Mental Health Disorders and Suicidal Behaviors Among Provincial Correctional Workers.

OBJECTIVE: We examined the prevalence ofmental health disorders and suicidal behaviors (ideation, planning, and attempts) among a sample of provincial correctional workers in Manitoba.
METHODS: Self-reported mental health data from a survey on correctional worker mental health and well-being were analyzed for 491 correctional workers.
RESULTS: Over half (57%) of respondents screened positive for mental health disorder, most commonly major depressive disorder, and over one-third of respondents (37%) screened positive for more than one disorder. Positive mental health screens for all mental health disorders were associated with statistically significantly increased odds of lifetime suicidal ideation, and positive screens for most disorders were associated with past year suicidal ideation.
CONCLUSIONS: Relative to other public safety personnel and the general public, correctional workers appear have a higher prevalence of mental health disorders and suicidal behaviors. The association between positive screens for mental health disorders and suicidal behaviors highlights the vulnerability of correctional workers in regards to mental well-being.

Rates of mental health disorders (eg, posttraumatic stress dis-order [PTSD], major depressive disorder [MDD], generalized anxiety disorder [GAD], panic disorder [PD], alcohol use disorder [AUD]), are considerably higher among public safety personnel (PSP; eg, border services personnel, correctional workers, firefighters, operational and intelligence personnel, paramedics, police, public safety communicators, search and rescue personnel) relative to the general Canadian population.[1-3] Higher rates of mental health disorders among PSP appear due to potentially psychologically traumatic event exposures in the context of their work[4] and other occupational stressors.[5] Correctional workers are PSP who appear at heightened risk for mental disorders, even relative to other PSP.[6] Existing research also suggests correctional workers are at a higher risk of suicide for suicidal behaviors (ie, ideation, planning, attempts) relative to the general public and other PSP.[7,8] The current study draws on self-reported mental health survey data from provincial correctional workers in the province of Manitoba to explore associations between mental health disorders and suicidal behaviors. We situate our current results within emerging scholarship examining mental health and well-being among PSP generally, and correctional workers more specifically.

Mental Well-Being and Suicide Risk Among Correctional Workers

Correctional workers report a high prevalence of adverse mental health and well-being outcomes, including mental health disorders, burnout, and work life conflict.[1,6,8-12] Research indicates that PSP are also at a higher risk of suicide. A systemic review by Stanley et al[13] examined international research on suicidal ideation, attempts, and fatalities among some PSP (ie, firefighters, police, paramedics). The review found that approximately 47% of PSP reported lifetime suicidal ideation, with estimates of prior suicide attempts ranging from 0.7% to 55%. Estimates of death by suicide among PSP also varied across studies, ranging from 11.7 to 32.9/100,000 per year. Notwithstanding the variability, findings suggest suicide risk is higher for PSP relative to the general population. Correctional workers were not included in the 2016 review, but results remain relevant in light of related duties and experiences.

A multitude of factors, some operational, some organizational, and others personal underpin death by suicide and suicidality.[14-16] Higher risk of suicide among PSP has been attributed to several factors, including reduced fear of death relative to the general population,[17] compromised social supports,[17] irregular sleep patterns,[18,19] access to and familiarity of lethal means,[13] and history of adverse childhood events.[20] Suicide risk also is linked to the presence of a mental health disorder among PSP.[13]

Studies in the United States have indicated that the prevalence of death by suicide among correctional workers may be double that of police officers.[7,21] Results from the United States National Occupational Mortality Surveillance database suggest correctional workers are at a significantly higher age-adjusted risk for death by suicide, with even higher risk for women correctional workers.[22] Research also suggests the death by suicide rates for correctional workers appear as high as 105 per 100,000, which is more than seven times higher than the US national rate for the general populations (ie, 14 per 100,000).[23]

Recent scholarship with a national sample of Canadian correctional workers placed lifetime suicide ideation at 35.2%, planning at 20.1%, and attempts at 8.1%. Regarding past year ideation, correctional workers screened positive for ideation at a prevalence of 11%, planning with 4.8%, and 0.4% had past year attempts.[24] More recently, a study conducted looking at suicidality among correctional workers in Ontario provincial correctional services found prevalence of life time ideation at 26.6%, 11.9% planning, and 5.2% attempts, with past year prevalence of 7.0% for ideation and 2.6% for planning (past year attempts were noted reported due to low numbers).[25]

The current study builds on previous research[25] by assessing lifetime and past-year suicidal behaviors (ie, thoughts, planning, attempts) among a sample of correctional workers in Manitoba. The study assesses for differences between correctional workers in Manitoba relative to other provincial correctional workers and federal correctional workers. The study also assesses relationships between suicidal behaviors, mental disorders, and demographic variables. It was expected that respondents would report rates comparable to other provincial correctional workers.

