“I HAVE A CHANCE TO MAKE IT DIFFERENT FOR THIS CHILD”: AN EXPLORATION OF MATERNAL CUMULATIVE RISK AND CHILD WELFARE OUTCOMES OF CHILDREN BORN DURING MATERNAL INCARCERATION By Natalie Blake Kasiborski A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Social Work - Doctor of Philosophy 2014 ABSTRACT “I HAVE A CHANCE TO MAKE IT DIFFERENT FOR THIS CHILD”: AN EXPLORATION OF MATERNAL CUMULATIVE RISK AND CHILD WELFARE OUTCOMES OF CHILDREN BORN DURING MATERNAL INCARCERATION By Natalie Blake Kasiborski Introduction: Children of women involved in criminal justice may be at greater risk for involvement in the child welfare system than children of fathers involved in the criminal justice system, since mothers are likely to be primary caregivers prior to jail or incarceration (Glaze & Maruschak, 2008). In addition, a greater proportion of women in the criminal justice system, as compared to men, experience multiple risk factors simultaneously. Purpose: The aim of this dissertation is to explore the intersection between the child welfare and criminal justice systems, as well as the role of maternal cumulative risk on the outcomes of involved mother/child dyads. Chapter One provides an introduction to the topic, background on the issue, and the organization of the dissertation. Chapter Two expands upon the introduction by exploring the literature on families who experience simultaneous maternal criminal justice and child welfare involvement. Chapter Three uses a cumulative risk scale to compare child welfare outcomes of high and low maternal cumulative risk groups, while Chapter Four utilizes a collective case study approach to explore the movement of children into and within the child welfare system. The fifth and final Chapter links together the themes and findings of preceding chapters and continues the discussion, elaborating on implications for both criminal justice and child welfare policy and practice. Methods: Chapters Three (N=47) and Four (N=4) build upon previous research using data from a sample of pregnant women sentenced to state prison and follow-up data obtained ten years later from multiple state-level administrative databases. Results: The mother-child dyads that intersect with criminal justice and child welfare systems face many risk factors. Application of a cumulative risk framework further differentiates those with the highest risk and greatest likelihood of negative outcomes. Results also suggest that the presence of systematic barriers make it difficult to simultaneously navigate criminal justice and child welfare systems. Discussion & Implications: Cross-system collaboration between criminal justice and child welfare systems is needed to be able to provide targeted intervention aimed at the high risk mother-child dyads who are dually navigating these systems. In addition, collaboration is essential to systematic data collection which will allow researchers to better understand the scope of this issue. Copyright by NATALIE BLAKE KASIBORSKI 2014 This dissertation is dedicated to: My parents and siblings, Jim, Carolyn, Terry, Kathy, Lindsay, Meghan, Kevin, Kate, and Michael. This would not be possible without your love, support, or childcare. My sweet sons, Connor and Blake. You inspire me and fill my days with joy. You have forever changed my context and transformed my life in the most incredible ways. Finally, to my husband Brian. I could not ask for a more supportive and encouraging partner. This is as much yours as it is mine. v ACKNOWLEDGEMENTS I would like to extend my appreciation to Dr. Angie Kennedy, Dr. Gary Anderson, and Dr. Kimberly Maier, for their guidance and support throughout the doctoral education and dissertation writing process. I am especially grateful to Dr. Sheryl Pimlott Kubiak, my committee chair, who provided feedback through multiple reviews and revisions. Her encouragement and guidance was critical to my success. vi TABLE OF CONTENTS LIST OF TABLES x LIST OF FIGURES xi CHAPTER ONE: AN OVERVIEW OF THE DISSERTATION RESEARCH Introduction Pregnancy and Incarceration Prison Nurseries and Residential Facilities Caretaking During Confinement Foster Care Child protective services Reunification Cycle of Child Welfare Involvement Purpose of the Study Theoretical Approach: Cumulative Risk Cumulative Risk and Child Maltreatment Cumulative Risk and Maternal Incarceration Defining Key Concepts Child Welfare Criminal Justice Organization of the Dissertation Chapter 2, Manuscript 1 Chapter 3, Manuscript 2 Chapter 4, Manuscript 3 Chapter 5 1 1 2 4 5 6 7 8 8 9 11 12 13 14 14 15 15 16 16 16 16 CHAPTER TWO: THE INTERSECTION OF MATERNAL CRIMINAL JUSTICE AND CHILD WELFARE INVOLVEMENT Abstract Introduction Background Criminally Involved Mothers: Need for Special Emphasis Review Rationale: Risk and CW Involvement Across the CJ Continuum Substance use disorders CJ continuum Arrest Arrest and Child Care Policy Arrest and Out-of-Home Placement Conviction Collateral Consequences Confinement in Jail or Prison Reasons for Incarceration Among Females vii 18 18 18 19 20 21 22 23 23 24 25 26 26 31 32 Care-giving During Confinement Kinship care placements Foster care Parole, Reintegration, and Reunification The Role of Policy The Adoption and Safe Families Act of 1997 ASFA and termination of parental rights Reunification bypass The “Policy Vacuum” Cycle of Child Welfare Involvement The Role of Race Future Research Needs and Recommendations The Need for Gender Distinction The Need for Systematic Data Collection Conclusion 32 32 33 34 36 37 38 38 39 40 40 42 42 44 45 CHAPTER THREE: MATERNAL CUMULATIVE RISK AND MALTREATMENT OF CHILDREN BORN DURING MATERNAL INCARCERATION Abstract Background Cumulative Risk Current Study Method Sample Maternal Risk Factors: Rationale and Location Within the Data Race Maternal education Previous children Relationship status Abused or neglected as a child Sexual assault Risk Groups Sample Descriptives Data Collection Department of Human Services data Vital statistics Child Welfare Involvement Analysis and Results Discussion Limitations of the Study Conclusion 46 46 46 47 49 49 49 50 51 51 51 52 52 52 53 54 55 56 56 57 58 59 61 62 CHAPTER FOUR: EXPLORING PATHWAYS THROUGH FOSTER CARE -- A COLLECTIVE CASE STUDY OF CHILDREN BORN DURING MATERNAL INCARCERATION 64 viii Abstract Introduction Incarcerated Mothers and Termination of Parental Rights Current Study Methods Sample Selection Inclusion Criteria Data Sources Criminal history Human services Analysis and Results Researcher As Instrument Cases Mother A Target Child A Mother B Target Child B Mother C Target Child C Mother D Target Child D Salient Themes Substance abuse Complexities of maternal relationship Efforts at reunification Worker subjectivity and system limitations Conclusions 64 64 65 67 68 69 69 70 70 71 72 72 73 73 74 76 77 78 79 80 80 81 81 83 85 87 88 CHAPTER FIVE: CONCLUSION Overview Major Findings Practice Correctional settings Community settings and interagency collaboration Policy Social Work Education Limitations of the Study Recommendations for Future Research Focusing on Children of Criminally Involved Mothers Identifying and Tracking Efforts 90 90 90 92 93 94 95 97 98 98 98 99 REFERENCES 101 ix LIST OF TABLES Table 3.1. Dichotomized Cumulative Risk Score Variables 50 Table 3.2. Cumulative Risk Factors 54 Table 3.3. Demographic Characteristics of Mothers 54 Table 3.4. Post-Birth Child Welfare Involvement 57 Table 3.5. Cumulative Risk and Protective Service Involvement 58 Table 4.1. Descriptive Case Characteristics 73 x LIST OF FIGURES Figure 2.1. Criminal Justice Continuum 19 xi CHAPTER ONE: AN OVERVIEW OF THE DISSERTATION RESEARCH Introduction Nearly 7.5 million children have a parent who is incarcerated in federal, state, or local jail, or who is supervised under a form of community corrections, such as probation or parole (Glaze & Maruschak, 2008; Mumola, 2006; Schirmer, Nellis, & Mauer, 2009). In the last 20 years, the number of children with an incarcerated mother has more than doubled, increasing 131% (Glaze & Maruschak, 2008). The majority of criminally involved women report having at least one minor child (Glaze & Maruschak, 2008; Schirmer, et al., 2009) and estimates indicate that up to 25% of women enter prison pregnant, or within one year postpartum (Knight & Plugge, 2005; Siefert & Pimlott, 2001; Wooldredge & Masters, 1993). Although a smaller proportion, the children of criminally involved women may be at greater risk for negative outcomes including maltreatment and involvement in the child welfare (CW) system than children of criminally involved fathers. Perhaps the primary reason for this increased risk is that women are more likely to be primary caretakers of children upon arrest or incarceration than men (Glaze & Maruschak, 2008). Criminally involved mothers experience many risk factors that potentially impact both criminal justice (CJ) and CW outcomes. When a mother experiencing these risk factors is the primary caregiver, her children likely experience these same risks. Criminally involved mothers are often poor (Myers, Smarsh, Amlund-Hagen, & Kennon, 1999), have limited education (Glaze & Maruschak, 2008), have experienced childhood abuse or trauma themselves (DeHart & Altshuler, 2009), and often suffer from physical and mental health problems, including substance abuse (Glaze & Maruschak, 2008; Hanlon, O’Grady, Bennett-Sears, & Callaman, 2005; Kubiak & Arfken, 2006). In fact, the presence of multiple, cumulative risk factors led authors Myers, Smarsh, Amlund-Hagan, & 1 Kennon (1998) to declare, “Children whose mothers are in prison or jail are among the riskiest of the high risk children in our nation” (p. 11). This dissertation focuses on children of incarcerated women, and more specifically, children born during maternal incarceration. Larger bureaucratic social systems govern the lives of these women and their children and, as such, this dissertation explores the overlap of the CW and CJ systems. Given the multiple risks that the mothers, and as a result, the children experience, this dissertation uses a cumulative risk theoretical perspective. This approach theorizes that it is the sum total of risk factors, rather than the individual risk which influences likelihood of negative outcomes, allowing for a holistic portrayal of their lives. Using a cumulative risk framework, this dissertation explores the relationship between maternal CJ involvement and simultaneous involvement of her children in the CW system. In addition to providing a detailed introduction and overview of existing literature on this group, both qualitative and quantitative methods are used to further explore cumulative risk in this context. Part of exploring the ‘riskiest of high risk’ children, those with an incarcerated parent, is determining what happens to them during maternal incarceration—particularly if a mother is a sole caregiver prior to entering prison or jail. There is no universal procedure used by states, nor is there a recognized and adhered to best practice; rather, there are different strategies used across states. Included among these is the use of the formal child welfare system to assume the responsibility of care for the child, often resulting in temporary or permanent separation from the mother. Below is information presented about some of these strategies and the consequences that result when there is an absence of options. Pregnancy and Incarceration 2 Estimates indicate up to 10% of women enter jail or prison pregnant (Clarke et al, 2006; Hotelling, 2008), and up to 25% have given birth within the last year (Bloom & Steinhart, 1993; Greenfeld & Snell, 1999; Grella & Greenwell, 2006; Sefiert & Pimlott, 2001; Wooldredge & Masters, 1993). It is necessary to note that it is likely that this number is an underestimation. Not only are some of these statistics, by the most recent measure, nearly 20 years old, but in addition were published before the number of incarcerated women increased dramatically, likely impacting these measures. The result of pregnancy during confinement on birth and developmental outcomes is a topic that needs further exploration. Using infant birth weight as a measure, some researchers argue that incarceration during pregnancy may actually benefit the health and wellbeing of the baby. Multiple research studies have examined infant birth weight as a birth outcome and found that infant birth weights of incarcerated mothers are similar to birth weights of low-risk comparison groups (Kyei-Aboage, Vragovic, & Chong, 2000; Martin, Kim, Kupper, Meyer, & Hays, 1997; Martin, Rieger, Kupper, Meyer, & Qaqish, 1997;; Moses & Potter, 2008). However, the majority of research indicates incarcerated women are more at risk for increased negative prenatal and postnatal outcomes than the general population of expectant mothers (Fertz & Clarke, 2012, Understanding Prison Health Care, 2002 as cited in Hotelling, 2008). In addition to the high-risk background many of these women come from, this increased risk may be a product of the stressful prison conditions in which these women live and give birth. Despite reports drawing attention to the practiced method of shackling women during labor and delivery (Amnesty International, 2005; National Women’s Law Center, 2010, Sussman, 2009), women continue to be shackled during labor and birth. Illinois was the first of ten states to make shackling during labor illegal; however, the State now faces a class action suit from former jail 3 inmates who allege they were restrained during birth (Mastony, 2010). A report published by the National Women’s Law Center (NWLC, 2010) graded individual states based on policies for prenatal care and delivery practices with incarcerated women, focusing on four specific areas including prenatal care, shackling, family based treatment, and prison nurseries. Among their findings: 38 states received failing grades for inadequate or absent policies related to prenatal care, 36 received failing grades related to shackling and restraints during labor/delivery and postpartum, and 38 received failing grades due to absence of prison nurseries which would allow for the development of an attachment relationship in a supported environment. Prison Nurseries and Residential Facilities After giving birth in prison, the majority of women must arrange for outside care of their newborn, and rely on a friend, family member, or formal foster care placement to care for their children during their confinement. Although rare, there are some programs that allow mothers to parent infant and young children while concurrently serving their sentence. These programs focus on facilitating attachment while enhancing parenting skills; such programs allow women to keep their children with them through either a prison nursery or a community-based setting. Prison nurseries are not a new model. In fact, there is evidence that indicates, despite a lack of formal programming, children may have been present in American prisons as early as 1800 (Craig, 2009; Zemans & Smith, 1964). In the early 1900s, when women prisoners were confined in “reformatories”, usually in the form of small cottages, children up to age two were allowed to remain with their mother (Allen, 1974; Craig, 2009; Kurshan, 1992). Craig (2009) speculates that support for these programs was interrupted by the Great Depression and World War II. However, it was in the 1960s, after two social workers declared prison to be an inappropriate setting for children, that prison nurseries became almost non-existent ( Heffernan, 4 1993). In recent decades as the number of incarcerated women began to increase, there has been a renewed interest in mother-child programming within correctional facilities. As of 1994, there were more than 230 programs for incarcerated mothers and their children (Craig, 2009); however, the majority of these are community-based programs. There are currently only nine states (California, Illinois, Indiana, Ohio, Nebraska, New York, South Dakota, Washington, and West Virginia) offering prison nursery programs (Villanueva Frome, & Lerner, 2009). The majority of these programs are relatively new; all but one was developed in the last 20 years, and four were developed in the last five years. The organization or format of these programs varies greatly. The duration that a mother-child dyad is allowed to remain in most prison nursery or community programs ranges from 30 days to three years; however, some community programs allow children up to school age (Villanueva Frome, & Lerner, 2009). (Note: For a detailed table listing all prison nursery programs and an overview of community programs see Mothers, Infants, and Imprisonment: A National Look at Prison Nurseries and Community-Based Alternatives (Villanueva Frome, & Lerner, 2009). Caretaking During Confinement When a mother is confined, her children are more likely to be displaced from their home, unlike when a father is incarcerated and children live with the other parent (mother) 88% of the time. Yet when a mother is incarcerated, the other parent (father) assumes the role of caregiver only 37% of the time (Glaze & Maruschak, 2008). Upon maternal incarceration, the most common type of caregiver arrangement for the child is a kinship placement with grandparents or other relatives (67%), most likely a maternal grandmother. This is in contrast to 17% of incarcerated fathers who have a child placed in kinship care setting (Glaze & Maruschak, 2008). 5 Maternal incarceration has been cited as a contributing factor to the increase in grandparents raising their grandchildren (Dressel & Barnhill, 1994; Glass & Honeycutt, 2002). The majority of children are placed with a relative, yet kinship placement is not necessarily synonymous with permanency or stability. Rarely do caretakers seek official guardianship through formal court proceedings; this creates a network of informal care giving placements (Kubiak, Kasiborski, & Schmittel, 2009). This occurs for a multitude of reasons, including a lack of trust in the legal system or a fear of foster care involvement. A comparison of kinship and foster care in a sample of children with incarcerated mothers found the quality of care equal for infants and young toddlers; but for young children ages 3-6 years, quality of care was significantly better in foster care than in kinship care (Gaudin, & Sutphen, 1993). This is attributed to the availability of resources, including a higher socio-economic status and more support networks. In addition, foster parents reported more developmentally appropriate expectations of the children. This research suggests an area that can be targeted through the use of tailored services. Placing children in formal kinship placements through the CW system, rather than informal placements, would provide caregivers with access to more resources to potentially improve the quality of care. These additional resources include financial support and social support, in addition to education about child development and healthy discipline techniques, all of which act as protective factors and aid healthy development of the child. Foster Care Not only are kinship care placements more likely when a mother is incarcerated compared to a father, foster care placements are also more likely. Children of an incarcerated mother are more than five times more likely to enter foster care than when father is incarcerated. According to the Bureau of Justice Statistics, 11% of children with an incarcerated mother are 6 currently in a foster home or agency placement, compared to 2% of children with incarcerated fathers (Glaze & Maruschak, 2008). This figure has gradually increased since 1997, when 8% of incarcerated mothers reported a child in foster care (Johnson & Waldfogel, 2002). However, it is likely that these numbers underestimate the actual overlap between the CW and corrections systems. While these figures represent children of incarcerated mothers currently in foster care, when we examine mothers who have ever had a children in foster care, this number increases to nearly one-third of women (Grella & Greenwell, 2006; Moses, 2006). This measure may be even greater when children who have moved from foster care into adoption are added into the calculation. When the overlap between foster care and maternal criminal involvement is examined from the perspective of the CW system, research shows that approximately one-third of children in foster care have a mother with a history of criminal involvement (Ehrensaft, Khashi, Ross, & Wamsley, 2003; Pecora, Kessler, Williams, O’Brien, Downs, et al, 2005). Mothers may have more than one child in foster care, therefore making up a larger percentage of the foster care system. Parental incarceration has been identified as a reason for growing foster care caseloads (Swann & Sylvester, 2006); however, the majority of children actually enter foster care prior to their mother’s arrest and incarceration (Ross, Khashu, & Wamsley, 2004), which implies that other behaviors trigger protective service involvement rather than incarceration. Child protective services. Prior to entering foster care, children are often brought to the attention of human service officials through a protective service referral. Families with a recent arrest most likely come to the attention of protective services for complaints related to neglect, rather than abuse include charges of failure to provide, failure to supervise, and abandonment (Phillips, Burns, Wagner, & Barth, 2004). 