Elsevier

Child Abuse & Neglect

Volume 81, July 2018, Pages 235-248
Child Abuse & Neglect

Research article
Understanding child protection decisions involving parents with mental illness and substance abuse

https://doi.org/10.1016/j.chiabu.2018.05.005Get rights and content

Abstract

Among children investigated for maltreatment, those with parents experiencing mental illness or substance abuse are more likely to be placed out-of-home; however, little is known about why these children are at greater risk. Using a sample of 2488 Structured Decision Making® assessments administered in San Francisco county from 2011 to 2015, we identified a profile of safety threats that accounts for why workers are more likely to determine children of parents with mental illness and/or substance abuse unsafe in the home. Eight percent of assessments in our sample involved parents with current mental illness only and 10% had comorbid substance abuse. The odds of an unsafe determination more than doubled among parents with mental illness (OR = 2.52, p < 0.001) and were nearly tenfold higher among parents with comorbid substance abuse (OR = 9.62, p < 0.001). Three safety threats accounted for all of the effect of parental mental illness on safety determination: caretaking impairment due to emotional stability/developmental status/cognitive deficiency (57%), failure to meet a child’s immediate needs (30%), and threats of harm (14%). Three safety threats accounted for 55% of the effect of comorbid mental illness and substance abuse on safety determination: failure to meet a child’s immediate needs (21%), presence of a drug-exposed infant (21%), and caretaking impairment due to emotional stability/developmental status/cognitive deficiency (14%). Results suggest that sustained linkage to effective mental health services and material resources at the outset of a child welfare case may help to promote faster and more likely reunification, and prevent future maltreatment.

Introduction

Ensuring the safety and wellbeing of children involves critical decisions about when and how to intervene with families reported for maltreatment. Although children of parents with mental illness and/or substance abuse are at increased risk of out-of-home placement (Park, Solomon, & Mandell, 2006; Westad & McConnell, 2012), no study has examined the child protection decision-making processes that lead to this determination. Our study addresses this critical gap in the literature by evaluating which safety threats, as assessed by child protection workers, account for the association between parental mental illness/substance abuse and safety determination. Understanding why child protection workers are more likely to deem children of parents with mental illness and/or substance abuse unsafe could help focus service planning for this population and prevent future maltreatment.

Section snippets

Maltreatment

Child maltreatment is a critical public health concern. In 2014, nearly 44 children per 1000 were referred for maltreatment, and rates of investigation or alternative response rose more than 7% from 2010 to 2014 (Administration for Children & Families, 2016). Of all children investigated or provided alternative response, 19% were victims—an average of 9.4 children per 1000 (Administration for Children & Families, 2016). Maltreatment is associated with increased social, emotional, and behavioral

Safety assessments and child protection decision-making

When maltreatment referrals are screened in for investigation, CPS workers conduct an in-home assessment to determine current child safety, often using an inventory of safety threats. Common threats include unmet basic needs, hazardous living conditions, and signs of physical abuse, among others (DePanfilis & Scannapieco, 1994; Pecora, 1991). Child safety in the home is determined largely, though not exclusively, based on this threat assessment.

Whereas some assessments include mental illness

The present study

We hypothesize that a specific profile of safety threats documented during assessment accounts for most of the effects of parental mental illness and substance abuse on safety determination. Identification of these threats may help inform the development of more targeted services. Over time, this may prevent more serious child welfare involvement, and promote family unity and healthy development among children of parents with mental illness and/or substance abuse.

Study context

In this study, we examined initial child maltreatment assessments administered by Family and Children’s Services (FCS)—San Francisco’s public child welfare agency. FCS administers all procedures involving child maltreatment referral, screening, and assessment, and also provides in-home services, out-of-home placements, and reunification and adoption services.

