Abstract
To fill a current gap in the literature on aging, mental health, and the criminal justice system, a content analysis of international peer-reviewed research studies was conducted. A comprehensive literature search was conducted to identify English language research studies published in 2018-19. Forty-four articles were located using keyword search terms, such as aging, mental health, and criminal justice. The methods and major findings were analyzed using deductive and inductive approaches. It was found that the majority of the research studies were conducted in the United States and England. The results of the inductive analyses revealed major themes related to mental health detection and access to services, comorbid conditions, and the relationship of age, mental health, criminal behavior, and the social determinants of mental health. Findings from this review have significant implications for advancing epidemiological research, practice, and policy, especially as it relates to the influence of the social determinants of health of aging on life course mental health and criminal justice involvement. Research findings about the correlates and consequences of the social determinants of health, especially as it relates to aging, mental health, and criminal justice involvement, can be used to inform prevention and intervention efforts that target the social determinants of life course health and criminal justice involvement. It also provides a comprehensive assessment of the m methods used in prior studies to help improve future studies in this important area of investigation.
Keywords
Background
The global prison system has been growing at a rapid rate, especially in the United States. Of the approximately 2.6 million people in United States’ prisons as of 2010, about 220,000 (16%) were aged 50 and older (American Civil Liberties Union [ACLU], 2012). The Bureau of Justice Statistics (BJA) reports that in the past two decades the state prison population aged 55 and older has increased from 3% (1993) to 10% (2013). This increase has largely been attributed to the growing segment of incarcerated people who are aged 55+ serving long-term prison sentences, mostly for serious violent offenses (Carson & Sabol, 2016). Disproportionality also is reflected in other characteristics of the U.S. aging prison population. In 2010, 8% of inmates were aged 55 and above, male (93%), and disproportionately racial and ethnic minorities (46%), who came from lower socioeconomic backgrounds (Human Rights Watch [HRW], 2012). There also is a growing body of prison research on health and justice disparities, in which racial, ethnic minorities have less access to quality care and justice than the majority Caucasians and are at an increased risk for experiencing more health and justice disparities (Adler & Rehkopf, 2008). For example, Black men have a lifetime likelihood of imprisonment of one in three, compared with one in 17 for White men (Sentencing Project, 2013).
Official statistics also suggest that as many as half of adults aged 50 and older in prison are diagnosed with some type of mental health problem, including serious mental illnesses, such as major depression, schizophrenia, and dementia (James & Glaze, 2006). However, there is a dearth of research available about the experiences of adults aged 50 and older with mental health problems involved in the criminal justice system, which includes at the point of arrest, court processing, probation, prison, and forensic psychiatric hospitalization, and parole or community supervision. In addition, very little is known about the social determinants of health and mental health among this population of older adults involved in the criminal justice system, often referred to throughout the article as the commonly used term justice-involved older adults. However, due to methodological limitations in prior research such as the use of cross-sectional designs, small sample sizes, and the varying use of mental health assessment tools, current studies may provide inaccurate prevalence and incidence mental health estimates among older adults in the criminal justice system (Maschi & Aday, 2014; Maschi & Leibowitz, 2017).
Available evidence also suggests that there is a subpopulation of older adults with histories of minor to serious mental illnesses. The types of disorders noted among current studies include posttraumatic stress disorder, substance abuse disorders, major depressive disorders, dementia, and schizophrenia, in the criminal justice system, especially among older adults in prison (e.g., Arndt, Turvey, & Flaum, 2002; Fazel, Hope, O’Donnell, & Jacoby, 2002; Murdoch, Morris, & Holmes, 2008). Scholars, practitioners, and advocates have noted a process of accelerated aging, in which their biological age is much older than their chronological age, among people in prison. That is, the literature suggests that adults aged 50 and older in prison age 10 to 15 years faster than their community counterparts (Maschi, Viola, & Sun, 2013). The literature also suggests that this accelerated aging process may be attributed to high-risk personal histories and further exacerbated by the stressful conditions of confinement and lack of adequate prison health services (Anno et al., 2004; HRW, 2012; Wilson & Barboza, 2010). The projected increase of dementia among prisoners as well as other commonly noted age-related physical and mental health decline of incarcerated older adults elevates this issue to one of a significant public health concern (ACLU, 2012). Because mental and physical health problems are intertwined among older adults coupled with other known psychosocial determinants of health (employment, family, housing, social security, and financial assistance), it is imperative that there is an interdisciplinary response to foster the well-being of arrested persons or those who serve time in the criminal justice system.
