Traumatic Brain Injury and Recidivism Among Returning Inmates

Abstract

In recent years, there has been a surge in research that examines the relationship between traumatic brain injury (TBI) and involvement in the criminal justice system. However, the bulk of this research has been largely retrospective and descriptive, comparing rates of TBI in the offending population with the rates of TBI in the general population. Although findings from these studies indicate a higher prevalence of TBI in the offending population, virtually no studies have examined whether those with TBI are more likely to recidivate. To address this gap, the present study examined rearrest post release from prison among a cohort sample of Indiana inmates who were screened using the Ohio State University Traumatic Brain Injury Identification (OSU-TBI-ID) instrument. Findings indicate that, net of control variables, those with TBI were more likely to recidivate sooner than those without TBI. Policy implications and directions for future research are discussed.

Keywords

recidivism traumatic brain injury mental illness returning inmate TBI

In recent years, traumatic brain injury (TBI) has been the subject of increasing scholarly attention. Although there have been many definitions of TBI proposed over the years, the most recent consensus suggests it is an alteration in brain function caused by an external force to the head (Menon, Schwab, Wright, Maas, & Demographics and Clinical Assessment Working Group of the International and Interagency Initiative Toward Common Data Elements for Research on Traumatic Brain Injury and Psychological Health, 2010). Thus, TBI refers to disruptions to the normal function of the brain caused by blows or jolts to the head. Moreover, recent definitions recognize that not all head injuries result in TBI but also that the severity of TBI can range from mild (a brief change in mental status or consciousness) to severe (a sustained period of unconsciousness or amnesia). In the United States, an estimated 1.7 million people sustain TBI each year (Faul, Xu, Wald, & Coronado, 2010). As a result of this large number, TBIs have been referred to as a “silent epidemic” as there is little public awareness and because the symptoms of TBI are not always readily evident (Goldstein, 1990). Most adults, at some point in their lives, experience a bump on the head but never experience adverse psychological or behavioral changes from the injury; however, for some individuals, TBI might result in impairments (Ferguson, Pickelsimer, Corrigan, Bogner, & Wald, 2012). Unless these individuals are identified early on, and diverted into appropriate treatment programs, the TBI can lead to a host of negative outcomes such as attention problems, increased aggression, hypersexual behavior, and a lack of impulse control, all of which are particularly salient if the injury occurs during childhood (Eghwrudjakpor & Essien, 2008; Fazel, Lichtenstein, Grann, & Langstrom, 2011; Leon-Carrion & Ramos, 2003; Turkstra, Jones, & Toler, 2003).

Involvement in the criminal justice system is another potentially negative outcome from TBIs that has received attention among researchers. However, much of this research consists of case studies or descriptive samples comparing the offending population with the general population. For example, the rate of TBI in the general population is estimated to be at about 8.5% (Silver, Kramer, Greenwald, & Weissman, 2001) while studies of offending populations have found rates between 25% and 87% (Barnfield & Leathem, 1998; DelBello et al., 1999; Langevin, 2006). Although these results illustrate higher rates of TBI among criminally involved populations, they are not able to assess a causal relationship between offending or reoffending rates as a function of TBI. Moreover, it is possible that offenders with TBI have repeated experiences with the criminal justice system, and so understanding how to screen for TBI is important in helping to deliver appropriate treatment while incarcerated and post release.

The Ohio State University Traumatic Brain Injury Identification (OSU-TBI-ID) instrument was designed for clinical purposes to capture the prevalence and severity of TBI and has been utilized in numerous studies and across a variety of populations, including military personnel, veterans, and prisoners (Bogner & Corrigan, 2009; Corrigan & Bogner, 2007a, 2007b; Corrigan, Bogner, & Holloman, 2012; Ferguson et al., 2012). However, although extant research suggests that the OSU-TBI-ID accurately captures the presence of TBI, it has not been used to determine whether having TBI is associated with subsequent involvement in the criminal justice system.

In this study, we examine data from a statewide cohort of adult male inmates who were screened for TBI, using the OSU-TBI-ID, over a 1-month period in 2012. From this cohort, we follow up a subsample of inmates who were released and returned to the same county (N = 150) to examine whether those with TBI were more likely to recidivate than those without.

TBI and Criminal Behavior

Much of the research establishing a relationship between TBI and criminal behavior comes from descriptive studies that compare rates of TBI in offender populations with those in the general population. Meta-analyses estimate that approximately 60% of adult offenders (Farrer & Hedges, 2011; Frost, Farrer, Primosch, & Hedges, 2013; Shiroma, Ferguson, & Pickelsimer, 2012) and 30% of juvenile offenders (Farrer, Frost, & Hedges, 2013) have TBI compared with 8.5% of the general population (Silver et al., 2001). However, many of these studies have targeted specific populations—homicide offenders (Blake, Pincus, & Buckner, 1995; Freedman & Hemenway, 2000; Lewis, Pincus, Feldman, Jackson, & Bard, 1986), sex offenders (Langevin, 2006), or those with mental health or a substance abuse problem (DelBello et al., 1999; Hawley & Maden, 2003; Martell, 1992; Walker, Staton, & Leukefeld, 2001)—rather than the general offender population. Moreover, sampling strategies have varied considerably with most relying on convenience samples (Blake et al., 1995; DelBello et al., 1999; Diamond, Harzke, Magaletta, Cummins, & Frankowski, 2007; Freedman & Hemenway, 2000; Lewis et al., 1986; Turkstra et al., 2003; Williams, Cordan, Mewse, Tonks, & Burgess, 2010).

Given that the above studies were cross-sectional and descriptive, they are unable to determine a direct causal relationship between TBI and involvement in the criminal justice system leading researchers to speculate about how TBIs lead otherwise law-abiding people to engage in criminal behavior. Biological explanations suggest that damages to certain parts of the brain—namely, the frontal lobe, prefrontal cortex, and temporal lobes¹—can impair judgment, reasoning, and impulse control, which might contribute to criminal behaviors (Chayer & Freedman, 2001). Damage to this area has been associated with changes in personality traits, attention deficits, short-term memory loss, and impaired learning, speech, and language functions (Barkley, Grodzinsky, & Dupaul, 1992; Bechara & Van der Linden, 2005; Fellows, 2006; Grafman et al., 1996; Pardini et al., 2011). Similarly, damage to the prefrontal cortex has been associated with violence, aggression, and antisocial behavior (Fabian, 2010; Martell, 1996; Raine, 2002), whereas damage to the temporal lobe is linked to unprovoked anger, memory problems, intellectual impairment, and the regulation of responses to cues that indicate threat (Fabian, 2010).

