Drug court as a potential intervention point to impact the well-being of children and families of substance-using parents

Abstract

Background

A high proportion of justice-involved individuals have a substance use disorder and many of those individuals serve in a caregiving role to a child under 18. Given the negative impact of substance use and justice-involvement on the wellbeing of children, the criminal justice system may offer a unique intervention point with high public health impact. This study describes characteristics of adult drug court participants (DCP) that affect the wellbeing of their children and families and compares the DCP parenting and mental health characteristics to their child’s other caregiver in order to understand how parenting differs within drug court families.

Method

Data were collected from a sample of 100 DCP; 58 had a matched other caregiver. Drug court data regarding substance use and criminogenic risk/need were collected. Analyses differentiated the parenting behaviors and mental health needs of DCP from other caregivers.

Results

The DCP were at moderate to high risk for recidivism and presented with multiple and significant criminogenic and psychosocial functioning needs. Risk for potential maltreatment and poor parenting behaviors were elevated, and significantly higher compared to other caregivers. DCP demonstrated clinically elevated mental health needs, and were significantly different across all indicators of mental health compared to other caregivers.

Conclusions

Adult drug courts address the occurrence of substance use disorders but there are additional needs to be intervened upon. Adult drug courts may be a viable intervention point to address issues of parenting and mental health to improve the wellbeing of criminal justice-involved individuals, their children, and families.

Background

Parents who misuse substances come into contact with the criminal justice system at significant rates. As an alternative to imprisonment, evidence-based treatment and service interventions present the most effective approach to disrupt the well-documented relationship between substance misuse and criminal behaviour (Andrews & Bonta, 2010b; Friedmann, Taxman, & Henderson, 2007; Marlowe, 2003; McCarty & Chandler, 2009). It is estimated that over 60% of people involved with the criminal justice system (CJS) have a substance use disorder (SUD; National Research Council, 2014). Adult drug courts are community-based interventions for individuals with SUD facing criminal charges. Adult drug courts include court-mandated and monitored substance use treatment involving regular court hearings, intensive judicial monitoring, drug screenings, and sanctions and rewards depending on program compliance (Guastaferro, Lutgen, & Guastaferro, 2016). In a meta-analysis of 92 adult drug court evaluations, Mitchell and colleagues (2012) documented a drop in recidivism ranging from 38 to 50% with effects lasting approximately 3 years post-completion compared to non-participants of drug court programs. Other meta-analyses have shown as much as a 20% reduction in recidivism (Shaffer, 2011; Wilson, Mitchell, & MacKenzie, 2006). Drug courts’ demonstrated effectiveness contributed to their rapid deployment: there are more than 3,400 drug courts across the United States that provide services to approximately 150,000 people annually (National Association of Drug Court Professionals, 2018a; National Drug Court Resource Center, 2018).

Individuals considered for and enrolled in drug courts often present with myriad needs that extend beyond substance misuse and criminal behavior. While drug courts must attend to these primary objectives, programs must also consider treatments and services for needs that could interfere with successfully engaging in, and completing, treatment and supervision requirements. Drug court participants (DCP) often have needs around their mental health, employment, education, problem-solving skills, anti-social cognitions, family conflict, parenting, and medical care and some drug courts have added these services either in-house or through community referrals (Green & Rempel, 2012; National Research Council, 2014; Underhill, Dumont, & Operario, 2014; Wenzel, Longshore, Turner, & Ridgely, 2001). Importantly, best practices and empirical evidence caution that programs should provide services to participants in need and not require unnecessary services as these may negatively affect participant success and are not an efficient use of scarce resources. Offering complementary treatment and services to those in need is a best practice standard identified by the National Association of Drug Court Professionals (2018b, 2018c). Further, this more comprehensive approach to DCP needs is in line with an integrated health model and the practice of treating the whole person, a best practice for substance abuse treatment (National Institute on Drug Abuse, 2018).

Drug Court Participants as Parents

Among the adult drug court population, a high proportion are parents. A multi-site drug court evaluation found that about half of adult DCP have children under 18-years old and 20% are primary caregivers (Rossman et al., 2011). Parental arrest and incarceration have negative effects on children; having a criminal justice-involved parent is stressful and disruptive to child wellbeing and family functioning (Arditti, 2015; National Research Council, 2014; Turanovic, Rodriguez, & Pratt, 2012; Uggen & McElrath, 2014). For example, parental incarceration increases children’s risk for depression, hyperactivity, withdrawal, and eating problems (Murray, 2005; Murray & Farrington, 2008). A parent’s CJS involvement creates the potential for harmful trauma for children related to parental absence, caregiver instability, and stigma (Turney, 2014). Additionally, parental substance use has the potential to negatively impact the home environment (e.g., inconsistent supervision, available resources), parent-child relationships (e.g., lack of consistent responses to promote social emotional development), and child problem behaviors (e.g., delinquency, performance in school; (Barnard & McKeganey, 2004). As a result, children of substance using adults are at increased risk for child maltreatment (Choi, Huang, & Ryan, 2012; Dube et al., 2001).