Despite over 40 years of research that has examined mental health among older adults in the criminal justice system, there has been no recent systematic assessment of the available research studies that examine the social determinants of mental health in later life among a justice-involved aging population. This article attempts to fill the gap by conducting a content analysis of the peer-reviewed empirical literature that examines older adults, mental health etiology, and the criminal justice system. The research question guiding this review was “What does the peer-reviewed literature report about the methods and major findings on justice-involved older adults, the social determinants of mental health, and the criminal justice system?” That is, in studies that examine this population, (a) what kinds of research design and sampling strategies are used, (b) how are samples selected, (c) how is mental health measured and data collected, and (d) how might these choices influence the findings that were found and how they were interpreted? In addition, what are the common factors or themes found in the literature that speaks to the correlates and consequences of social determinants of health?
Findings from this review have significant implications for advancing epidemiological research, practice, and policy. Research findings about the correlates and consequences of the social determinants of health and justice can be used to inform prevention and intervention efforts, especially as it relates to the older vulnerable population. It also provides a comprehensive assessment of the methods used in prior studies to improve future studies in this important area of investigation.
Method
To answer the research question, a content analysis of peer-reviewed journal articles was conducted. The research team located the sample of articles that examined older adults, the social determinants of mental health, and the criminal justice system using the EBSCO HOST-Academic Search Premier research database. A comprehensive literature search was conducted to identify English-language research studies published as of May 2018. All databases were selected, which included MEDLINE, PsycINFO, and Sociolit. The following keyword search term combinations were used: older adults, old, elderly, mental health or mental illness, and criminal justice system (including arrest, courts, probation, jails, prisons, and parole). Two members of the research team also manually searched article references lists to identify any additional articles not found in the archives of the electronic research databases. Articles were included in the sample if they met the following criteria: (a) was a research study published in a peer-reviewed journal, (b) targeted mental health among adults aged 50 and older at some stage of the criminal justice process (i.e., arrest, court processing, probation, prison, or parole), and (c) was available in the English language. Articles were excluded from the sample if they (a) did not examine mental health assessment or treatment, or (b) were studies on older adults with mental health issues in noncriminal justice settings.
Of the 73 articles located during the initial search, 44 were determined to meet the study inclusion criteria, and all were quantitative studies and mostly descriptive studies. A data extraction form was developed by the research team to extract the following data into an Excel spreadsheet: publication characteristics (see Table 1 and its summary in the “Method” section), study research methods (which included research design, sampling strategies, sample characteristics, diagnostic assessment methods), and summaries of major findings across studies (see Tables 1-6). Three trained researchers extracted and coded the data. The data were reviewed weekly for a 12-week period with the lead researcher until 100% consensus was reached for all categories of data extracted.
Aging, Mental Health, and Criminal Justice: Characteristics of Peer-Reviewed Journal Articles (N = 44).
Note. DSM = Diagnostic and Statistical Manual of Mental Disorders.
Several studies used multiple methods of data collection.
Several studies used multiple methods of mental health assessment.
Aging, Mental Health, and Criminal Justice: Overview of Sample Size, Age, Gender, and Race/Ethnicity Across Studies (N = 44).
Note. AA = African American; Asian = Asian American, NP = Native American; PI = Pacific Islander; — = not reported; FPH/U = forensic psychiatric hospital/unit.
African/Caribbean.
Frequencey of Mental Health Diagnoses Across Studies on Aging, Mental Health, and Criminal Justice (N = 44).
Note. MH = mental health; MDD = major depressive disorder; PTSD = Posttraumatic Stress Disorder; BPD = Borderline Personality Disorder; ASP = Antisocial Personality Disorder; — = not reported; OMD = Organic Mood Disorder.
Percentage of Psychosocial and Legal Histories Across Participants of Studies on Aging, Mental Health, and Criminal Justice (N =44).
Note. MH = mental health; Tx = treatment; Hx = history; HS = high school; VOP = violation of probation; PV = parole violation.
Methods of Empirical Studies That Examined the Social Determinants of Health or Epidemiology of Mental Health/Psychiatric Disorders in Justice-Involved Older Adults (N = 11).
Note. DSM-IV = Diagnostic and Statistical Manual of Mental Disorders (4th ed.; American Psychiatric Association, 1994); PTSD = Posttraumatic Stress Disorder.
Overview of Major Findings on Aging, Mental Health, and Criminal Justice (N = 49).
Note. CST = competent to stand trial; IST = incompetent to stand trial.
Subscript of “e” represents studies that examine the etiology of mental health disorders in elderly prisoners.