7 Many of the families brought to the attention of protective services overlap within the CJ system. A nationally representative sample found that 1 in 8 children who were the subjects of recent protective service investigations had a parent who was recently arrested (Phillips, Burns, Wagner, & Barth, 2004). In the same sample, when a parent had been recently arrested, the likelihood of out-of-home placement was almost three times more likely (Phillips, Burns, Wagner, & Barth, 2004). Of all children who remain in the home following a CPS investigation, 5% have a parent (95.6% mothers) sentenced to probation within the previous year (Phillips, Leathers, & Erkanli, 2009). Other risk factors commonly present in this population, such as maternal depression and substance abuse, may also be influencing factors in protective service investigations. The most common risk factor noted by caseworkers among recently arrested parents was impaired parenting, likely a consequence of substance abuse issues (Phillips, Burns, Wagner, & Barth, 2004). Reunification. Despite protective service and/or foster care involvement, incarcerated mothers have expectations that they will reunite with their child post-release (Foster & Hagan, 2009). They are often unprepared, however, for the barriers to successful reunification. Out-ofhome placement in foster care makes reunification more difficult following incarceration, (Hayward & DePanfilis, 2007). Children of incarcerated mothers in foster care are less likely than other children in care to be reunited, and are four times more likely than their counterparts to remain in care (Moses, 2006). In samples of incarcerated parents with children in the CW system, African American children were both overrepresented and less likely to reunify (Courtney, 1995; Hayward & DePanfilis, 2007; Wulczyn, 2004) than Caucasian children. Cycle of Child Welfare Involvement 8 It is important to note the presence of a cycle of CW involvement within this population. Many mothers themselves were involved in the CW system as children. Current rates indicate that approximately one-quarter of incarcerated mothers spent time in foster care during their own childhood (Dalley, 2002; Glaze & Maruschak, 2008). Within one sample of female substance abusing offenders, those who had parental rights terminated were more likely than their counterparts to have been in foster care or were adopted themselves (Grella & Greenwell, 2006). Placement in foster care during childhood shapes s a child’s developing attachment (Schofield & Beek, 2005) and may impact later parenting ability. This lack of skills, and perhaps resources, combined with an absence of healthy attachment and absence of positive parenting to model may increase the likelihood of formal CW involvement in this population. These factors may also act as barriers to reunification, eventually contributing to the termination of parental rights. While the use of targeted services, such as counseling, parenting classes, and treatment of substance abuse, have been connected with successful reunification (D'Andrade, & Nguyen, 2013), the navigation of these systems is often difficult. Purpose of the Study Despite the myriad risk factors present in the lives of these women and their children, the majority of criminally involved mothers continue to parent and provide care for their children. According to the Bureau of Justice Statistics (Glaze & Maruschak, 2008), prior to incarceration 75% of incarcerated mothers were primary caregivers for their children and nearly half were living in a single parent household. A recent study of incarcerated mothers found that the majority of women (67%) retained their legal parental rights ten years after their initial incarceration (Kubiak, Kasiborski, Karim, & Schmittel, 2012). Although criminally involved mothers may be temporarily separated from their children due to confinement in state or federal 9 prison or local jail, research has shown that despite confinement not only do the majority of women retain their legal rights, many mothers also retain their maternal identity and desire to parent, continuing to show concern about the well-being of their child(ren) (Brown & Bloom, 2009; Moe & Ferraro, 2007). This dissertation explores the intersection between maternal CJ involvement and the entanglement of her infant/child in the CW system. This investigation is primarily interested in the relationship between maternal cumulative risk factors and negative child welfare outcomes. Many of the risk factors associated with maternal criminality, such as poverty, substance abuse, and mental health issues are also risk factors related to CW involvement (Phillips & Detlaff, 2009). The presence of risk factors like poverty (Berger, 2004; Connell, Bergeron, Katz, Saunders, & Tebes, 2007; Sedlak & Broadhurst, 1996), substance abuse (Smith, Johnson, Pears, Fisher, & DeGarmo, 2007; Young, Boles, & Otero, 2007), and intimate partner violence (IPV) (Appel & Holden, 1998; Tajima, 2000) increase a child’s risk for maltreatment and foster care involvement. Despite the presence of many overlapping risk factors that indicate the population of children of criminally involved mothers may be at elevated risk for abuse and neglect, there is an absence of empirical research conducted on the relationship between maternal criminal involvement and CW overlap. Previous research examining this interface often groups mothers and fathers together, examining “criminal parents” as a whole (see Phillips, Burns, Wagner, & Barth, 2004; Phillips, Detlaff, & Baldwin, 2010; Phillips, Leathers, & Erkanli, 2009) or focuses primarily on foster care involvement (see Ehrensaft, Khashu, Ross, & Wamsley, 2003; Johnson & Waldfogel, 2002; Ross, Khashu, & Wamsley, 2004) without attention devoted to maltreatment or to the trajectory of protective service investigation that precedes foster care involvement. 10 The purpose of this study is to address this gap in the research by using prospective, longitudinal administrative data to explore the relationship between maternal criminal involvement and connection to the child welfare system in a sample of children born to women during maternal incarceration. The research utilizes a cumulative risk framework, illustrating how the presence of concurrent risk factors affects the likelihood of CJ and CW systems interface. Moreover, the research explores the CW outcomes of the mother-child dyads within these systems. This knowledge will address a gap in the literature and will advance the small but growing body of research that can inform both CJ and CW policymakers and practitioners alike. Theoretical Approach: Cumulative Risk This dissertation draws from cumulative risk theory, using a cumulative risk model to examine how the multiple risk factors in the lives of criminally involved women influence the long-term interface with the CW system. The dissertation explores these risk factors both quantitatively and qualitatively in two separate chapters. In one chapter a scale of risks is used to predict child welfare outcomes and in a subsequent chapter an in-depth case study is used to provide context to the risk factors. To fully understand the relationship between maternal criminal and CW involvement, the context of these risk factors within the family system must be explored as well. Ongoing research on the impact of risk factors on individual and family outcomes has led researchers to conclude that it is not the specific risk that is the best predictor of negative outcomes, but rather the number of risk factors present, or “cumulative risk” (Sameroff, Seifer, Baldwin & Baldwin, 1993). Within this approach, no single risk factor is responsible for unfavorable outcomes, but rather it is the accumulation of multiple risk factors that negatively impact outcomes. Using cumulative risk as a guiding theory, the more risk factors that are 11 present, the more likely the individual is to experience adverse or unfavorable outcomes. Cumulative risk theory, applied to the topic of this dissertation, would imply that the greater the number of risk factors experienced by the mother, the more likely the presence of adverse CW outcomes. The notion of cumulative risk originated from the Rochester Longitudinal Study (RLS) (Sameroff, Seifer, & Zax, 1982; Sameroff, Seifer, Zax, & Barocas, 1987, Sameroff, et al.,1993; Sameroff, Bartko, Baldwin, Baldwin, & Seifer, 1999). Designed to study intergenerational transmission of psychiatric disorders, the RLS followed children from prenatal though adolescence. Data showed that high cumulative risk was associated with lower verbal IQ scores (Sameroff, Seifer, Barocas, Zax, & Greenspan, 1987) and that scores were inversely related to social competence (Sameroff, Seifer, Zax, & Barocas, 1987). Cumulative risk models are commonly used to study child development outcomes such as behavior, mental health, and academic performance (Appleyard, Egelans, van Dulman, & Sroufe, 2005; Atzaba-Poria, Pike, & Deater-Deckard, 2004; Fursentenberg, Cook, Eccles, Elder, & Sameroff, 1999). Cumulative risk has also been used to examine various CW outcomes including maltreatment and foster care. Cumulative Risk and Child Maltreatment The total number of risk factors has also been linked to adverse CW outcomes. A recent study of caregivers of children between ages 3 and 6 explored child maltreatment potential using various risk frameworks (Begle, Duman, & Hanson, 2010). Results indicated that cumulative risk model was a significant predictor of child maltreatment potential. In a sample of substance abusing women, Nair and colleagues (2003) concluded that parenting stress and child abuse potential were higher for women with more cumulative risk. In addition to maltreatment potential, when maltreatment recidivism was examined using a cumulative risk framework, a 12 higher number of caregiver risks were associated with post-reunification re-entry into foster care (Festinger, 1996). Larrieu et al. (2008) examined risk factors as predictors of foster care outcomes, and concluded that maternal cumulative risk was a stronger predictor of custody loss than individual maternal risk factors. Cumulative Risk and Maternal Incarceration Cumulative risk has also been used to explore adverse outcomes in children of incarcerated parents. Poehlmann (2005) studied the role of caregiver cumulative risk on intellectual and cognitive outcomes of children of incarcerated concluding that sociodemographic risks predicted negative outcomes. Using a nationally representative sample, Johnson and Waldfogel (2002) studied the role of cumulative risk and how it relates to placement during parental incarceration. The authors used the 1997 Survey of Inmates in State and Federal Correctional Facilities (US Department of Justice, Bureau of Justice Statistics, 2000), creating a cumulative risk score based on the following risk factors: 1) Less than 12th grade education, 2) parental substance abuse, 3) parental mental health, 4) socioeconomic status, 5) parental physical or sexual abuse history, 6) parent previously incarcerated, 7) parent ever in foster care as a youth, and 8) parent’s own parents had ever been incarcerated. While scores ranged from 0-8, the average risk score was higher for mothers (2.7) than fathers (1.9). When these risk scores were compared to child living arrangements, the authors found that, as the number of risk factors increases, children were more likely to be placed in foster care or agency settings. Although this study (Johnson & Waldfogel, 2002) provides a useful framework, the previous work did not examine a) these associations over time; b) pathways into foster care, or c) case outcomes. Moreover, their cross-sectional data examined children’s whereabouts at the time of incarceration and did not consider the unique situation of birth during incarceration. 13 Defining Key Concepts Although the terms “child welfare” and “criminal justice” will be used throughout this dissertation it is important for the reader to understand that each of these systems is multifaceted with several distinct divisions within each. Due to the dearth of research in this area, we tend to refer to them as monolithic, but it should be understood that the intersections between the two systems can occur at multiple points. Child Welfare The term “child welfare”, abbreviated as “CW” in this dissertation is used to describe the formal government administered system designed to protect the emotional and physical wellbeing of minors. The CW system is organized into specific areas (i.e., protective services, foster care, adoption), as well as organizationally with state, regional, often county-specific, offices which respond to allegations of abuse or neglect and coordinate foster care services. While the majority of criminally involved women retain their parental rights, despite retention of these legal rights, many still interface with the CW system through protective services or foster care involvement. Recent estimates indicate nearly 5.9 million children were referred to child protective services through an estimated 3.3 million referrals of child maltreatment (US Department of Health and Human Services, 2011a). Maltreatment, defined as, “Any act or series of acts of commission or omission, by a parent or other caregiver that results in harm, potential for harm, or threat of harm to a child” (Leeb, 2008, p.11) includes abuse, neglect, and abandonment. Allegations of maltreatment may result in protective service investigations, removal of the child, and potentially, foster care placement. As of 2010, there were more than 400,000 children in foster care (US Department of Health and Human Services, 2011b). Within this 14 dissertation, CW involvement is defined as both protective service and/or foster care involvement, often resulting from child maltreatment allegations. Criminal Justice Within this dissertation, the term ‘criminal justice’, abbreviated as ‘CJ’ refers to the formal system that is responsible in society for processing the violation of criminal laws and the resulting consequence. The CJ continuum begins with arrest following a violation of law. Following an arrest, an individual may be arraigned and charged and held, or charges may be dismissed. If charged, guilt is determined through a bench or jury trial and then the individual is sentenced. Sentencing is dependent on many factors including criminal history and severity of the crime. Sentencing can include community supervision in the form of probation, confinement in the form of jail or prison, or costs and fines. Sentencing could also include a combination of methods. Similar to child welfare, criminal justice system processes can be organized at several governmental levels; federal, state, county, and municipal. Each governmental entity is likely to have its own law enforcement, courts, and detention facility. Organization of the Dissertation This dissertation contributes to the growing body of research exploring intersection of maternal criminal involvement and CW using a multiple-manuscript approach. This framework allows for varied approaches including a comprehensive literature review (Chapter Two), a quantitative analysis (Chapter Three), and a qualitative case study (Chapter Four). Each individual manuscript explores how individual and cumulative risk influence CW involvement while contributing to the dissertation’s broader focus on the impact of these factors on CW outcomes. 15 Chapter 2, Manuscript 1 The first manuscript provides an extensive review of the literature on maternal criminal involvement, CW involvement, and contextual risk factors surrounding the individuals in these systems. The organization of this chapter follows that of the CJ continuum and examines various aspects of the child welfare system as they relate to women who are CJ involved. First, it explores the role of arrest, followed by probation or confinement, concluding with parole. It establishes the need for distinction between criminally involved mothers and fathers, presents the literature to date, and examines how policy contributes to both CW and CJ outcomes. Chapter 3, Manuscript 2 The second manuscript uses a quantitative approach to explore the following research question: Are there differences in protective services involvement between mothers with low and high cumulative risk? The creation of a cumulative risk scale allows comparison of CW outcomes of high and low cumulative risk groups. Chapter 4, Manuscript 3 The third manuscript is qualitative in nature, using a collective case study approach to explore in depth the pathways of entry into and movement within the CW system. The research question posed in Manuscript 3 is: How do children move into and within the CW system after they are born to incarcerated women? Which risk factors influence the termination of parental rights? Chapter 5 The fifth and final chapter connects the individual manuscripts and links together the themes and findings of the individual chapters. It begins with a review of major findings, 16 followed by an application of these findings to both CJ and CW practice and policy. The chapter concludes with recommendations for future research. 17 CHAPTER TWO: THE INTERSECTION OF MATERNAL CRIMINAL JUSTICE AND CHILD WELFARE INVOLVEMENT Abstract The aim of this manuscript is to review the literature on maternal criminal involvement, the intersection of child welfare involvement, and the contextual risk factors surrounding the individuals within these systems. The chapter is organized to mirror the continuum of criminal justice involvement, beginning with arrest, moving to probation, confinement, and then parole. It examines the collateral consequences associated with conviction and how they act as barriers to successful re-entry. This manuscript emphasizes the need for a body of research focused solely on mothers, rather than researching parents as a mixed-gender group. The need for collaboration between criminal justice and child welfare systems is also presented. Introduction Criminal justice (CJ) supervision implies a continuum from community supervision (i.e., probation) to confinement (jail/prison). Of the more than a million women in the United States under some form of CJ supervision, including those confined to prison and local jail as well as those on parole or probation, the vast majority are mothers (Glaze & Bonczar, 2005; Harrison & Beck, 2006). On any given day, there are approximately 106,410 women confined to state or federal prison (West & Sabol, 2009); however, this number nearly doubles, to 208,300 when considering women housed in local jails (Sabol & Couture, 2008). Further, there are an additional 100,000 women on parole, but the large majority of women under CJ supervision, nearly 750,000, are on probation (Glaze & Bonczar, 2009). These numbers are cross-sectional rather than total annual counts, meaning that the actual number of women who cycle in and out 18 of the system annually is far greater, and as a result, the number of children affected is also far greater. There are multiple phases across the CJ continuum, each affecting children differently. Involvement in the CJ system begins with arrest when there is a suspicion of criminal activity. Arrest generally removes someone from the community and begins the process of determining whether there is enough evidence to formally charge the person with the offense. Once charged, there are multiple options for a judge or magistrate to choose, based upon the offense and the individual’s history. They can be remanded to jail until a trial/sentencing or released on bond. Once a trial takes place, the convicted person will be ordered to pay fines, sentenced to jail (short term), prison, (long term), probation (community supervision), or a combination of confinement and community supervision, such as jail and probation or prison and parole. This chapter focuses on the intersection of CJ involvement of women and child welfare (CW) involvement of their children. It begins with a description of criminally involved mothers and highlights the differences between criminally involved mothers and fathers. The paper then uses the CJ continuum as an organizing framework (see Figure 1), focusing specifically upon arrest, probation, or confinement (in jail or prison), and ending with parole, while examining CW intersection at each of these points: Figure 2.1. Criminal Justice Continuum Arrest Probation Confinement  Parole Background Current estimates indicate that nearly 7.5 million children have a parent in the CJ system including those incarcerated in state, federal, or local penal institution and those supervised 19 under a form of community corrections, such as probation or parole (Mumola, 2006). Dallaire and Wilson (2009) give perspective to this number, indicating that 3.5 times more children are affected by parental incarceration than by autism spectrum disorders, and more than ten times the number of children affected by juvenile diabetes. While more than 1.7 million children have a parent in prison (Glaze & Maruschak, 2008; Schirmer, Nellis, & Maurer, 2009); the majority of these children, just over five million, have a parent under some form of community supervision — either probation or