Despite the behavioral changes that can result from TBI, the causal relationship between TBI and crime is likely complex with multiple causal pathways. For example, several studies have found high co-occurring rates of TBI and substance abuse among juvenile and adult offenders (Barnfield & Leathem, 1998; Perron & Howard, 2008; Sacks et al., 2009; Walker, Hiller, Staton, & Leukefeld, 2003; Williams et al., 2010) as well as nonoffender adolescent populations (Ilie, Boak, Adlaf, Asbridge, & Cusimano, 2013). Also, a recent retrospective study among inmates found that TBI early in life was associated with earlier onset of substance abuse and the severity of drug use (Fishbein, Dariotis, Ferguson, & Pickelsimer, 2014). However, such studies are unable to determine a causal relationship, and so it is still unclear whether substance abuse is the result of TBI or, as other studies have suggested, substance abuse is what places one at a higher risk for TBI (Kolakowsky-Hayner et al., 2002; Vickery et al., 2008). A similar line of research which explores a potentially indirect mechanism in the relationship between TBI and crime focuses on mental illness following brain injury. These studies suggest that those with TBI, including both adolescents and adults, are more likely to have symptoms of a psychiatric disorder (e.g., attention deficit hyperactivity disorder [ADHD], bipolar disorder, major depression, panic disorder, depression) as well as increased aggressive behaviors at follow-up (Bloom et al., 2001; Cole et al., 2008; Dinn, Gansler, Moczynski, & Fulwiler, 2009; Gerring et al., 2009; Hesdorffer, Rauch, & Tamminga, 2009; Koponen, 2005; Max, Castillo, Lindgren, & Arndt, 1998; Max, Robertson, & Lansing, 2001; McKinlay, Grace, Horwood, Fergusson, & MacFarlane, 2009; Van Reekum, Bolago, Finlayson, Garner, & Links, 1996). Thus, having TBI might also result in symptoms of mental illness, substance abuse, or violence, which could exacerbate the likelihood of arrest.

TBI and Recidivism

The above studies have documented higher rates of TBI among inmates than the general population and established that TBIs are associated with behaviors that could lead to involvement in the criminal justice system (i.e., violence and substance abuse). However, there have been few prospective studies that examine the likelihood of criminal justice involvement following TBI. The only studies to prospectively look at arrest following TBI use Finnish birth cohort data and find that adolescents with TBI, based on hospital diagnosis, were significantly more likely to have been arrested than those without (Luukkainen, Riala, Laukkanen, Hakko, & Rasanen, 2012; Rantakallio, Koiranen, & Möttönen, 1992; Timonen et al., 2002). Thus, further research is needed to determine not only whether TBI is associated with the initial involvement in the criminal justice system but also whether TBI is associated with subsequent recidivism.

The present research seeks to advance the TBI crime literature by providing the first study to prospectively examine whether inmates with TBI are more likely to be rearrested post release than those without TBI. We used the OSU-TBI-ID—a well-established and validated instrument—to measure the presence of TBI among a cohort of male inmates and link this to follow-up data to assess rearrest post release. Although we are unable to determine whether TBI was associated with initial involvement in the criminal justice system, our study examines an equally important question regarding barriers to prisoners reentering their community, which is the association between TBI and subsequent involvement in the criminal justice system.

Method

Setting

The short version of the OSU-TBI-ID was employed as a screening instrument for all incoming male inmates in Indiana for approximately 1 month. This instrument was designed to provide both a valid and reliable method to assess lifetime exposure to TBI. Indeed, prior research has shown acceptable to high levels of reliability, including both interrater and test/retest reliability (Bogner & Corrigan, 2009; Corrigan & Bogner, 2007a, 2007b; Corrigan et al., 2012; Ferguson et al., 2012). Similar levels of predictive validity have been demonstrated by examining the relationship between lifetime history of TBI and effects that are commonly associated with TBI, including cognitive performance, interpersonal functioning, and violence and aggression (Bogner & Corrigan, 2009; Corrigan & Bogner, 2007a, 2007b; Corrigan et al., 2012; Ferguson et al., 2012). In addition, the OSU-TBI-ID is becoming a widely recognized valid instrument as evidenced by its adoption in a variety of research and clinical practices such as the “TBI Model Systems National Database Syllabus” (National Data and Statistical Center, 2016) and the National Institutes of Health’s “National Institute of Neurological Disorders and Stroke Common Data Elements” (NINDS; 2016). Thus, there is considerable evidence that the OSU-TBI-ID instrument is a reliable and valid method to assess lifetime exposure to TBI.

The instrument is administered through a structured interview lasting approximately 10 min. Individuals are asked to recall injuries involving the head or neck that resulted in altered consciousness or a loss of consciousness. For example, questions include the following: “Have you ever been hospitalized or treated in an emergency room following an injury to your head or neck?”; “Have you ever injured your head or neck in a car accident or from some other moving vehicle accident?”; “Have you ever injured your head or neck in a fall or from being hit by something?”; “Have you ever been injured playing sports or on the playground?”; “Have you ever injured your head or neck in a fight, from being hit by someone, or from being shaken violently?”; “Have you ever been shot in the head?”; and “Have you ever been nearby when an explosion or a blast occurred?” “If you served in the military, think about any combat-related incidents.” For each of the instances, the respondent is also asked how long they were unconscious and at what age the injury occurred. Based on the age of the first injury, number of injuries, severity (i.e., a loss of consciousness), and the length of time unconsciousness, the OSU-TBI-ID places respondents in one of five TBI categories: improbable, possible, mild, moderate, or severe. TBIs that resulted in losing consciousness (excluding, for example, loss of consciousness due to drug overdose) for greater than 30 min are classified as “severe,” while those resulting in periods of unconsciousness lasting less than 30 min are coded as “mild.”