Data Analysis Methods
Content analysis strategies as outlined by Krippendorff (2004) and Neuendorf (2002) were used to analyze the data. Content analysis is a systematic procedure that codes and analyzes qualitative data, such as the content of published articles, and a combination of deductive and inductive approaches can be used (Bernard & Ryan, 2010; Krippendorff, 2004; Neuendorf, 2002). For example, the current study used deductive analysis, which consisted of preexisting categories for journal article characteristics and research methods (e.g., country of study, study setting, research design, measures, data collection) to extract the data. Counts of textual variables were then calculated to identify frequencies and percentages using the descriptive statistics function of SPSS 21.0. For example, the following preexisting categories (i.e., police, parole, arrest, court, jail probation, prison, reentry, forensic psych unit) were created to represent the different stages of the criminal justice trajectory. No reentry studies were found, at the time of this review, that addressed all three factors (i.e., aging, mental health, and criminal justice).
Next, the narrative data based on the major findings were content analyzed using a qualitative, inductive approach in which no preexisting categories were used, and all categories and themes emerged directly from patterns found in the data. More specifically, the authors applied the Tutty, Rothery, and Grinnell (1996) four-step qualitative data analysis strategy, which is described next. Step 1 involved identifying “meaning units” (or in-vivo codes) from the data (Tutty et al., 1996). For example, the assignment of smaller “meaning units” included assigning codes (e.g., competency to stand trial, dementia diagnosis, access to services). In Step 2, second-level coding and first-level “meaning units” were sorted and placed in their emergent categories (e.g., mental health detection and access to service). These meaning unit codes, such as the example noted above, were arranged by clustering similar codes into a category and separating dissimilar codes into separate categories. The data were then analyzed for relationships, themes, and patterns (e.g., mental health detection and access to services and group differences). In Step 3, the categories were examined for meaning and interpretation (e.g., social determinants of health and justice which histories of earlier onset or prolonged mental illness and housing status (e.g., homelessness; education, gender). In Step 4, a conceptually clustered matrix was constructed to illustrate the patterns and themes found in the data (see top section of Table 6; Miles & Huberman, 1994).
Findings
Studies
Table 1 provides an overview of the research methods used across studies. This included study publication year, countries of study, study setting, study design, research design, sampling strategies, data collection procedures, and sample characteristics. They are reviewed in that order, respectively.
Article characteristics, study setting, and research designs
As shown in Table 1, the 44 studies were published between 1980 and 2018. Of the 44 studies, the majority were conducted in the United States. Slightly less than half were conducted in Europe and the Middle East, with a majority from England. Only one study was conducted in Australia.
The criminal justice setting which examined mental health among older adults in the criminal justice system also varied. Although the settings spanned police departments, courts, jails, prisons, and forensic psychiatric hospitals or units, over half of the studies were conducted in prisons followed by forensic psychiatric hospitals and the courts. Only one study was conducted at the point of arrest, jail, and probation. Interestingly, only one study was located that examined older adults with mental health issues on parole.
In general, there were methodological limitations that plagued the sample of studies. All of the studies were determined to have used cross-sectional research designs, thus limiting their ability to establish causality. Over half of the studies used probability sampling (random sampling) and about half of the studies used two or more comparison groups to examine age, gender, or race differences. Overall, these studies often had small sample sizes, which, in turn, limited their representativeness and generalizability to draw causal conclusions for justice-involved older adults with diverse backgrounds.
In answering our research question, we also examined methods used to determine a variety of mental health diagnoses (e.g., traumatic stressors, PTSD, and addiction disorders), commonly found in criminal justice populations as well as access to treatment for them. Mental health status was determined most commonly using case records followed by self-reported surveys, clinical assessment based on the Diagnostic and Statistical Manual of Mental Disorders criteria, and a variety of other methods. Interestingly, most of the studies did not report other relevant psychosocial factors and service use patterns. Also, aspects of mental health diagnoses related to trauma (e.g., PTSD), addictions (e.g., substance abuse), or prior or current access to mental health or addiction treatments were reported in only a few studies (e.g., Maschi, Viola, & Koskinen, 2015). The failure to report this type of additional information is a concern given those involved in the criminal justice system are disproportionately affected by trauma, mental health, and substance abuse which may affect their mental health status.
Data collection methods among the studies had questionable methodological rigor. Data were most often collected using only one data collection method, which was predominantly case record reviews. Only 10 of the studies used multiple forms of data collection to triangulate, that is, to confirm their findings were consistent (i.e., Barak, Perry, & Elizur, 1995; Cima, Merckelbach, Klein, Shellbach-Matties, & Kremer, 2001; Fazel, Hope, O’Donnell, Piper, & Jacoby, 2001; Fazel, Hope, O’Donnell, & Jacoby, 2002; Fazel, Hope, O’Donnell, & Jacoby, 2004; Hunt et al., 2010; Murdoch et al., 2008; and Paradis, Broner, Maher, & O’Rourke, 2000). The lack of triangulation of data collection methods limits the validity of mental health diagnosis and assessment, especially in studies that used only case record reviews based on clinical notes.
