Elsevier

Journal of Criminal Justice

Volume 53, November 2017, Pages 34-45
Journal of Criminal Justice

A little early risk goes a long bad way: Adverse childhood experiences and life-course offending in the Cambridge study

https://doi.org/10.1016/j.jcrimjus.2017.09.005Get rights and content

Highlights

  • The impact of Adverse Childhood Experiences were assessed in a unique sample.

  • Protective factors that would buffer the impact of ACEs on crime were also studied.

  • Several protective factors were identified that decreased crime through adulthood.

  • These protective factors should be targeted in offending intervention programs.

Abstract

Purpose

While Adverse Childhood Experiences (ACEs) have been found to predict an increased prevalence and seriousness of offending, these findings are based on a sample from one U.S. state. To increase the generalizability of these findings, the impact of ACEs was investigated using a geographically-distinct sample. The current study also sought to identify potential protective factors that may buffer the impact of ACEs on offending.

Methods

Using the Cambridge Study in Delinquent Development, the prevalence and impact of ACEs on offending through age 56 were investigated. In order to identify protective factors, a large set of early childhood measures were examined to assess the degree to which they decreased offending among those with ACEs.

Results

Similar to prior studies, ACEs were found to increase the likelihood of offending throughout the life course. Across two operationalizations of risk, a number of protective factors were identified including low troublesomeness, low daring, and low hyperactivity.

Conclusions

Though prior research has focused on identifying offending risk factors, equally important is the identification of protective factors. This comprehensive approach allows interventions to strengthen these factors in order to buffer the deleterious impact of ACEs on offending.

Introduction

One of the key features underlying many developmental/life-course criminology perspectives is the extent to which risk factors, such as low socioeconomic status and poor parental supervision, affect delinquency and criminal behavior over the life course (see Farrington, 2003; Loeber & Farrington, 1998). One strategy has been to focus attention on a larger, more expansive set of negative experiences that have been found to be associated with an increase in the likelihood of offending as well as a wider variety of other negative life outcomes—a problem that becomes magnified when the exposure to such experiences is frequent and cumulative (see e.g., Agnew, 1992). These negative experiences, which have been referred to as Adverse Childhood Experiences (ACEs), were initially identified by Felitti et al. (1998) as ten distinct events before age 18 that were found to predict a high prevalence of poor health outcomes. Given the relationship between poor health outcomes and antisocial behavior more generally (Moffitt et al., 2011; Pajer, 1998; Piquero, Daigle, Gibson, Piquero, & Tibbetts, 2007; Piquero, Shepherd, Shepherd, & Farrington, 2011; Reingle, Jennings, Piquero, & Maldonado-Molina, 2014; Vaughn, Salas-Wright, DeLisi, & Piquero, 2014), it is no surprise that researchers have linked ACEs to criminal offending, especially serious and chronic offending (Fox, Perez, Cass, Baglivio & Epps, 2015) and victimization in the form of human trafficking (Reid, Baglivio, Piquero, Greenwald, & Epps, 2017).

However, much of this research on the relationship of ACEs to offending has centered on studies based on samples of adjudicated delinquents from the state of Florida (Baglivio & Epps, 2015; Craig, Baglivio, Wolff, Piquero, & Epps, 2017; Fox, Perez, Cass, Baglivio, & Epps, 2015; Wolff, Baglivio, & Piquero, 2015). As a result, the generalizability of the relationship between ACEs and offending to non-justice involved youth remains an important and little-investigated research question. Accordingly, the goal of the current study is to assess the generalizability of this relationship by investigating the prevalence and impact of ACEs among a sample that is different not only in geography but also in temporal context. Additionally, as developmental criminologists have also argued for the importance of identifying potential protective factors that may buffer the impact of risk factors on later offending (see Ttofi, Farrington, Piquero, & DeLisi, 2016), an additional goal of the study is to identify factors that may buffer the relationship between ACEs and later offending.

Prior to presenting the results of the current study, we first offer an overview of the impact that ACEs have on later offending. Following that material, we will review the literature on protective factors for criminal behavior, and close with a focus on potential protective factors for ACEs.

As described by Felitti et al. (1998), the ten negative childhood events found to be positively related to later chronic disease among a sample of privately-insured adults were: (1) physical abuse, (2) emotional abuse, (3) sexual abuse, (4) physical neglect, (5) emotional neglect, (6) household substance abuse, (7) violent treatment towards mother, (8) parental separation or divorce, (9) household mental illness, and (10) having a household member incarcerated. Each ACE event is measured dichotomously, so that an individual's ACE score can range from 0 to 10 and represents the cumulative number of exposures the individual has experienced. For instance, if an individual ever experienced emotional abuse prior to turning eighteen years old, then they were given a score of 1 for that ACE regardless of the number of times this form of abuse occurred. Thus, individuals with higher scores have been exposed to more distinct types of ACEs.