For approximately 1 month, all male inmates entering the Indiana state prison system were screened for TBI at the Reception Diagnostic Center prior to incarceration. Intake specialists gather information on criminal history and medical needs. Based on this information, inmates are then placed into one of several Indiana Department of Corrections (hereafter IDOC) facilities. All intake specialists were trained on how to administer the short version of the OSU-TBI-ID. The instrument was then incorporated into the electronic IDOC intake data collection system from November 5, 2012, to December 3, 2012, in an attempt to determine the overall rate of TBI among 831 male inmates.

In this study, we examine data from a subsample of 155 who were released to Marion County (Indianapolis), Indiana, within 18 months (June 2014) following the OSU-TBI-ID intake survey. Indiana does not have a statewide jail database system so we were only able to capture reliable jail data from Marion County; therefore, for this subsample, we collected prior and follow-up data on all arrests that resulted in jail in Marion County. After listwise deletion of missing cases, our final sample included 151 inmates.

Analytic Procedure

Follow-up incarceration data for the subsample released by June 2014 (N = 151) were collected the following year in August 2015. Our sample, then, had a minimum of 12 months at risk for recidivism. However, most inmates were released earlier than that with the earliest being 2 months following intake screening. Thus, inmates’ risk period for rearrest ranged from 12 to 29 months with an average of 19.9 months (SD = 6.6). We examined time to recidivism using Cox regression survival analysis. One of the strengths of survival analysis techniques is the management of censored data: Left censoring occurs when data on the starting point are not available and right censoring when there are no data on the ending point, which often occurs when studying recidivism. In short, survival analysis methods are able to correct for the unequal distribution of follow-up time by using censored and uncensored (i.e., those that did recidivate) cases to calculate the probability of surviving (i.e., not recidivating) for each time point (Box-Steffensmeier & Jones, 2004). Therefore, it is able to examine both the likelihood of and time to recidivism while also being able to control for the effect of covariates on these outcomes. All analyses were conducted using IBM SPSS 21.

Measures

The dependent variable used in this study was recidivism, which was operationalized as any rearrest that resulted in jail, following discharge from the IDOC. The OSU-TBI-ID screening instrument provided data on the presence and severity of TBI (improbable, possible, mild, moderate, or severe), the likely age of the first TBI injury, and the source of the injury (accident, hit by an object, violence, or explosion). IDOC also provided sociodemographic information on offenders, including age at intake, race/ethnicity categories, and education level. Offense type (person, property, controlled substance, or other) and the presence of a psychiatric disorder (1 = yes, 0 = no) were also provided; unfortunately, additional information on diagnosis was not available nor were data on substance abuse. Inmates were given a mental health screening prior to being sent to a prison facility. This determined whether the inmate had a psychiatric disorder that required additional services and whether this disorder caused functional impairment at the time of intake. Data were only available on whether the inmate had a psychiatric disorder, not on the specific diagnosis. Finally, the criminal history data also provided a total number of lifetime arrests in Marion County as well as a total number of incarceration in IDOC.

Results

Sample Description

The descriptive characteristics are provided in Table 1. All of the participants in this study are males, and the average age at release was 31.2 (SD = 10.0). More than half (60.26%) of the population was identified as Black, 36.42% was White, and 3.1% was Hispanic. However, given the overall small sample size, the small number of individuals identified by IDOC as Hispanic, and well-documented race/ethnic disparities in the criminal justice system, we dichotomized the race/ethnicity categories into “White” and “race/ethnicity minority status” for subsequent analysis. Over half of the sample (53.64%) had either a high school (HS) diploma or General Educational Development (GED) at the time of intake, with 41.72% not having completed HS and 4.64% having some secondary education beyond HS; however, because of the small sample size for the purposes of statistical analysis, we dichotomized the education variable (1 = HS degree or more, 0 = no HS degree). The most common offense type among this cohort was property crimes (41.72%), followed by person crimes (24.50%), controlled substances (16.56%), and other (17.22%). Lifetime prior arrests ranged from 0 to 39 with an average of 6.56 (SD = 6.93). Approximately half (51.66%) of the sample had five or fewer arrests (11.26% had no prior arrest and 4.6% had one arrest). A psychiatric diagnosis was present in 14.57% of the population which is consistent with studies on the prevalence of mental illness among prisoners (Fazel & Danesh, 2002; Steadman, Osher, Robbins, Case, & Samuels, 2009). Finally, over half (53.64%) of the sample had an arrest that resulted in jail following release from prison.

Table 1:

Sample Characteristics (N = 151)

Variable	n	%	M	SD	Range
Age			31.2	10.0	18-60
Race/ethnicity
White	55	36.42
Black	91	60.26
Hispanic	5	3.31
Education
Less than HS education	63	41.72
HS or GED	81	53.64
Any postsecondary	7	4.64
Offense type
Person	37	24.50
Property	63	41.72
Controlled substance	25	16.56
Other	26	17.22
Psychiatric diagnosis
Yes	22	14.57
No	129	85.43
TBI
Improbable	97	64.24
Possible	5	3.31
Mild	35	23.18
Moderate	7	4.64
Severe	7	4.64
Prior jail bookings			6.56	6.93	0-39
Recidivism post exit
Yes	81	53.64
No	70	46.36

Note. HS = high school; GED = General Educational Development; TBI = traumatic brain injury.

TBI Status

The results of the OSU-TBI-ID screening instrument’s classifications are presented in Table 1 and show that approximately one third of the sample screened positive for TBI. Of those, 9.26% were coded as possible, 64.81% as mild, 12.96% as moderate, and 12.96% as severe. Among those who screened positive for TBI (n = 54), we were also able to examine the age, self-reported prior hospitalization, and indictors of the TBI injury from the OSU-TBI-ID instrument. The self-reported age of the first TBI incident ranged from 3 to 42 years old with an average age of 18.76 years (SD = 9.47); over a third (35.2%, n = 19) had the first TBI incident from age 14 or below. Over three quarters (77.80%, n = 42) reported that they had been hospitalized or seen in an emergency following an injury to the head or neck. For the sources of head- or neck-related TBIs, 45.10% reported being in a car accident; 46.30% reported being hit by something and/or injured playing sports; 33.30% reported being in a fight, being shaken, or shot; and 7.40% reported being in an explosion or blast. Although these were not mutually exclusive categories (respondents can identify multiple sources of injury), over two thirds (68.5%) identified only one, 22.2% identified two sources, and 9.26% three.