Sample Characteristics
The sample size and sociodemographic characteristics of participants (gender, age, and race/ethnicity) were examined next (see Table 2).
Sample size, gender, and age
As illustrated in Table 2, the sample sizes varied across studies and ranged from seven to 2,478. Out of the 44 studies, only 20 had sample sizes of 150 or more. As for age, 14 of the studies used the age of 60 to qualify adults as elderly and eight of the studies used age 55 and older. The age of 50 was the youngest age used to classify older adults as elderly and was used in 10 studies. In addition, over one third of the studies used gender specific (male or female only) samples. Most studies included men, compared with 22 studies that included women. Of the 22 studies that included women, female participants made up at least 25% of the study population in only four studies. Interestingly, in nine studies the gender of participants was not reported. Given that only one out of four women were represented in half of the studies, the relevance of these findings as representative of women is questionable.
Race/Ethnicity
Race/ethnicity varied in how it was reported across studies. Strikingly, almost half of the studies provided no information on the race and ethnicity of the participants. Of those studies that did report it, Whites were the most common and largest racial ethnic group sampled. White people represented over 50% of the sample in 14 of the studies. Minorities were minimally represented across the 31 studies. African-Americans were included in 15 studies and Latinos in only 11 of the studies. Nine other studies reported an “other” ethnicity category. These collective findings suggest that what we know about social determinants and mental health among justice-involved older adults is based mostly on White majority participants as opposed to racially diverse minority participants. Given the statistics that provide evidence of the disproportionate representation of racial ethnic minorities in the criminal justice system, their underrepresentation in research studies is a notable limitation (e.g., Sentencing Project, 2013).
Findings on Mental Health Diagnoses, Psychosocial and Legal Histories
The next stage of the analysis involved content analyzing the descriptive findings on mental health diagnoses, types of diagnoses, and psychosocial and legal histories among the study samples (see Table 3).
Mental health diagnoses
As shown in Table 3, of the 44 studies, schizophrenia, major depressive disorder, and dementia were the most widely diagnosed mental illnesses, particularly in prisons, forensic psychiatric hospitals/units, and the courts. Older adults diagnosed with dementia represented 44% to 46% of the sample in court settings (Frierson, Shea, & Shea, 2002; Lewis, Fields, & Rainey, 2006) and 5% in prison settings. The percentage of older adults with schizophrenia ranged from 2% to 91% among the studies and this diagnosis was most prevalent in forensic psychiatric hospitals/units (91%). There was a wide range in the percentage of older adults who had major depressive disorder (MDD; 4%-59%), with the highest prevalence of older adults with MDD being in the prison system. Interestingly, there was an absence of other commonly noted mental health disorders among the general criminal justice population (i.e., James & Glaze, 2006). For example, common disorders especially related to trauma and addictions (e.g., alcohol and substance use) found in the criminal justice system that were least reported in the sample of studies on the aging criminal justice population included posttraumatic stress disorder, anxiety, and dissociation (e.g., Maschi, Morge, Zgoba, Courtney, & Ristow, 2011; Maschi et al., 2015). This finding is interesting as there is a high prevalence of trauma-related mental health symptoms, such as PTSD, depression, and anxiety disorders commonly found in younger criminal justice populations (e.g., James & Glaze, 2006).
Psychosocial and legal histories
Table 4 presents the psychosocial and legal histories of the study population. Across the 44 studies, substance use disorders were the most commonly reported mental disorder. Only three of the 24 studies that reported substance use disorder also provided information on substance abuse treatment history.
Comorbid physical conditions among older adults with criminal justice involvement were reported in 15 of the studies. In five of these 15 studies, at least 80% of the study population reported one or more physical ailments. Only 13 of the studies provided data on the educational level of participant. Among these studies, between 11% and 90% of the participants reported having received a high school diplomas or equivalent. Available data on the criminal justice histories of older adult participants also varied across studies. For example, only four studies reported parole and probation violations and less than half of the studies (n = 6) reported violent offense history.
The Social Determinants of Mental Health among Incarcerated Older Adults
We next examined a subsample of 11 studies that described any association or causal link of the social determinants of mental health and criminal justice involvement of older adults using a variety of quantitative research designs and were cross-sectional studies from both the United States and Europe. Varying sample sizes ranged from 52 to 9,741 participants. Self-report surveys, interviews, and Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR; American Psychiatric Association [APA], 2000) criteria were the primary measures used for the studies with varying reliability and validity. Also, the data were collected primarily through case reviews followed by one-on-one interviews and self-report surveys. Given that most studies use cross-sectional research designs and one measure/assessment tool or data collection method (such as case record reviews), any conclusion about a causal link between the social determinants of mental health among justice-involved older adults reviewed below are made cautiously.