Our understanding of how psychosocial needs, including interpersonal relationships with children and families, impact DCP outcomes is not well understood (Christian, 2009). Drug courts generally gather little information about the individuals’ parenting behaviors. Despite a documented link between substance use and family dysfunction, few evaluations of adult drug courts have studied how drug courts positively impact families and in turn how families impact an individual’s treatment success in a drug court (Green & Rempel, 2012). Family drug treatment courts have demonstrated a positive impact in child well-being outcomes (Gifford, Eldred, Sloan, & Evans, 2016; Gifford, Eldred, Vernerey, & Sloan, 2014), this it is plausible that the incorporation of interventions that target adult DCP who are parents, their child(ren), and family could have important public health and criminal justice implications. The drug court setting presents a potential opportunity to effectively and efficiently intervene, thus informing a public health strategy.

We know that justice-involved parents typically do not parent their minor children alone. Other caregivers are likely to become involved at the point of arrest, if not before (Guastaferro, Guastaferro, & Stuart, 2015; National Research Council, 2014). However, little is known about the family environment surrounding the DCP, their child, and other caregivers of that child. The DCP’s relationships with other caregivers, as well as the other caregivers’ behaviors, are not well understood thereby limiting the knowledge of the DCP’s children’s health and wellbeing. Other caregivers may be similar or different from DCP with respect to drug use, criminal justice involvement, mental health, and parenting practices. Other caregivers may also encompass a variety of relationships with the DCP including but not limited to spouses or ex-spouses, boy/girlfriends, parents, or siblings. Understanding the family environment risk and protective factors is an important aspect of designing a potential intervention strategy.

Poor parenting behaviors, such as a lack of supervision or inconsistent discipline, common among parents with SUD, have been associated with adverse outcomes for children (Calhoun, Conner, Miller, & Messina, 2015; Fals-Stewart, Kelley, Fincham, Golden, & Logsdon, 2004). However, we do not know specifically how characteristics of the DCP substance use, addiction severity, and criminogenic risk impact parenting behaviors. There is evidence that certain substances (heroin, cocaine, crack cocaine, methamphetamine) have a robust, direct relationship with criminal behavior (Bennett, Holloway, & Farrington, 2008), which has implications for treatment and supervision practices (Taxman, Caudy, & Pattavina, 2013; Taxman, Pattavina, Caudy, Byrne, & Durso, 2013). Explanations for the increased likelihood of engagement in criminal behavior among users of crack cocaine, cocaine, methamphetamine, and heroin (i.e., criminogenic drugs) include pharmacological factors related to addiction and withdrawal issues, types of crime committed by these individuals, and individual-level/demographic factors, such as race and gender (Bennett et al., 2008). The relationship between criminogenic drug use, addiction severity, and parenting behaviors among DCP may be important in intervention design. Primary criminogenic needs include the constellation of antisocial attitudes and behaviors, antisocial peers, and substance misuse and have a clear link to criminal behavior. Psychosocial functioning needs can support or distract from treatment engagement and recovery. These needs include mental health, education/employment, family and intimate relationships, housing, and financial. Treatment and supervision in the adult drug court setting commonly target these areas because of their relationship with criminal behavior and because they are amenable to change. However, we do not know if there is an association between criminogenic needs and parenting behaviors.

Parental mental health is also associated with child wellbeing (Chaffin, Kelleher, & Hollenberg, 1996; Stith et al., 2009). Among those under correctional supervision, including drug courts, it is estimated that half to nearly three-quarters of individuals have a serious mental illness and a co-occurring SUD (Osher, D’Amora, Plotkin, Jarrett, & Eggleston, 2012). Not only do we not know how DCP and other caregivers differ with regard to mental health, but also how a DCP’s mental health is associated with substance use characteristics, addiction severity, or criminogenic needs.

These gaps in knowledge exist because addressing the needs of children and the family unit more broadly is generally outside the purview of CJS goals. It may be advantageous to leverage involvement with the CJS as a result of SUD to prevent adverse outcomes for the children of justice-involved parents, such as preventing child maltreatment and improving mental health outcomes, but a description of need is necessary to design an effective and efficient intervention strategy. Thus, the objectives of the present paper are twofold: (1) to examine the characteristics of DCP that affect their children and families including parenting and mental health and (2) to compare these characteristics to the other caregiver. This description of need, as it relates to child and family wellbeing, may inform public health intervention strategies that leverage involvement in the CJS to address other public health priorities.

Method

Participants

Participants were recruited from two Metro-Atlanta felony-level adult drug courts. Data were collected from family units consisting of up to three participants: a DCP who was a caregiver to at least one child under 18; a child under 18; and a co-parent or another adult caregiver of that child (other caregiver). Data collected from the child are not presented as it is outside the scope of the current paper. To be eligible, the DCP had to be enrolled in one of two adult drug court programs and function as a caregiver for at least one child under 18 years old. Between the two courts, there were 529 drug court clients; 164 met eligibility criteria. A total of 100 DCP enrolled in the study and 64 declined to participate, a participation rate of 61%. Complete sociodemographic characteristics are presented in Table 1. DCP were primarily male (60%), Black (53%), and unmarried (55%). The mean age of DCP was 34.8 years (SD = 8.52). Most DCP worked ≥30 hours per week (59%) and had at least some college education (51%).

