TABLE 1.

Characteristics of Studies Included in Meta-analysis of Graded Effects of ACEs on Justice System Contact

Author, y	Data Source	Population	N	Study Design and/or Analysis	Study Time Period	No. ACEs Captured (Age Range): ACEs Captured (Exposure Assessment)	Outcome and Definition	Outcome Ascertainment	Covariates	Key Finding(s)
Baglivio et al,47 2015	FDJJ archival data records	All youth in Florida with an arrest history who were administered the full C-PACT and turned 18 during the study period	64 329	Cohort, logistic regression	January 1, 2007–December 31, 2012	10 (ever): emotional abuse, physical abuse, sexual abuse, emotional neglect, physical neglect, family violence, household substance abuse, household mental illness, parental separation or divorce, household member incarceration score transformed from C-PACT items (administered through a semistructured interview conducted by a trained juvenile probation officer or contracted assessment staff; additional review of case file and education and child abuse records)	Membership in juvenile offending trajectory group; we extracted data on early starters compared with mid- to early starters who later desist.	JJIS data extracts were used to gather every instance of arrest for youth at each age up to age 17.	Race, sex	Each 1-point increase in ACE score was associated with a 28.7% increase in the odds of being an early starter relative to the odds of being a mid- to early starter who later desists.
Craig,44 2019	FDJJ archival data records	3-y cohort of all youth with an arrest history who completed some form of a community-based placement during the study period	25 461	Cohort, logistic regression	July 1, 2009–June 30, 2012	10 (ever): emotional abuse, physical abuse, sexual abuse, emotional neglect, physical neglect, household substance abuse, violent treatment toward mother, parental separation or divorce, household mental illness, household member incarceration score transformed from C-PACT items (administered through a semistructured interview conducted by a trained juvenile probation officer or contracted assessment staff; additional review of case file and education and child abuse records)	Reoffending, defined as rearrest within 12 mo after completion of community-based placement	Arrest records were from FDJJ.	Race, sex, age, disadvantage, additional covariates related to antisocial peer associations, impulsivity, social bonds, youth criminal history and criminal attitudes	Each 1-point increase in ACE score was associated with a 3% increase in the odds of 12-mo rearrest.
Fagan and Novak,39 2018: (1) Black participant; (2) white participants	LONGSCAN study of child maltreatment (Baltimore, Chicago, San Diego, Seattle, Chapel Hill)	High-risk sample of children ages 4–6 and caregivers (based on having a history of maltreatment or considered at risk for based on parents’ low SES and maternal substance use)	620	Cohort, logistic regression	1990–2002	10 (before age 12): emotional abuse, physical abuse, sexual abuse, failure to provide (ie, physical neglect), lack of supervision, caregiver intimate partner violence victimization, caregiver depression, caregiver substance use or abuse, caregiver criminality, family trauma transformed from child protective services agency records from states participating in LONGSCAN (based on child and caregiver responses to Modified Maltreatment Classification System and Conflict Tactics Scale; cumulative ACE score was winsorized at 7)	Past-year arrest at age 16	Self-reported, primary caregiver reports	Age, sex, single parent, geographic region, poverty, other neighborhood and community covariates; analyses stratified by race	Among Black participants, each 1-point increase in ACE score was associated with a 23% increase in the odds of past-year arrest at age 16. Findings were not significant for white participants (aOR 0.91; 95% CI [0.69–1.20]).
Fleming and Nurius,49 2019	Washington state BRFSS survey (2011)	State implementation of nationally representative survey conducted in collaboration with the CDC	13 803	Cross-sectional, Wald difference test	2011	8 (before age 18): sum of participant responses to 8 CDC categories of ACEs: emotional abuse, physical abuse, sexual abuse, household incarceration, living with someone with serious mental illness, living with someone with substance use issues, parents divorced or separated, parents who physically hurt one another	Adult incarceration (after age 18)	Self-reported	Race, sex, education	Findings after data transformation: each 1-unit increase in ACE score was associated with an 18% increase in the odds of adult incarceration.
Giovanelli et al,40 2016: (1) outcome: juvenile arrest; (2) outcome: adult felony charge	Chicago Longitudinal study	Low-income, minority sample born in high-poverty neighborhoods in Chicago from 1979 to 1980	1200	Cohort, logistic regression	1986 (start of study) to 2002 (age 22–24 follow-up survey)	9 (before age 18): physical abuse, sexual abuse, neglect, prolonged absence of parent or divorce of parents, death of family member or close friend or relative, frequent family conflict, problems of substance abuse of parent, witness to a shooting or stabbing, violent crime victimization (assessed in survey at 22–24 y; physical abuse, sexual abuse, and neglect items obtained from administrative records)	Two outcomes: (1) juvenile arrest (ages 10–18) and (2) felony charge (ages 18–24)	Juvenile arrest records were obtained from petitions to Cook County Juvenile Court and 2 other Midwestern locations. Felony charges were taken from federal prison records as well as documented histories in state, county, and circuit courts.	Race, sex, family ecology of risk, CPC intervention status	Findings after data transformation: each 1-unit increase in ACE score was associated with a 13% increase in the odds of juvenile arrest. Findings were not significant for felony charge (aOR 1.04; 95% CI [0.97–1.12]).
Kowalski,45 2019: (1) male participants; (2) female participants	Archival records from juvenile justice agency in Washington state	Youth on probation in Washington state who completed the PACT full assessment during the study period	35 442	Cohort, logistic regression	December 2003–June 2017	10 (ever): emotional abuse, physical abuse, sexual abuse, emotional neglect, physical neglect, domestic violence, household substance abuse, household mental health problems, parental separation or divorce, incarceration of a household member score transformed from C-PACT items (administered through a semistructured interview conducted by a trained juvenile probation officer or contracted assessment staff; additional review of case file and education and child abuse records)	Reoffending, defined as a new felony, misdemeanor, violent, property, drug, or sex offense at 12 mo	Records were from Washington state (standard 18-mo follow-up period).	Race, sex, age, mental health status, substance use, risk level	Among male participants, each 1-point increase in ACE score was associated with a 7% increase in the odds of 12-mo reoffending. Among female participants, each 1-point increase in ACE score was associated with a 4% increase in the odds of 12-mo reoffending.