As shown in Table 5, about 25% of the studies explored possible social determinants of mental health and criminal justice involvement among older adults. These social determinants included gender (e.g., a higher risk for men compared with women), histories of earlier onset or prolonged mental illness, housing status (e.g., homelessness), education and employment history, the level of past trauma and chronic stress (including life course cumulative trauma), family and social support level, criminal history, and prior access to mental health assessment and treatment (e.g., Curtice, Parker, Schembri-Wismayer, & Tomison, 2003; Farragher & O’Connor, 1995; Haugebrook, Zgoba, Maschi, Morgen, & Brown, 2010, Maschi, Viola, & Koskinen, 2015). For example, religion and spirituality was noted as a potential protective factor in one study. In a prison study, Koenig (1995) found that older adults who reported being raised by someone with a religious affiliation were significantly associated with lower depressive symptoms. The study also found that incarcerated older adults who reported attending religious services more frequently reported lower levels of depressive symptoms that attended less frequently (Koenig, 1995). Allen (2008) also found that for older male inmates in Alabama, having a greater number of daily spiritual experiences and religion associated with better emotional health (Allen, 2008).
Inductive Analysis of Major Findings
Directly following the deductive analysis of studies, we next conducted an inductive analysis of the major findings of 44 studies. For more details about the inductive data analysis methods, please see the detailed description reviewed earlier in the “Method” section. As illustrated at the top of Table 6, the results of the inductive analysis yielded four major themes or categories: (a) mental health detection and access to services, (b) group differences, (c) comorbid conditions, and (d) the relationship of age, mental health, and criminal behavior, particularly related to social determinants of mental health and criminal justice involvement. The findings about these four themes are detailed below.
Mental health detection and access to services
It was found that the detection of mental health issues and access to services was identified at different stages of the criminal justice trajectory (e.g., courts vs. prison) for older adults. For example, in the early stages of a court setting, serious mental illnesses, such as dementia and schizophrenia and other psychotic or personality disorders, were commonly identified and reported. Detection of serious mental illness was often conducted as part of an assessment of older adults’ competency to stand trial. The findings here suggest that it is important to determine whether older adults are found to have serious mental illnesses (e.g., dementia and schizophrenia) that influence their behavior. These studies suggest that these illnesses may be an issue that can be flagged and possibly dealt with differently than sentencing them to prison. In comparison, the literature suggests that substance abuse problems are most commonly detected later in the criminal justice trajectory, during the prison intake process. Interestingly, in one study participants reported having never received substance abuse treatment even after 40 years of use (Arndt et al., 2002). This suggests that in the earlier stages (e.g., arrest or the courts) of the criminal justice trajectory, substance abuse is routinely included as part of the court assessment.
Other studies found that detection of a mental illness in the criminal justice system increased an older adult’s access to mental health services. For example, some court studies referred older adults for a forensic psychiatric evaluation to assess competency to stand trial. Other studies, especially in prison studies, found that most older adults diagnosed with serious mental illnesses, such as schizophrenia or dementia and/or personality disorders, were more likely to be referred for psychiatric care, including transfer to forensic psychiatric units (Curtice et al., 2003; Heinik, Kimhi, & Hes, 1994; Shah, 2006). Some studies found that older prisoners oftentimes have difficulty accessing mental health services (Maschi, Viola, & Sun, 2013; Maschi et al., 2015). Many studies recommended that prison systems need to allocate their resources to programs to improve the overall well-being of incarcerated older adults, perhaps by creating facilities that specifically target the health of older adults in prison (Maschi, Viola, & Sun, 2013; Sterns, 2008; Williams et al., 2010).
Group differences
A few studies examined group differences based on characteristics, such as age, gender, or gender/ethnicity, or offense histories. Some of the studies compared age group differences (i.e., older adults in prison compared with older adults in the community). For example, Needham-Bennett, Parrott, and MacDonald (1996) found that the prevalence of mental disorders of arrested older adults were higher when compared with community samples in Togliatti City, Russia. Other studies examined the differences between groups (older compared with younger male or female offenders), or racial differences (i.e., White vs. non-White) (Needham-Bennett et al., 1996). In court settings, Fazel and Grann (2002) found that justice-involved older adults were more likely to be diagnosed with serious mental illnesses, such as dementia and schizophrenia, and personality disorders when compared with justice-involved younger counterparts (Fazel & Grann, 2002). Similarly, Hunt and colleagues (2010) found justice-involved older adults (aged 65 and above) were more likely to be diagnosed with a mental illness compared with justice-involved younger adults.