To examine potential differences in the sample by TBI status, we dichotomized the categories so that improbable indicated “No TBI” (64.2%, n = 97) and possible, mild, moderate, and severe indicated “TBI” (35.76%, n = 54). Means tests found no differences in TBI status by age, and chi-square tests found no significant differences by race/ethnicity minority status or education by TBI status. However, those with TBI were more likely to have a psychiatric diagnosis than those without TBI (24.1% and 9.3%, respectively, χ² = 6.10, p = .014), had a greater number of prior lifetime arrests than those without (8.04 [SD = 6.63] and 5.73 [SD = 6.99], t = 2.67, p = .047), and had significantly different offense types. Specifically, those with TBI were more likely to have been incarcerated on person offense type than those without TBI (42.6% and 14.4%, respectively, χ² = 21.24, p < .001).

Survival Analysis Predicting Recidivism

During the 12 to 30 months of the follow-up period, 46.36% (n = 70) of the sample remained out of the criminal justice system while slightly over half (53.64%, n = 81) of the sample was rearrested. The number of days to recidivism ranged from 9 to 700 days with an average of 221.70 days (SD = 166.75). Survival analyses were used to examine time to rearrest, which produces a life table that describes duration distributions for the full sample or by key variable levels. One way to describe the life table is to plot the cumulative proportion surviving at each time interval by a key covariate. Figure 1 shows the survival function for those with and without TBI, and illustrates that those without TBI went longer until recidivism event. For example, at 6 months, the cumulative proportion of those without TBI who had not yet recidivated (i.e., surviving) was 74.23% compared with 64.81% of those with TBI. Similarly, at 12 months, the survival rate for those without TBI was 62.89%, and among those with TBI, almost half (48.15%) had recidivated.

Figure 1:

Time to Rearrest Post Prison Release by TBI Status

To assess the effect of TBI on criminal recidivism, while controlling for covariates, we used Cox regression survival analysis. Table 2 shows the results of Cox regression models predicting the time to recidivism. Model 1 includes only the dichotomous TBI variable and suggests that those returning inmates with TBI have a hazard rate of recidivism that is 1.57 times greater than those who did not have TBI (Exp(B) = 0.79).² However, we want to assess whether there is an association between TBI and recidivism after controlling for other theoretically important predictors of recidivism. To do this, we included additional measures in subsequent Cox regression models. Model 2 examines the effect of TBI on recidivism while controlling for several sociodemographic variables, including age, race/ethnicity minority status, and education. The results suggest that TBI status and minority status are associated with recidivism. Net of other factors, the hazard of recidivism increased about 69% for those with TBI. Whites were at a decrease hazard of recidivism when compared with minorities (Exp(B) = 0.63).

Table 2:

Cox Regression Predicting Recidivism Post Exit

Variable	Model 1		Model 2		Model 3		Model 4
Variable	B (SE)	Exp(B) (95% CI)	B (SE)	Exp(B) (95% CI)	B (SE)	Exp(B) (95% CI)	B (SE)	Exp(B) (95% CI)
TBI (1 = yes)	0.45 (0.23)*	1.57 [1.01, 2.44]	0.52 (0.23)*	1.69 [1.07, 2.67]	0.64 (0.27)*	1.89 [1.12, 3.18]	0.61 (0.27)*	1.85 [1.08, 3.15]
Age			0.00 (0.01)	1.00 [0.97, 1.02]	−0.01 (0.01)	0.99 [0.97, 1.02]	−0.02 (0.01)	0.98 [0.96, 1.01]
Race/ethnicity (1 = White)			−0.47 (0.23)*	0.63 [0.40, 0.99]	−0.46 (0.24)*	0.63 [0.40, 1.00]	−0.54 (0.24)*	0.58 [0.36, 0.94]
Education (1 = HS degree or above)			−0.26 (0.25)	0.77 [0.47, 1.26]	−0.30 (0.26)	0.74 [0.45, 1.22]	−0.30 (0.26)	0.74 [0.44, 1.23]
Person (1 = yes)					−0.36 (0.45)	0.70 [0.29, 1.69]	−0.49 (0.45)	0.61 [0.25, 1.49]
Property (1 = yes)					−0.11 (0.40)	0.90 [0.41, 1.98]	−0.36 (0.41)	0.70 [0.31, 1.57]
Controlled (1 = yes)					−0.16 (0.46)	0.86 [0.35, 2.12]	−0.23 (0.47)	0.80 [0.32, 1.99]
Weapons (1 = yes)					−0.06 (0.57)	0.94 [0.31, 2.85]	−0.29 (0.57)	0.75 [0.24, 2.31]
Other (reference category)					—	—	—	—
Psychiatric diagnosis (1 = yes)							−0.31 (0.32)	0.74 [0.39, 1.39]
Number of prior jail bookings							0.05 (0.01)***	1.05 [1.02, 1.08]
−2 log likelihood χ²	755.37*		750.47*		749.49*		738.12**

Note. TBI group includes possible, mild, moderate, and severe. Race/ethnicity reference group includes Black and Hispanic. CI = confidence interval; TBI = traumatic brain injury; HS = high school.

p < .05. **p < .01. ***p < .001.

Model 3 includes four additional dichotomous variables measuring each offense type: person, property, and controlled substance, with other offenses as the reference group. Although none of the offense type measures are significant, both TBI status and the dichotomous race/ethnicity variables remain significant. Moreover, adding these variables again increased the effect size of the TBI status variable. Finally, Model 4 includes a dichotomous variable measuring the presence of a psychiatric diagnosis as well as the number of prior arrests. In this final model, TBI status, minority status, and prior arrests are all associated with recidivism. Model 4 suggests that among this sample, controlling for the other variables in the model, the hazard of recidivism increases by 5.2% for each additional prior arrest, is 1.71 times greater for race/ethnicity minority groups and 1.85 times greater for those with TBI.

