![]()
Caregiver Substance Use and Child Outcomes: A Systematic Review
MICHELE STATON-TINDALL, PHD
Associate Professor, College of Social Work, University of Kentucky, Lexington, Kentucky, USA
GINNY SPRANG, PHD
Professor, College of Medicine, University of Kentucky, Lexington, Kentucky, USA
JAMES CLARK, PHD
Professor, School of Social Work, University of Cincinnati, Cincinnati, Ohio, USA
ROBERT WALKER, MSW, LCSW
Assistant Professor, College of Social Work, University of Kentucky, Lexington, Kentucky, USA
CARLTON D. CRAIG, PHD
Associate Professor, College of Social Work, University of Kentucky, Lexington, Kentucky, USA
In spite of widespread concern that children living with substance-misusing caregivers are experiencing greater risk for maltreatment, little research examines the direct effects of caregiver substance use on child outcomes. This systematic review investigates the work done within and across disciplines of adult substance abuse, child welfare, and child mental health, including the measurement of key terms, conceptualization of primary variables, and suggested implications for translational science to practice. The findings of the review show considerable shortcomings for examining this complex problem. To move research forward, we suggest ways to improve measures and methods to provide more robust support for inferences about child maltreatment and mental health outcomes.
The National Center on Addiction and Substance Abuse (CASA) estimates that about 50% of U.S. children live with a caregiver who uses alcohol, illicit substances, or tobacco (CASA, 2005). Tobacco is the most common substance used in many of these homes (approximately 27 million), but an estimated 17 million children are exposed to caregivers engaging in binge drinking, and 9.2 children million have caregivers using illicit substances (CASA, 2005). The Substance Abuse and Mental Health Services Administration (SAMHSA) recently reported that more than 8 million children between 2002 and 2007 lived with at least one caregiver who abused alcohol or used illicit substances during the past year or was dependent on those substances (National Survey on Drug Use and Health [NSDUH], 2009).
The relationship between caregiver substance misuse and child maltreatment has been the target of research in the clinical and empirical literature for many years (Dunn et al., 2002; Johnson & Leff, 1999; Magura & Laudet, 1996; Testa & Smith, 2009; Wells, 2009). A number of reasons have been offered for why substance misuse adversely affects children. The abuse of substances can diminish caregivers’ ability to respond to children’s cues for nurturing, and can impair judgment of priorities related to care, supervision, and guidance (Testa & Smith, 2009). Caregiver involvement with drug-using lifestyles often leads to unstable and chaotic home environments for children and can even involve exposure to crime or toxic substances. Caregiver substance misuse can also seriously affect caregiver expression of emotion, which can interfere with secure caregiver–child attachments. These problems then can lead to negative consequences for children’s development of emotional regulation, confidence, social skills, and trustful relationships with others (Department of Health and Human Services [DHHS], 2009). Research indicates that these children disproportionately suffer from other mental health, behavioral, and academic problems (DHHS, 2009). Caregiver substance misuse has also been documented as a powerful predictor of severity of maltreatment of children (Sprang, Clark, & Bass, 2005).
In addition, caregiver substance misuse is among the most powerful and robust predictors of children developing addiction disorders as adolescents (e.g., Widom, White, Czaja, & Marmorstein, 2007), and this suggests that substance use among adolescents is a coping strategy to deal with their chaotic and violent family lives (Kilpatrick, Acierno, Saunders, Resnick, & Best, 2000). Such adolescents are more likely to become involved with the criminal justice system (Huebner & Gustafson, 2007; Murray, Janson, & Farrington, 2007).
Given the findings about the effects of caregiver substance misuse on children, research that concurrently examines these effects from both the caregiver and child perspective is very limited. Typically, the research has either examined maltreatment among children with some consideration for caregiver substance misuse, or it has been focused on adult substance abuse, with peripheral examination of parenting behaviors. Thus, one body of literature has taken a child protective interest as it examines caregiver substance misuse (Johnson & Leff, 1999; Kroll, 2004). Another very large body of research has looked at adult substance misuse patterns and the interests and needs of these individuals for treatment and rehabilitation (Magura & Laudet, 1996; Testa & Smith, 2009). The difference in perspectives between child protective interests and adult rehabilitation mirrors the tension in public policy between goals of child protection versus family reunification. These dual foci are often at odds in cases when both cannot be satisfied and one interest must predominate. Thus far, the literature has done little to bridge the gap between these two perspectives by examining how integrated treatment (for substance misuse and parenting) or other interventions might serve both goals. These distinct research perspectives have yielded important information for the substance misuse and child maltreatment fields, but they pose challenges to practitioners and policymakers when converting scientific information into either policy or practice. It is reasonable to speculate that the relationships among these phenomena are not simple and universal across all caregiver–child configurations, and that their effects will vary for different children living in various contexts.
The purpose of this article is to review relatively current literature that examined caregiver substance misuse as it impacts or influences child maltreatment with consideration for the specificity of substance use measures, child maltreatment measures, and the degree to which studies could link caregiver substance use with child outcomes. Essentially, the question revolved around how well the emerging literature steps outside the two research silos of child maltreatment and substance abuse.
To explore this problem, a review was conducted to examine relevant, empirical work that has been accomplished within and across disciplines of adult substance misuse, child welfare, and child mental health. The objectives were to include literature that (a) examined measurement of caregiver substance use as an independent variable, (b) examined child maltreatment as a dependent variable, and (c) examined the implications of the existing research for addressing this problem.