Findings from prison settings include identifying that suicidal ideation was more common among White jail detainees (37%) than among minority prison detainees (8%) (Hunt et al., 2010). In contrast, minority jail detainees (69%) were more likely to be diagnosed with personality disorders than White jail detainees (37%; Paradis et al., 2000) were. Findings from prison studies include one prison where older women were found to have lower levels of personality disorder compared with younger women (Hurt & Oltmanns, 2002). Findings from forensic psychiatric settings demonstrated that patients with schizophrenia below the age of 50 were more likely to have committed a violent offense than those without schizophrenia, above the age of 50 (Wong, Lumsden, Fenton, & Fenwick, 1995).
Comorbid conditions
Some studies reported comorbid health and mental health conditions among older adults in the criminal justice system. Some studies reported comorbid major health and mental health issues (Fazel et al., 2002). Several studies reported comorbid mental health and substance use issues as well as histories of life course and cumulative trauma (Haugebrook et al., 2010; Maschi et al., 2015; Taylor & Parrott, 1988). For example, Haugebrook et al. (2010) found that approximately four of five older adult prisoners reported an occurrence of traumatic experiences and/or life event stressors with an average of 2.59 different traumas or stressful life events occurring. Mental health issues were present in 36% of the sample. Furthermore, Taylor and Parrott (1988) found that about half of older adult prisoners had active psychiatric symptoms. Psychosis (37%) and alcoholism (27%) were the major psychiatric disorders. Finally, Maschi et al. (2015) reported that seven of 10 older adults in prison reported some type of medical problem, often having histories of trauma experience.
The relationship of age, mental health, and criminal behavior (including the social determinants of health)
Fourteen of the studies explored to varying extents the relationship of age, mental health, and criminal behavior, including social determinants, and found mixed results. As shown in Table 6, some studies found that justice-involved older adults were less likely to commit serious offenses compared with justice-involved younger adults. For example, other studies found that a small percentage of older adults with serious mental illness committed homicides prior to incarceration (Coid, Fazel, & Kahtan, 2002; Regan, Alderson, & Regan, 2003). However, other studies reported correlations related to age, mental health, and violence. For example, Paradis and colleagues (2000) found that older adults who reported experiencing delusions were more likely to have committed violent crimes prior to incarceration (Paradis et al., 2000). Several studies also found that older males were more likely to commit sex offenses compared with younger offenders (e.g., Regan et al., 2003). Finally, McShane and Williams (1990) found that a combination of mental health issues, crimes committed, and low social support (i.e., infrequent family visits) predicted problem behaviors in an older adult prison population (McShane & Williams, 1990). These findings suggest possible social correlates of mental health among justice-involved older adults include age, criminal history, substance use, and level of social support.
Discussion
Summary of Major Findings
This study sought to build upon the literature by systematically examining and evaluating the methods and major findings of the peer-reviewed empirical literature of articles, which were mostly cross-sectional and descriptive studies. However, research in this area is plagued with methodological flaws, such as the use of cross-sectional designs and nonprobability sampling (purposeful sampling), which makes any correlational or causal conclusions drawn from them inconclusive. In addition, many of the studies used samples of White males in European and American prisons, forensic psychiatric hospitals, or courts. There is a dearth of studies conducted in police, probation, or community reentry or parole settings. These findings also may not be representative or generalizable to other racial/ethnic or minority group populations, which constitute a disproportionate number of people involved in this system.
The major findings from this body of research suggests that prior research also shows that serious mental illness is a significant issue among the older prison population. As noted in the findings section, there is a growing body of empirical literature that targets the mental health of older people involved in the criminal justice system. In total, we located 45 peer-reviewed journal articles published between 1980 and 2018 that addressed aging, mental health, criminality, and/or the criminal justice system. The majority of the articles most commonly documented the prevalence of serious mental health illnesses, such as schizophrenia, major depressive disorder, and dementia, particularly among older adults who came in contact with the courts, prisons, and psychiatric hospitals for committing alleged or being convicted of one or more criminal offenses. For example, several studies found that older adults who were diagnosed with dementia were more likely to be detected early in the criminal justice trajectory, particularly prior to a court trial as opposed to jails or prisons (e.g., Coid, 2002; Fazel & Grann, 2002; Heinik, 1994).