Discussion

This study uses the OSU-TBI-ID to assess the relationship between TBI and recidivism among male offenders in Marion County, Indiana. Our findings suggest that those with TBI were more likely to recidivate post release than those without TBI. The results of our bivariate analysis indicate that those with TBI did not differ in terms of age, race/ethnicity minority status, or education. However, those with TBI did appear to be more likely to have a psychiatric diagnosis and a greater number of prior arrests than those without TBI. The majority of persons with TBI reported having been hospitalized or seen by a medical professional following their TBI. In addition, individuals with TBI experienced a larger number of arrests for person offenses. This is supported by prior research that suggests that brain injuries are associated with violence and aggression (e.g., Bufkin & Luttrell, 2005; Fabian, 2010; Martell, 1996; Raine, 2002). We then used survival analysis to adequately adjust for the right censored nature of our recidivism measure. Our results suggest that those with TBI were more likely to recidivate, and this relationship held when controlling for other theoretically relevant covariates, including age, race/ethnicity, education, and type of offense. Thus, those with TBI tended to experience an arrest sooner than those without TBI. In addition to TBI, race/ethnicity minority status and the total number of prior arrests were also associated with time to recidivism. These findings are consistent with much of the criminal justice research on recidivism. Prior criminal activity and involvement in the justice system are among the strongest predictors of future criminal behavior in the initial years following release (Gendreau, Little, & Goggin, 1996; Huebner & Berg, 2011; Langan & Levin, 2002; Ulmer, 2001); however, some studies suggest that the effect of prior arrests is no longer significant several years post release (Huebner & Berg, 2011; Kurlychek, Brame, & Bushway, 2007). Similarly, although research finds that recidivism is highest among African Americans (particularly males under the age of 18; see Beck & Shipley, 1989; Langan & Levin, 2002), more recent studies suggest that this pattern is the result of the social context and structural conditions (i.e., racial inequality and poverty rate) to which African American ex-prisoners return (Kaufman, 2005; Kubrin & Stewart, 2006; Reisig, Bales, Hay, & Wang, 2007; Visher & Travis, 2003). Unfortunately, the administrative data used in this study were limited and we were unable to look at long-term recidivism patterns or the social contextual factors to which inmates returned; however, it is important to note the potential explanations for these results.

Our study has several strengths over existing research investigating the association between TBI and involvement in the criminal justice system. First, we use prospective data which allow us to establish the temporal precedence of the variables. Prior research investigating the relationship between TBI and crime has largely been descriptive or retrospective, and compared rates of TBI in the offender population with rates of TBI in the general population. In this study, we were able to first assess the presence of TBI and then determine postincarceration recidivism. However, we were not able to determine whether the TBI occurred prior to their first criminal justice experience. In fact, those with TBI had a greater number of prior lifetime arrests than those without. Although it is possible that prior arrests is also a driving factor in the TBI and recidivism, we found that even when controlling for prior arrests in a multivariate model the presence of TBI was still associated with recidivism. Second, inmates in this study were screened using the OSU-TBI-ID, a validated clinical screening instrument (Bogner & Corrigan, 2009). Prior studies have typically used nonstandardized methods to obtain information on TBIs. Moreover, the OSU-TBI-ID was used as a screening instrument among all incoming inmates in this study. Thus, our study may be more generalizable to the incarcerated population given that much of the prior research has only examined TBI in specific offense types or used convenience samples. Finally, many of the prior studies failed to control for other well-known predictors of criminal behavior, including age, race/ethnicity, and prior criminal behavior.

Although there have been virtually no prospective studies examining the relationship between TBI and involvement in the criminal justice system, our study is consistent with many retrospective, descriptive studies (Barnfield & Leathem, 1998; Langevin, 2006; Ray, Sapp, & Kincaid, 2014; Shiroma et al., 2012) though adds to this body of literature by suggesting that there is a relationship between having TBI while incarcerated and involvement in the criminal justice system following release. However, it is important to note that this association could operate through several mechanisms not measured in this study. For example, recent biological theories have shown disruptions in key areas of the brain posthead injury are responsible for impulse control, the regulation of emotions, and planning and judgment (Raine, 2014). TBI can lead to disruptions in executive functioning such as impulse control affecting levels of self-control, which is a consistent predictor of antisocial behavior and crime (Gottfredson & Hirschi, 1990), or attachments that restrain individuals from engaging in criminal behavior (Hirschi, 1969; Sampson & Laub, 1993), neither of which were measured in the present study. It is also possible that those with TBI become involved in the criminal justice system, not directly because of their TBI but because their social background and TBI serve as barriers to prosocial activities with family and friends. This may lead to difficulties finding sufficient employment or may lead them to self-medicate with alcohol and drugs. On reentering the community, many inmates are ill-prepared for life outside an institution and too often return to disorganized, high-crime neighborhoods lacking supports for housing, employment, social activities, and mental health and substance abuse treatment where they resume antisocial behavior patterns (see Fisher et al., 2014; Hartwell, 2004). Moreover, it is also possible that offending patterns are similar across groups, but those without TBI are better at concealing crimes than those with TBI.

Although the current study fills an important gap in the literature, it is not without limitations. First, we do not have postrelease supervision data on the inmates. In other words, we do not know whether the offenders in our analysis had court-ordered supervisions upon their release from IDOC (i.e., parole), which would affect postrelease criminal behavior and the discovery of those behaviors. For those under supervision, their criminal behavior may be more easily discovered. Second, the sample used in this analysis only consisted of males who returned to Marion County, Indiana. Moreover, our measure of recidivism extended only to this county; while this might adjust for potential regional variations in recidivism, it limits the generalizability of our findings. Extending this study beyond Indiana and including female offenders could increase the generalizability of the findings. Third, while the OSU-TBI instrument is a reliable instrument, there is a potential for recall bias as it relies on retrospective reporting of TBIs from the inmates during the intake screening. There is also concern that inmates may self-report TBIs more or less often than those in the general population. However, recent research shows the validity of TBI self-reports to be fairly consistent when comparing self-reports in the incarcerated population with the general population (Bogner & Corrigan, 2009). Last, we were unable to directly test potential underlying causal mechanisms linking TBI to initial involvement in the criminal justice system or subsequent recidivism. For example, we were unable to measure personality traits or disorders (such as substance abuse, psychopathic traits, or other personality disorders) that might be associated with an increased risk of TBI, arrest, and recidivism. Future research should attempt to examine these specific mechanisms longitudinally to determine whether there is a direct or indirect relationship between TBIs and self-reported criminal behaviors as well as involvement in the criminal justice system.