Levenson et al,42 2017: (1) male participants; (2) female participants	FDJJ archival data records	Youth who aged out of the juvenile justice system (turned 18 y old) and who were assessed with the C-PACT full assessment during the study period	89 045	Case control, logistic regression	January 1, 2007–December 31, 2015	10 (ever): emotional abuse, physical abuse, sexual abuse, emotional neglect, physical neglect, family violence, household substance abuse, household mental illness, parental separation or divorce, household member incarceration score transformed from C-PACT items (administered through a semistructured interview conducted by a trained juvenile probation officer or contracted assessment staff; additional review of case file and education and child abuse records)	Juvenile sexual offending (misdemeanor or felony offenses), defined as arrest ≥1 time for a sexual offense before age 18	Arrest records were from FDJJ.	None	Findings after data transformation: among male participants, each 1-point increase in ACE score is associated with a 1% increase in the odds of juvenile sexual offending versus nonsexual offending. Among female participants, each 1-unit increase in ACE score is associated with a 4% increase in the odds of juvenile sexual offending versus nonsexual offending.
Naramore et al,43 2017	FDJJ archival data records	All youth in Florida ages with an arrest history who were administered the full C-PACT and were 11.4–22.5 y at the time of their last assessment	64 329	Cross sectional, logistic regression	December 14, 2005–December 30, 2012	10 (ever): emotional abuse, physical abuse, sexual abuse, emotional neglect, physical neglect, family violence, household substance abuse, household mental illness, parental separation or divorce, household member incarceration score transformed from C-PACT items (administered through a semistructured interview conducted by a trained juvenile probation officer or contracted assessment staff; additional review of case file and education and child abuse records)	Arrest for trading sex (ie, “offer to commit, or to commit, or to engage in, prostitution, lewdness, or assignation” or “aid, abet, or participate in any of the acts or things enumerated in this subsection”43)	Arrest records were from FDJJ.	None	Findings after data transformation: each 1-point increase in ACE score is associated with a 68% increase in the odds of being arrested for trading sex compared to arrest for other offenses.
Perez et al,46 2018	FDJJ archival data records	Youth who aged out of the juvenile justice system (turned 18 y old) and who were assessed with the C-PACT full assessment during the study period	64 329	Case control, logistic regression	January 1, 2007–December 31, 2012	9 (ever): emotional abuse, physical abuse, sexual abuse, emotional neglect, physical neglect, witnessing household violence, household substance abuse, household mental illness, household member incarceration score transformed from C-PACT items (administered through a semistructured interview conducted by a trained juvenile probation officer or contracted assessment staff; additional review of case file and education and child abuse records)	SVC delinquency, defined as committing ≥3 serious felony offenses, with at least 1 violent offense	Arrest records were from FDJJ.	Race, sex, SES	Each 1-point increase in ACE score is associated with a 30% increase in the odds of a juvenile offended being classified as an SVC offender.
Stinson et al,41 2016	Inpatient forensic psychiatric facility in the Midwestern United States	Selected participants had commitments for violent or sexual offending and a length of admission ≥1 y at the time of data collection of 2 nonoverlapping time samples in 2007 and 2012.	381	Cross sectional, logistic regression	2007, 2012	6 (during developmental years): verbal and/or emotional abuse, physical abuse, intrafamilial sexual abuse, extrafamilial sexual abuse, neglect, parental substance abuse (coded from the social service reports generated at admission and annually by facility personnel; experiences were self-reported by clients, reported by corroborating family members, and/or records obtained from state investigations of reported maltreatment)	Juvenile arrest, defined as arrest before age 19	Available social service records	None	Each 1-point increase in ACE score is associated with 34% higher odds of juvenile arrest among violent and sexual offenders.
Wolff et al,48 2020	FDJJ archival data records	Youth who aged out of the juvenile justice system (turned 18 y old) and who were assessed with the C-PACT full assessment during the study period but who were not involved with a gang at time of first assessment and who had information on race and/or ethnicity	104 266	Cohort; rare-events logistic regression	January 1, 2007–December 31, 2017	10 (ever): emotional abuse, physical abuse, sexual abuse, emotional neglect, physical neglect, family violence, household substance abuse, household mental illness, parental separation or divorce, household member incarceration score transformed from C-PACT items (administered through a semistructured interview conducted by a trained juvenile probation officer or contracted assessment staff; additional review of case file and education and child abuse records)	Verified gang involvement; only youth for whom there exists written documentation from law enforcement certifying them as gang involved (as per state statute) were classified as verified.	Law enforcement documentation	Race, sex, age	Each 1-point increase in ACE score is associated with 14% higher odds of gang association among juvenile offenders.

BRFSS, Behavioral Risk Factor Surveillance System; CDC, Centers for Disease Control and Prevention; CI, confidence interval; C-PACT, Community Positive Achievement Change Tool; FDJJ, Florida Department of Juvenile Justice; JJIS, xxx; LONGSCAN, Longitudinal Studies on Child Abuse and Neglect; PACT, Positive Achievement Change Tool; SES, socioeconomic status; SVC, serious, violent, and chronic.