Many of the studies documented evidence of serious mental illnesses among the aging prison population. For example, older adults with schizophrenia in the criminal justice system represented anywhere from 2% to 91% of a criminal justice setting and were primarily housed in forensic psychiatric hospitals/units inside a correctional system. As for major depressive disorder, most older adults were commonly diagnosed with this mood disorder after they were in a prison setting. Interestingly, there was a significant unexplored research and practice gap in the literature about justice-involved older adults with respect to the prevalence of other common mental health disorders found among the general criminal justice population, such as posttraumatic stress disorder, anxiety, dissociation, suicide risk, and co-occurring substance abuse and other addictions.
These collective findings also show to a much lesser extent, other comorbid, less serious mental health issues and other biopsychosocial factors, such as physical health, homelessness, trauma and chronic stress, and social support were examined. Of significant concern is serious mental illness, such as dementia, which is much more prevalent among older adults in prison compared with community-dwelling older adults (Maschi, Kwak, Ko, & Morrissey, 2012). Clearly, more research is needed to better understand how older adults are assessed and treated while they are being detained. Moreover, in light of the proposed changes to the Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5; APA, 2013), significant attention must be given to how the assessment of mental disorders are translated to research and practice with older adults in the criminal justice system, especially prisons. The collective findings also do suggest that mental health/psychiatric disorders are detectable at all different stages of the criminal justice process from the point of initial police contact, court processing, probation, courts, prison, and parole.
Furthermore, the etiology behind the prevalence of mental illness among justice-involved older adults was examined in the extant literature. Though it is difficult to pinpoint exactly why older prisoners have a higher prevalence of illness, a history of trauma, substance abuse, homelessness, unemployment, and other stressful life events were common reasons associated with mental health issues among older adults. However, the prevalence of such health issues only seems to be growing given the lack of resources in the prison systems, lack of access to care, inability to assimilate in the prison environment, and difficulty reentering society (Maschi et al., 2015; Maschi, Suftin, & O’Connell, 2012; Maschi, Kwak, Ko, & Morrissey, 2012). Given that prisons and communities must attend to the needs of justice-involved older adults with mental health needs, the data seem to suggest that there is a clear necessity to better prepare prison and community service providers to provide culturally responsive mental health, health, and social services to this vulnerable population with complex care needs.
Results of the inductive analysis showed interesting findings regarding the potential social determinants of mental health and criminal justice involvement among samples of older adults, which include gender (e.g., a higher risk for men compared with women), histories of earlier onset or prolonged mental illness, housing status (e.g., homelessness), education and employment history, and the level of past trauma and chronic stress, family support, criminal history, and prior access to mental health assessment and treatment.
Implications
The collective findings of these studies especially related to the social determinants of health and mental health among the aging prison population suggest social and economic interventions are necessary and complementary to mental health interventions. These findings also suggest that legal and clinical professionals, especially forensic psychiatrists, psychologists, lawyers, and social workers, serve a key role in detecting minor to serious mental illness and providing referrals to services. What was not clearly discernable from these studies is the extent to which individuals enter the criminal justice system with mental disorders and/or develop them as part of age-related mental health decline or due to the often overcrowded and stressful conditions of confinement. Additional research is needed in this area. There is some evidence that mental illness is prevalent prior to prison, during prison, and/or is exacerbated during prison. Court studies show mental health illness is prevalent prior to the prisoner’s institutionalization but the extent of mental health illnesses prior to prison could be underestimated due to the first comprehensive assessment taking place in prison. In addition, there is little evidence about how age-related changes confound mental illness even further, especially related to cognitive impairment and depression.
Given the potential correlates and consequences of the social determinants of health and mental health of justice-involved older adults, it is imperative that a community-wide approach is adopted. As for a professional response, it is essential that mental health, health, social service, and other professionals work together, especially in interprofessional teams. Forensic professionals, such as psychiatrists, psychologists, lawyers, social workers, nurses, geriatricians, education, and labor specialists are key components of professional prevention and intervention efforts for this vulnerable population. Forensic psychiatrists can play a key role at every stage of the criminal justice trajectory, which include providing comprehensive assessments for older adults at every point of the criminal justice trajectory. They also can make an important collaborative contribution in building the research and practice base; provide expert testimony in local, state, and/or federal courts; and even engage in advocacy with and on behalf of this vulnerable population. Prison and community programs would better serve this population if they provide opportunities for geriatric mental health training for forensic professionals, especially forensic psychiatrists. Cross-sector coordination between the different agencies serving this population, such as mental health, aging, health, and criminal justice, would assist in streamlining the service provision process, which in turn may reduce the risks such as stress, psychiatric relapse, lack of access to needed mental health and other services, and homelessness and recidivism.