parole (Mumola, 2006). Community supervision status may not separate a child from their parent; however, the increased surveillance with the constant threat of confinement for non-compliance with the terms of supervision (e.g., employment, restitution, drug free) can increase stress and affect the parent/child relationship. Criminally Involved Mothers: Need for Special Emphasis Incarcerated parents and their children are commonly studied as one population, grouping mothers and fathers together. However, the descriptive characteristics, experiences, and parental relationship of criminally involved mothers differ from the population of criminally involved fathers. Generally, incarcerated mothers face more risk factors than incarcerated fathers, as described by Glaze and Maruschak (2008) in a Bureau of Justice Statistics Special Report, Parents in Prison and Their Minor Children. When comparing mothers and fathers in state prison who lived with their minor children prior to incarceration, mothers are more likely to report homelessness (8.5 % v. 4.0%), physical or sexual abuse (59.7% v. 16.1%), a current medical problem (50.0% v. 39.7%) and a mental health problem (72.8% v. 54.5%) than incarcerated fathers. Similar distinctions between jailed men and women have been observed as well. A representative sample of jailed individuals found that women were more likely than men 20 to report homelessness, mental health, substance abuse, and previous trauma (Fedock, Fries, & Kubiak, 2013). These gender differences support the premise that criminally involved women are one of the most socially and economically disadvantaged populations. Not only do the characteristics and experiences of women in the CJ system differ from those of men, mothers more frequently than fathers report caretaking responsibilities prior to incarceration (Glaze & Maruschak, 2008). Despite the aforementioned gender differences, research often collapses male and female “parents” into one group (e.g., Bocknek, Sanderson, & Britner, 2009; Hairston, 2007: Hayward & DePanfilis, 2007; LaVigne, Davies, & Brazzell, 2008; Miller, 2006; Phillips, Burns, Wagner, & Barth, 2004; Phillips, Leathers, & Erkanli, 2009). However, examination of child-focused outcomes needs to differentiate between a criminally involved male and female parent due to the gender differences in social/psychological needs as well as variation in care-taking roles. Incarcerated mothers were three times more likely than incarcerated fathers to report they had provided most of the daily care for their children prior to their incarceration (Glaze & Maruschak, 2008). Moreover, mothers were frequently providing this daily care as a single parent. Incarcerated mothers, compared to incarcerated fathers, are more likely to have headed a single parent-household prior to their confinement (Glaze & Maruschak, 2008, Nieto, 2002). This suggests that maternal involvement in the CJ system may have a more direct impact on the child than paternal involvement does, since mothers report being present in the daily lives of their children, thus making it more likely that children experience a disruption in their daily routine when a mother is arrested or confined. The frequency with which mothers assume the primary caregiver role provides additional context to examine the overlap between maternal CJ and CW involvement. Review Rationale: Risk and CW Involvement Across the CJ Continuum 21 In addition to the imperative to differentiate between male and female parents, the purpose of this review is to illuminate the intersection between CJ involvement and the CW system. Many experiences of criminally involved mothers, such as poverty (Child Welfare League of America, 2005; Myers, Smarsh, Amlund-Hagen, & Kennon, 1999), violence exposure (Dallaire, 2007a; DeHart & Altshuler, 2009), and substance abuse (Dallaire, 2007a; Hanlon, O’Grady, Bennett-Sears, & Callaman, 2005; Myers, et al., 1999), are risk factors for CW system involvement of the child (Courtney, McMurty, & Zinn, 2004; DeBellis et al, 2001; Dhillon, 2005; Phillips & Detlaff, 2009) of the child. The presence of multiple risk factors led Myers and colleagues (1999) to declare, “Children whose mothers are in prison or jail are among the riskiest of the high risk children in our nation” (p. 11). Risk factors of the mother are experienced within the same family system and can lead to negative child outcomes. Substance use disorders. The high prevalence of substance use disorders (SUD) within both the CW and CJ systems deserves specific attention. Due to changing drug policy legislation, SUDs and drug related crimes have played a major role on the increase of women in the CJ system, leading to a 888% increase in the number of women incarcerated for drug-related offenses (Lapidus, Luthra, Verma, Small, Allard, & Levingston, 2005). A survey of women in state prison concluded that nearly two-thirds (63.6%) of women who lived with their minor children in the month before their arrest or just prior to their incarceration had a substance use disorder, as measured by criteria in the Diagnostic and Statistical Manual of Mental Disorders (Glaze & Maruschak, 2008). Not only a risk for those associated with the CJ system, there is also an association between SUDs and child maltreatment (Young, Boles, & Otero, 2007). Children with a parent who abuses alcohol and/or illegal drugs are at a greater risk of abuse or neglect than children of parents who do not abuse drugs or alcohol (DeBellis et al, 2001). SUDs 22 are prevalent within both CJ and CW systems, and it is likely that underlying substance issues influences involvement in both systems. CJ continuum. The majority of research exploring the intersection between CW and CJ focuses on incarcerated mothers. There has been less research on the intersection with CW at other phases of the CJ continuum (arrest, probation, or parole). This is unfortunate given the changing status that a person often occupies along this continuum. Identifying a woman’s specific phase along the CJ continuum is an important distinction to make, as their current CJ involvement has implications for prevention and intervention efforts. Moreover, CJ continuum status has implications for the daily care of children and, therefore, the potential intersection with the CW system. Women who have been arrested and are awaiting trial or who have been placed on probation retain the ability to be the primary caregiver for their children; with the multitude of risks, however, children may be already involved in protective services or as a result of arrest. Children of women in jail or prison require an alternative primary caregiver during confinement and the caregiver can be either formal (i.e., established through the CW system) or informal (i.e., privately arranged, such as a family or friend). In addition, women who were formerly incarcerated and who are currently on parole may have had children removed during incarceration, and may struggle with reunification. Arrest A mother’s arrest begins her pathway into the CJ system. The most recent data indicate nearly 3.5 million women are arrested annually (Synder, 2011). However we can only estimate the percentage of these arrests that involve mothers of minor children since this information is not routinely collected. Using data from a nationally representative sample, approximately 12.5% of children brought to the attention of protective services due to maltreatment investigations have 23 a parent who was arrested within the six months prior to the beginning of the investigation (Phillips, Burns, Wagner, & Barth, 2004). Arrest and Child Care Policy At the time of arrest, a person is usually brought into custody until a formal arraignment can be held. At the arraignment, there is a decision as to whether to formally charge the person with the offense they were arrested for or charges are dropped due to lack of evidence. The person usually waits in jail until an arraignment can be arranged. Because a person may be with their children at the time of arrest, there is frequently an issue of deciding who will care for minor children. The majority of criminally involved women have children, yet the majority of local law enforcement agencies do not have a policy to guide officers how to respond when a parent is arrested. If children are physically present at the time of arrest, law enforcement officers will generally contact protective services, which then determine how to proceed. If officers learn that the arrested woman has children, there may also be differences in how (or if) they follow-up on this information based on lack of policy. Preliminary survey findings in one Midwestern state found that less than half of responding law enforcement officers indicated that their department had a written policy or procedure related to assuming responsibility of minor children when their caretaker is arrested (Neville, 2009). An absence of policy likely influences the fact that law enforcement officers do not routinely ask the arrested person about their children. A survey of law enforcement officials found that in only 13% of agencies do officers ask about dependent children when an arrest is made (Nieto, 2002). Not asking about children both prohibits an accurate count of the number of arrested mothers and prevents unidentified children from being connected with the appropriate services, perhaps also causing them to remain in vulnerable situations. 24 Caregivers with an arrest history are also associated with increased risk factors. Compared to households with caregivers who have never been arrested, the households with caregivers with an arrest history have elevated risk including higher rates of substance abuse, domestic violence, mental health and extreme poverty (Phillips, Burns, Wagner, & Barth, 2004; Phillips & Detlaff, 2009; Phillips & Erkanli, 2008). The most common risk factor noted by caseworkers among recently arrested parents was impaired parenting, defined as “unreal expectations of the child, poor parenting skills, or inappropriate or excessive discipline” (Phillips, Burns, Wagner, & Barth, 2004, p. 176). Families with a recent arrest most likely come to the attention of protective services for complaints related to neglect, rather than abuse; such complaints include failure to provide, failure to supervise, and abandonment (Phillips, Burns, Wagner, & Barth, 2004). Arrest and Out-of-Home Placement Arrest history is also related to child out-of-home foster care placement. An increase in number of times a mother is arrested is associated with decreased likelihood they live with their child (Harm & Phillips, 2001; Gabel & Johnson, 1995). This indirect correlation has implications for additional CW involvement in the form of foster care. If a parent had been recently arrested, the out-of-home placement was almost three times more likely (Phillips, Burns, Wagner, & Barth, 2004). Not only is the number of times a woman has been arrested a significant indicator, so too is the type of arrest (ordinance, misdemeanor, or felony). One study found that felony arrest was a significant predictor of foster care placement (Kubiak, Kasiborski, Karim, & Schmittel, 2012). This suggests that providing additional parenting support or services at the time of arrest may prevent additional instances of maltreatment. One study found that only 16.7% of mothers who had contact with criminal authorities eventually were incarcerated 25 (Phillips, Erkanli, Costello, & Angold, 2006). The majority of maternal arrests do not result in jail or prison, but rather in fines and costs, probation, or dismissal of charges or a finding of innocence. Therefore intervention at the time of arrest may be opportune. Although intervention at the time of arrest may be beneficial for some women, for others it may be too late to connect her to appropriate CW prevention or parenting services. A review of CW foster care cases in New York City found that arrest occurred more frequently after children entered care, suggesting a “downward spiral” pattern of arrest and conviction which contributes to the removal and placement of the child, and often results in incarceration (Ehrensaft, et al, 2003). Acknowledging this pattern, if support is provided to mothers when children are removed, it may serve to prevent additional CJ encounters and aid with reunification. Whether intervention occurs at arrest or at removal, there should be systematic methods to identify children at risk and to provide these at-risk families with additional supports. Conviction Irrespective of the type or length of criminal sentence received (i.e., probation versus incarceration), having a conviction can result in stigma, discrimination and collateral consequences. Individuals with a conviction, particularly those with a felony conviction, experience informal social stigma and discrimination through the attitudes of others and routine practices in housing and employment. Applications for employment, and often rental agreements for housing, request information on criminal history that often decreases opportunities for those who admit to convictions. In addition to these less formal sanctions, more formalized sanctions are present for those with a conviction, including both misdemeanor and felony, in terms of what are referred to as collateral consequences. Collateral Consequences 26 In addition to the explicit sanctions associated with the sentence one receives after being convicted of a crime (i.e., incarceration, probation), other implied sanctions occur, in conjunction with the sentence, as a result of state or federal law that restricts social, financial, and citizenship benefits to those with a conviction. Many of these collateral consequences are a result of federal law, and are therefore applicable to all with felony convictions, regardless of the circumstances surrounding their crime or the details of their re-entry. However, states also have the ability to impose additional consequences of crimes through administrative law; most states average 1,000 or more collateral consequences (Love, Roberts, & Klingele, 2013). The National Inventory of Collateral Consequences of Conviction (2013), a project of the National Institute of Justice and the American Bar Association, examine each federal and state jurisdiction individually as they related to 15 separate collateral consequence categories. Examples of these categories include employment, government financial benefits (welfare, food stamps), education, housing, family/domestic rights (such as custody, visitation and name changes) and voting. The implementation and monitoring of ‘collateral consequences’ is often controlled by state, versus federal, regulations, however, they are not universal for everyone with a conviction. They may vary by state as well. Women who have lost their right to vote as a result of collateral consequence are not able to elect officials who will advocate for policy change on their behalf, or who will critically examine policies to determine if they result in unintended consequences. Collateral consequences are often tied to a specific type of offense, particularly drugrelated crimes. For example, women convicted of drug-related crimes are likely to face collateral consequences due to the nature of their crime including lost public benefits like cash assistance and food stamps (Allard, 2002). Section 115 of the Personal Responsibility and Work Opportunity Act of 1996 mandates a lifetime ban on welfare benefits for those convicted of a 27 drug-related felony. States had the option of ‘opting in’ to this regulation. One study exploring the impact of this ban found that the majority of women convicted of felonies were charged with drug use or possession and received a sentence of probation (Kubiak, Siefert, & Boyd, 2004). Women in this sample averaged two children, and it is likely that many of these women were caring for their children prior to their arrest. A sentence of probation likely means these women returned to their home and the children in their custody. Had these mothers been reliant on welfare benefits prior to conviction, and then banned from receiving them as a collateral consequence of their conviction, their ability to provide for their children would be directly affected. The Housing Opportunity Program Extension Act of 1996 in which the federal government authorized local public housing agencies to deny housing to persons convicted of or suspected of drug offenses was designed to be “tough on crime” with the intent of ridding public housing of violent offenders, gangs, and drug dealers (Renzetti, 2001). However, this policy disproportionately impacts women, since women head at least 75% of the 1.27 million public housing units (US Department of Housing and Urban Development, 2000). Without the availability to secure stable housing, women are unlikely to be able to properly care for or reunify with their children. In addition, women in poverty who are not able to care for the basic needs of their children including food and housing, are likely more susceptible to situation that resemble neglect. Another collateral consequence that may act as a barrier to successful re-entry or successful caretaking is the ban on federal student financial aid (US Government Accountability Office, 2005). Without access to federal financial aid, women are unlikely to be able to fund higher education or career development including community college or trade school. Not only is 28 there are correlation between education and economic outcomes, there is also a correlation between education and social mobility, or the movement out of poverty ((US Bureau of Labor Statistics, 2013)). Further, there is a negative correlation between education and both criminal justice (Lochner & Moretti, 2001) and child welfare (Sedlak et al, 2010) involvement, indicating the more education one has, the less likely they are to interface with either of these systems. The collateral consequences highlighted here provide only a snapshot of the way women continue to pay for their crime, long after they complete their sentence. The need for further research on female probationers is a valuable opportunity for researchers to explore the role of collateral consequences on the relationship between CJ and subsequent CW. After an arrest, an arraignment occurs to determine whether there is enough evidence to officially charge the person with the offense. If so, a trial (bench or jury) determines guilt or innocence. If a person is found guilty, or convicted of the crime, they may be given a community sentence or sanction which does not always include physical confinement. Some women may be sentenced to probation depending on severity of the crime committed and previous criminal history, in lieu of jail or prison, or sometimes as part of a “split-sentence” which includes a shortterm jail sentence. Probation is a form of community supervision where the individual must abide by specific conditions and rules; non-compliance may result in confinement (Bureau of Justice Statistics, 2012). As the rates of incarceration began climbing in the 1980s, community supervision and treatment became a more frequent substitute to incarceration as a mechanism to prevent overcrowding in jails and prisons, and to reduce associated costs (Petersilia, 1985). Those with minor offenses may be sentenced to inactive, or non-reporting, probation initially, and in some states successful completion of probation can result in discharge of the original crime. For example, in Michigan there are opportunities for either deferred judgment or delayed 29 sentence (State of Michigan, 2013). A deferred judgment results when an individual does not enter a guilty plea but is sentenced to probation, sometimes in combination with specialty court participation (e.g. drug court, mental health court). Upon successful completion of probation the charge is dismissed. With a delayed sentence, a guilty judgment occurs however incarceration in jail or prison is postponed for up to 1 year pending successful completion of probation or specialty court participation (State of Michigan, 2013). For women with children, deferred judgment and delayed sentence provide opportunities to remain in the home and prevent additional family disruption. For women with first time and non violent offense, these alternatives to confinement offer many positives. For women charged with substance related offenses, these sentencing practices provide an opportunity to connect a mother to drug-court and the substance treatment necessary to prevent future criminal involvement and to provide appropriate parenting. In a nationally representative sample of children who remained in an in-home setting following a protective services investigation, approximately 5% (1 in 20) had a primary caregiver (95.6% mothers) on probation (Phillips, Leathers, & Erkanli, 2009). Despite the presence of a conviction, a probation sentence provides a woman with the opportunity to physically remain in the home and to continue to provide daily care for her child. However, the opportunity to resume her care giving role may be affected if she faces additional collateral consequences of conviction in addition to her formal probation sentence. Conditions of probation require different types of reporting dependent on probation status. Those with an active