Conclusion

Only recently have researchers started to examine TBI within a criminal justice context as studies consistently showed higher rates of TBI within incarcerated populations than the general population. Although the results of this study suggest a link between TBI and rearrest following incarceration, perhaps more importantly it highlights both the need and manner by which criminal justice institutions can screen for the presence of TBI. The short version of the OSU-TBI-ID can be easily implemented into existing screening instruments to detect the presence of TBI among inmates. Once inmates are identified, they can be placed into programs that address their individual needs and develop proper intervention strategies. If inmates get the appropriate treatment for brain injuries while incarcerated, they may be less likely to engage in criminal behavior after release. Moreover, detecting TBI as early as possible in the criminal justice process could reduce the burden of TBI on initial and repeat offending.

Footnotes

We would like to thank Brooke Hiltbold, Spencer Lawson, and Jennifer Lutz for their help in collecting and preparing the data for analysis.

Notes

Bradley Ray is an assistant professor in the School of Public and Environmental Affairs at Indiana University–Purdue University, Indianapolis. His research focuses on mental illness and substance use disorder, particularly where these populations connect with the criminal justice system. He also examines policies and interventions at the intersection of behavioral health and criminal justice systems.

Nicholas J. Richardson is a research analyst in the Policing Science and Investigative Services Program in the Center for Justice, Safety, and Resilience at RTI International. His primary research interests include substance abuse and addiction as well as examining the role of biology in predicting violence and aggression.

References

Barkley

R. A.

Grodzinsky

Dupaul

G. J.

(1992). Frontal lobe functions in attention deficit disorder with and without hyperactivity: A review and research report. Journal of Abnormal Child Psychology, 20, 163-188.

Barnfield

T. V.

Leathem

J. M.

(1998). Incidence and outcomes of traumatic brain injury and substance abuse in a New Zealand prison population. Brain Injury, 12, 455-466.

Bechara

Van der Linden

(2005). Decision-making and impulse control after frontal lobe injuries. Current Opinion in Neurology, 18, 734-739.

Beck

Shipley

(1989). Recidivism of prisoners released in 1983. Washington, DC: Bureau of Justice Statistics.

Blake

P. Y.

Pincus

J. H.

Buckner

(1995). Neurologic abnormalities in murderers. Neurology, 45, 1641-1647.

Bloom

D. R.

Levin

H. S.

Ewing-Cobbs

Saunders

A. E.

Song

Fletcher

J. M.

Kowatch

R. A.

(2001). Lifetime and novel psychiatric disorders after pediatric traumatic brain injury. Journal of the American Academy of Child & Adolescent Psychiatry, 40, 572-579.

Bogner

Corrigan

J. D.

(2009). Reliability and predictive validity of the Ohio State University TBI Identification method with prisoners. Journal of Head Trauma Rehabilitation, 24, 279-291.

Box-Steffensmeier

J. M.

Jones

B. S.

(2004). Event history modeling: A guide for social scientists. New York, NY: Cambridge University Press.

Bufkin

Luttrell

(2005). Neuroimaging studies of aggressive and violent behavior: Current findings and implications for criminology and criminal justice. Trauma Violence Abuse, 6, 176-191.

10.

Chayer

Freedman

(2001). Frontal lobe functions. Current Neurology & Neuroscience Reports, 1, 547-552.

11.

Cole

W. R.

Gerring

J. P.

Gray

R. M.

Vasa

R. A.

Salorio

C. F.

Grados

. . . Slomine

B. S.

(2008). Prevalence of aggressive behaviour after severe paediatric traumatic brain injury. Brain Injury, 22, 932-939.

12.

Corrigan

J. D.

Bogner

(2007a). Initial reliability and validity of the Ohio State University TBI Identification method. Journal of Head Trauma Rehabilitation, 22, 318-329.

13.

Corrigan

J. D.

Bogner

(2007b). Screening and identification of TBI. Journal of Head Trauma Rehabilitation, 22, 315-317.

14.

Corrigan

J. D.

Bogner

Holloman

(2012). Lifetime history of traumatic brain injury among persons with substance use disorders. Brain Injury, 26, 139-150.

15.

DelBello

M. P.

Soutullo

C. A.

Zimmerman

M. E.

Sax

K. W.

Williams

J. R.

McElroy

S. L.

Strakowski

S. M.

(1999). Traumatic brain injury in individuals convicted of sexual offenses with and without bipolar disorder. Psychiatry Research, 89, 281-286.

16.

Diamond

P. M.

Harzke

A. J.

Magaletta

P. R.

Cummins

A. G.

Frankowski

(2007). Screening for traumatic brain injury in an offender sample: A first look at the reliability and validity of the Traumatic Brain Injury Questionnaire. Journal of Head Trauma Rehabilitation, 22, 330-338.

17.

Dinn

W. M.

Gansler

D. A.

Moczynski

Fulwiler

(2009). Brain dysfunction and community violence in patients with major mental illness. Criminal Justice and Behavior, 36, 117-136.

18.

Eghwrudjakpor

P. O.

Essien

A. A.

(2008). Hypersexual behavior following craniocerebral trauma: An experience with five cases. Libyan Journal of Medicine, 3, 192-194.

19.

Fabian

J. M.

(2010). Neuropsychological and neurological correlates in violent and homicidal offenders: A legal and neuroscience perspective. Aggression and Violent Behavior, 15, 209-223.

20.

Farrer

T. J.

Frost

R. B.

Hedges

D. W.

(2013). Prevalence of traumatic brain injury in juvenile offenders: A meta-analysis. Child Neuropsychology, 19, 225-234.

21.