METHODS

Data and Sample

Provincial correctional staff in Manitoba staff were invited to participate in an online survey on mental health and well-being in 2019. The research project received ethical clearance by research ethics boards at the University of Regina and Memorial University of Newfoundland and Labrador. Emails were sent by a representative of the provincial correctional ministry and/or union representatives inviting participation in the study. Participation was voluntary and there were no incentives for participation. Staff were provided with the option of completing the survey during paid work hours. Upon accessing the survey, respondents were provided with a unique login access code. This code enabled respondents to close and reopen the survey from any computer, allowing for the survey to be completed over more than one sitting if desired. Survey responses were hosted on a secure server through Qualtrics.

Upon accessing the survey, respondents were required to read an Informed Consent section that detailed the purpose of the survey, the potential benefits and risks, and the estimated time required for completion. Measures were taken to offset risks to participation including sensitive word choice, warnings prior to sensitive questions, and the identification of support resources available to respondents. Within the survey, we asked respondents to contact their Employee Assistance Program, family doctor, or another qualified mental health professional for help if respondents were experiencing suicidal/self-harming thoughts. Respondents were also provided with a source for immediate support and directed to call emergency services for immediate assistance as necessary.

In total, 772 participants commenced the survey. Responses were filtered for accuracy using a statement at the end of the survey: “I believe I have answered throughout such that I would allow my responses to be part of a dataset that will inform health care and policy for someone I care about.” Data were excluded if the respondent specified “false” or did not provide an answer to the question. In total, 491 respondents completed the survey and responded “true” to this statement and were thus included in the analyses.

MEASURES

The survey included established screening measures for several mental health disorders, as well as questions pertaining to demographics, history of potentially psychologically traumatic event exposure, corrections-specific event exposure, and other items tied to work experiences and well-being. Questions included a mixture of both structured and open-ended items. For the current analyses, close-ended responses for mental health screening measures and questions pertaining to demographics and suicidal experiences were examined.

Positive screens for mental health disorders were based on self-reported symptoms in reference to established measures and associated cut-off scores: PTSD was assessed using the PTSD Check List 5[26-30]; MDD was assessed using the nine-item Patient Health Questionnaire[31-34]; GAD was assessed using the seven-item GAD-7[32,35,36]; PD was assessed using the PD Symptoms Severity scale–Self-Report (PDSS-SR)[37-39]; and AUD was assessed using the AUDs Identification Test (AUDIT).[40,41] Respondents were also asked about their own history of formal mental health disorder diagnoses to better inform prevalence rates.

Suicidal ideation, planning, and attempts were assessed through a series of yes/no questions that have been used in prior PSP research.[24] For suicidal ideation, questions included: “have you ever contemplated suicide?” and, if a response of yes was indicated, “has this happened in the past 12 months?” For suicidal planning, questions included: “have you ever made a serious plan to attempt suicide?” and, if yes, “has this happened in the past 12 months?” For suicidal attempts, questions included: “have you ever attempted suicide?” and, if yes, “did this happen in the past 12 months?”

Socio-demographic variables included sex (male, female; the category of other was excluded due to low numbers), age (19 to 29, 30 to 39, 40 to 49, 50 to 59, 60+), marital status (single, partnered, separated/divorced/widowed), education (high school or less, some post-secondary, completed degree), and years of service (1 to 5 years, 6 to 10 years, 11 to under 15 years, 16 to 20 years, more than 20 years). Occupational groups were originally categorized as: Correctional Officers, Institutional Management, Institutional Healthcare, Institutional other (eg, program, training, religious/spiritual, and administrative staff), Probation Officers, and Community other (eg, managers, program staff, administrative staff). Due to asymmetries in counts within occupational groups, only two categories are used in written analysis (namely Correctional Officers and all other occupational groups).

Statistical Analysis

Descriptive statistics were computed to examine the prevalence of mental health disorders, lifetime and past-year suicidal behaviors (ideation, planning, and attempt), and sociodemographic covariates. Respondents were grouped into categories based on positive screens for mental health disorders, namely respondents who screened positive for PTSD, MDD, GAD, PD, and AUD. We computed logistic regression models to examine the association between mental health disorders and each of lifetime and past-year suicidal behaviors. Results at P < 0.05 were considered statistically significant.