It is important to note that scholars have reported ACEs to be highly interrelated and have strong cumulative effects on an individual's brain development (Anda et al., 2006; Anda, Butchart, Felitti, & Brown, 2010; Cicchetti, 2013; Teicher et al., 2003). This underscores the importance of taking a “cumulative stressor approach”, where one considers the summative impact of ACEs as opposed to each exposure in isolation (Anda et al., 2006; Anda et al., 2010; Baglivio & Epps, 2015).

Recent analyses using data of adjudicated delinquents from the Florida Department of Juvenile Justice (FL DJJ) have revealed that higher exposure to ACEs are associated with an increased risk of rearrest and a decreased length of time to rearrest (Wolff et al., 2015). Youth with more ACEs have also been found to have a higher likelihood of becoming serious, violent, and chronic juvenile offenders (Fox et al., 2015) as well as having a higher likelihood of being diagnosed as having oppositional defiant disorder and ADHD (among males) (Baglivio, Wolff, DeLisi, Vaughn, & Piquero, 2016). Additionally, higher ACE exposure was found to distinguish both early-onset and chronic offending trajectories from other offending patterns (Baglivio, Wolff, Piquero, & Epps, 2015).

ACEs have also been found to be more prevalent among those at risk for encountering the criminal justice system. Analyses of the FL DJJ adjudicated delinquents found that these individuals were more likely to have higher ACE prevalence rates as well as higher ACE scores than the original ACE Study participants, consisting of privately-insured adults (Baglivio & Epps, 2015). Further analyses also revealed that children from disadvantaged environments were more likely to experience ACEs than their counterparts in affluent neighborhoods (Baglivio, Wolff, Epps, & Nelson, 2015). In sum, these findings highlight the importance of not only continuing the study of the impact of ACEs with a different sample but also investigating potential protective factors that may help prevent or otherwise buffer the negative impact of ACEs on later outcomes.

Our focus in the current study is centered largely on the relevant material related to ACEs within a risk factor paradigm. Criminologists, of course, will likely observe that many of the ACEs originally identified and assessed in the empirical literature are common stressors that Agnew (1992) has identified as being relevant in his General Strain Theory (GST). Although we do not consider our work to be a test of GST, it is important to bear in mind that ACEs occupy much common ground within the theory.

In the field of criminology, the 1990s brought with it a new focus to identify risk factors related to criminal behavior and then gear intervention and prevention efforts towards these factors (Farrington, 2000). Though risk factors are more commonly studied, factors that may protect an individual from later antisocial behavior are beginning to receive more empirical attention (Farrington, Ttofi, & Piquero, 2016). Not only is this important to increase our understanding of the etiology of criminal behavior, but it also has important policy implications as interventions can seek to bolster protective factors as well as reduce risk factors. Further, there have been recent efforts to move away from strictly risk-based interventions and focus more on those that promote an individual's or community's strengths, a goal necessitating the identification of protective factors (Jolliffe, Farrington, Loeber, & Pardini, 2016; MacKinnon-Lewis, Kaufman, & Frabutt, 2002).

Unfortunately, the term ‘protective factor’ has been used ambiguously in the literature leading some scholars to argue that it is simply the opposite of a risk factor (White, Moffitt, & Silva, 1989) while others define it as a variable that moderates the impact of a risk factor (Rutter, 1985). In order to clarify this ambiguity, the current study utilizes the definition offered by Loeber, Farrington, Stouthamer-Loeber, & White, 2008; see also Farrington et al., 2016 and Jolliffe et al., 2016, for similar applications. Inspired by Sameroff, Bartko, Baldwin, Baldwin, and Seifer (1998), a promotive factor is defined as a factor that is at the positive end of the risk dimension. In other words, being on the promotive end of a particular variable would predict a low probability of offending. A variable is considered mixed when it has both promotive and risk effects. A risk-based protective factor, on the other hand, is a variable that is promotive in situations of risk. In short, a variable that moderates the impact of a risk factor on crime is a protective factor and a variable that on its own predicts a low probability of offending is promotive factor. Based upon these definitions, the current study will focus on protective factors specifically as its goal is to identify factors that would decrease the harmful effects of ACEs on offending. While promotive factors are an important area for further research, they are beyond the scope of the current study.

