daily counts of violent assaults in the Mississippi prison system as the heat index exceeded

of different forms of violence. Similar to our findings, Mukherjee et al8 observed an association of

theory and empirical evidence showing a positive association between heat exposures and incidence

health emergencies, and incidence of suicidality in other contexts.11,15,37-39 This study expands on

linking extreme heat to increases in psychiatric morbidity, utilization of clinical protocols for mental

among a sample of imprisoned people. Our observations align with evidence from prior studies

To our knowledge, this is the first study to link extreme heat and an indicator of suicidality

potential seasonality effects.

climbing into the 90 to 103 °F range, after controlling for relevant facility-level covariates and

August 11, 2023

8/11

Extreme Heat and Suicide Watch Incidents Among Incarcerated Men

29% greater on days when the maximum heat index reached 80 to 89 °F and 36% greater on days

JAMA Network Open | Public Health

c..G

Downloaded from jamanetwork.com by guest on 12/27/2023

JAMA Network Open. 2023;6(8):e2328380. doi:10.1001/jamanetworkopen.2023.28380 (Reprinted)

4. Skarha J, Peterson M, Rich JD, Dosa D. An overlooked crisis: extreme temperature exposures in incarceration
settings. Am J Public Health. 2020;110(S1)(suppl 1):S41-S42. doi:10.2105/AJPH.2019.305453

3. Motanya NC, Valera P. Climate change and its impact on the incarcerated population: a descriptive review. Soc
Work Public Health. 2016;31(5):348-357. doi:10.1080/19371918.2015.1137513

2. Skarha J, Dominick A, Spangler K, et al. Provision of air conditioning and heat-related mortality in Texas prisons.
JAMA Netw Open. 2022;5(11):e2239849. doi:10.1001/jamanetworkopen.2022.39849

REFERENCES
1. Colucci AR, Vecellio DJ, Allen MJ. Thermal (in) equity and incarceration: a necessary nexus for geographers.
Environ Plan E Nat Space. 2023;6(1):638-657. doi:10.1177/25148486211063488

Data Sharing Statement: See the Supplement.

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection,
management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and
decision to submit the manuscript for publication.

Funding/Support: The Criminal Justice Research Training Program at Brown University and Miriam Hospital
awarded the corresponding author a pilot grant for this study.

Conflict of Interest Disclosures: Dr Cloud reported having been formerly employed by the Vera Institute of
Justice, which provided the secondary data for this analysis. No other disclosures were reported.

Supervision: Cloud, Cooper.

Administrative, technical, or material support: Cloud, Williams, Brinkley-Rubinstein.

Obtained funding: Cloud.

Statistical analysis: Cloud, Haardörfer.

Critical review of the manuscript for important intellectual content: Haardörfer, Brinkley-Rubinstein, Cooper.

Drafting of the manuscript: Cloud, Brinkley-Rubinstein.

Acquisition, analysis, or interpretation of data: Haardörfer, Cooper.

Concept and design: All authors.

Author Contributions: Dr Cloud had full access to all of the data in the study and takes responsibility for the
integrity of the data and the accuracy of the data analysis.

Author Affiliations: Department of Behavioral, Social, and Health Education Sciences, Rollins School of Public
Health, Emory University, Atlanta, Georgia (Cloud, Haardörfer, Cooper); Amend at the School of Medicine,
University of California, San Francisco School of Medicine (Cloud, Williams); Department of Population Health
Sciences, Duke University School of Medicine, Durham, North Carolina (Brinkley-Rubinstein).

Corresponding Author: David H. Cloud, PhD, JD, Department of Behavioral, Social, and Health Education
Sciences, Rollins School of Public Health, Emory University, 1518 Clifton Rd, Atlanta, GA 30322 (dcloud@
emory.edu).

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2023 Cloud DH
et al. JAMA Network Open.

Published: August 11, 2023. doi:10.1001/jamanetworkopen.2023.28380

ARTICLE INFORMATION
Accepted for Publication: June 29, 2023.

pursuit of environmental justice, human rights, and the abolition of carceral spaces.

underscores the importance of seeking long-term solutions through collective movement building in

conditioning and other heat-mitigation protocols in carceral spaces in the short-term and

and self-injury. It offers evidence to enhance the public health rationale for urgent calls for air

likely to unfold. This study illuminates the connection between solitary confinement, extreme heat,

than any sovereignty in modern history, a host of humanitarian and public health emergencies are

August 11, 2023

9/11

Extreme Heat and Suicide Watch Incidents Among Incarcerated Men

As the warming of the planet escalates and the US continues to incarcerate more of its population

Conclusions

JAMA Network Open | Public Health

c..G

Downloaded from jamanetwork.com by guest on 12/27/2023

JAMA Network Open. 2023;6(8):e2328380. doi:10.1001/jamanetworkopen.2023.28380 (Reprinted)