Implications for Mental Health Assessment and Intervention
These findings also suggest that conducting comprehensive mental health assessments for all possible disorders among older adults commonly found among a younger justice population is warranted. For example, in addition to assessing for serious mental illness, other possible associated traumatic stress disorders, mood disorders related to anxiety and depression, personality disorders associated with trauma histories, such as borderline personality disorder, and addictions (e.g., alcohol, drug, nicotine, gambling) should also be assessed. In addition, the incorporation of a detailed biopsychosocial assessment, often prepared by a social worker, also can assist with identifying many of the historical and current social environmental risk and protective factors that will have an influence on the mental and overall well-being of justice-involved older adults.
In addition to assessing for the “mental health problem,” the literature also underscored the importance that it leads to access to mental health treatment and other related services when older adults are in contact with the criminal justice system. Several studies found that the use of comprehensive mental health assessments significantly increased the use of referrals for older adults to receive mental health services whether it be in the courts, jails, or prisons. For example, studies of court settings were often of assessments of older adults’ competency to stand trial or increasing referrals or diverting them to mental health services. Prison studies also found that most older adults diagnosed with serious mental illnesses, such as schizophrenia or dementia or personality disorders, were more likely to be referred for psychiatric care, including transfer to forensic psychiatric units.
Implications for the Social Determinants of Mental Health and Criminal Justice Involvement
In addition to providing mental health services “as usual” to justice-involved older adults, it is equally as important to link them to wraparound services that address the common social determinants of these mental health among older adults in the justice system. The body of research studies on justice-involved older adults included in this review have generated empirical evidence for social determinants of mental health that include gender/race (e.g., a higher risk for males and racial ethnic minorities); histories of earlier onset or prolonged mental illness; prior access to mental health assessment and treatment; housing status (e.g., history of homelessness or solitary confinement); education and employment history; the frequency, magnitude, and duration of past interpersonal trauma and chronic stress (i.e., life course cumulative trauma); level of family and social support; spirituality/religious practices; and criminal justice history. The accumulation of social determinants of mental health (e.g., trauma and stress histories, low levels of education, access to mental health and social services) was described by them as a risk factor for the onset of prolonged mental health issues, criminal behavior, and justice system involvement (e.g., Maschi, Sutfin, & O’Connell, 2012).
Interventions that address their social determinants of mental health, such as access to housing, employment, education, and social supports while also in mental health treatment, is a paramount concern for many older adults released from prison. Based on the case histories, the risk factors, such as untreated childhood traumas, poverty, and lack of access to steady employment or services, placed these individuals at risk for criminal justice involvement, including recidivism.
It is also recommended that forensic professionals advocate across criminal justice settings (e.g., the courts, prisons) to incorporate comprehensive mental health assessments and interventions. On a final note, community placement challenges for these three cases with violent and/or sexual offense histories often are more challenging to obtain basic needs post release. In many cases, case managers often must educate and advocate on older adults’ behalf.
Current Study Research Limitations
This content analysis has methodological limitations that temper how these findings can be applied to practice and policy development. First, although all attempts were made to identify all of the studies that met the inclusion criteria, there may have been some studies that were not identified. Other mental health conditions, such as obsessive–compulsive and sexual disorders as well as personality disorders like paranoid, schizoid, schizotypal, and passive aggressive, were not extracted. In addition, substance abuse was reported globally, and alcohol and drug use were not reported separately in this analysis. In addition, although the data were extracted using a systematic and two coders, there is no way to ensure that the data extracted are completely reliable.
Future Directions for Research
Despite these limitations, these findings suggest areas for the future research that can be used with an eye toward the development or improvement of evidence-based prevention, assessment, and intervention with older adults in the criminal justice system. For example, future studies would be most useful if they gathered a more comprehensive portrait of the mental health, physical health, and other psychosocial factors impacting this population. These additional variables may include physical health and functional status as well as psychosocial and environmental factors that include race, gender, age, offense histories, trauma and chronic stress histories, education, and prior and current access to and quality of services. In addition, studies that examine professional bias in diagnostic assessment based on characteristics, such as age, race, and gender are important because of their potential significant implications for treatment and placement in prison or the community. Future studies also may examine the impact of stricter sentencing policies and long-term incarceration, including solitary confinement, on short- and long-term mental health, especially among older adults. Future studies that are designed to be more methodologically rigorous than prior studies can be used to inform culturally competent assessment and intervention with older adults at each stage of the criminal justice process. Perhaps a motto such as no elder left behind is quite fitting when examining the plight of older adults with or at risk of mental illness in the criminal justice system.
Footnotes
Acknowledgements
The authors would like to thank Adriana Kaye for their assistance with the preparation of the manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