status may be required to report in person to their probation agency on a weekly basis (Bureau of Justice Statistics, 2012). Women may be required to travel significant distances dependent on where the probation office is located. In addition, probation sentences can include mandates for 30 treatment, drug testing, employment, community service and/or restitution. This can be difficult for women who work hourly, minimum wages jobs and are required to take unpaid time to travel to and from a probation visit or treatment session. In addition, women may face barriers as a result of transportation and child care issues, however missing appointments may result in a probation violation which can lead to jail. Confinement in Jail or Prison There are differing opinions regarding whether the jail and prison populations should be studied independently or together. Jail is administered locally (e.g., county) whereas prison is state or federal level confinement. Pre-conviction, a woman may be temporarily housed in a local jail while waiting for her trial or sentencing, or she may be ordered to serve a short-term sentence in jail. Typically, time in jail is for misdemeanor or low-level felony crimes with sentences of less than a year. Confinement in a state or federal prison is usually for a longer duration of time and is typically reserved for more serious felony offenses. Although this paper ultimately groups together women confined in jail and prison as one group, given the limited body of research on either, it is also important to distinguish differences between the jailed and incarcerated populations and how these differences impact the opportunity for care giving and the potential for CW prevention and intervention. Both jail and prison require separation from children and prevent continuation of daily care giving; however, there are some differing implications for CW involvement due to the length of time a woman is serving. A woman in jail is more likely to be housed closer to her home than a woman who is incarcerated in state prison, although smaller states may combine their women’s jail and prison facilities. This proximity makes visitation more feasible than when a woman is confined to a state prison, which could be hundreds of miles from her children. In addition, because a jail sentence is typically less than a year, while a prison 31 sentence is generally for more than a year, there may also be a greater likelihood that formal guardianship is pursued when a mother has a longer sentence, like those typically encountered in state prison. Despite the identified differences, there is a dearth of research on both jailed and incarcerated mothers. Therefore, for purposes of this paper, maternal confinement will be explored grouping jail and prison together. Reasons for Incarceration Among Females In the last 30 years, the number of incarcerated women has increased at nearly double the rate of their male counterparts (Harrison & Beck, 2005). In fact, since 1977 the number of women confined in state and federal prison rose 832% (West & Sabol, 2009). Research has indicated that the increasing number of incarcerated women is not a result of increasing criminal activity, but rather is a consequence of legislative and policy decisions such as harsh drug laws, mandatory minimum sentencing guidelines, and punitive attitudes regarding punishment (BushBaskette, 2000; Covington & Bloom, 2003; Mauer, Potler, & Wolf, 1999). Current sentencing practices, such as mandatory minimums, ignore the presence of gendered criminal involvement and contextual factors about women and their role in crime. Instead, laws are based on the assumption that all who engage in criminal behavior are also dangerous, and therefore threats to public safety (Covington & Bloom, 2003). This assumption proves false for any individual, but especially for women, who primarily commit drug and property offenses rather than violent offenses (West, Sabol, & Greenman, 2010). Care-giving During Confinement Kinship care placements. Women are more likely to be primary, and often single, caretakers prior to incarceration; therefore, when a mother is confined her children are more likely to be displaced (Glaze & Maruschak, 2008; Johnson & Waldfogel, 2002). When a father is 32 incarcerated, children primarily reside with their mother; however, when a mother is incarcerated, the most common type of caregiver arrangement for the child is a kinship placement with a grandparent or other relative. Compared to 17% of incarcerated fathers who have a child placed in kinship care setting, 67% of incarcerated mothers have a child placed in a kinship care arrangement (Glaze & Maruschak, 2008), most commonly with the maternal grandmother. Maternal incarceration is cited as one of the main reasons for the sharp increase in grandparents raising their grandchildren (Dressel & Barnhill, 1994; Glass & Honeycutt, 2002). Commonly, these placements are informal, and are not official kinship foster care placements through the state human services agency (Kubiak, Kasiborski, & Schmittel, 2010). Additional research is needed to fully understand this informal network of care. Foster care. Just as child caretaking responsibility during incarceration is influenced by gender of the incarcerated parent, rates of child foster care placement vary based on the gender of the incarcerated parent as well. Likely a collateral consequence of a mother’s primary (often single) caregiver role prior to incarceration, children of an incarcerated mother are more than five times more likely to be in foster care than children of an incarcerated father. Of fathers incarcerated in state prison, 2% had a child placed in a foster care setting, compared to 11% of mothers (Glaze & Maruschak, 2008). This number has gradually increased since 1997, when 8% of incarcerated mothers reported a child in foster care (Johnson & Waldfogel, 2002). These numbers likely underestimate the actual overlap between the CW and CJ systems. While these numbers represent children of incarcerated mothers currently in foster care, when mothers who have ever had a children in foster care are included, this number increases to almost one-third of women (Grella & Greenwell, 2006; Moses, 2006). 33 Children of incarcerated mothers face multiple cumulative risks, of which foster care is one. Research indicates that foster care placement is ten times more likely for children in families with the highest cumulative risk compared to children in families with low risk (Kohl, Edleson, English, & Barth, 2005). Paralleling the rising rates of women involved in the CJ system, more incarcerated women are reporting a child placed in foster care (Johnson & Waldfogel, 2002) and maternal incarceration has been identified as a reason for growing foster care caseloads (Swann & Sylvester, 2006) . Although children may be in foster care during maternal incarceration, incarceration is infrequently the cause of foster care placement. Research indicates that children are often placed in foster care prior to their mother’s arrest and subsequent incarceration (Pecora, Kessler, Williams, Downs, et al., 2005; Ross, Khashu, & Wamsley, 2004), implying that protective service involvement, rather than CJ, may be the initial reason for CW involvement. Parole, Reintegration, and Reunification Mothers with a criminal history face many challenges in regard both to reintegrating into the community after confinement and reunifying with their children. Women may be released from incarceration on parole, a condition which allows them to complete their sentence in a community setting (Bureau of Justice Statistics, 2013). Parole may be issued through discretionary release decided by a parole board, or through mandatory release as a result of legal obligation (Bureau of Justice Statistics, 2013). As incarceration rates have continued to climb, some states have amended their parole considerations by up to 90 days resulting in short term population decreases (Maurer, 2013). Once paroled, women often return to the same environment where previously they engaged in criminal behavior (Covington, 2003; Taylor, 1996) and to home situations that may include 34 domestic violence and/or illegal activity (Richie, 2001). Securing housing and employment may create major barriers to re-unification efforts. As previously discussed, conviction is accompanied by many collateral consequences and may prevent women from collecting welfare benefits or accessing federal housing programs, which become barriers to successful re-entry. In addition to establishing a lifestyle that allows for proper care of children, women on parole must also balance the reporting requirements of their parole. Women on parole can be categorized as active or inactive supervision status which impacts the type (in person, by mail, or by phone) and frequency of reporting (Bureau of Justice Statistics, 2013). In addition, parole may mandate involvement in treatment, drug testing, employment, and/or restitution. This may have implications for reintegration; a women who has to attend bi-weekly treatment and has in person reporting requirements for parole, may be unable find a job which allows for such routine absences in addition to her child care-giving responsibilities. When a mother returns home, she must also work to resume her relationship with her children. Mothers may idealize their return home and their role as a mother, yet their children may be angry or distant. Mothers must also navigate the caretaking situation that developed during their confinement. Caregivers may have become attached to the children and are wary to return them to the care of a mother exiting jail or prison. Alternatively, caregivers who have struggled with their role due to behavior or financial burden may be anxious to return the children to their mother. This situation may force women to immediately resume parenting despite being unprepared psychologically. In addition, the decreased access to resources as a result of collateral consequences, and the high-risk environment to which they return, may add burden and stress. Returning to the home environment that may have originally resulted in incarceration may reintroduce women on parole to the risks that originally resulted in 35 incarceration. On parole, women must adhere to the conditions of community release; rules that may be more strict than the actual law and limit their choices and social supports (e.g., not using alcohol, not socializing or living with other parolees, etc.) (Bureau of Justice Statistics, 2013). If children were placed in foster care while a mother was away, reunification may be even more difficult following incarceration (Hayward & DePanfilis, 2007). Children of incarcerated mothers in foster care are less likely than other children in care to be reunited and are four times more likely than their counterparts to remain in care (Moses, 2006). Not only is the mother responsible for meeting the conditions of her release, often through a period of probation or parole that requires formal monitoring by the CJ system, but she must now simultaneously navigate the formal CW system in order to regain custody, all while readjusting to life outside of prison or jail. Once formally released from jail or prison, a woman may still have restrictions including curfews, and may continue be monitored closely through the use of both reporting to a parole officer and electronic tether (Bureau of Justice Statistics, 2013). An electronic tether is a physical monitoring device usually worn around an ankle. A land line telephone within the home is usually required for operation of the tracking mechanism. Mothers with these specific time/place restrictions may not have the freedom of movement to take care of child emergencies (e.g, take a child to the Emergency Room), as this would be a violation of conditions of parole and she would then be subject to possible re-incarceration. While some parole officers may take into account circumstances associated with a parole violation, this introduces additional complexity to issue of parole---worker subjectivity. The Role of Policy Navigation through the CJ and CW systems may be a difficult task in an environment with multiple risk factors. These collateral consequences of incarceration make it difficult to reunite 36 with children and resume parenting, potentially lengthening foster care placement or contributing to the higher rates of terminated parental rights. Mothers have expectations that they will reunite with their child post-release (Foster & Hagan, 2009), but are often unprepared for the psychological, social and financial barriers they may face. Current CW policy adds additional complexity to the reunification process. Research has found elevated rates of termination of parental rights in this population when compared to national averages. However, these higher termination rates may not be due solely to poor parenting, but rather to an unintended consequence of broader social policy, including current CW policy designed to promote permanency for child, the Adoption and Safe Families Act (ASFA) of 1997.. As articulated below, the mandated timeframe and requirements placed by ASFA may contribute to the system navigation problems these women face. The Adoption and Safe Families Act of 1997 The Adoption and Safe Families Act (ASFA) was designed to alleviate the problem of children “stuck” in foster care without a permanent home (Halperin & Harris, 2004) and to curtail the rising rates of foster care occurring with AACWA. The ASFA focused on increasing permanency for children within the CW system, removing the “reasonable efforts” guidelines previously established through the AACWA which stated that CW agencies and caseworkers must make “reasonable efforts” to not only prevent the removal and placement of a child into foster care, but also to promote the return of a child to their home (PL 96-272m AACWA). Where previously, the rights of the biological parents were put first and emphasis was placed on family preservation, the ASFA changed this prioritization (Stein, 2003), placing new emphasis on the safety and permanency of the child. 37 ASFA and termination of parental rights. Following guidelines stipulated within this policy, concurrent reunification and termination plans are made. A permanency hearing must occur within the first 12 months, although there is some flexibility to this. ASFA also requires mandatory petition of termination of parental rights once a child has been in foster care for a total fifteen of twenty two months as a way to prevent children from lingering in foster care (PL 105-89, Adoption and Safe Families Act of 1997), this is commonly known as “15 of 22.” Although designed to facilitate permanency, and targeted at children who appear to be lingering in foster care, the “15 of 22” mandatory petition does not have considerations for extenuating circumstances, such as incarcerated mothers. Given that the average sentence length of an incarcerated woman is approximately 60 months (Glaze & Maruschak, 2009), a majority of the 11% of incarcerated women with a child in foster care will face a mandatory petition for termination of their parental rights. The “15 of 22” mandatory petition to terminate rule can be appealed by the caseworker if they believe or have evidence that termination of parental rights is not the best avenue, but there is evidence that this exception is rarely pursued. The CW League of America (CWLA, 1998) studied termination of parental rights in instances of parental incarceration and found that termination is often pursued first, prior to explored visitation or reunification. In cases where mothers were incarcerated, termination was granted in 92.9% of cases and in cases where both parents are incarcerated, termination occurred 100% of the time (Lee, Genty, & Laver, 2005). Reunification bypass. The high rates of termination within this population may result from the ASFA allowance for “aggravated circumstances” which allows states to deny reunification services to parents who fall into certain categories. Berrick, Choi, D’Andrade, and Frame (2008) examined the use of reunification bypass in one state and found that among the 38 conditions used to bypass reunification are: “Parent convicted of a violent felony” and “Extensive, abusive, chronic history of substance use, and has resisted court-ordered treatment within last three years, or failed care plan compliance for substance abuse treatment twice” (p. 166). These researchers found that nearly half of the cases they examined were eligible for a reunification bypass, and the most common reason recommended was extensive substance abuse, of which the court approved nearly two-thirds of recommendations. Cases recommended for bypass due to parental conviction of a violent felony were approved 100% of the time. Given the previously mentioned rates of substance abuse issues within this population, many women would fall into the reunification bypass category. The mother-child dyads who experience the termination of parental rights are also often the most disadvantaged and experience the most risk factors. When compared to other incarcerated mothers who retained rights, those with terminated rights faced more stressors. Mothers who lost parental rights were younger, less likely to have completed high school, less likely to be partnered, and less likely to hold employment or have any vocational training (Grella & Greenwell, 2006). In addition, mothers who lost parental rights initiated risky behaviors at a younger age, including first arrest and onset of drug use. The resulting termination of parental rights, the most extreme of CW responses, can be attributed to the cumulative risk factors present within this group. The “Policy Vacuum” Halperin & Harris (2004) refer to the absence of policy as a “policy vacuum”; they write: “The failure of CW and corrections bureaucracies to formulate policy to address these problems results in a serious compromise of these women’s rights as parents” (p. 339). As previously discussed, the majority of surveyed law enforcement agencies do not have a policy to address 39 children encountered during an arrest (Nieto, 2002). The lack of policy addressing children of incarcerated parents may be a reflection of the need for a broader social understanding of the collateral consequences of CJ policy and the effect such absence of policy has, not only on the person committing the crime, but on their family and children as well. Current drug policy continues to focus on enforcement rather than treatment, and mandatory minimums result in harsh sentences for first-time offenders. In addition, the construct of these policies disproportionately impacts women, one of the major reasons why this population is increasing so steadily. Legislative decisions that produce unintentional collateral consequences and punish women with “lifetime sentences” also create additional risk factors for the children of these women. Cycle of Child Welfare Involvement It is important to highlight the presence of a cycle of CW involvement within this population. Many mothers themselves have experienced the CW, or foster care, system as children. Current rates indicate that approximately one-quarter of incarcerated mothers spent time in the foster care system during their childhoods (Dalley, 2002; Glaze & Maruschak, 2008). Placement in foster care during childhood may impact later parenting skills, making CW involvement more likely and possibly acting as a barrier to reunification or eventually leading to termination of parental rights. Grella and Greenwell (2006) found that within their sample of female substance abusing offenders, those who had parental rights terminated were themselves more likely to have been in foster care or adopted than female substance abusing offenders whose rights remained intact. The Role of Race 40 When examining the CJ and CW systems, we must acknowledge the disproportionate rate of minorities present in each. Often viewed as a form of institutional racism, racial disparity is seen throughout the CJ system. Women of minority status, especially African Americans, are incarcerated at disproportionate rates (West & Sabol, 2008). In 2000 the ratio of incarcerated African American women compared to incarcerated White women was a 6 to 1(Maurer, 2013). While there was a decrease in this ratio, as of 2009 2.8:1 (Maurer, 2013), it remains an alarming difference given the proportion of African Americans to Whites in the general population. Moreover, incarceration rates of Hispanic women rose 23% between 2000 and 2009 (Maurer, 2013). This racial disparity not only impacts the mothers, but their children. Of the nearly 150,000 children who had a mother in prison, 30% had an African American mother, and 19% had a Hispanic mother in prison (Glaze & Marushack, 2009), an over-representation of minorities. African American children are nine times more likely to have a parent in prison than White children (Mumola, 2000). Thus, it is important to consider the potential negative implications for African American and Latino familes and their communities given their disproportionately greater likelihood of incarceration. This disproportionally in incarceration rates and criminal justice involvement trickles down to child welfare. African American children are over-represented within the foster care system (United States Government Accountability Office, 2007). Despite this disproportion of foster care rates, there is not a higher rate of maltreatment by African Americans within the general population (Sedlak & Shultz, 2005), suggesting other factors contribute to this discrepancy. This over-representation is true for children with parents in the criminal justice system. In a sample of children of incarcerated parents in the CW system, African American children are both over-represented and less likely to reunify (Courtney, 1995; Hayward & 41 DePanfilis, 2007; Wulczyn, 2004). One in five African American children brought to the attention of protective services have a recently arrested parent, compared to 1 in 10 White children and 1 in 20 Hispanic children (Phillips, Burns, Wagner, & Barth, 2004). There are broader societal conditions which influence both CJ and CW and outcomes, namely the correlation with poverty. Family disruption, like the disruption that occurs when a mother is in jail or prison, decreases stability, attachment and weakens relationships. The absence of a strong family support network has been identified as a risk factor for child welfare involvement (Shin & Lee, 2011). The rate of minority children living in poverty is two times higher than the rate of white children, a result of cumulative disadvantage that persists throughout generations (Lin & Harris, 2009). Low income families are more likely to be single parents, and African American families are most likely to be a single parent household (Simms, Fortuny, & Henderson, 2009). Minority families in poverty face barriers to improving their socioeconomic status. Research has identified States with a greater number of minorities on welfare are also more likely to have lifetime limits and to enforce stricter sanctions (Lin & Harris, 2009). These social policies produce unintended consequences which contribute to the over-representation of minorities in both the child welfare and criminal justice systems. Future Research Needs and Recommendations The Need for Gender Distinction Currently, much of the available research focuses on parental incarceration, collectively studying children of incarcerated mothers and fathers. While this research provides general information about this population of vulnerable children, the differences between these populations, many of which are directly related to the gender of the criminally involved or incarcerated parent, create the need for distinguished bodies of research. 