28. Climate Central. America’s Preparedness Report Card 2015: Louisiana extreme heat. Accessed July 10, 2023.
https://reportcard.statesatrisk.org/report-card/louisiana/extreme_heat_grade

27. Zhong S, Senior M, Yu R, et al. Risk factors for suicide in prisons: a systematic review and meta-analysis. Lancet
Public Health. 2021;6(3):e164-e174. doi:10.1016/S2468-2667(20)30233-4

26. Larney S, Farrell M. Prisoner suicide: a multilevel problem. Lancet Psychiatry. 2017;4(12):894-895. doi:10.
1016/S2215-0366(17)30435-2

25. Huey MP, McNulty TL. Institutional conditions and prison suicide: Conditional effects of deprivation and
overcrowding. Prison J. 2005;85(4):490-514. doi:10.1177/0032885505282258

24. Fazel S, Ramesh T, Hawton K. Suicide in prisons: an international study of prevalence and contributory factors.
Lancet Psychiatry. 2017;4(12):946-952. doi:10.1016/S2215-0366(17)30430-3

23. Fazel S, Cartwright J, Norman-Nott A, Hawton K. Suicide in prisoners: a systematic review of risk factors. J Clin
Psychiatry. 2008;69(11):1721-1731. doi:10.4088/JCP.v69n1107

22. Lee S, Lee H, Myung W, Kim EJ, Kim H. Mental disease-related emergency admissions attributable to hot
temperatures. Sci Total Environ. 2018;616-617:688-694. doi:10.1016/j.scitotenv.2017.10.260

21. Vida S, Durocher M, Ouarda TB, Gosselin P. Relationship between ambient temperature and humidity and visits
to mental health emergency departments in Québec. Psychiatr Serv. 2012;63(11):1150-1153. doi:10.1176/appi.ps.
201100485

20. Carlsen HK, Oudin A, Steingrimsson S, Oudin Åström D. Ambient temperature and associations with daily
visits to a psychiatric emergency unit in Sweden. Int J Environ Res Public Health. 2019;16(2):286. doi:10.3390/
ijerph16020286

19. Pan R, Zhang X, Gao J, et al. Impacts of heat and cold on hospitalizations for schizophrenia in Hefei, China: an
assessment of disease burden. Sci Total Environ. 2019;694:133582. doi:10.1016/j.scitotenv.2019.133582

18. Almendra R, Loureiro A, Silva G, Vasconcelos J, Santana P. Short-term impacts of air temperature on
hospitalizations for mental disorders in Lisbon. Sci Total Environ. 2019;647:127-133. doi:10.1016/j.scitotenv.2018.
07.337

17. Trang PM, Rocklöv J, Giang KB, Kullgren G, Nilsson M. Heatwaves and hospital admissions for mental disorders
in northern Vietnam. PLoS One. 2016;11(5):e0155609. doi:10.1371/journal.pone.0155609

16. Florido Ngu F, Kelman I, Chambers J, Ayeb-Karlsson S. Correlating heatwaves and relative humidity with
suicide (fatal intentional self-harm). Sci Rep. 2021;11(1):22175. doi:10.1038/s41598-021-01448-3

15. Nori-Sarma A, Sun S, Sun Y, et al. Association between ambient heat and risk of emergency department visits
for mental health among US adults, 2010 to 2019. JAMA Psychiatry. 2022;79(4):341-349. doi:10.1001/
jamapsychiatry.2021.4369

14. Noelke C, McGovern M, Corsi DJ, et al. Increasing ambient temperature reduces emotional well-being. Environ
Res. 2016;151:124-129. doi:10.1016/j.envres.2016.06.045

13. Thompson R, Hornigold R, Page L, Waite T. Associations between high ambient temperatures and heat waves
with mental health outcomes: a systematic review. Public Health. 2018;161:171-191. doi:10.1016/j.puhe.2018.
06.008

12. Lõhmus M. Possible biological mechanisms linking mental health and heat—a contemplative review. Int J
Environ Res Public Health. 2018;15(7):1515. doi:10.3390/ijerph15071515

11. Burke M, González F, Baylis P, et al. Higher temperatures increase suicide rates in the United States and Mexico.
Nat Clim Chang. 2018;8(8):723-729. doi:10.1038/s41558-018-0222-x

10. Dixon PG, Kalkstein AJ. Where are weather-suicide associations valid? an examination of nine US counties with
varying seasonality. Int J Biometeorol. 2018;62(5):685-697. doi:10.1007/s00484-016-1265-1

9. Dixon PG, McDonald AN, Scheitlin KN, et al. Effects of temperature variation on suicide in five U.S. counties,
1991-2001. Int J Biometeorol. 2007;51(5):395-403. doi:10.1007/s00484-006-0081-4

8. Mukherjee A, Sanders NJ. The causal effect of heat on violence: social implications of unmitigated heat among
the incarcerated. National Bureau of Economic Research; 2021. doi:10.3386/w28987

7. Holt D. Heat in US prisons and jails: corrections and the challenge of climate change. SSRN. Preprint posted
online September 30, 2015. doi:10.2139/ssrn.2667260

6. Haney C, Williams B, Lobel J, Ahalt C, Allen E, Bertsch L. Consensus statement from the Santa Cruz summit on
solitary confinement and health. NW UL Rev. 2020;115:335.