Table 1.

Demographic and Family Structure Characteristics of Drug Court Participants (DCP) and Other Caregivers

	DCP (N=100)			Other Caregivers (N=59)
	n		%	n		%
Male	52		60a	14		24
Marital Status, married	45		45	40		59
Race
White	45		45	30		51
Black	53		53	24		41
Other	2		2	5		8
Educational Status
< High School	17		17	7		12
HS Graduate	32		32	25		42
Some College	51		51	27		46
Employment Status
Unemployed	8		8	22		37
< 30 hrs/ wk	33		33	11		19
≥ 30 hrs/ wk	59		59	26		44
Annual Household Income
< $25,000	50		50	28		51a
$25–49,000	28		28	16		29
≥ $50,000	16		16	11		20
Number of Adults in Home
1	14		14a	12		21a
2	37		37	36		63
3+	22		22	9		16
Family Structure
Total # of Children < 18y		179			109
Age
0 – 2	28		14	16		15
3 – 5	21		11	10		9
6 – 11	65		36	43		39
12 – 18	65		39	40		35
Custody Status
Non-custodial	65		36	19		17
Shared or Partial	46		26	16		15
Full	53		30	67		62
Other	14		8	109		6
# Children Living with Adult	81		45	84		77
Frequency of Seeing Children
Daily	93		52	87		80
Weekly	39		22	12		11
Monthly	19		11	2		2
Annually	19		11	2		2
Never	7		4	6		5

^aIndicates missing data due to participants option to not respond; percentages are of those who answered

The other caregiver was identified by the DCP at study enrollment and was eligible to participate if they served in a caregiving role for a child under 18-years old. To our knowledge, none of the other caregivers were currently enrolled in a drug court program. Not all DCP were able to identify another caregiver, and not all other caregivers consented to participate. The total number of other caregivers enrolled was 59. Relationships of the other caregivers to the DCP included friend, sibling, parent, romantic partner, or spouse. As described in Table 1, the other caregivers were primarily female (76%) and married (59%). The mean age of other caregivers was 34.9 (SD = 12.16). There was more variability in race among other caregivers: 51% White, 41% Black, and 8% other. Forty-six percent of other caregivers had at least some college education and 44% worked ≥30 hours per week. There were 58 matched DCP and other caregiver dyads (one other caregiver participated without a DCP). Matched pairs answered survey questions referencing the same child.

Procedure

Presented here are cross-sectional, baseline data collected from a larger, longitudinal project, funded by the Administration for Children and Families, designed to implement family focused interventions in the adult drug court setting. All participants completed an audio computer-assisted self-interview overseen by a research assistant at a location of the participant’s preference, which included their home, at the drug court or treatment facility, or at a community location (e.g., library or coffee shop). Participants were able to skip questions, resulting in varying sample sizes for different questions, including demographics. All participants were provided headphones to hear the questions read aloud as an accommodation for varying reading levels and to maximize privacy. Research assistants were present during survey administration for technical assistance. The DCP and other caregiver often participated at the same time but not always. Each participant who completed the survey received a $75 gift card as compensation for their time. Participation in the research was voluntary and had no bearing on the DCPs drug court status. The university Institutional Review Board approved all research procedures.

Measures

Family environment

Family characteristics including the number of children, family structure, and custody status were collected for the DCP and other caregiver in the computer-assisted interview. The other caregiver was asked about their substance use using the Alcohol, Smoking, and Substance Involvement Screening Test (WHO ASSIST Working Group, 2002) which was designed to detect and manage substance use and related problems in primary and general medical care settings. To provide context of the environment, in the present analysis tobacco, alcohol, and illicit drug use frequencies were dichotomized to indicate any substance use in the past year.

Substance use and addiction severity

DCP addiction severity was assessed at intake into the drug court and is used by treatment staff to determine appropriate treatment services. These analyses used self-reported substance use information including IV drug use, polydrug use, viewed drug problems as serious, the perceived importance of treatment, and history of prior treatment, each measured as dichotomous variables. These variables were extracted from validated screeners used by the drug court program at intake to assess drug use severity. The DCP-specified drug of choice identified at intake was used in the present analysis as a dichotomous variable to represent whether a criminogenic drug (e.g., crack, cocaine, heroin, methamphetamine, amphetamine) was indicated as the most problematic drug.

Criminogenic risk and needs

The DCP’s risk for recidivism was evaluated prior to drug court enrollment using the Level of Service Inventory- Revised (LSI-R), a broad risk/needs assessment of criminogenic needs designed for use as an indicator for recidivism and to guide treatment strategies (Andrews & Bonta, 2010, 2003; Guastaferro, 2012; Kelly & Welsh, 2008). The 54-items are scored in a “yes-no” format or on a “0–3” scale indicating satisfaction with a given situation; the higher the total score, the higher the level of risk for rearrest and criminogenic need (Andrews & Bonta, 2003). Static and dynamic risk factors are operationalized by 10 domains: criminal history, employment/education, financial needs, family/marital relationships, accommodations, leisure/recreation, companions, alcohol/drug problems, emotional/personal, and attitudes/orientation. For the present analysis, the number of criminogenic and psychosocial functioning needs were calculated and the LSI-R total risk score (α= .83) was categorized into three levels of risk aligned with guidelines for community-based supervision and services: low (score 1–18), moderate (score 19–23), and high (score 24–54). LSI-R total scores were also dichotomized to low vs. moderate or high for statistical models.