FCS uses the Structured Decision Making® (SDM) system (National Council on Crime & Delinquency; Children’s Research Center, 2015) to

SDM® Safety assessment

SDM® safety assessment version 2 assesses five factors influencing child vulnerability, 13 safety threats, 10 caretaker and child protective capacities, and 10 safety interventions (see Fig. 2). All of these safety assessment indicators are dichotomous (0 = no, 1 = yes). We disaggregated the first of the 13 safety threats (which comprised five sub-threats,) resulting in a total of 17 unique safety threats. Based on safety threat assessment, the child protection worker determines the child’s

Mediation with dichotomous outcomes and mediators

Mediation analysis estimates the proportion of the variance in outcome y predicted by exposure variable x that is accounted for by mediating variable m—the indirect effect of x on y. When mediators and outcomes are dichotomous (as in this analysis), conventional approaches to mediation analysis often provide unreliable estimates of indirect effects (Mackinnon & Dwyer, 1993; VanderWeele, 2015, p. 33). The statistical data analysis program STATA (StataCorp, 2017) offers several mediation packages

Exposure strata

Table 1 provides summary descriptive statistics of the sample. Eight percent of assessments involved a parent with MI, 13% involved parents with SA, and 10% involved parents with MISA. Fifty-six percent of parents had no MI or SA, and 12% had insufficient information to assign an exposure stratum.

Caretaker relationship to child

Roughly 80% of all referrals documented two or more caretakers. Overall and by exposure, birth mothers (96%) and fathers (79%) were the most common caretakers documented on referrals. We found no

Discussion

Our results indicate that four safety threats (caretaking impairment due to emotional/developmental/cognitive deficiency, caretaking impairment due to substance abuse, failure to meet immediate needs, and having a drug-exposed infant) accounted for much of the total effects of all three exposures on safety determination. For MI, threats of harm had a significant, though small, indirect effect. Although MI accounted for a small proportion of the total variance in safety determination (3%), all

Conclusions

Our findings suggest that the majority of the effect of parental mental illness (with or without substance abuse) on safety determination is due to a parsimonious profile of safety threats that tends to involve impaired caretaking capacity, failure to meet a child’s immediate needs, and (in the case of comorbid or primary substance abuse) prenatal exposure to substances. These findings have practice implications. When working with parents affected by mental illness, we must respond effectively

Compliance with ethics requirements

This research does not involve any studies with human or animal subjects.

Funding source declaration

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflicts of interest

None.

Authorship declaration

All authors approved of this manuscript. This manuscript has not been previously published. Aspects of this paper were previously presented at the 2017 Society for Social Work and Research conference in New Orleans and the 2017 Society for Prevention Research conference in Washington, DC.

Acknowledgements

We wish to thank the San Francisco Human Services Agency for their support of our work and Dr. Stacey Alexeeff for statistical consultation.

References (32)

  • P. Ender

    Binary mediation: A new command to compute mediations with multiple mediators and binary and continuous variables in stata

    (2010)
  • L.D. Hollingsworth

    Child custody loss among women with persistent severe mental illness

    Social Work Research

    (2004)
  • N. Jewell

    Statistic for epidemiology

    (2004)
  • W. Johnson

    Effectiveness of California's child welfare structured decision making (SDM) model: A prospective study of the validity of the California family risk assessment

    (2004)
  • U. Kohler et al.

    Khb: Stata module to decompose total effects into direct and indirect via khb-method

    (2015)
  • J.E. Lansford et al.

    A 12-year prospective study of the long-term effects of early child physical maltreatment on psychological, behavioral, and academic problems in adolescence

    Archives of Pediatrics & Adolescent Medicine

    (2002)
  • Cited by (32)

    • Multisystemic therapy for child abuse and neglect: Parental stress and parental mental health as predictors of change in child neglect

      2022, Child Abuse and Neglect
      Citation Excerpt :

      In a more recent study with a larger sample size and more precise mediation models, the authors found evidence that failure to meet immediate needs of the child was the major mediator between current and chronic mental health problems on the one hand and assessments of child safety on the other (Roscoe et al., 2021). The authors concluded that providing high-quality mental health services for parents might make it possible to recover their caregiving and parenting skills and prevent the recurrence of maltreatment (Roscoe et al., 2018). On the level of parent-child interaction, parental stress plays a crucial role for child neglect.

    View all citing articles on Scopus
    View full text