42 One of the main reasons for this need for distinction is the parent-child relationship, both before and after confinement. Within this context, relationship refers to the mechanical caretaking role rather than an attachment or emotional connection. Multiple stressors such as poverty may mean that parents are not emotionally available despite being physically present. Exhaustion from the daily stress of their lives may be a barrier to assuming an active parenting role. These relationships between children and their incarcerated parent differ based on the gender of the parent, specifically regarding direct caregiving to the child before and after confinement. Mothers are more likely to be primary caregivers prior to incarceration (Glaze & Maruschak, 2008); therefore, their children may be more at risk for negative outcomes, including CW involvement, due to this primary caregiver relationship and subsequent proximity of the child to the maternal risk environment. The need for gender distinction between incarcerated fathers and mothers also arises as a limitation within the existing research. One prominent researcher commonly groups mothers and fathers together to examine “parents.” For example, in a sample of 87 children in the CW system with probationer parents, the study reported results on “parents” even though the sample of “parents” was 95.6% mothers (Phillips, Leathers, & Erkanli, 2009). Another study examining parental arrest and CW involvement uses a nationally representative sample that is almost 91% mothers (Phillips, Burns, Wagner, & Barth, 2004). Grouping men and women together may possibly skew the results when looking at the risks for CW involvement. From existing research we know that incarcerated mothers are more likely to be single parents, to be victims of domestic assault, and face economic instability, among other characteristics that distinguish this group from incarcerated fathers. Grouping children of incarcerated mothers and children of incarcerated fathers together to examine “children of incarcerated parents” is rationalized by the 43 fact that the majority of children of incarcerated mothers also have incarcerated fathers (Phillips, et al, 2004). However the majority of children of incarcerated fathers do not have incarcerated mothers, and therefore, this logic is problematic. Including fathers in a sample of mothers possibly skews or masks the unmet needs of children of incarcerated mothers. The Need for Systematic Data Collection The children of criminally involved women have been referred to as an “invisible population.” No government entity is responsible for collecting data on the children women leave behind when incarcerated (Gentry, 1998; Lee, Genty, & Laver, 2005; Wright & Seymour, 2000), so we must rely on prisoner self-report or estimates, a gap in the current literature. Estimates indicate more than 5 million children have a parent under community supervision such as probation or parole (Mumola, 2006) and more than 1.7 million have a parent in prison (Glaze & Maruschak, 2008; Schirmer, Nellis & Maurer, 2009), however these numbers are not further broken down by parent gender. Additionally, these numbers may underestimate actual rates since some participants may choose not to disclose information about their children, but it is one of the only measures available. Identification is the first step to ensuring that children of incarcerated mothers are connected with the proper and necessary services. The Rhode Island Department of Corrections has started collecting detailed information on family members and children as part of their intake policy (Brazzell & LaVigne, 2008). However, they are in the minority. The Child Welfare League of America (1997) indicated that only 6% of states had policies in place specific to children with an incarcerated parent, a survey that has not yet been re-administered to measure progress. Without state leadership, the lack of policy is present at the agency level as well. The majority of law enforcement and corrections agencies lack specific policies about placing 44 children when mothers are involved in the CJ system (Nieto, 2002; Smith & Elstein, 1994). Although the ASFA may have created a renewed focus on the safety and permanency of the child, the intended goals of ASFA are not being met, and children are lingering in foster care without being adopted (Lowry, 2004). In addition, as previously discussed, an unintended consequence of ASFA is a higher rate of termination of parental rights in this population. Conclusion Additional attention to this risk-laden group of women and children is necessary to decrease cycles of both CJ and CW familial involvement. Certainly data alone will not provide the impetus for system changes. However, accurate data may influence action. Understanding the dynamic between the CW and CJ systems, the number of families dually navigating these systems, and how families move within and in-between these system is critical knowledge to obtain. Once our foundational knowledge is advanced, we can revisit current policies and their unintended consequences, begin to target interventions for this group, and ideally, increase positive outcomes for those in both CJ and CW systems. 45 CHAPTER THREE: MATERNAL CUMULATIVE RISK AND MALTREATMENT OF CHILDREN BORN DURING MATERNAL INCARCERATION Abstract Purpose: The intersection of families dually navigating the child welfare and criminal justice systems is likely connected to the many risk factors they experience. Previous research has demonstrated cumulative risk is the best predictor of negative outcomes, rather than specific individual risks. Methods: Using a combination of self report and archival administrative criminal justice and child welfare data, this study examines the relationship between maternal cumulative risk and child welfare outcomes in a sample (N=47) of children born during maternal incarceration. Individual risk factors (race, maternal education, previous children, relationship status, and history of abuse as a child) are summed to create a cumulative risk score. Based on cumulative risk score mothers are then categorized as “low” or “high” cumulative risk. A series of independent samples t-tests compares number of protective service allegations and time to first allegation. Results: Mother-child dyads categorized as high cumulative risk have significantly more protective service allegations and significantly more substantiated protective service allegations. Conclusions: Results indicate that application of cumulative risk theory further differentiates those with highest risk, even in a traditionally high-risk group. Background The literature confirms there are families simultaneously involved in both the criminal justice (CJ) and child welfare (CW) systems. As the number of incarcerated mothers has increased in recent years, so has the number of children brought to the attention of the CW system through protective service reports or foster care placements (Ehrensaft et al, 2003; Glaze 46 & Maruschak, 2008; Johnston & Waldfogel, 2002). Approximately, 1 in 8 children who were the subjects of recent protective services investigations had a parent who was recently arrested (Phillips, et al, 2004). Analyzing a sample of children w ho remained in the home following a CPS investigation, 5% had a parent who was sentenced to probation within the previous year. The parent sentenced to probation among this 5% is almost exclusively mothers (95.6%) (Phillips et al., 2009). Research indicates that families with a recent arrest history are more likely than those without a recent arrest history to come to the attention of protective services for complaints related to neglect (failure to supervise, failure to provide, abandonment) rather than to abuse (Phillips, et al., 2004). The overlap of affected families across these two systems is likely connected to the many risk factors they experience. These risk factors associated with both CJ and CW system involvement, include poverty (Berger, 2004; Myers et al., 1999; Nikulina, Widom, & Czaja, 2011; Sedlak et al., 2010), low educational attainment (Glaze & Maruschak, 2008; Murphey & Braner, 2000; Sidebotham & Heron; 2006; Wu et al., 2003), history of childhood abuse or trauma (DeHart & Altshuler, 2009; Thompson, 2006), and mental health and substance abuse issues (Conron, Bearslee, Koenen, Buka, & Gortmaker, 2009; Glaze & Maruschak, 2008; Hanlon, O’Grady, Bennett-Sears, & Callaman, 2005; Kubiak & Arfken, 2007; Sedlak et al., 2010). While the presence of an individual risk factor may result in CJ or CW involvement, research has demonstrated that the best predictor of negative outcomes is actually the quantity of risk factors present rather than the specific risk factor itself (Begle, Dumas, & Hanson, 2010). The accumulation of various risk factors is also referred to as cumulative risk. Cumulative Risk 47 The theory of cumulative risk was born from the Rochester Longitudinal Study (Sameroff, Seifer, & Zax, 1982; Sameroff, Seifer, Zax, & Barocas, 1987; Sameroff, Seifer, Baldwin, & Baldwin, 1993; Sameroff, Bartko, Baldwin, Baldwin, & Seifer, 1998), a study originally intended to explore intergenerational psychiatric disorders. Cumulative risk theory has been furthered studied in the context of negative child developmental outcomes (Appleyard, Egelans, van Dulman, & Sroufe, 2005; Atzaba-Poria, Pike, Deater-Deckard, 2004; DeaterDeckard, Dodge, Bates, Pettit, 1998), and also specifically to explore CW outcomes (Larrieu, Heller, Smyke, & Zeanah, 2008; Nair et al., 2003). Begle and colleagues (2010) found that using a cumulative risk model was significant in predicting child abuse potential; however, they did not study actual instances or occurrence rates. Additional research by Mackenzie and colleagues (2011) individual and cumulative risk factors as predictors of maltreatment, concluding that cumulative risk is the strongest predictor of later maltreatment. Larrieu and colleagues (2008) explored cumulative risk versus individual risk factors leading to permanent loss of custody for mothers of infants and toddlers, and concluded that cumulative risk was a stronger predictor than individual risk factors. Cumulative risk has also been used as an approach within CJ research. Johnson and Waldfogel (2002) found that the likelihood of children of incarcerated parents to be placed in an agency or foster care setting, rather than with a parent, increase as the number of risk factors increase. The current manuscript extends the existing research on cumulative risk, maternal CJ, and CW outcomes in two ways: First, it focuses on particularly high risk mother-child dyads – women who gave birth in prison and the resulting child. Secondly, the study uses actual maltreatment incidents (substantiated and unsubstantiated maltreatment allegations) as the 48 measure of child maltreatment, rather than child abuse potential scores as used by Begle and colleagues (2010). Current Study There is limited research focusing on this small, but risk-laden population of incarcerated mothers and their children, and even less on children born during maternal confinement. This manuscript is the first to examine the relationship between maternal cumulative risk and child maltreatment using longitudinal administrative data in sample of children born during maternal incarceration. Moreover, this manuscript explores the relationship between maternal cumulative risk factors and subsequent child protective service involvement, and hypothesizes that mothers with greater cumulative risk scores will interface more frequently with protective services, and will interface sooner following the birth of the child. The primary research question is: Are there differences in protective service involvement between mothers with low and high cumulative risk? Method Using the initial sample and self-report data from the Women and Infants at Risk (WIAR) study, a project focused on the needs of pregnant and post-partum women within a state prison (Kubiak et al., 2010; Siefert & Pimlott, 2001), this study uses ten years of archival data from state-level human services administrative databases to assess CW outcomes over time. Sample Eligibility criteria for participation in the original study among incarcerated pregnant women required women to have a non-violent offense, a sentence of less than two years, no serious mental illness, agreement to participate, and a desire to parent their child. Between 1996 and 1999, a total of 97 women were part of the original data collection and agreed to participate 49 in the study; however, only 57 completed a trauma screening at intake – an inclusion criterion for the current study. Women who were deceased (n=2) and women who could not accurately be categorized into high or low risk groups (n=8) due to missing data were removed from the sample, resulting in the final sample (N=47). Maternal Risk Factors: Rationale and Location Within the Data The following risk factors were examined and then summed to create a cumulative risk score (see Table 3.1) for determining whether mothers fit into the low or high risk group: race, maternal education, previous children, relationship status, abused or neglected as a child, and sexual assault history. All women who participated in the study had a history of substance abuse as part of eligibility criteria for program participation. Therefore, due to lack of variance, substance use is not included as part of the individual or cumulative risk factors. Definitions of individual risk factors are based upon empirical data and are derived from the data of each case through a review of materials. An index of scores based on the presence/absence of each risk Table 3.1. Dichotomized Cumulative Risk Score Variables Risk Factors Low risk (0) High risk (1) Race White African American/Latina 47.4 Maternal Education >High school graduation or beyond Less than high school completion 59.6 1 (1.5) Previous Children Less than 3 previous children 3 or more previous children 33.3 1 (1.8) Relationship Status Significant other No significant other 52.6 3 (5.5) Abused or Neglected as a No history of child History of abuse/neglect abuse/neglect 29.8 5 (9.1) 50 % with risk Missing (#, %) n/a Table 3.1 (cont’d) Child Sexually Assaulted No history of rape or molestation Indicated history of rape or molestation 56.1 9 (16.4) Race. Just as in the CJ system, a disproportionate number of minority children interface with the CW system (Lee & George, 1999; Magruder & Shaw, 2008; Nedell & Barth, 1998). According to Chibnall et al. (2003), when compared to White children, twice as many African Americans had substantiated abuse or neglect allegations. Therefore, using program intake materials, mothers who identified as African American or Latina were given a score of 1 (high risk), and mothers who identified as White were scored as 0 (low risk). Maternal education. Low maternal education has been associated with increased child abuse potential (Murphey & Braner, 2000; Sidebotham & Heron; 2006; Wu et al., 2003). A score of 1 (high risk) was given to mothers who reported less than a high school education on program intake materials. A score of 0 (low risk) was given to mothers who indicated they had completed high school/GED or beyond. Previous children. Research on family structure has concluded that additional dependent children increase the likelihood of maltreatment. Findings from the National Incidence Study (Sedlak & Broadhurst, 1996; Sedlak, et al., 2010) concluded that children from families with four or more children had the highest incidence rates of abuse and neglect. Based on this research, and previous cumulative risk models that categorized families of children with four or more children as high risk (Sameroff et al, 1987; Nair et al., 2003), women who indicated on program intake materials that they had three or more previous children were categorized as 1 (high risk), since the target child would be at least the 4th child in those families. 51 Relationship status. There is risk associated with both single and partnered status. While research indicates that the presence of a partner is a risk factor within CJ literature (Griffin & Armstrong, 2003) single parent status is been associated with protective service overlap (Needell & Barth, 1998; Paxton & Waldfogel, 1999; Sedlak et al., 2010;Wu et al., 2003,). Since the focus of our analysis is CW outcomes, mothers who reported on program intake materials that they were single or did not have a significant other were categorized as 1 (high risk), while mothers who reported their relationship status as married or indicated the presence of a significant other were categorized as 0 (low risk). Abused or neglected as a child. Research supports the notion of a cycle of abuse, meaning adults who themselves were victims of childhood abuse or neglect are more likely to abuse or neglect their own children (Craig & Sprang, 2007; DelViccio, Erlanger, & Slep, 2013; Hall, Sachs, & Rayens, 1998; Milner et al., 2010). Many incarcerated women experience childhood abuse; approximately 60% report a history of childhood physical or sexual abuse (Glaze & Maruschak, 2008). A score of 1 (high risk) was given to mothers who affirmed any of the following questions asked at intake: You were physically abused as a child, You were seriously neglected as a child or, Even if you did not think of it as ‘abuse’, someone struck you, beat you, attacked you, repeatedly slapped you, or otherwise hurt or harmed you when you were a child. Sexual assault. A score of 1 (high risk) was given to mothers whose answers indicated they were raped or molested by answering yes to either of the following questions, You were raped (someone having intercourse with you against your will using some type of threat or degree of force), and/ or You were molested (someone felt or touched your genitals when you did not want them to). In addition, women were given a score of 1 if they answered yes to the 52 following questions originally administered as part of a trauma inventory, “Even if you did not think of it as ‘rape’, someone (parent, caregiver, spouse, romantic partner, acquaintance, or stranger) forced you to have oral or anal sex or intercourse against your will, or when you were drugged or helpless or Even if you did not think of it as being ‘molested’, someone touched private parts of your body, or made you touch their private parts, against your wishes, or when you were asleep, drugged or helpless.” Risk Groups Women were grouped into ‘Low Cumulative Risk’ and ‘High Cumulative Risk’ groups based on the total number of risk factors identified as present (See table 3.2). Mackenzie and colleagues (2011) used framework established by Sameroff et al. (1998) to subdivide cumulative risk into three categories based on 10 dichotomous risks: low (0-2), medium (3-5), and high (6 or greater). Building on Mackenzie et al. (2011), the current study explores maltreatment using six risk factors in a specified group. The established 0-2 risks definition is applied to categorize the low risk group. Given the smaller number of risk factors within this study (n=6), those with 3 or more risk factors are categorized as high risk. A total of 12 mothers were missing at least one cumulative risk data point. Further exploration of missing data indicated that 4 of the 12 women with missing data would be categorized as high risk (3 or more risk factors) based on the data that were present. Therefore, despite the missing data these women were included in analysis. The distribution of the missing data of the remaining eight women negated our ability to confidently categorize them as either high risk (3 or more) or low risk (2 or fewer). Therefore, the remaining women with missing data (n=8) were excluded from further analysis, resulting in a sample of n=47. 