Parenting

We examined the potential for maltreatment, parenting behaviors, and parent-child communication. Though these characteristics are related, they measure different aspects of parenting. The Brief Child Abuse Potential Inventory ( Ondersma, Chaffin, Mullins, & LeBreton, 2005) examined the potential for maltreatment. The 34-item actuarial scale is comprised of 7 content-specific subscales (e.g., lack of happiness, feelings of persecution, loneliness, family conflict, rigidity, distress, and poverty) as well as lie and random response scales to indicate a potentially invalid response. Respondents endorsed whether they agreed or disagreed with a given statement and earned one point for each positive endorsement. A total score ≥9 indicated at-risk levels for child maltreatment and a score ≥12 indicated high-risk. Descriptive statistics are presented for the total score of the seven content-specific subscales (α = .80).

The Alabama Parenting Questionnaire was used to examine self-reported parenting practices related to children’s externalizing problem behaviors. Across 35-items the instrument assesses parental involvement (10 items; α = .83), positive parenting (6 items; α = .77), poor monitoring or supervision (9 items; α = .72), inconsistent discipline practices ( 6 items; α = .75), and corporal punishment (3 items; α = .43; Shelton, Frick, & Wootton, 1996). Seven other discipline practices (e.g., yelling, timeout) were summarized to provide greater detail about the parent-child relationship and to reduce implicit negative bias toward corporal punishment items (Frick, Christian, & Wootton, 1999). All 42 items are rated on a 5-point frequency scale (1= never to 5=always) with higher scores indicating a higher frequency of the behavior.

The Parent Child Communication scale (Conduct Problems Prevention Research Group, 1992, 2002) was used to assess respondents’ perceptions of their communication skills with their child (e.g., “Does your child let you know what is bothering him/her?”). The PCC includes 20 items, which are answered on a 5-point scale (1 = “Almost Never” to 5 = “Almost Always”). We did not replicate the original factor structure of the scale; instead, principal components analyses showed that one factor solution that included 14 of the 20 items accounted for 26.4% of the variance. Those 14 items were summed to produce a single communication score (α = .85).

Adult mental health

The Brief Symptom Inventory assesses mental health status across domains of psychoticism (5 items; α = .67), somatization (7 items; α = .77), depression (6 items; α = .87), hostility (5 items; α = .77), phobic anxiety (5 items; α = .73), obsessive-compulsive (6 items; α = .85), anxiety (6 items; α = .76), paranoid ideation (5 items; α = .72), and interpersonal sensitivity (4 items; α = .81) (Derogatis & Melisaratos, 1983). Raw scores were normed with a non-patient sample provided by the developer (Derogatis, 1993). Three global indices are reported in addition to the subscales: General Severity Index (the overall severity based on the combination of symptoms and disruption of activities of daily life), Positive Symptom Total (the number of items endorsed with a positive response) and Positive Symptom Distress Index (divides the sum of item values by the positive symptom total). Adult trauma symptoms were assessed with the Posttraumatic Diagnostic Scale (Foa, Cashman, Jaycox, & Perry, 1997), which provides diagnostic criteria for posttraumatic stress disorder (PTSD) and severity of PTSD symptoms in clinical and research settings. The total number of symptoms reported by adults was averaged and used to calculate symptom severity (mild, moderate, severe, or none) as outlined by the developers.

Child mental health

The Behavior Assessment System for Children measures parent perception of adaptive and problem behaviors in children over two-years old (Reynolds & Kamphaus, 2004; Sandoval & Echandia, 1994). Participants answered these questions on only one child under 18 regardless of the number of children in their family. Delivery of the assessment has age specific versions of the measure for ages 2 to 5 (α = .85), 6 to 11 (α = .85), or 12+ (α = .79). T scores for each participant using age- and sex-specific norms were computed from raw scores using the developer-provided manual. Results of composite scales (externalizing problems, internalizing problems, behavioral symptoms index, and adaptive skills) are presented.

Analytic Plan

The first objective was to describe the characteristics of DCP that may affect the wellbeing of their child(ren) and family including the family environment, substance use and addiction severity, criminogenic risk and needs, parenting, adult mental health and child mental health. The second objective was to compare characteristics of DCP to other caregivers as it pertained to parenting behaviors, reported adult mental health, and perceived child mental health. Differences between matched DCP and other caregiver dyads (n=58) were examined using Wilcoxon matched-pairs signed rank test for continuous variables and McNemer’s test for categorical variables.