METHOD
Systematic Literature Review and Analysis
This review was based on literature searches of peer-reviewed articles in scholarly databases. One primary search engine was Academic Search Premier (ASP), which enabled searches in Medline, PsychINFO, and Social Work Abstracts. Given the medical implications of substance misuse and child maltreatment, PubMed (PM) was also utilized. Primary search terms with the initial number of displayed articles for those terms in parentheses were caregiver substance use and child maltreatment (7 articles ASP, 9 articles PM), caregiver substance use and child neglect (3 articles ASP, 31 articles PM), caregiver substance use and child abuse (10 articles ASP, 30 articles PM), parental substance use and child neglect (22 articles ASP, 332 articles PM), and parental substance use and child abuse (60 articles ASP, 331 articles PM). Given the large body of research on child trauma in relation to caregiver behavior, we also included “caregiver substance use and child trauma” in the search terms, but curiously, this only yielded four articles across both search engines (ASP and PM). One of the articles was repeated under the other search terms, and the other three articles did not meet search criteria for the study. Therefore, for consistency and clarity of operationalization, the primary dependent variable search term focused on issues related to child maltreatment rather than child trauma.
From the initial display of articles for each of these terms, final articles were selected for systematic review based on meeting the targeted criteria for this review based on the following decision rules:
1. Articles had to be empirical, data-driven studies (no literature reviews or meta-analyses).
2. Parental or caregiver substance use must have been used as an independent variable. For example, articles that included substance use as a covariant or mediating variable were not included in this analysis.
3. The primary dependent variable for the study must have focused on the child—including any form of child maltreatment such as neglect or abuse, accidental injury, or behavioral or psychological outcome.
4. The analytic approach was intended to examine the relationship between caregiver substance use and child outcomes.
5. Articles with a publication date since 2000 were included to garner the most recent approaches in measurement and the most current thinking in implications for practice and research.
Twenty-three articles met the inclusion requirement for this study. A content analysis approach was used to address each of the primary objectives of the article. For Objective 1, the articles were organized to define measurement of key constructs, including substance misuse, child outcomes, and potential covariants used in the analysis. Other key elements of the sample were also outlined for descriptive purposes. For Objective 2, the articles were organized by their outcome relative to child maltreatment, including (a) acts of maltreatment (physical abuse, sexual abuse, neglect, and frequency and severity of harm) or (b) other child outcomes (any and all emotional, cognitive, or behavioral response, typically to either a maltreatment event or response to caregiver substance misuse), and primary findings were described. The table was then reorganized so that child maltreatment outcomes were grouped together and other child outcomes were together to easily compare study findings. For Objective 3, discussion sections of each article were carefully examined and organized into tabular format and compared for their implications for particular types of professions. To show the practice relevance for findings, we identified potential user groups broadly into child welfare, child mental health, or adult caregiver substance misuse professions.
RESULTS
Complexity of Measurement
CAREGIVER SUBSTANCE MISUSE
As shown in Table 1, some articles used standardized and widely used substance use assessment tools such as the Addiction Severity Index (Connors-Burrow, Johnson, & Whiteside-Mansell, 2009; Van de Mark et al., 2005), Composite International Diagnostic Interview Short Form (Barth, Gibbons, & Guo, 2006), Michigan Alcohol Screening Test (Jester, Jacobson, Sokol, Tuttle, & Jacobson, 2000; Ondersma, 2002), and the Timeline FollowBack Method (Jester et al., 2000). These tools allowed for assessment of a number of variables to determine indicators of problem severity associated with alcohol or illicit substance use. Using either these or other tools, a few articles examined their primary measure of substance use in terms of frequency of drug and alcohol use (Bailey, Hill, Oesterle, & Hawkins, 2009; Berger, 2005). Although these articles reflect consistency with the overall substance use literature in assessment of problem severity and frequency of use, they were limited in comparison to the larger body of literature on adult substance misuse independent of relationship to child care or maltreatment. The majority of articles in this review included only a single item indictor of “any substance use,” and the primary method of data collection included secondary analysis from child protective service (CPS) records. Other measures to note included perceptions of caregiver substance misuse by the child or adolescent or the CPS worker. Thus across these studies, validated and extensive self-reports, brief self-reports, investigation records, and provider and professional opinions were used, making across-study comparisons impossible.
CHILD MALTREATMENT
Measures of child maltreatment were more clearly defined. Standardized tools such as the Child Behavior Checklist (CBCL; Achenbach, 1991) were commonly used to measure parent’s report of child internalizing behavior and child externalizing behavior (Bailey et al., 2009; McNichol & Tash, 2001; Ondersma, Delaney-Black, Covington, Nordstrom, & Sokol, 2006; Ostler et al., 2007; Van de Mark et al., 2005). Other well-validated tools such as the Conflict Tactics Scale (Straus & Gelles, 1990) were used to examine exposure to specific violent acts between adult caregivers from which inferences of traumatic child witnessing could be made (Berger, 2005; Jester et al., 2000). In addition, other scales such as the Things I Have Seen and Heard Scale (Richters & Martinez, 1992) were used to assess exposure to violent activities among children, again allowing for inferences about child trauma or maltreatment (Connors-Burrow et al., 2009; Ondersma et al., 2006). However, the most consistent and widely used dependent variable was the investigator opinion about the presence of suspected or substantiated abuse and neglect charges. As earlier, the primary method of data collection for these variables was extraction from CPS records. This review also found more articles that focused on child welfare and safety issues than issues specific to psychosocial or behavioral outcomes such as trauma or specific developmental consequences.
TABLE 1 Measurement Summary of Substance Use and Child Outcome Variables
Notes. CBCL = Child Behavior Checklist; IPV = intimate partner violence; NSCAW = National Survey of Child and Adolescent Well-Being; CIDI-SF = Composite International Diagnostic Interview Short Form; CPS = child protective services; DSM–IV = Diagnostic and Statistical Manual of Mental Disorders (4th ed.; American Psychiatric Ass...