RESULTS

Sociodemographic Variables

Sample demographics are presented in Table 1. Approximately 60% of the 491 respondents self-identified as men and 40% identified as women. The average age was 41.77 (median = 42), with 8% between 19 and 29, 33% between 30 and 39, 37% between 40 and 49, 17% between 50 and 59, and 3% 60 or older. Most reported being married or in a common law relationship (74%), with the remaining identifying as single (13%), or divorced/separated/widowed (12%). Regarding educational achievement, 37% reported completing a 4-year university/college program or more, 41% reported completing some post-secondary education (but less than a 4-year degree), and 19% reported a high school diploma or less. Respondents reported an average of 11.40 (median = 10.00) years of service. Correctional Officers represented the largest occupational group (73%). Other institutional groups included Institutional Managers (7%), Institutional Healthcare Staff (<1%), and other types of institutional staff (eg, program, administrative, and spiritual staff; 8%). Community correctional workers included Probation Officers (7%) as well as other types of community staff (eg, program and administrative staff; 3%).

TABLE 1

Sociodemographic Information (n = 491)

Sociodemographic Information	% (n)
Sex
Female	39.7 (195)
Male	59.7 (293)
Age
19–29	8.1 (40)
30–39	33.4 (164)
40–49	36.9 (181)
50–59	17.1 (84)
60 and older	2.9 (14)
Marital status
Single	12.6 (62)
Married/Common-law	74.3 (365)
Separated/Divorced/Widowed	11.8 (58)
Education level
High school graduate or less	19.1 (94)
Some post-secondary school	41.1 (202)
University/college degree or higher	36.5 (179)
Years of service
1 to 5 years	17.3 (85)
6 to 10 years	36.9 (181)
11 to under 15 years	15.7 (77)
16 to 20 years	16.7 (82)
More than 20 years	10.8 (53)
Occupational group
Correctional Officer	73.1 (359)
Institutional Management	7.3 (36)
Institutional Healthcare	0.2 (10)
Institutional other*	7.5 (37)
Probation Officer	6.7 (33)
Community other†	2.9 (14)

*Examples include program/vocational staff, training staff, spiritual services staff, and administrative staff.

†Examples include managers, supervisors, healthcare staff, program staff, and administrative staff.

Mental Health Disorder Screens

Based on self-reported measures, a majority of respondents (57%) screened positive for at least one mental health disorder (see Table 2). By occupational group, the percentage of respondents who screened positive for at least one mental disorder was 59% for Correctional Officers (the percentage was 64% for female Correctional Officers and 57% for male Correctional Officers) and 51% for other all other occupational groups. Respondents reporting a high school education or less were slightly more likely to screen positive than those with more education (ie, 65% vs 55%). Non-partnered respondents who reported being single or divorced/separated/wid-owed more likely to screen positive than those who reported being married or in a common law relationship (ie, 69% vs 52%). Many respondents (37%) screened positive for more than one disorder, including several (ie, 26%) who screened positive for three or more disorders. In the case of Correctional Officers, 42% screened positive for more than one disorder (42% for women and 41% for men).

TABLE 2

Prevalence of Mental Health Disorders (n = 491)

Disorder	% (n)
PTSD (PCL-5)	29.9 (147)
MDD (PHQ-9)	42.0 (206)
GAD (GAD-7)	29.5 (145)
PD (PDSS-SR)	20.6 (101)
AUD (AUDIT)	9.2 (45)
Any other self-reported mood disorder	3.1 (15)
Any positive screen for any mental disorder	56.8 (279)

AUD, alcohol use disorder; GAD, generalized anxiety disorder; MDD, major depressive disorder; PD, panic disorder; PTSD, posttraumatic stress disorder.

Positive screens were most common for MDD (ie, 42% overall; 47% for Correctional Officers; 30% for other all other occupational groups). Positive screens for PTSD (30%) and GAD (30%) were the second most common. By occupational category, positive screening for PTSD was 33% for Correctional Officers and 23% for other all other occupational groups; for GAD, 33% for Correctional Officers and 19% for other all other occupational groups; for PD, 21% for all occupational groups; and for AUD, 10% for Correctional Officers and 8% for other all other occupational groups.