Results

Family Structure

The majority of DCP (59%) and other caregivers (79%) reported living in a home with at least one other adult (Table 1). The 100 DCP reported being a caregiver to a total of 179 children under 18 (45% of whom live with the DCP), and the 59 other caregivers reported being a caregiver to 109 children under 18 (77% of whom live with the other caregiver). The majority of all children reported by the DCP and other caregiver respondents in this study were older than 6 years old (75% and 74%, respectively). DCP reported having full or shared/partial custody of 56% of the 179 reported children and the majority of children are seen by the DCP on a daily basis. The mean number of days from enrollment in drug court to the baseline assessment for DCP was approximately 374 days (SD=232).

Most of the other caregivers (N = 59) reported not using any substances (83%). Ten other caregivers reported using any illicit drug at least once in the past 12 months (17%). The most common reported drug used was cannabis, reported by all 10 other caregivers. The following substances were used once or twice in the past year: cocaine (2 people), amphetamines (2 people), sedatives/sleeping pills (3 people), hallucinogens (2 people), heroin (2 people), methamphetamines (2 people), and inhalants (1 person). Two other caregivers reported weekly use of sleeping pills. One person reported monthly use and two people reported daily or almost daily use of other opioids. More than half of the other caregivers reported tobacco use (53%), 24 of whom reported smoking every day. Nearly half (n = 30) of the other caregivers reported using alcohol at least once in the prior 12-months; 13 indicating use 1–2 times, 9 indicating use once per month, 5 indicating use once per week, and 3 indicating daily or almost daily use.

Substance Use and Addiction Severity

The majority of DCP (76%) indicated their primary drug of choice as a criminogenic drug: cocaine or crack cocaine (n=41), heroin (n=13), or methamphetamines and other amphetamines (n=23). It was common (87%) for DCP to report poly drug use (Table 2). Four in ten (41%) reported prior substance use treatment. The majority of DCP believed their drug problems were serious (84%) and that receiving treatment was important (85%).

Table 2.

Risk-Needs and Substance Use Characteristics of Drug Court Participants (N=100)

Primary Indicated Drug of Choice	n	%
Alcohol	5	5
Marijuana	14	14
Cocaine/Crack Cocaine	41	41
Heroin/Other Opiates	13	13
Methamphetamines/Other Amphetamines	23	23
Other	3	3
Substance Use Behavior
Intravenous drug use in the past 12 months	13	13
Drug of choice is a criminogenic drug	76	76
PolyDrug Use	87	87
Views drug problems as serious	84	84
Views treatment considerably important now	85	85
Number of times in prior substance use treatment
Never	59	59
Once	26	26
Two or more times	15	15
Risk-Needs	M	SD
Mean days in treatment to study interview	374.05	231.94
LSI-R Total Risk/Needs Score	25.07	7.14
LSI-R Subscales—High or moderate scores	n	%
Criminal History	72	75.8
Education/Employment	62	65.3
Financial	64	67.4
Family/Marital	42	51.2
Accommodations	41	43.2
Leisure/Recreation	90	94.7
Companions	78	82.1
Alcohol/Drugs	93	97.9
Emotional/Personal	34	35.8
Attitudes/Orientations	31	32.6
LSI-R Risk-Needs Categories	n	%
Low (1–18)	17	17
Moderate (19–23)	23	23
High (24–54)	60	60

Note: LSI-R = Level of Service Inventory- Revised

Criminogenic Risk and Needs

Eight in 10 DCP are at moderate or high risk for recidivism as indicated by the LSI-R at drug court program intake: the mean total risk-needs score was 25.07 (SD=7.14). The percentage of DCP who scored moderate or high on the LSI-R subscales are displayed in Table 2; six participants were missing item-level data and are not included in the subscale findings. Eighty-five percent (85%) of DCP had moderate or high needs in multiple criminogenic areas (e.g., companions, attitudes, leisure time, alcohol or drug problems). The majority (68%) had moderate or high needs in multiple psychosocial functioning areas (e.g., education/employment, financial, accommodations, emotional/personal, family/marital). Nearly all respondents (93%) indicated prior convictions with 71% reporting three or more adult convictions. The majority of DCP (61%) were incarcerated upon conviction at least once and 28% were arrested under the age of 16. More than half (61%) were unemployed at the time of enrollment in the drug court program; 71% had been fired at least once. Over half (57%) of the DCP reported a family member or spouse who was involved in criminal behavior. Nearly two-thirds (63%) reported being satisfied with their marital (or equivalent) situation. Over one third (35%) of the DCP indicated receiving mental health treatment in the past and a psychiatric assessment was indicated for 39% of DCP, however only 11% were receiving treatment at program enrollment. Alcohol or drug use was causing family or marital problems and problems with work or school for nearly all DCP (97% and 90%, respectively); 19% reported medical problems as a result of their alcohol or drug use.

Parenting

Two sets of means for the parenting outcomes are presented in Table 3 for the DCP: the full sample (n = 100) and the sample of DCP that had a matched other caregiver (n = 58). Over one quarter (26%) of all DCP scored above the at-risk cutoff (≥9) for child maltreatment potential, as measured by the Brief Child Abuse Potential inventory. In contrast less than 14% (n=8) of all other caregivers were identified as at-risk. Among matched dyads, DCP were overall at a higher risk for child maltreatment compared to other caregivers (d=.39, p = .03). Statistically, there was no difference in meeting the high-risk (≥12) criterion between DCP and other caregivers; though, the percent of DCP exceeding the high-risk criterion was nominally higher than that of other caregivers.