Farrer

T. J.

Hedges

D. W.

(2011). Prevalence of traumatic brain injury in incarcerated groups compared to the general population: A meta-analysis. Progress in Neuro-Psychopharmacology & Biological Psychiatry, 35, 390-394.

22.

Faul

Wald

M. M.

Coronado

V. G.

(2010). Traumatic brain injury in the United States: Emergency department visits, hospitalizations and deaths 2002-2006. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control.

23.

Fazel

Danesh

(2002). Serious mental disorder in 23,000 prisoners: A systematic review of 62 surveys. The Lancet, 359, 545-550.

24.

Fazel

Lichtenstein

Grann

Langstrom

(2011). Risk of violent crime in individuals with epilepsy and traumatic brain injury: A 35-year Swedish population study. PLoS Medicine, 8(12), e1001150. doi:10.1371/journal.pmed.1001150

25.

Fellows

L. K.

(2006). Deciding how to decide: Ventromedial frontal lobe damage affects information acquisition in multi-attribute decision making. Brain, 129, 944-952.

26.

Ferguson

P. L.

Pickelsimer

E. E.

Corrigan

J. D.

Bogner

J. A.

Wald

(2012). Prevalence of traumatic brain injury among prisoners in South Carolina. Journal of Head Trauma Rehabilitation, 27(3), E11-E20.

27.

Fishbein

Dariotis

J. K.

Ferguson

Pickelsimer

E. E.

(2014). Relationships between traumatic brain injury and illicit drug use and their association with aggression in inmates. International Journal of Offender Therapy and Comparative Criminology, 60, 65-77.

28.

Fisher

W. H.

Hartwell

S. W.

Deng

X. G.

Pinals

D. A.

Fulwiler

Roy-Bujnowski

(2014). Recidivism among released state prison inmates who received mental health treatment while incarcerated. Crime & Delinquency, 60, 811-832.

29.

Freedman

Hemenway

(2000). Precursors of lethal violence: A death row sample. Social Science & Medicine, 50, 1757-1770.

30.

Frost

R. B.

Farrer

T. J.

Primosch

Hedges

D. W.

(2013). Prevalence of traumatic brain injury in the general adult population: A meta-analysis. Neuroepidemiology, 40, 154-159.

31.

Gendreau

Little

Goggin

(1996). A meta-analysis of the predictors of adult offender recidivism: What works! Criminology, 34, 575-607.

32.

Gerring

J. P.

Grados

M. A.

Slomine

Christensen

J. R.

Salorio

C. F.

Cole

W. R.

Vasa

R. A.

(2009). Disruptive behaviour disorders and disruptive symptoms after severe paediatric traumatic brain injury. Brain Injury, 23, 944-955.

33.

Goldstein

(1990). Traumatic brain injury: A silent epidemic. Annals of Neurology, 27, 327.

34.

Gottfredson

M. R.

Hirschi

(1990). A general theory of crime. Stanford, CA: Stanford University Press.

35.

Grafman

Schwab

Warden

Pridgen

Brown

H. R.

Salazar

A. M.

(1996). Frontal lobe injuries, violence, and aggression: A report of the Vietnam Head Injury Study. Neurology, 46, 1231-1238.

36.

Hartwell

S. W.

(2004). Comparison of offenders with mental illness only and offenders with dual diagnoses. Psychiatric Services, 55, 145-150.

37.

Hawley

C. A.

Maden

(2003). Mentally disordered offenders with a history of previous head injury: Are they more difficult to discharge? Brain Injury, 17, 743-758.

38.

Hesdorffer

D. C.

Rauch

S. L.

Tamminga

C. A.

(2009). Long-term psychiatric outcomes following traumatic brain injury: A review of the literature. Journal of Head Trauma Rehabilitation, 24, 452-459.

39.

Hirschi

(1969). Causes of delinquency. Berkeley: University of California Press.

40.

Huebner

B. M.

Berg

M. T.

(2011). Examining the sources of variation in risk for recidivism. Justice Quarterly, 28, 146-173.

41.

Ilie

Boak

Adlaf

E. M.

Asbridge

Cusimano

M. D.

(2013). Prevalence and correlates of traumatic brain injuries among adolescents. Journal of the American Medical Association, 309, 2550-2552.

42.

Kaufman

(2005). Explaining the race/ethnicity-violence relationship: Neighborhood context and social psychological processes. Justice Quarterly, 22, 224-251.

43.

Kolakowsky-Hayner

Gourley

E. V.

Kreutzer

J. S.

Marwitz

J. H.

Meade

M. A.

Cifu

D. X.

(2002). Post-injury substance abuse among persons with brain injury and persons with spinal cord injury. Brain Injury, 16, 583-592.

44.

Koponen

(2005). Personality disorders in patients with traumatic brain injury. European Psychiatry, 20, S203-S204.

45.

Kubrin

C. E.

Stewart

E. A.

(2006). Predicting who reoffends: The neglected role of neighborhood context in recidivism studies. Criminology, 44, 165-198.

46.

Kurlychek

M. C.

Brame

Bushway

S. D.

(2007). Enduring risk? Old criminal records and predictions of future criminal involvement. Crime & Delinquency, 53, 64-83.

47.

Langan

Levin

(2002). Recidivism of prisoners released in 1994. Washington, DC: Bureau of Justice Statistics.

48.

Langevin

(2006). Sexual offenses and traumatic brain injury. Brain and Cognition, 60, 206-207.

49.

Leon-Carrion

Ramos

(2003). Blows to the head during development can predispose to violent criminal behaviour: Rehabilitation of consequences of head injury is a measure for crime prevention. Brain Injury, 17, 207-216.

50.

Lewis

D. O.

Pincus

J. H.

Feldman

Jackson

Bard

(1986). Psychiatric, neurological, and psychoeducational characteristics of 15 death row inmates in the United States. American Journal of Psychiatry, 143, 838-845.

51.

Luukkainen

Riala

Laukkanen

Hakko

Rasanen

(2012). Association of traumatic brain injury with criminality in adolescent psychiatric inpatients from Northern Finland. Psychiatry Research, 200, 767-772.

52.

Martell

D. A.

(1992). Estimating the prevalence of organic brain dysfunction in maximum-security forensic psychiatric patients. Journal of Forensic Sciences, 37, 878-893.