53 Table 3.2. Cumulative Risk Factors Total Number of Risk Factors 0 1 2 3 4 5 N 0 4 13 11 15 4 % 0 8.5 27.7 23.4 31.9 8.5 Risk Group Low Risk (2 or less) n=17; 36% High Risk (3 or more) n=30; 64% Sample Descriptives It should be noted that as part of the WIAR study, some women served their sentence through participation in a community-based program, which allowed them to actively parent; others remained in the state correctional facility. Previous analysis determined there were no differences between groups initially or over time (Kubiak, Kasiborski, & Schmittel, 2010), so they will be analyzed as one group for this study. A child who was born during maternal incarceration is referred to as the target child (TC), since many of the women had previous children. At the time the woman gave birth to the target child, women ranged from 19 to 40 years old (Table 3.3), averaging 28.0 years old (SD 5.4). Property crimes were the most common charge (40%) and nearly a third (28%) of women were charged with parole or probation violations. While all women identified drug and/or alcohol issues at program intake, only 23% Table 3.3. Demographic Characteristics of Mothers Variable All n=47 M(SD) Range High Risk n=30 M(SD) Range Low Risk n=17 M(SD) Range Maternal Age at Birth (Range) 28.6(5.4) 18-40 28.7(5.6) 18-40 28.5(5.4) 21-38 Previous # of Children (Range) 2.19(1.9) 0-9 2.8(2.1) 0-9 1.1(1.1) 0-4 54 Table 3.3 (cont.’d) All n=47 n(%) High Risk n=30 n(%) Low Risk n=17 n(%) 23(48.9) 51.1 18(60.0) 12(40.0) 5(29.4%) 12(70.6%) 59.6 38.3 2.1 8 (26.7) 21(70.0%) 1(3.3%) 10 (58.8%) 7(41.2%) 0 Initial Charge Drug Property Assaultive Prob/Parole Violator 23.4 40.4 8.5 27.7 7(23.3%) 15(50%) 3(10%) 5(16.7%) 4 (23.5%) 4(23.5%) 1 (5.9%) 8 (47.1%) Highest Education Completed 9th grade or less 10th or 11th grade 12th or GED Some college/post HS 21.3 44.7 23.4 10.6 10 (33.3%) 12(40%) 6 (20%) 2 (6.7%) 0 9 (52.9%) 5 (29.4%) 3(17.6%) 25.5 40.4 25.5 4.3 4.3 7(23.3%) 15(50.0%) 5 (16.7%) 2(6.7%) 1(3.3%) 5(29.4%) 4(23.5%) 7(41.2%) 0 1(5.9%) Race African American/Latina White Relationship Status* Significant Other Single Missing/Unknown Drug of Preference Alcohol Cocaine Marijuana Other Unknown were incarcerated for drug related offenses. Looking only at formal drug charges may be a misrepresentation of actual drug-related offenses, as it is likely that property crimes were committed to obtain money to purchase substances, or that probation/ parole violations were due to testing positive for drugs or alcohol. The majority of women in the sample (66%) did not complete high school and the sample was disproportionately African American/Latina (49%). Data Collection 55 In 2008, ten years after the birth of all TC in the original study, approval was obtained from the University Institutional Review Board. The data were collected on the mother and the TCs who was born during incarceration between the years of 1996-1999, from multiple statelevel data sets. Collected data were entered and analyzed using IBM SPSS STATISTICS (Ver. 20.0). Administrative data related to this study are as follows: Department of Human Services data. Following a formal request and approval, administrative data was collected from the Department of Human Services (DHS). DHS is the state agency responsible for administering and retaining information on all CW services (including protective service, foster care, adoption, and termination of parental rights) and financial assistance (such as food benefits, Medicaid, and cash assistance) programs. Both CW and financial assistance data were of interest to researchers. Involvement in either type of program verified the presence of the child within the state. Researchers provided identifying information (name, date of birth, and social security number) about the mother and their target children to a DHS employee, who then searched the DHS electronic records in an attempt to locate either or both within the system. Once the preliminary search was completed, the DHS employee then acted as a consultant to our research team. Two research team members were taught about the state system and various databases within it, including one which contained protective service information and two (one archived, one current) which contained foster care information. Data was then extracted from both the archived and current data bases by research team members and transferred to forms that were designed to capture key information. If a child was not located within any of the CW or financial aid databases, it was assumed that the child did not encounter either system within this state. 56 Vital statistics. Using names and social security numbers, state-level public health records were searched for indicators of morbidity of mother or child. We did not find any evidence of death among the target children; however, we found evidence of two maternal deaths. Cause of death was not available. The deceased mothers and their children were removed from the sample prior to data analysis. Child Welfare Involvement All but one target child (98%) were located within the DHS system (Table 3.4) in either the protective services or financial assistance databases. Table 3.4. Post-Birth Child Welfare Involvement Variable Overall (n=47) Low Cumulative Risk Group (n=17) High Cumulative Risk (n=30) Child present in any DHS database 46 (97.9%) 16 (94.1%) 30 (100%) Child present within protective service database 39 (83.0%) 12 (70.5%) 27 (90.0%) Child has substantiated protective service allegation 29 (61.7%) 5(29.4%) 24(80%) Child present in foster care 20 (43.5%) 2 (11.8%) 18 (60%) Maternal rights terminated 14 1 (5.9%) 13 (43.3%) Protective service allegations ranged from 0 to 12. One mother-child dyad had a total of 28 allegations, more than nine times the mean. For purposes of data analysis, truncation was applied (Osborne & Overbay, 2004); the total number of allegations was truncated at 12, the next greatest number of allegations against a mother-child dyad. The average number of days from the target child’s birth to the first protective service allegation was 830 days (SD 906), which 57 equates to approximately 2.3 years. The average number of days from the target child’s birth to the first substantiated protective service allegation was 1115 days (SD972), or approximately three years. Not quite half of children were present in the foster care database (43%), and just under one-third of mothers lost their parental rights (30%). Analysis and Results A series of independent samples t-tests were conducted to explore the relationship between total cumulative risk and protective service reports (Table 3.5) including the total number of reports and the total number of substantiated reports. Of the 47 mother/child dyads, 7 (17%) had no protective service reports. Table 3.5. Cumulative Risk and Protective Service Involvement Variable Overall Group Total Low Risk High Risk M(SD) M(SD) M(SD) Range Total allegations if 3.9 (3.3) 2.4 (2.4) 4.7 (3.4) present in DHS database 0-12 Total substantiated allegations if present in PS database 1.3 (1.2) 0-6 .5 (.67) Time to first PS allegation 830 (906) 1-393 971 (1171) 1.7 (1.2) 768 (779) t df 2.41* 44 -3.12*** 37 .64 37 Time to first 1115 (972) 1583 (1223) 1018 (912) 1.19 substantiated PS 1-3164 allegation Note. *p =≤.05, ***=p ≤.001. Standard Deviations appear in parentheses below means. 27 Those with high risk had a mean 4.7 (SD=3.4) protective service allegations, significantly more than those who were categorized as low risk group (M=2.4, SD=2.4; t(44)=-2.41, p=.02). The effect size was moderate (eta squared = .11). Those categorized as high risk (M=1.67; 58 SD=1.2) also had significantly more substantiated allegations than those in the low risk group (M-.5, SD=.67; t(37)=-3.12, p=.004). The magnitude of the differences in the means was large (eta squared=.18). In addition to exploring the quantity of protective service reports, the time in days to the first protective service allegation and the time in days to the first substantiated allegation were also explored. There was no significant difference in the time in days to the first protective service allegation between the high risk (M=768, SD=779) and the low risk group (M=971, SD=1171; t(37)=.640). A post-hoc power analysis had power=0.1, indicating that the lack of statistically significant results may be a result of power issues. The high risk (M=1018, SD=912) and low risk groups (M=1583, SD=1223) also lacked a significant difference in the number of days to the first substantiated allegation (t(27)=1.19). Discussion Consistent with previous findings on cumulative risk and child maltreatment (Begle et al., 2010; Larrieu et al, 2008; Johnson & Waldfogel, 2002; Mackenzie et al., 2001), results of this study found differences in child maltreatment occurrences based on categorization of the mother as high or low risk. This analysis of CW outcomes of children born during maternal incarceration results confirm the high risk nature of the mother; more than half were categorized as high risk with three or more risk factors. These numbers reflect the similarity of this population within other groups of incarcerated women nationally. Using a cumulative risk framework further distinguishes risk within a group already categorized as high risk. Mother-child dyads categorized as high risk had significantly more protective service allegations (2.4 compared to 4.7) and significantly more substantiated protective service allegations (0.5 compared to 1.7) than mother-child dyads categorized as low-risk. 59 Additionally, the high cumulative risk group was more likely to be present within the protective service database (90% compared to 71%) and was more likely to have a substantiated allegation (80% compared to 29%). The high cumulative risk group was also five times more likely to be present in the foster care database (60% compared to 12%), and more than seven times more likely to have maternal rights terminated (43% compared to 6%) than the low cumulative risk group. These findings are congruent with the theory that the best predictor of negative outcomes is the quantity of risk factors rather than a specific, individual risk factor. Although all the women in this study would likely be considered high risk, the application of cumulative risk theory to this group further differentiates the highest of high risk. Application of cumulative risk theory to service provision and intervention may provide new insight. Reduction of overall risk factors may be a mechanism to reduce negative outcomes; however, further research is needed in the area. This has potential to shift how parenting and reintegration is approached while women are confined. Reintegration preparation should acknowledge the maternal role that many women assume and recognize that while a woman may not be actively parenting while confined, it is likely she will return to a care-giving role at some point, and therefore, may need specific parenting skills and education. At the same time, policies and practices should also attempt to help women to reduce their overall cumulative risk when possible, in addition to providing educational parenting courses, given the connection between cumulative risk and negative outcomes. While cumulative risk theory is based on the sum of risk factors rather than on specific individual risk factors, results indicate that one notable risk factor, ‘history of sexual assault’, deserves further exploration due to the high percentage of missing data on this specific indicator. While more than half of women in this sample reported experiencing sexual assault in their 60 lifetime (56%), an additional 16% had missing data for this question. Even with more than half of women disclosing their experience, it is likely that the reported number is low, as victims of childhood assault often do not disclose even as adults often due to perceived response others will have (Goodman-Brown, Edelstein, Goodman, Jones, & Gordon, 2003). This frequency is more than three times the national rate (18%) of lifetime prevalence of sexual violence against women (Black et al., 2011), but aligns more closely with the percentage of mothers in prison who have experienced sexual assault within their lifetimes. A survey of mothers in state prison indicated 66% reported ever experiencing sexual or physical abuse (Glaze & Maruschak, 2008). It is likely our frequency would increase if we included physical assault in addition to sexual assault within our analysis. Sexual violence is well-documented as a contributor to substance abuse issues (Brown & Finkelhor, 1986; Kendler et a.l, 2000; Wilsnack, Vogeltanz, Klassen, & Harris,1997), and substance abuse is a risk factor for child maltreatment (Young Boles, & Otero, 2007) and CJ involvement (CITE). The majority of women in this study were incarcerated for property or probation violations likely related to substance use, or for drug related offenses like possession or distribution. These themes of sexual trauma and substance abuse which contribute to negative CJ and CW outcomes reinforces the need for gender informed trauma and substance abuse treatment approaches. Limitations of the Study Since the research in this area is still developing, this study uses a specific group of highrisk mother-child dyads for exploratory research. This approach resulted in a small sample size that is not sufficient for certain analytical approaches. For example, survival analysis was not a feasible analytical approach, although a comparison of time to a specific event would be 61 informative. In addition, this study focuses on the CW outcomes of children who were born during maternal incarceration; therefore, results may not generalize to all children of incarcerated mothers. Future research on this topic should be designed to advance the current study through the use of a larger, perhaps multistate sample, and additional analysis. The current study uses a total of six risk factors to explore the application of cumulative risk to this group. Future research should include additional risk variables, such as maternal depression and maternal self-esteem, to further establish the role of cumulative risk within this group and to continue to establish the application of cumulative risk framework to differentiate even groups that are traditionally considered high risk. Alternatively, smaller studies that gather additional information on specific situations would also broaden our knowledge of specific barriers to successful parenting upon reintegration. Building on our current findings and high rate of sexual assault history, it would be interesting to design future research to look more in-depth at this phenomenon and to explore differences in CW outcomes based on sexual assault history of mothers in the CJ system. Conclusion This study introduces the application of a cumulative risk framework to a sample of women and their children who were born during incarceration. The ability to categorize women into a risk category and the knowledge that women in the high cumulative risk category are more likely to have negative outcomes has implications for social work practice. Both CJ and CW caseworkers should be aware of all of a woman’s identified risk factors, regardless of direct implications for care-giving, as reduction of total risk may lead to more favorable CW outcomes. Poehlmann (2005) studied the role of cumulative risk on cognitive 62 outcomes of children’s of incarcerated mothers, concluding that cumulative risk is mediated by home and family environment. This research confirms the importance of child caretaking during maternal incarceration, whether in a formal foster care placement or a kinship placement. This knowledge also suggests the need for an individualized, hierarchical model of reunification which first assesses the strengths and risks of each individual family unit. Once risks have been assessed, and ideally, cumulative risk is reduced, then interventions such as parenting education can be delivered with a greater likelihood of success. Despite the previously discussed limitations of this study, results indicate that the application of a cumulative risk framework further differentiates those with the greatest risk and likelihood of negative outcomes, even in a high-risk sample. Significant differences in both the quantity of protective service allegations and the quantity of substantiated protective service allegations between high and low cumulative risk groups suggest that reducing overall cumulative risk may be an approach to reducing child maltreatment. 63 CHAPTER FOUR: EXPLORING PATHWAYS THROUGH FOSTER CARE -- A COLLECTIVE CASE STUDY OF CHILDREN BORN DURING MATERNAL INCARCERATION Abstract Purpose: Women in the criminal justice system face many risk factors which put their children at risk for involvement in the child welfare system. This study aims to explore the pathways into and within the child welfare system. Methods: This study uses a collective case study methodology with purposefully selected participants (N=4). Themes are analyzed using an inductive approach. Results: Mother-child dyads faced extensive protective service involvement. Themes of substance abuse and sexual trauma emerged, contributing to the complex maternal relationships between the mothers and their children. Conclusions: Included among factors that contribute to termination are the many risk factors of the mother like substance abuse. Despite similar circumstances, the studied cases had different termination outcomes. Pathways to these outcomes are not transparent, suggesting worker subjectivity contributes to these outcomes. Introduction The most recent estimates indicate there are more than a million women under criminal justice (CJ) supervision including those on probation, in jail or prison, and on parole (Glaze & Bonczar, 2009, Harrison & Beck, 2006). Since 1977, the number of women confined in state and federal prison rose 832%, to the most recent estimate of 106, 410 (West & Sabol, 2009). However, this number is a cross-sectional count, and likely underestimates the actual number of women who cycle in and out of the CJ system each year. In the last two decades, there has been a 131% increase in the number of children with an incarcerated mother (Glaze & Maruschak, 2008). The majority of incarcerated women are not only mothers but were also primary 64 caregivers of their children prior to their arrest (Ehrensaft, Khashu, Ross, & Wamsley, 2003; Glaze & Maruschak, 2008). As the number of women incarcerated in state and federal prisons continues to grow, so does the number of children in out-of-home placements as a result of maternal incarceration. While incarcerated fathers most frequently reported the mother of the child was currently caring for the child, incarcerated mothers most frequently reported grandparents, specifically the maternal grandmother, was caring for the child (Glaze & Maruschak, 2008). Compared to 2% of incarcerated men who report a child in foster care, approximately 11% of incarcerated mothers report a child currently in foster care (Glaze & Maruschak, 2008). These numbers reflect women who currently have a child in a foster care placement; the percentage of those who have ever had a child in foster care is likely greater. Between 1985 and 2000, foster care caseloads more than doubled, an increase that has been linked to the parallel increase of incarcerated women (Swann & Sylvester, 2006). Yet, research indicates that most children actually enter foster care prior to maternal arrest (Ross, Khashu, & Wamsley, 2004), which suggests that foster care involvement may be a result of protective service involvement, rather than a direct product of maternal absence due to incarceration. Incarcerated Mothers and Termination of Parental Rights Once in foster care, children of incarcerated mothers may experience termination of parental rights (TPR) more quickly than parents with children in out-of-home placement because, due to their incarcerated status, mothers may not be afforded opportunities for visitation to assist with reunification (Lee, Genty, & Laver, 2005). One study of women who gave birth while incarcerated found that placement of children in foster care is more likely to lead to TPR (Schmittel, Kubiak, & Kasiborski, 2010), rather than to reunification. Current policies