Table 3.

Drug Court Participants (DCP) and Other Caregiver Characteristics on Parenting, Adult Mental Health, and Child Mental Health Domains

	All DCP N=100		Matched Dyads n=58
			DCP		Other Caregiver		p-value (α=.05)
	M	SD	M	SD	M	SD
Parenting
BCAP
Total Risk	7.38	4.60	6.41	4.36	4.79	3.90	.03
Total Risk Greater than 12, %	21.00		12.00		3.40		.59
APQ
Involvement	37.53	6.83	37.80	6.32	41.00	5.55	<.01
Positive Parenting	26.50	3.04	26.70	2.79	27.44	2.67	.23
Poor Monitoring	14.36	4.62	14.48	4.68	14.82	5.15	.81
Inconsistent Discipline	13.07	4.19	13.90	4.51	12.40	3.31	.02
Corporal Punishment	4.30	1.54	4.38	1.72	4.42	1.49	.93
Other Discipline	17.08	2.82	17.30	2.96	16.08	2.89	<.01
PCC	3.17	.42	3.91	.65	4.06	.64	.16
Adult Mental Health
BSI
Global Severity Index	56.87	11.13	60.18	10.59	48.80	10.61	<.001
PST Sum	58.10	10.57	54.77	9.03	53.14	9.91	.02
Positive Symptom Distress Index	54.76	9.05	60.23	11.05	50.81	7.53	<.01
Psychoticism Subscale	60.05	11.14	62.12	10.73	56.27	9.89	<.01
Somatization Subscale	52.21	9.19	54.96	9.37	49.02	8.90	<.001
Depression Subscale	54.90	10.42	56.98	10.88	50.38	10.03	<.01
Hostility Subscale	54.40	9.96	56.49	10.08	52.21	10.45	.03
Phobic Anxiety Subscale	55.70	9.74	57.60	9.83	50.70	8.48	<.001
Obsessive-Compulsive Subscale	57.68	10.32	58.74	10.00	51.8	10.35	<.001
Anxiety Subscale	53.77	11.08	56.11	10.92	48.11	10.07	<.001
Paranoid Ideation Subscale	57.57	10.47	58.58	10.47	53.63	9.04	.02
Interpersonal Sensitivity Subscale	56.06	11.15	57.47	10.11	50.82	9.37	.001
PDS
Total Symptoms	5.55	5.23	5.71	5.58	2.71	4.01	<.001
Total Severity	9.23	9.10	9.74	10.00	5.00	5.88	.001
Child Mental Health
BASC Composite Scores
Externalizing	50.74	9.96	51.52	11.70	49.14	10.66	<.01
Internalizing	48.39	9.07	49.15	9.48	45.54	8.62	.06
Behavioral Symptoms Index	50.01	8.89	51.41	9.78	48.59	8.99	.02
Adaptive Skills	47.30	11.12	47.54	9.83	49.57	10.41	.24

Note: BCAP = Brief Child Abuse Potential Inventory; APQ = Alabama Parenting Questionnaire; PCC = Parent-Child Communication; BSI = Brief Symptom Inventory; PST = Positive Symptom Total; PDS = Posttraumatic Stress Disorder Scale; BASC = Behavior Assessment System for Children

DCP mean ratings for the parental involvement subscale on the Alabama Parenting Questionnaire was 37.5 on a 10 to 50 scale (the equivalent of a 3.7 on the 5-point scale). Matched dyads (DCP and other caregivers) significantly differed in reports of parental involvement, with other caregivers reporting higher levels of involvement (d = .54, p = <.01). DCP and other caregivers did not differ in mean ratings for positive parenting, monitoring, and corporal punishment subscales. DCP reported being more inconsistent in their discipline than did other caregivers (d=.38, p = .02) and were more likely to report using non-corporal punishment or other discipline strategies (e.g., yelling, privilege restriction, time out; d=.42, p= <.01). There were no differences in self-ratings of parent-child communication between DCP and other caregivers.

We examined whether any of the parenting variables differed by LSI score and whether the client reported a criminogenic drug as their drug of choice. No differences were found for any of the parenting variables for either LSI level or criminogenic drug as drug of choice.

Adult Mental Health

DCP reported significantly higher rates of mental health distress when compared to other caregivers on the Brief Symptom Inventory (Table 3). DCP reported significantly more symptoms of mental health distress (d = .99, p = <.01) and more severe symptoms as indexed by the global severity index score (d = 1.07, p < .001) when compared to the other caregiver. DCP scores across all subscales (>50) indicated elevated risk though not necessarily to the level of clinical significance. There were statistically significant differences between matched dyads. The DCP had significantly higher scores (p < .05) compared to their matched other caregiver on all of the nine subscales. Related specifically to PTSD, DCP reported exposure to more traumatic events on the (d=.58, p = <.001), and more severe trauma symptoms (d=.61, p = .001). Tests of significance were conducted to examine the differences on adult mental health outcomes between levels of risk on the LSI-R and use of criminogenic drug criminogenic drug as drug of choice. The only significant difference found was between LSI level of risk and the somatization subscale of the BSI (p =.03).