Scholars from developmental criminology and related disciplines have identified several potential protective factors among youth considered at-risk (though not necessarily based on ACE scores). These can generally be categorized into one of several domains. At the individual level, several factors such as low hyperactivity, being shy/withdrawn, low extraversion, high nonverbal IQ, high verbal IQ, and low neuroticism have been identified as protective against offending among at-risk juveniles (Farrington et al., 2016; Farrington, Gallagher, Morley, St Ledger, & West, 1988). Having good parental supervision, good quality housing, high family income, low parental stress, parents with attitudes against antisocial behavior, parental harmony, a mother with a full-time job, good maternal discipline, small family size, and parents with a high interest in the youth's education have been identified as family-level protective factors (Farrington & Ttofi, 2011; Fontaine, Brendgen, Vitaro, & Tremblay, 2016; Kim, Gilman, Hill, & Hawkins, 2016; Stouthamer-Loeber, Loeber, Stallings, & Lacourse, 2008). School-level protective factors include exhibiting a strong school commitment and high educational attainment (Farrington et al., 2016; Herrenkohl, Tajima, Whitney, & Huang, 2005; Jennings et al., 2016; Jolliffe et al., 2016; Kim et al., 2016). At the peer-level, having peers that do not hold antisocial attitudes as well as having very few friends have been reported as important protective factors (Farrington et al., 1988; Farrington & Ttofi, 2011; Herrenkohl et al., 2005). Finally, community-level protective factors include having strong social support, strong attachment to others, perceived legitimacy of authority figures, and religious participation (Fontaine et al., 2016; Herrenkohl et al., 2005; Kim et al., 2016; Lodewijks, de Ruiter, & Doreleijers, 2010).

Given its wide variety of early childhood measures, the Cambridge Study in Delinquent Development (CSDD) has often been used to identify protective factors. Among juveniles considered to be at risk for later criminal behavior based on early childhood factors such as low family income, large family size, and low IQ, having few or no friends, being shy or withdrawn, having siblings without behavioral issues, and having parents that were never convicted were found to be protective against offending up to age 32 (Farrington et al., 1988). In a later analysis predicting protective factors against convictions up to age 50, Farrington and Ttofi (2011) considered two specific criminogenic risk factors: being troublesome and coming from poor housing. Low extraversion, low neuroticism, having few or no friends, living in a home characterized by parental harmony, a mom with a full-time job, and good parental supervision were found to be protective among troublesome boys (Farrington & Ttofi, 2011). Interestingly, these factors were not the same when poor housing was considered. Instead, being rated low in impulsivity, having good maternal discipline, a small family size, good paternal discipline, having parents with a high interest in the boy's education, and having parents that exhibited good child-rearing were found to be protective against later convictions.

More recent analyses investigated potential protective factors against youth offending among the CSDD (Farrington et al., 2016). Similar to the study by Farrington and Ttofi (2011), these scholars based offending risk on the boys being considered troublesome or having a convicted parent. While readers may note that having a convicted parent is related to one of the ACEs (having an incarcerated household member), the study did not consider this within the ACE framework specifically. Further, the study only considered delinquent offending, not offending throughout the life course. The scholars reported that at-risk boys with high nonverbal IQ, high verbal IQ, high educational attainment, coming from a family with a high parental interest in education, good parental supervision, and high family income were less likely to engage in delinquency than other boys.

To the best of our knowledge, there has only been two empirical studies to investigate potential moderating effects on the ACEs-crime relationship specifically. Craig et al. (2017) assessed how social bonds, measured as attachment to prosocial others, may buffer the negative relation of ACEs to recidivism. Using a sample of adjudicated delinquents from the Florida Department of Juvenile Justice (FL DJJ), it was found that while those with stronger social bonds were less likely to recidivate than their counterparts, social bonds did not moderate the impact of ACEs on recidivism. Regardless of social bonds, those with more ACEs were more likely to recidivate than those with fewer such exposures. Most recently, Craig, Intravia, Wolff, and Baglivio (2017) examined how substance non-use buffered the impact of ACEs on crime among the FL DJJ data. While ACEs were more likely to increase the chances of recidivism among those who used a moderate-to-high amount of substance use, this relationship did not exist among those who did not use substances, indicating a buffering effect. However, both studies were limited in that the sample consisted of known delinquents, thus leaving open the question as to whether similar conclusions would be reached in non-justice-involved samples. Further, only two potential protective factors were considered in these studies leaving many of those factors described earlier un-investigated.

In order to expand upon the work done thus far mainly on justice-involved youth, the current study assesses both the prevalence and impact of ACEs on a different sample, the Cambridge Study in Delinquent Development (CSDD). This community-based sample of boys from south London will help to increase the generalizability of our understanding of the effect of ACEs on crime. Additionally, as the CSDD is rich with potential protective factors, the extent to which these factors reduce the impact of ACEs on offending will also be investigated. While there have been a few prior studies to use the CSDD in the investigation of protective factors, only one of them used a risk factor that is closely related to an ACE (Farrington et al., 2016) and none of them utilized the ACE framework for conceptualizing risk (Farrington et al., 1988; Farrington & Ttofi, 2011). Finally, our analyses captures involvement in offending throughout late middle-adulthood, where most studies investigating ACEs have been limited to adolescence.