influence 65 this trend. The emphasis on permanency placed by the Adoption and Safe Families Act (ASFA) requires a mandatory petition for TPR when a child has been in out-of-home placement for 15 of 22 months. A study comparing outcomes of all mandatory petition court cases (both those that did and did not lead to TPR) found that those mandatory petitions which resulted in TPR were more likely to involve parental incarceration than the mandatory petition cases that did not result in TPR (Meyer, McWey, McKendrick, & Henderson, 2010). Although children of mothers with a CJ history may be more likely to experience TPR, it is not clear how these children move into and within the CW system or how risk factors influence outcomes. Within this group, multiple risk factors such as experiencing and witnessing violence, maternal substance abuse, and poverty are known to contribute to both CJ and CW involvement (Appel & Holden, 1998, Berger, 2004; Connell, Bergeron, Katz, Saunders, & Tebes, 2007; Glaze & Maruschak, 2008; Myers, Smarsh, Amlund-Hagen, & Kennon, 1999; Sedlak & Broadhurst, 1996; Smith, Johnson, Pears, Fisher, & DeGarmo, 2007; Tajima, 2000; Young, Boles, & Otero, 2007). While individual risks may increase the likelihood of unfavorable child development outcomes, previous research has indicated that it is the total number of risk factors present, or cumulative risk, that is the best predictor of negative outcomes, rather than any specific risk itself (Sameroff, Seifer, & Zax, 1982; Sameroff, Seifer, Baldwin, & Baldwin, 1993; Sameroff, Seifer, Barocas, Zax, & Greenspan, 1987). Total cumulative risk may be more salient than any individual risk factor because it indicates the presence of multiple factors may be more difficult to manage. The relationship between cumulative risk and CW outcomes has also been examined and cumulative risk was found to be a significant predictor of potential child maltreatment (Begle, Duman, & Hanson, 2010). Johnson and Waldfogel (2002) assessed risk factors such as race, 66 education, abuse history, and mental health history in a study of cumulative risk and child placements with a sample of incarcerated parents. They concluded that children with high cumulative risk scores were more likely to be placed in foster care. Current Study The current study is an in-depth exploratory case study designed to further explore the role of risk as it relates to CW involvement within the population of criminally involved women. This collective case study examines the context and circumstances that result in foster care placement within a purposefully selected, small (N=4) group of mothers who gave birth while incarcerated. The purpose of the study is to better understand pathways into and within the CW system and to derive knowledge that could assist in the planning and development of efforts aimed toward prevention and reunification. Once points for intervention are identified, further research can be designed and an appropriate prevention plan can be piloted and implemented. The current research builds upon previous studies, using a sample of pregnant women that entered state prison with low-level offenses (2-year minimum sentence) between 1996 and 2000. Then, 10 of years post-birth data using multiple state-level administrative databases is explored. All live births were verified via hospital records. Of the 83 women who met all of the identified study criteria, 48 women (Treatment Group) were part of a community-based intervention (Women and Infants at Risk - WIAR), which allowed them to leave prison prior to giving birth. The community residential program provided substance abuse treatment, education/employment support, and day care for up to one year before women and their infants were released into the community on parole. Although eligible, the remaining 35 women (Comparison Group) did not enter the community-based program due to program capacity issues. 67 Previous analysis had examined the role of subsequent maternal CJ involvement as a predictor of foster care and termination of parental rights for both groups. When a participant’s first arrest post-release was for a felony offense, the odds of the child entering foster care were five times greater than if the first arrest was for a lesser crime (misdemeanor or ordinance offense). However, felony as the first type of arrest post-release was not a predictor of TPR (Kubiak, Kasiborski, Karim, & Schmittel, 2012). This previous research therefore supports the need for further qualitative inquiry around foster care involvement of this sample, particularly with regard to how children, once they enter foster care, they move within the system to either reunification or termination. The following exploratory research questions guided the current study: How do children move into and within the CW system after they are born to incarcerated women? Which risk factors influence the termination of parental rights? Methods Analyzing data using a purely quantitative approach does not allow us to see the complexities and nuances of the data. In order to begin to better understand the complex relationships and specific details of the lives of these women and their children, this study used a collective case study methodology, with purposefully selected participants. The collective case study methodology allows for the integration of multiple data sources and is ideal for examining complex issues within context (Baxter & Jack, 2008; Creswell, 2007; Yin, 2003). The availability of multiple data types (e.g. interviews, surveys, and administrative data) and sources (participant, state police, Department of Human Services) lends itself to this methodology (Creswell, 2007). Within this collective case study, inductive analysis allows for immersion in details as a mechanism for identification of themes and relationships. 68 Sample Selection A collective case study consists of multiple cases that illustrate an identified phenomenon. In this instance, the phenomenon of interest was the movement of children into and within the CW system after they are born to incarcerated mothers. Within a case study, a case is defined as the object of study or examination within a bounded system (Creswell, 2007; Stake, 1995). For our purposes, a case was defined as a mother-child dyad bounded by time and participation in the WIAR program. Purposive sampling was used, a method suggested when the objective is to explore a phenomenon in context rather than to generalize results to a greater population (Patton, 2002). According to Creswell (2007), “[Purposive sampling] means that the inquirer selected individuals and sites for study because they can purposefully inform an understanding of the research problem and central phenomenon in the study” (p. 125). The recommended case study sample parameters of no more than 4-5 cases were followed (Creswell, 2007), allowing for thorough individual case descriptions and theme identification, as well as cross-case analyses. Inclusion Criteria The final sample (n=4) was selected from the Treatment Group (n=48) as an intentional mechanism for maximizing available data. In addition, obtaining our data from the Treatment Group also controlled for potential surveillance bias introduced through program participation. Surveillance bias is the theory that higher rates of protective service involvement occur with some groups as an unintended consequence of heightened interaction with mandated reporters in program settings (Chaffin & Bard, 2006; Kubiak, Kasiborski, & Schmittel, 2010). Selecting women from the treatment group eliminated differential rates of surveillance that might be introduced by more frequent interaction with mandated reporters such as doctors, nurses and 69 social workers, since all women in the program had similarly increased encounters with these professionals. Additional inclusion criteria included both the successful completion of the program and subsequent foster care placement after protective service referral, narrowing the sample of 48 to nine. The additional requirement of availability of post-treatment CJ data narrowed our sample further (n=8), as we were unable to located one participant within the state police database. As a way to prevent confounding the results, cases where the mother was convicted of a post-birth felony were excluded. The final inclusive criteria is the presence of a status plan, resulting in the final sample (n=4). A status plan is a document obtained from the Department of Human Services, detailing aspects of foster care placements and involvement. Although all children with foster care involvement should have a status plan available, some data was missing possibly due to caseworker or system error. Data Sources Data were collected from various sources at multiple time points. Initial program intake data was collected during program participation (1996-1998) including general demographic information, health history, mental health measures and structured interviews at multiple time periods. These structured interviews include uniform questionnaires administered to women as part of the intake process. In addition to program intake data, I also focus on two types of follow-up data, criminal history and human services. Beginning in 2008, follow-up data were obtained from multiple state-level administrative databases: Criminal history. CJ data were obtained through the Law Enforcement Information Network (LEIN), a database which houses all municipal and county CJ information operated by 70 the state police. Identifying information about the mothers including name, date of birth, social security number and prison record number was submitted to a data specialist. The data specialist queried the LEIN database, providing information on arrests, convictions, and sentences that occurred after program participation and after the birth of the target child. Human services. In addition to CJ data, human services data were collected from the State Department of Human Services (DHS), including both CW (protective service, foster care, and termination of parental rights) and financial assistance (food benefits, Medicaid, cash assistance) data. Identifying information about the mother and the target child was submitted to a DHS analyst who initially located the mother-child dyads within their multiple systems. The analyst then trained graduate research assistants on the structure and definitions of the database, which included foster care information from two time periods as well as protective service information. Using the structure of the database, data extraction forms were created. Data was captured using the extraction forms and then entered into SPSS. In addition, case files were created for each target child and narrative data was printed and filed. This data included information extracted from reports required by the Department of Human Services (State of Michigan, 2014) Children’s Foster Care Manual including initial service plans (ISP), updated service plans (USP) and Permanent Ward Status Plans (PWSP). The ISP documents information about the family including their functioning, the initial permanency goal, and the services that are needed to meet the goal. The ISP should also include the needs and strengths of the family (State of Michigan, 2014). The USP reexamines the situation and any progress made towards the permanency goals, including compliance of the parent (State of Michigan, 2014). Once rights are terminated and a child becomes a ward of the State, a PWSP is used to document goals and services for the child (State of Michigan, 2014). 71 Analysis and Results The data was analyzed using an inductive approach. A detailed description of each case and the chronology of events within their context (Stake, 1995; Yin, 2003) were created. Thomas (2003) describes this approach, “Inductive approaches are intended to aid an understanding of meaning in complex data through the development of summary themes or categories from the raw data (‘data reduction’)” (p. 2). Following Yin’s (2003) suggested analytic approach for collective case study methodology themes were identified within each case, a form of within-case analysis, followed by a cross-case analysis of themes. Through exploration and immersion in the details and nuances of the data, inductive analysis was used to explore the predetermined risk factors and then to identify and confirm themes (Patton, 2002). A composite of each mother and child is provided based on synthesis of data obtained through the multiple sources. Quotes used within the analysis are extracted from the various data sources including the intake questionnaires and the status plans. They are reported exactly as written by case workers, however it should be noted they were not a result of direct verbal interviews with the women. Researcher As Instrument The role of the researcher during the creation of the case descriptions and analysis shaped the trajectory of this paper. During initial data collection, the author was not yet a mother; however, during the course of the research project she welcomed her first child. Then, during the period of data analysis and manuscript writing, she became pregnant again. This newly discovered aspect of the author’s identity, and the shift that occurred from non-mother to mother during the course of this study, provided a new lens from which to view the data. This prompted 72 reflection and debriefing with other researchers during the period of analysis and results writing (Hill, Thompson, &Williams, 1997). Cases Table 4.1 provides a summary of the descriptive risk characteristics of each mother. Table 4.1. Descriptive Case Characteristics Mother A Mother B Mother C Mother D Age at birth 42 28 31 22 Maternal Race W AA AA AA Child Race AA AA AA AA Education 9 12/GED 12/GED Some College Alcohol Alcohol Cocaine Cannabis 6 4 1 1 No Significant Other No Significant Other No Significant Other No Significant Other Drug Preference Number of Previous Children Relationship Status The following are brief descriptions of the selected cases: Mother A. At the time of the WIAR program, Mother A was a 42-year-old White female with a 9th grade education, enrolled in a part-time GED program. She had a total of 6 children ranging in age from 3 to 25. The target child was her 7th. She indicated that her children knew where she was and that she saw them, “a few times” while incarcerated. When asked to describe her relationship with her children, Mother A wrote, “Great! I love my children very much” While in school she indicated that she “got along with teachers okay”, however she “quit school to get married and have a baby.” Although she did not work the year prior to enrollment in WIAR, she stated she “really enjoy[s] working the only problem is reliable and reasonable child care.” Her longest period of employment was 2 years, during which she worked as a waitress. 73 When asked about her drug usage, Mother A reported, “Not a problem any longer because I quit on my own way before my arrest.” Her most recent living situation was with her three younger children. When asked to describe her feelings about being incarcerated, she said, “A place I never want to see again in life.” While growing up, Mother A was raised by her mother and stepfather. Her father was deceased. She had two siblings and indicated that they got along “okay.” She denied substance abuse or criminal involvement by either of her siblings. When asked how she got along with her step-father she said, “Not at all”, that he was “demanding, overbearing”, but described her “real father” as “happy, great.” She described her mother’s personality as “Quiet withdrawn but good to us.” She disclosed that her stepfather sexually assaulted her, a memory she suppressed until three years prior to the WIAR program. She described her childhood as, “always in trouble, no matter what.” After the target child was removed at 8 years 2 months old, Mother A was receiving outpatient substance abuse treatment prior to being incarcerated. She was had a felony drug arrest when the child was 8 yr. 10 mos. and was later convicted on a misdemeanor drug charge. She was sentenced to 365 days in jail. Following her time in jail, she entered inpatient treatment. During this time she also completed a parenting course. Caseworker notes indicate that Mother A has “demonstrated an anger management problem” and “can become argumentative quickly if provoked”, however “[she] is working on her emotional issues and has demonstrated new skills in managing her anger when confronted” Target Child A. Target Child A first came to the attention of Child Protective Services at 1 month old due to an allegation of physical neglect against Mother A; the allegation was investigated and unsubstantiated. This allegation occurred during the program, which suggests it 74 may have been a referral that was triggered automatically due to a previous protective service or foster care case with other children which remained open. Had this occurred during the program, the mother likely would have been removed. At 1 year and 5 months, a neglect allegation of improper supervision and abandonment listing Mother A as the perpetrator was substantiated and eventually closed. Additional neglect allegations at ages 2 yr. 11mos.; 3 yr. 6 mos.; 3 yr. 9 mos.; and 7 yr. 1 mo. allegedly perpetrated by Mother A were unsubstantiated, bringing the total number of protective service allegations to 6. A 7th protective services allegation of “physical neglect” and drug abuse by the parent, presumably the mother, led to foster care placement of the child at 8 yr. 2 mos. During the two and a half years spent in foster care, Target Child A had a total of 6 placements. The most recent placement began when the child was 10 yr., 8 mos. The child remained in this placement at the conclusion of data collection (age 11 yr., 0 mo.). An additional protective service allegation of medical neglect occurred when the child was 9 yrs., 2mos. However, the target child is not listed as victim nor is the mother listed as the perpetrator. Likely, the target child was present in the home of the victim and is therefore associated with this allegation. The child was in foster care when this allegation occurred. When the child first entered care, the initial goal was to return home/reunification. While the child was in care, Mother A was receiving outpatient substance abuse treatment until she was re-incarcerated. After release, she started inpatient treatment. When the target child was 10 yrs., 6 mos. “…[Mother A] relinquished her parental rights….” Termination proceedings were determined to be in the target child’s best interest because “…[Mother A] has not maintained abstinence and is not following court orders.” 75 As of data collection in 2008, the child was a State Ward living with an unrelated caregiver. Target Child A has a sibling in a different placement, and Target Child A has expressed wanting to be placed with this sibling. Despite a court order, the siblings have not visited since their mother was incarcerated. When the child was 9 years, 4 mos., the case worker noted that although a mandatory petition for termination was not required at this time, a petition was filed anyway; however, the caseworker wrote, “It was not in the child’s best interest to proceed.” In an updated service plan report, goal had changed from return home/reunification to termination of parental rights/adoption and caseworker notes indicated that DHS was ordered to file a TPR petition. Parental rights were terminated when the child was 10 years, 7 mos. old. At that time it was noted that it is was in the child’s best interest to proceed with termination because, “[Mother A] has not maintained abstinence and is not following court orders.” Mother B. At the time of the WIAR program, Mother B was a 28-year-old African American female with a high school diploma. Prior to entering the WIAR program, she was on probation for possession of cocaine charge, but was sent to prison due to a probation violation. She described her school years and relationship with teachers, “I did well and got along well.” The longest period she ever held a job was 3 months and she described the job as “industrial.” She has 4 previous children, ranging in age 13 to 7. She describes her relationship with her children as ”great, they are my best friends. Open/honest and strong.” While in the WIAR program, her children resided with her grandmother. When she found out she was pregnant with Target Child B, she reports being, “happy”; when asked how she feels now about being pregnant, she reports being, “very happy.” She reports one brother who also has a history of substance abuse and criminal involvement. She was raised by her grandparents. She described her mother’s personality as “unpredictable, volatile.” 