Child Mental Health

Overall, DCP reported greater externalizing behaviors (d=.21, p = <.01) and more behavioral symptoms (d = .30, p = .02) for their children than the other caregiver (note: these two metrics are related and include some of the same subscales). No differences between the DCP and other caregiver were found for internalizing behaviors and adaptive behaviors (Table 3). It is worth noting that DCP and caregiver ratings of mental health symptoms were within or close to the normal range, with T score means around 50. Similar to other outcomes, there were no significant differences on any of the child mental health variables by LSI group or criminogenic drug use group (all p > .05).

Discussion

This study explored the family environment of DCP related to parenting children under the age of 18. At program enrollment the majority of DCP were at high risk for rearrest and presented with multiple psychosocial functioning needs. Determining one’s risk and need is essential to the provision of effective programming and is a best practice in interventions for individuals involved in the criminal justice system (see Andrews & Bonta, 2010). Justice-related intervention and evaluation research focuses a great deal on how risk and to a lesser extent how need vary in their impact on outcomes such as recidivism and program completion (Taxman & Pattavina, 2013). Researchers have called for a closer examination of individual needs to determine their role in well-being and reduced criminal behaviour and to pursue a more sophisticated approach to intervention development (Hannah-Moffat, 2016; Long, Sullivan, Wolldredge, Pompoco, & Lugo, 2019; Polaschek, 2011). While the assessment instrument used by the drug courts is not designed to assess specific family-related strengths and needs, it provides important client information. Two-thirds of the DCP reported being satisfied with their marital (or equivalent) situation and nearly all (97%) reported their substance use caused problems in their familial relationships. These findings speak to areas of focus for treatment.

The risk-need-responsivity model posits that individuals who are at differential risk for recidivism are distinct from one another and thus need different levels of supervision and types of services in order to optimize outcomes. Here we sought to measure various facets of parenting in a drug court population and explored differences by risk. We found no significant differences by DCP risk/need level in parental monitoring (e.g. knowing where your child is), positive parenting (e.g. giving rewards or compliments), parent-child communication (e.g. discuss child-related problems with child), parental involvement (e.g. play games together, child involved in planning family activities), or parent-reported child mental health. These similarities could be because this group of DCP have developed their insight and problem-solving skills given their on-going treatment exposure at the time of interview. Another obvious possibility is that individual differences in justice-related measures of risk/need are not as meaningful or applicable to family-related matters. Finding significant differences in parenting behaviours would have implications for the type and intensity of family intervention needed at the individual level. Similarities in parenting behaviours do not, however, translate to lack of need: 15% of DCP reported poor monitoring practices, and 13% reported using inconsistent discipline practices. Our findings speak to several other areas relevant to parenting in drug court families.

First, the majority of DCP fell below the high-risk criteria for potential child maltreatment and had low frequency of corporal punishment. These findings differ from previous research that indicated substance-using parents reported harsher discipline practices than non-substance using parents placing the child at greater risk for maltreatment (Kelley, Lawrence, Milletich, Hollis, & Henson, 2015; Kepple, 2018). One-quarter of DCP, however, scored above the at-risk cut-off for child maltreatment potential thus indicating need for improving communication, problem-solving skills, and nurturing capacity. While strong substance abuse treatment curricula address these issues in part they are not designed to address effective parenting. Second, a unique aspect of this research is the inclusion of the other caregiver in these families. Our data include the DCP as parent and their parenting counterpart, each reporting on their own parenting behaviors and their perceptions of their child’s mental well-being. Collectively these DCP report less parental involvement, more inconsistent discipline practices, and were more likely to report using non-corporal punishment or other discipline strategies (such as time out, privilege restriction, and extra chores) compared to the other (matched) caregivers. Next we examined the parenting and mental health characteristics of DCP and then compared these characteristics to those of the child’s other caregiver. Both caregivers also reported their perceptions of their child’s adaptive and problem behaviors. To our knowledge this is the first examination of how parenting differs within adult drug court families.

Parenting and adult mental health status outcomes among DCP largely did not differ between different levels of risk for recidivism (LSI-R) or between type of drug used (e.g., criminogenic drug or not). There was one notable area of difference by risk among DCP: those at higher risk for recidivism reported greater distress arising from perceptions of bodily dysfunction (somatization). This is in keeping with a long line of research documenting the disparities in health and mental health in justice-involved populations (National Research Council, 2014). An estimated 75% of individuals under correctional supervision have a co-occurring mental illness and SUD (Osher et al., 2012). Here, DCP reported more and more severe symptoms related to mental health distress. The DCP were at elevated risk across all subscales of the BSI. Example feelings and thoughts measured in the BSI include lack of motivation, nervousness and tension, thoughts and impulses that are persistent and irresistible, isolation, and feelings of personal inadequacy. There is a clear need for the availability of adult mental health services for DCP with 57% reporting symptoms at a clinical risk threshold level when compared to non-patient norms. The DCP were at significantly greater risk than the other caregivers on the global severity index (measure of symptom severity) and each of the BSI subscales. Here the comparison between parents has implications for child well-being and family dynamics that in turn could impact the DCP’s recovery. Further, these comparisons demonstrate the other caregivers’ need for support as it relates to co-parenting with the DCP. Implementation of family-focused mental health interventions and the involvement of family members improves child outcomes and family functioning and could provide important lessons for drug court programming (Foster et al., 2016; Reupert & Mayberry, 2016).