August 11, 2023

10/11

Extreme Heat and Suicide Watch Incidents Among Incarcerated Men

5. Cloud DH, Drucker E, Browne A, Parsons J. Public health and solitary confinement in the United States. Am J
Public Health. 2015;105(1):18-26. doi:10.2105/AJPH.2014.302205

JAMA Network Open | Public Health

c..G

Downloaded from jamanetwork.com by guest on 12/27/2023

JAMA Network Open. 2023;6(8):e2328380. doi:10.1001/jamanetworkopen.2023.28380 (Reprinted)

SUPPLEMENT.
Data Sharing Statement

44. Ball v LeBlanc, 881 346 (5th Cir 2018).

43. Schuck K, Backes EP, Western B, Wang EA, eds; National Academies of Sciences, Engineering, and Medicine;
Division of Behavioral and Social Sciences and Education; Committee on Law and Justice; Committee on the Best
Practices for Implementing Decarceration as a Strategy to Mitigate the Spread of COVID-19 in Correctional
Facilities. Decarcerating Correctional Facilities During COVID-19: Advancing Health, Equity, and Safety. National
Academies Press, 2020.

42. Hawks L, Lopoo E, Puglisi L, et al. Community investment interventions as a means for decarceration:
a scoping review. Lancet Reg Health Am. 2021;8:100150. doi:10.1016/j.lana.2021.100150

41. Prins SJ, Story B. Connecting the dots between mass incarceration, health inequity, and climate change. Am J
Public Health. 2020;110(S1):S35-S36. doi:10.2105/AJPH.2019.305470

40. Glade S, Niles S, Roudbari S, et al. Disaster resilience and sustainability of incarceration infrastructures:
a review of the literature. Int J Disaster Risk Reduct. 2022;80(1):103190. doi:10.1016/j.ijdrr.2022.103190

39. Liu J, Varghese BM, Hansen A, et al. Is there an association between hot weather and poor mental health
outcomes? a systematic review and meta-analysis. Environ Int. 2021;153:106533. doi:10.1016/j.envint.2021.
106533

38. Kim Y, Kim H, Gasparrini A, et al. Suicide and ambient temperature: a multi-country multi-city study. Environ
Health Perspect. 2019;127(11):117007. doi:10.1289/EHP4898

37. Charlson F, Ali S, Benmarhnia T, et al. Climate change and mental health: a scoping review. Int J Environ Res
Public Health. 2021;18(9):4486. doi:10.3390/ijerph18094486

36. Allison PD, Waterman RP. Fixed–effects negative binomial regression models. Sociol Methodol. 2002;32(1):
247-265. doi:10.1111/1467-9531.00117

35. Allison PD. Fixed Effects Regression Models. SAGE Publications; 2009.

34. Solitary Watch. Louisiana on lockdown: a report on the use of solitary confinement in Louisiana state prisons,
with testimony from the people who live it. June 2019. Accessed July 10, 2023. https://solitarywatch.org/
louisianaonlockdown/

33. Cloud D, LaChance J, Smith L, Galarza L. The safe alternatives to segregation initiative: findings and
recommendations for the Louisiana Department of Public Safety and Corrections, and progress toward
implementation. Vera Institute of Justice. May 2019. Accessed July 10, 2023. https://www.vera.org/downloads/
publications/safe-alternatives-segregation-initiative-findings-recommendations-ldps.pdf

32. Heutel G, Miller NH, Molitor D. Adaptation and the mortality effects of temperature across US climate regions.
Rev Econ Stat. 2021;103(4):740-753.

31. Deschênes O, Greenstone M. Climate change, mortality, and adaptation: Evidence from annual fluctuations in
weather in the US. Am Econ J Appl Econ. 2011;3(4):152-185. doi:10.1257/app.3.4.152

30. Anderson GB, Bell ML, Peng RD. Methods to calculate the heat index as an exposure metric in environmental
health research. Environ Health Perspect. 2013;121(10):1111-1119. doi:10.1289/ehp.1206273

August 11, 2023

11/11

Extreme Heat and Suicide Watch Incidents Among Incarcerated Men

29. National Weather Service. What is the heat index? Accessed July 13, 2023. https://www.weather.gov/ama/
heatindex

JAMA Network Open | Public Health

c..G