Section snippets

Methods

The CSDD is a sample of 411 males, mostly White and British (87%), born in or around 1953 and originally from a working class neighborhood in south London. Beginning when the boys were 8 years old, they were interviewed several times until their 18th birthday. The males were subsequently interviewed at ages 32 and 48. Surveys of the boys' parents and teachers were also conducted to gain additional information on the boy's background, behavior, and school performance. Official criminal records

Presence of ACEs

As can be seen in Fig. 1, it was not uncommon for the boys in the CSDD to have experienced at least one ACE. The ACE score distribution was bi-modal with modes at both 0 and 1 with 99 boys having experienced 0 ACEs and 99 experiencing only 1 ACE. However, there is a drop after that with each subsequent ACE score being less prevalent. It should be noted that as only 7 of the respondents experienced six ACEs, the graphical representations combined those who experienced five and six ACEs. In sum,

Discussion and conclusion

While it is evident that ACEs are associated with an increased likelihood of offending (Fox et al., 2015), what is less known is what can be done to buffer this impact. Two recent studies have revealed that substance non-use, but not social bonds, moderate the relationship between ACEs and offending (Craig et al., 2017; Craig et al., 2017). However, only these two potential buffering mechanism have been studied. The current study sought to expand prior research by considering a wide variety of

References (60)

  • D. Jolliffe et al.

    Protective factors for violence: Results from the Pittsburgh Youth Study

    Journal of Criminal Justice

    (2016)
  • B.E. Kim et al.

    Examining protective factors against violence among high-risk youth: Findings from the Seattle Social Development Project

    Journal of Criminal Justice

    (2016)
  • D. Laible et al.

    Negative emotionality and emotion regulation: A person-centered approach to predicting socioemotional adjustment in young adolescents

    Journal of Research in Personality

    (2010)
  • C. MacKinnon-Lewis et al.

    Juvenile justice and mental health: Youth and families in the middle

    Aggression and Violent Behavior

    (2002)
  • M.H. Teicher et al.

    The neurobiological consequences of early stress and childhood maltreatment

    Neuroscience & Biobehavioral Reviews

    (2003)
  • M.M. Ttofi et al.

    Protective factors against offending and violence: Results from prospective longitudinal studies

    Journal of Criminal Justice

    (2016)
  • R. Agnew

    Foundation for a general strain theory of crime and delinquency

    Criminology

    (1992)
  • R.F. Anda et al.

    Adverse childhood experiences and smoking during adolescence and adulthood

    JAMA

    (1999)
  • R.F. Anda et al.

    The enduring effects of abuse and related adverse experiences in childhood

    European Archives of Psychiatry and Clinical Neuroscience

    (2006)
  • R.F. Anda et al.

    Adverse childhood experiences, alcoholic parents, and alter risk of alcoholism and depression

    Psychiatric Services

    (2002)
  • M.T. Baglivio et al.

    The interrelatedness of adverse childhood experiences among high-risk juvenile offenders

    Youth Violence and Juvenile Justice

    (2015)
  • M.T. Baglivio et al.

    The prevalence of adverse childhood experiences (ACE) in the lives of juvenile offenders

    OJJDP Journal of Juvenile Justice

    (2014)
  • M.T. Baglivio et al.

    Effortful control, negative emotionality, and juvenile recidivism: An empirical test of DeLisi and Vaughn's temperament-based theory of antisocial behavior

    The Journal of Forensic Psychiatry & Psychology

    (2016)
  • M.T. Baglivio et al.

    Predicting adverse childhood experiences: The importance of neighborhood context in youth trauma among delinquent youth

    Crime & Delinquency

    (2015)
  • M.T. Baglivio et al.

    Racial/ethnic disproportionality in psychiatric diagnoses and treatment in a sample of serious juvenile offenders

    Journal of Youth and Adolescence

    (2016)
  • Centers for Disease Control and Prevention

    Injury prevention and control: Adverse childhood experiences (ACE) study

  • D. Cicchetti

    Annual research review: Resilient functioning in maltreated children—Past, present, and future perspectives

    Journal of Child Psychology and Psychiatry

    (2013)
  • J.M. Craig et al.

    Do social bonds buffer the impact of adverse childhood experiences on reoffending?

    Youth Violence and Juvenile Justice

    (2017)
  • J.M. Craig et al.

    What can help? Examining levels of substance (non)use as a protective factor in the effect of ACEs on crime

    Youth Violence and Juvenile Justice

    (2017)
  • L.J. Cronbach

    Coefficient alpha and the internal structure of tests

    Psychometrika

    (1951)
  • Cited by (113)

    View all citing articles on Scopus
    View full text