53.

Martell

D. A.

(1996). Organic brain dysfunctions and criminality. In Schlesinger

L. B.

(Ed.), Explorations in criminal psychopathology: Clinical syndromes with forensic implications (pp. 66-84). Springfield, IL: Charles C Thomas.

54.

Max

J. E.

Castillo

C. S.

Lindgren

S. D.

Arndt

(1998). The neuropsychiatric rating schedule: Reliability and validity. Journal of the American Academy of Child & Adolescent Psychiatry, 37, 297-304.

55.

Max

J. E.

Robertson

B. A. M.

Lansing

A. E.

(2001). The phenomenology of personality change due to traumatic brain injury in children and adolescents. Journal of Neuropsychiatry and Clinical Neurosciences, 13, 161-170.

56.

McKinlay

Grace

Horwood

Fergusson

MacFarlane

(2009). Adolescent psychiatric symptoms following preschool childhood mild traumatic brain injury: Evidence from a birth cohort. Journal of Head Trauma Rehabilitation, 24, 221-227.

57.

Menon

D. K.

Schwab

Wright

D. W.

Maas

A. I.

Demographics and Clinical Assessment Working Group of the International and Interagency Initiative Toward Common Data Elements for Research on Traumatic Brain Injury and Psychological Health. (2010). Position statement: Definition of traumatic brain injury. Archives of Physical Medicine and Rehabilitation, 91, 1637-1640.

58.

National Data and Statistical Center. (2016). Traumatic brain injury model systems. Englewood, CO: Craig Hospital Research Department. Available from https://www.tbindsc.org

59.

National Institute of Neurological Disorders and Stroke Common Data Elements. (2016). Common data elements. Washington, DC: Department of Health and Human Services, National Institutes of Health. Retrieved from https://www.commondataelements.ninds.nih.gov/#page=Default

60.

Pardini

Krueger

Hodgkinson

Raymont

Ferrier

Goldman

. . . Grafman

(2011). Prefrontal cortex lesions and MAO-A modulate aggression in penetrating traumatic brain injury. Neurology, 76, 1038-1045.

61.

Perron

B. E.

Howard

M. O.

(2008). Prevalence and correlates of traumatic brain injury among delinquent youths. Criminal Behavior and Mental Health, 18, 243-255.

62.

Raine

(2002). Biosocial studies of antisocial and violent behavior in children and adults: A review. Journal of Abnormal Child Psychology, 30, 311-326.

63.

Raine

(2014). The anatomy of violence: The biological roots of crime. New York, NY: Pantheon/Random House.

64.

Rantakallio

Koiranen

Möttönen

(1992). Association of perinatal events, epilepsy, and central nervous system trauma with juvenile delinquency. Archives of Disease in Childhood, 67, 1459-1461.

65.

Ray

Sapp

Kincaid

(2014). Traumatic brain injury among Indiana state prisoners. Journal of Forensic Sciences, 59, 1248-1253.

66.

Reisig

M. D.

Bales

W. D.

Hay

Wang

(2007). The effect of racial inequality on black male recidivism. Justice Quarterly, 24, 408-434.

67.

Sacks

A. L.

Fenske

C. L.

Gordon

W. A.

Hibbard

M. R.

Perez

Brandau

. . . Spielman

L. A.

(2009). Co-morbidity of substance abuse and traumatic brain injury. Journal of Dual Diagnosis, 5, 404-417.

68.

Sampson

R. J.

Laub

J. H.

(1993). Crime in the making: Pathways and turning points through life. Cambridge, MA: Harvard University Press.

69.

Shiroma

E. J.

Ferguson

P. L.

Pickelsimer

E. E.

(2012). Prevalence of traumatic brain injury in an offender population: A meta-analysis. Journal of Head Trauma Rehabilitation, 27(3), E1-E10.

70.

Silver

J. M.

Kramer

Greenwald

Weissman

(2001). The association between head injuries and psychiatric disorders: Findings from the New Haven NIMH Epidemiologic Catchment Area Study. Brain Injury, 15, 935-945.

71.

Steadman

H. J.

Osher

Robbins

Case

Samuels

(2009). Prevalence of serious mental illness among jail inmates. Psychiatric Services, 60, 761-765.

72.

Timonen

Miettunen

Hakko

Zitting

Veijola

von Wendt

Räsänen

(2002). The association of preceding traumatic brain injury with mental disorders, alcoholism and criminality: The Northern Finland 1966 Birth Cohort Study. Psychiatry Research, 113, 217-226.

73.

Turkstra

Jones

Toler

H. L.

(2003). Brain injury and violent crime. Brain Injury, 17, 39-47.

74.

Ulmer

J. T.

(2001). Intermediate sanctions: A comparative analysis of the probability and severity of recidivism. Sociological Inquiry, 71, 164-193.

75.

Van Reekum

Bolago

Finlayson

M. A. J.

Garner

Links

P. S

. (1996). Psychiatric disorders after traumatic brain injury. Brain Injury, 10, 319-327.

76.

Vickery

C. D.

Sherer

Nick

T. G.

Nakase-Richardson

Corrigan

J. D.

Hammond

. . . Sander

(2008). Relationships among premorbid alcohol use, acute intoxication, and early functional status after traumatic brain injury. Archives of Physical Medicine and Rehabilitation, 89, 48-55.

77.

Visher

Travis

(2003). Transitions from prison to community: Understanding individual pathways. Annual Review of Sociology, 29, 89-113.

78.

Walker

Hiller

Staton

Leukefeld

C. G.

(2003). Head injury among drug abusers: An indicator of co-occurring problems. Journal of Psychoactive Drugs, 35, 343-353.

79.

Walker

Staton

Leukefeld

C. G.

(2001). History of head injury among substance users: Preliminary findings. Substance Use & Misuse, 36, 757-770.

80.

Williams

W. H.

Cordan

Mewse

A. J.

Tonks

Burgess

C. N. W.

(2010). Self-reported traumatic brain injury in male young offenders: A risk factor for re-offending, poor mental health and violence? Neuropsychological Rehabilitation, 20, 801-812.