Suicidal Behaviors

Just over one-third of respondents (ie, 35%) reported having lifetime suicidal ideation (36% for men and 33% for women), with similar percentages for Correctional Officers compared with other occupational groups (ie, 35% and 34% respectively). Suicidal ideation within the past year period was reported by 15% of respondents (16% for men and 13% for women), with a somewhat higher percentage for Correctional Officers relative to other occupational groups (ie, 17% vs 9%). Gender differences within the Correctional Officer group category were minimal; lifetime suicidal ideation was 36% for men, and 35% for women, and past year suicidal ideation was 17% for men and 16% for women (see Table 3).

TABLE 3

Prevalence of Suicidal Behaviors (n = 491)

Suicidal Behavior	% (n)
Suicidal ideation
Life time	34.8 (171)
Past-year	15.1 (74)
Suicide plan
Life time	20.2 (99)
Past-year	7.7 (38)
Suicide attempt
Life time	9.6 (47)
Past-year	2.0 (10)

In regards to prior suicide planning, 20% of respondents reported lifetime planning; the percentage was 22% for men and 17% for women and was the same for Correctional Officers compared with other occupational groups (ie, 20%). Suicide planning in the past year period was reported by 8% of respondents, with similar prevalence for Correctional Officers compared with other occupational groups (ie, 8% and 7% respectively) and for men and women (9% and 6% respectively). Regarding suicide attempts, 10% of respondents reported a prior lifetime attempt; the percentage was 8% for men and 11% for women, and 8% for Correctional Officers compared with 13% for other occupational groups. In 2% of cases, respondents reported a suicide attempt in the past year (the percentage was the same for men and women and for Correctional Officers and other occupational groups). Across items, higher levels of educational and being married or in a common law relationship were inversely associated with lifetime and past-year suicidal behaviors, which is consistent with prior research.[24]

Association Between Mental Health Disorders and Suicidal Behaviors

The associations between mental health disorders and lifetime and past-year suicidal behaviors are provided in Table 4 Respondents who screened positive for a mental health disorder, particularly those screening positive for more than one disorder, were more likely to endorse suicidal behaviors. Differences were noted for prior lifetime suicidal ideation (ie, 16% for those who did not screen positive, 37% for those who screened positive for one disorder, and 56% for those who screened positive for more than one disorder), lifetime suicidal planning (6%, 22%, 36%), and lifetime suicide attempts (3%, 8%, 18%).

TABLE 4

Associations Between Past-Year and Lifetime Suicidal Behaviors and Mental Disorders (n = 491)

Suicidal Behavior	PTSD	MDD	GAD	PD	AUD
Lifetime
Ideation
OR (95% CI)	3.61 (2.40, 5.43)***	4.66 (3.12, 6.96)***	3.64 (2.42, 5.47)***	2.98 (1.89, 4.69)***	2.72 (1.45, 5.10)**
AOR (95% CI)	3.49 (2.22, 5.50)***	4.87 (3.10, 7.65)***	3.65 (2.28, 5.85)	2.76 (1.65, 4.62)***	3.02 (1.53, 5.96)***
Plan
OR (95% CI)	1.51 (0.81, 2.80)	2.23 (1.16, 4.29)*	1.42 (0.77, 2.62)	1.04 (0.53, 2.04)	0.74 (0.31, 1.77)
AOR (95% CI)	1.69 (0.81, 3.53)	2.76 (1.28, 5.96)**	2.08 (0.96, 4.51)	1.68 (0.71, 3.99)	0.74 (0.28, 1.95)
Attempt
OR (95% CI)	1.86 (0.94, 3.70)	1.92 (0.89, 4.15)	–	2.60 (1.27, 5.32)**	–
AOR (95% CI)	3.02 (1.17, 7.83)*	3.88 (1.32, 11.41)*	–	3.74 (1.37, 10.23)**	–
Past-Year
Ideation
OR (95% CI)	4.95 (2.55, 9.59)***	5.91 (2.70, 12.90)***	6.96 (3.51, 13.79)***	3.25 (1.63, 6.48)***	0.79 (0.33, 1.91)
AOR (95% CI)	4.51 (2.01, 10.13)***	5.52 (2.18, 14.02)***	11.30 (4.24, 30.09)***	4.00 (1.60, 10.03)**	1.47 (0.50, 4.33)

AOR, adjusted odds for socioeconomic status (sex, age, marital status, education, years of service, and correctional categories); AUD, alcohol use disorder; CI, confidence interval; GAD, generalized anxiety disorder; MDD, major depressive disorder; OR, odds ratios; PD, panic disorder; PTSD, posttraumatic stress disorder.