When DCP who are parents are engaged in their children’s lives, there are important implications for prevention and intervention around adult and child health and wellbeing. This research indicates that DCP play an active role in their children’s lives, but they also rely heavily on other caregivers. Six in 10 DCP live with at least one other adult in the home and 58 DCP had a matched other caregiver in our sample. More than half of the DCP’s minor children did not live with the DCP. This is not surprising given the stressful and disruptive nature of CJS involvement and parental substance use on child and family functioning (Arditti, 2015; Turanovic et al., 2012; Uggen & McElrath, 2014). Thus, interventions introduced in a drug court setting targeting the children of DCP must consider the importance of including or involving other caregivers of these children. This also suggests that the relationship between parents may need to be addressed as a function of the toll their addiction has had on relationships. This research identifies some of the ways family-centered work in adult drug courts, or other justice settings, face unique challenges.

The 100 DCP in this sample were caregivers to 179 children under the age of 18 and 74% saw their children on a daily or weekly basis. Some adult drug courts address parenting (Carey, Finigan, Crumpton, & Waller, 2006; Cissner et al., 2013) though such services are not universal and the national landscape of this is unknown. Given the well-established negative relationship between parental substance use and criminal justice involvement on child outcomes (Kopak & Smith-Ruiz, 2016; Murray, Loeber, & Pardini, 2012; Phillips, Erkanli, Costellow, & Angold, 2006), drug courts have an opportunity to impact family and child well-being in positive and proactive ways.

There are several limitations worth noting before universally implementing public health interventions in the CJS setting. The sample size limits the precision and generalizability of the findings. Though a strength of this study is the information about family structure, maltreatment, and parental mental health, the sample only included 58 matched pairs, which precludes the ability to design a universal intervention strategy that would benefit all drug court parents, their children, and families. The outcome measures have not been validated in the adult drug court population, which limits the ability to generalize these findings to other populations or to other drug court participant parents. Further, conclusions rely on retrospective self-report for many outcomes of interest. Integration of administrative data or health records might offer a way to validate self-reports. As these data are from a larger ongoing trial, no tests of significance were done between DCP with and without a matched other caregiver as this study focused solely on the risk parental substance use and criminal justice involvement has on child wellbeing. While we included the highly co-morbid risk factor of parental mental health, this is likely not sufficient in characterizing the potential risk on child wellbeing. For example, intimate partner violence is also highly correlated with both substance use and maltreatment (Barth, 2009; Kohl, Edleson, English, & Barth, 2005). Another limitation of the findings presented here is the reliance on parent report of child behavior and symptoms related to child mental health outcomes. Future research should include child self-report when possible and age appropriate, as neither parent can know everything that the child feels or experiences. It remains an empirical question as to how the introduction of these services might impact children’s mental and behavioral health, and how these services might impact the success of DCP in drug court. Future research should examine the effect of interventions related to mental health and parenting in a drug court setting not only in regard to child and family outcomes but also in relation to recidivism or lapses in sobriety.

Conclusion

As children and families are not often the focus of CJS interventions, this paper described DCP as parents and compared their parenting characteristics with other caregivers to identify related needs and potentially inform public health intervention strategies leveraging involvement in the CJS to address other public health priorities. The promise of integrating the CJS and public health approach is supported by the successful integration of psychosocial, behavioral, and medical interventions to address ancillary DCP needs (Binswanger, Redmond, Steiner, & Hicks, 2012) and is supported by improved child outcomes, such as reunification, resulting from parental involvement in family treatment courts (Bruns, Pullmann, Weathers, Wirschem, & Murphy, 2012; Gifford et al., 2014; B. L. Green, Furrer, Worcel, Burrus, & Finigan, 2007; Worcel, Furrer, Green, Burrus, & Finigan, 2008). Though, it is important to note that referral to a family treatment court is through the child welfare system as compared to the CJS for an adult drug court. Therefore, these two court programs serve distinct populations and there is limited redundancy in developing intervention approaches in either setting.

Inquiry into the families and children of DCP necessitates the collaboration between criminal justice and public health systems (Guastaferro et al., 2016). In this particular case, the CJS provides access to an underserved, hard-to-reach parenting population (DiPietro & Klingenmaier, 2013; Hammett, Gaiter, & Crawford, 1998) with regard to the primary and secondary prevention of public health priorities affecting adult drug court parents, their child(ren), and other caregivers of their child(ren). It is likely the collective action of both systems will be more effective than independent efforts.