– Not presented due to insufficient sample size (ie, n < 5) to run the analysis.

Statistically significant:

*P < 0.05.

**P < 0.01.

***P < 0.001.

Positive mental health screens for all mental health disorders were associated with statistically significantly increased odds of lifetime suicidal ideation (see Table 4). Most positive screens for mental health disorder (ie, all except AUD) were also associated with increased odds of past-year suicidal ideation (see Table 4). There were insufficient cell counts to compute lifetime suicide attempts for the GAD and AUD groups. Due to low cell counts, we were unable to examine the association between mental health disorders and past-year suicide planning and attempts.

DISCUSSION

The current results highlight key themes tied to mental health and suicidality among correctional staff. Consistent with prior research, a considerable high percentage (57%) of correctional workers screened positive for at least one mental health disorder, most commonly MDD, followed by PTSD and GAD. The current prevalence rates appear much higher than the diagnostic rates of mental health disorders among the general public in Canada (ie, 10%), but similar to positive screening rates (ie, 55%) from federal correctional workers.[1] Many respondents (37%) screened positive for more than one disorder, including several (ie, 26%) who screened positive for three or more disorders. The current results suggest that mental health profiles among correctional workers are complex and impacted by exposures to diverse stressors.

Many respondents in the current sample reported having lifetime experiences of contemplating suicide (35%), planning to attempt suicide (20%), or attempting suicide (10%). By comparison, prevalence among Canadian PSP more generally is 28% for ideation, 13% for planning, and 5% for attempts.[24] As such, the current results appear comparable to previous research, suggesting correctional workers are at high risk for suicidal behaviors, even relative to other PSP. The past-year and life time prevalence among Manitoba correctional workers is higher than reports from provincial correctional workers in Ontario (ie, 27%, 12%, and 5% for lifetime ideation, planning, and attempts respectively[25]), which may indicate the need for specific additional research.

Positive screens for mental health disorders were associated with increased likelihood of reporting lifetime and past-year suicidal behaviors. The associations were particularly pronounced for respondents who screened positive for more than one mental health disorder. All five mental health disorders examined (ie, MDD, PTSD, GAD, PD, AUD) were associated with statistically significantly increased odds of lifetime suicidal ideation and most were associated with past-year suicidal ideation, except for AUD. The associations further clarify vulnerabilities for correctional workers with respect to suicidal behaviors, and underscore the need for proactive and ongoing mental health supports for correctional workers.

Limitations

The current study has several limitations that offer important directions for future research. Respondents self-selected to participate after being invited through multiple listservs; as such, we do not have a well-defined sampling frame. Self-selection also prohibits a definitive understanding of the actual response rate, which caveats data interpretations and generalizability. The study results necessarily exclude individuals who have died by suicide, which suggests the current results may represent an underestimation of suicidal behaviors and relationships with mental health disorders in the population of correctional workers. The current study was designed using questions from the Statistics Canada’s 2013 Canadian Community Health Survey and previous PSP research[1,24]; nevertheless, comparisons between the current results and previous results may be inhibited by methodological differences. Not all respondents proceeded far enough in the survey to answer the questions on suicidal behaviors. We do not have information about why respondents did not complete the full survey, but survey length may have been a factor. Responses may have also been impacted by lingering stigma surrounding mental disorders and suicidal behaviors[42-44]; however, the anonymous survey format may have offset challenges with stigma by reducing concerns about possible social repercussions.[45] The current results were still based on self-report, rather than clinical interviews, which may also limit generalizability.[46] Respondents may under (or over) report symptoms, even when anonymous, which could impact results.[47,48] Finally, we were unable to identify causal associations between correctional work and the development of mental health disorders or suicidal behaviors; however, previous research does imply a link between conditions of work and adverse mental health outcomes among correctional workers.[49,50]

CONCLUSION

The current study unpacks the associations between prevalence of mental health disorders and suicidal behaviors among provincial correctional workers in Manitoba. Our results contribute to an emerging body of literature examining the well-being of PSP and, consistent with prior results,[25] highlights the vulnerability of correctional workers regarding mental well-being. In light of the association between positive screens for mental health disorders and suicidal behaviors, the results lend weight to the basis for current advocacy, policy, and academic efforts to enhance the mental health supports available to correctional staff, and point to the need for tailored multimodal mental health programs and services focused on mitigating suicidal behaviors.