76 She denies any childhood sexual assault; however, she reports that she was raped as an adult. She describes her childhood as, “childhood was pleasurable and fun. All fun memories. Happy.” Her identified substances of choice are alcohol and crack. She began using alcohol at age 14 and cocaine at age 20. She indicates that the longest she has been drug free in the past two years was 6 months. After being discharged from the WIAR program, she indicated she resided with her mother, and both her mother and grandmother helped her with childcare. When asked in a follow-up interview about coping with any issues with drugs or alcohol, she responded, “Great, not problem at all.” When the target child was 8 yr. 11 mos., Mother B was arrested for a felony drug charge, and later convicted of a misdemeanor which did not result in any jail time. Parental rights were already terminated at this time. Target Child B. Target Child B was first brought to the attention of protective services at 2 yr. 6 mos. due an unsubstantiated neglect allegation of abandonment against Mother B. Approximately two years later when the child was 4 yr. 10 mos., a physical neglect allegation against Mother B was made and again was found to be unsubstantiated. When the child was 5 yr. 10 mos., an allegation of neglect against the mother, specifically improper supervision, was substantiated; however, it does not appear that the child was removed at this time. Target Child B’s initial placement began at 6 years, 4 months. Case notes report that, the child, “…was in a guardianship and when her guardians passed away, she was brought into foster care.” Case notes report that the child was placed in a relative placement with the maternal grandmother since entering foster care, with a plan to remain in that placement for the purpose of adoption. Termination of Parental rights occurred when the child was 7 years, 9 months. 77 Mother C. At the time of the WIAR program, Mother C was a 31-year-old African American woman. She was paroled to the WIAR program from an original charge of Uttering & Publishing and Forgery. Intake indicates drugs were used in the commission of the crime. She reports marijuana and crack/cocaine as her substances of choice. When asked to describe her feelings about being incarcerated she indicates, “I was disappointed in myself.” She first used marijuana at age 14, cocaine at age 23, and crack at age 25. She reports that she was raped. She reports that she has 1 year of college education. She has been pregnant twice, resulting in 1 child who resides out of state with his father. When asked about her satisfaction with this living arrangement, she writes, “I wish he was here but I know he is well taken care of.” She reports that he knows she is currently incarcerated. She describes their relationship, “It is strained because he has not responded to my letters I’ve written him.” When she found out she was pregnant, she reports “Mixed feelings… because I knew my mom would flip out.” However, at the time of WIAR intake she reports feeling, “very happy…This is a new start for me and I love feeling my baby move inside of me.” She has two siblings and denies that they have a history of substance abuse or criminal involvement. She was raised by a single mother and describes their relationship, “She was my best friend when I was younger” and describes her mother’s personality as, “beautiful.” When asked to describe her childhood, she writes, “I had a beautiful childhood. I remember most the dancing lessons.” When asked to describes her feelings about being incarcerated she indicates, “I was disappointed in myself.” Mother C was discharged from the WIAR program when TC was 5 months old. A follow-up survey conducted when the child was 9 mos. indicated that Mother C and TC 3 were living with Mother C’s mother and brother. Mother C had recently been fired from a retail position due to unexcused absences and was looking for new employment. When asked about 78 coping with drugs and alcohol, Mother C indicated it hasn’t been a problem because she was involved in church activities and that she was able to maintain sobriety. Target Child C. Target Child C was first brought to the attention of protective services at 5 yrs., 3 mos. due to a substantiated allegation of physical abuse and a neglect allegation of failure to protect. The listed perpetrator of that allegation was not Mother C, and additional details of relationship to perpetrator were not available. The first allegation against Mother C occurred when the target child was 6 years old. A neglect allegation of abandonment was found unsubstantiated. When Target Child C was 8 yr. 3mos., an allegation against Mother C of abuse (mental injury) and neglect (physical, improper supervision, and failure to protect) was transferred to another agency (such as law enforcement or child care licensing) for further investigation. Six months later, when the child was 8 yr. 9 mos., an allegation of neglect (abandonment) listed Mother C as the perpetrator; however, Target Child C was not the alleged victim. The child was placed in foster care at 10 yr. 1 mo. following a substantiated neglect allegation that occurred at 10 yr. 0 mo. The substantiated allegation against Mother C was for physical neglect, improper supervision, and abandonment. According to the Initial Status Plan, “The mother abandoned the children with a paternal relative.” The child remained in the initial relative placement at the time of data collection with a listed goal of, “return home.” The ISP also indicates that an intensive home based prevention program and several substance abuse treatment facilities were tried prior to the removal of the child. There were a total of 5 protective service allegations; 3 of the 5 allegations listed Mother C as the perpetrator and Target Child C as the victim. Although not officially in foster care, the ISP also notes, “Target Child C was in the care of [an] aunt [at approx. age 2] while Mother C was in and out of jail.” 79 Mother D. At the time of the WIAR program, Mother D was a 22-year-old African American female. She has one child in addition to the target child. Prior to her incarceration and entry into the WIAR program she lived with her mother and an older child. She describes her mother’s personality as, “loving and caring for others.” She describes her relationship with the older child, “Very good, I love him very much” and she visits with her older child approximately 1 time per month. She describes her childhood as “disciplined and loving” and reports that her family is supportive during her incarceration, “they were supportive with my child and money and emotionally.” She had “mixed feelings” when she found out she was pregnant because she “was going in and out of court”, but later reported feeling “very happy.” When asked what happened to make her change her mind, she responded, “Because I have a chance to make it different with this child.” Crack-cocaine and marijuana are her drugs of choice. She began using marijuana at age 14 and crack-cocaine at age 19. She denied any sexual assault history. Target Child D. Target Child D was initially brought to the attention of protective services at 10 mos. due to an unsubstantiated physical neglect allegation against Mother D. The Initial Service Plan references this first allegation and indicates it was “for selling food stamps for marijuana and no food in the home.” At age 6 yr. 10 mos., another allegation of physical neglect, specifically, “drug activity in her home and unsanitary conditions”, against Mother D was found unsubstantiated. Three months later, when Target Child D was 7 yr. 1 mo., an allegation of physical neglect against Mother D was substantiated; however, Target Child D was not the listed victim. It is likely the child was in the home at the time of the incident. At this time, Target Child D was petitioned for removal due to “drug abuse of parent.” One month later, at 7 yr. 2 mos., the child was returned home. An unsubstantiated improper supervision neglect allegation occurred when Target Child D was 7 yr., 5 mos.; however, TC 4 was not listed as the 80 victim. Then, due to “non-compliance,” the child was again removed from the parental home at 7 yr. 9 mos. While in foster care at 8 yr. 1 mo., an allegation of sexualized behavior by a sibling occurred. Target Child D was not listed as the victim and the allegation was unsubstantiated. While Target Child D was in foster care, Mother D completed a substance abuse program, but continued to test positive for marijuana. Target Child D was in a total of 3 placements before returning to the care of Mother D at 8 yr., 7 mos. Two additional unsubstantiated allegations occurred following the return of the child to the parental home. At 9 yr. 2 mos., the allegation was for improper supervision, listing Mother D as the perpetrator. Target Child D was not listed as the victim. At 9 yr. 3 mos., an allegation of abuse (maltreatment), physical neglect, improper supervision, threatened harm, and drugs in the residence listed Target Child D as the victim and Mother D as the perpetrator. At the time of data collection, the status of the foster care case was closed, and the living arrangements were listed as “parental home.” Salient Themes At the conclusion of data collection, two mothers had lost parental rights and two retained parental rights. However, all four of the mother-child dyads had extensive protective service involvement (M=5.8 (2.2); Range 3-8), including the presence of both substantiated and unsubstantiated allegations and foster care placement. Congruent with national trends, the majority of protective service allegations were for neglect, rather than abuse (Child Welfare Information Gateway, 2012). Substance abuse. Regardless of termination outcome, substance abuse was a recurring theme identified during analysis. All mothers initiated drug use early in their teens, and all mothers indicated cocaine was their primary or secondary drug of preference. Details of 81 protective service referrals and foster care case notes cite substance use as a reason for removal of the child and as an identified issue for the mother. Mother A utilized both inpatient and outpatient substance abuse treatment after TC A entered out-of-home care. According to case notes, substance abuse eventually supported the decision that proceeding with termination was in the best interest of the child, since, “…[Mother A] has not maintained abstinence.” Mother B, who also eventually lost parental rights, was arrested on drug charges. Both mothers who lost their parental rights also had subsequent CJ involvement following the WIAR program. Both were arrested on felony drug charges and later convicted of misdemeanor offenses. The identified charges, however, suggest that substance abuse rather than criminal behavior may be the risk that contributes to the identified outcome. Perhaps arrest is a measure of the severity of the substance misuse; those who are arrested are engaging in riskier behaviors to obtain drugs due to the severity of their addiction. Although Mother C and Mother D retained parental rights, substance abuse was a pervasive issue for these women as well. It was reported that Mother C had tried several substance abuse treatment facilities prior to the removal of the child. In addition, case notes reveal that TC 3 has an awareness of the issue of maternal substance abuse. The worker reports, “[TC 3] writes about [Mother C’s] addiction…” The first allegation against Mother D was “for selling food stamps for marijuana…” Six years later, another allegation against Mother D cited, “drug activity in her home.” When TC 4 was petitioned for removal from the home, the reason was “drug abuse of parent.” Case notes indicate that Mother D is addressing her substance abuse issues and how they relate to her ability to care for her children. The worker notes, “She understands she put her children at risk but 82 denies living in a drug house. She states that she struggles but is a good mother. [She] wants to get help for her past substance abuse issues and to have a better living environment for her children.” Previous research has indicated a connection between parental substance use disorders and child maltreatment (Young Boles, & Otero, 2007). A review of substantiated maltreatment allegations in one county found that more than 2/3 of mothers abused alcohol or drugs (Jones, 2005). Despite the prevalence of these issues, treatment options may not be sufficient for parents with substance abuse issues given the presence of multiple risk factors. A comparison of appellate court foster care cases involving substance abuse concluded, “Considering the number of cases under appeal in this study which included non compliance, multiple treatment attempts, and relapse, perhaps the treatment offered is not meeting the critical needs of parents” (Meyer, McWey, McKendrick, & Henderson, 2010, p. 646). While this may be true, it is also known that relapse and multiple treatment attempts can be common as one initiates and maintains sobriety; this suggests that progress or “compliance” may not be as clearly defined as sober or using. A well-documented precursor to substance abuse, particularly for women, is experience of sexual abuse (Brown & Finkelhor, 1986; Kendler et a.l, 2000; Wilsnack, Vogeltanz, Klassen, & Harris, 1997). Within this case study, 3 of the 4 women reported a history of sexual assault. Both substance abuse treatment and CJ systems are moving toward gender- responsive treatment mechanisms that identify and address the core underlying issues of the behavior. Complexities of maternal relationship. The data revealed complex relationships between mother and child within all four dyads. Although the presence of risk factors may jeopardize a mother’s ability to provide daily care for her child, the data suggests that this does not correlate with the absence of a relationship. When asked how she felt about being pregnant 83 prior to the birth of the target child, Mother C wrote, “Very Happy…This is a new start for me and I love feeling my baby move inside of me.” When asked what would enable her to resist drugs, Mother A wrote, “Being with my kids and watching there [sic] little faces light up when we do things together.” Both of these descriptions reflect the joy that pregnancy and the maternal relationship can evoke when a woman becomes a mother. These complexities stood out to the author. She was surprised by her connection to these mothers based on their descriptions of how they felt about motherhood and their children, and noted that their words mirrored so similarly her own feelings of pregnancy and motherhood. Along with evidence that the mothers in this study care for their children, there is also an indication that the children care for their mothers. Caseworker notes regarding TC A report, “[TC A] talked about his mother in jail. He is very upset and wants to see her.” Commenting on the relationship between Mother C and Target Child C, the caseworker reports, “[Target Child C] has a great relationship with her mother. During her weekly visits with her mother, [Target Child C] is always hugging and kissing her mother. She never wants the visits to end…she always wants to make sure that her hair is pretty and during her weekly visits with her mother, she is always brushing and combing her mother’s hair.” Later, she writes, “[Target Child C was] happy and excited to see [Mother C]. [Target Child C] was jumping and hugging her as soon as she walked into the door.” Case 4 also had evidence that both mother and child were upset by the child’s removal and placement into foster care. The caseworker wrote, “[Mother D] was upset that her children were removed from her care. She was deeply sad. She is eager to resolve the issues that brought them into care.” The caseworker also commented on the children, “The children were upset to be removed from their mother. The children are very close to their mother.” Despite the multiple risk factors present and the placement of children into foster care, 84 both mother and child verbalized caring feelings towards each other. This complex relationship may be difficult for caseworkers as they navigate the tension between ensuring children are safe while also supporting existing attachment. Ideally, if attachment to a mother is secure, or has the potential to be, it should be the goal to strengthen the existing relationship while enhancing safety. Efforts at reunification. In addition to the identified themes regarding the complex nature of the mother child relationships, themes of reunification and the lack of visitation between both the mother and siblings emerged. TC A has a sibling in a different placement, and expressed a desire to be placed with the sibling. The status plan indicates that, despite a court order, the siblings did not visit since their mother was incarcerated. TC C was placed with her sibling, and the caseworker commented, “[The siblings] love being together as opposed to separate foster homes.” Although visitation appears to have occurred regularly, “[Mother C] has not missed any scheduled visits with the children during this reporting period”, there is still evidence suggesting that more could be done to facilitate the existing relationship. The status plan twice references TC C “sneaking to call her mother.” In addition to maintaining sibling relationships during placement, efforts at reunification should include regular visitation with parents. At one point prior to Mother A’s termination, the caseworker noted, “[Mother A] visited with her children frequently and always requests more time. She wants to be an appropriate parent and is willing to attend more parenting classes.” Visitation was not always facilitated however, as the caseworker noted, “Visitation with the mother was not accomplished before she moved to the drug treatment facility.” It is unclear if this is due to the mother’s inability to comply, worker constraints, or simply timing. 85 Previous research has demonstrated that maintaining the mother-child relationship during confinement in prison or jail has positive outcomes for both the mother and the child. Contact with children has been connected with lower levels of depression in incarcerated mothers (Houck & Loper, 2002; Poehlmann, 2005). Despite visitation with the mother often mandated by court order, only 15% of mothers report monthly visits from at least one child (Glaze & Maruschak, 2008). This low rate is a result of multiple barriers that prevent the facilitation of routine visits. Often, there is only one women’s correctional facility within a state, meaning that women may be housed hundreds of miles away from their children (Glaze & Maruschak, 2009). This distance is costly to caregivers who must pay for transportation, caregivers who may already struggle financially due to caring for the inmate’s child. The logistics and costs associated with traveling to the correctional facility may be why some caregivers discourage visitation (Bloom & Steinhart, 1993). Nearly 60% of women in state prison report they have never had a personal visit with their minor child (Glaze & Maruschak, 2008). These same issues may also be prohibitive to CW workers. Regardless, visits are mandated as part of reunification planning. When face-to-face visits are not feasible, advances in modern technology may make facilitation of mother-child communication easier. As a supplement to in-person visits, visitation utilizing video may be a way to increase the frequency of contact and communication between the child and the incarcerated mother (Phillips, 2012). Modern technologies, like Skype and Apple’s Facetime program, allow real-time video conferencing at minimal costs however currently internet capability is very limited for incarcerated women and cell phones are prohibited. Utilization of these or similar software may be a mechanism to increase the frequency of contact between incarcerated mothers and their children. There remains a need for 86 additional research to identify the appropriate frequency and type of contact based on the child’s developmental status (Poehlmann, Dallaire, Loper, & Shear, 2010). Worker subjectivity and system limitations. The absences of documented benchmarks that result in termination imply that worker subjectivity may be an integral part of whether or not a mother retains her rights. Despite common risks (evidence of substance abuse, substantial interface with protective services, and foster care placement), two mothers retained their parental rights and two mothers eventually lost their parental rights. It is well established that women within the CJ system are a high-risk population. Ultimately, the women in this study have similar circumstances but different outcomes, yet it is not completely clear why these cases result in different outcomes. Although there may have been additional strengths that contributed to the workers decision, the reports did not include mandatory strengths-assessments as part of their reports. In addition to the absence of uniform strengths- assessment within the child welfare records, the data included in the initial and updated status plans is not uniformly collected or reported. Without this information, it is not clear whether workers were weighing additional child-specific or environmental factors when making their recommendations. Narrative information and descriptive detail included in status plans vary widely among the four selected cases. This variation is not only a limitation to the study which prevents further analysis of trends, but it also supports the notion of subjective differences among workers. Reflecting on the termination process, the researcher was struck by the lack of information about attempted interventions prior to proceeding to legal judgment. Although the court ultimately makes the final decision regarding parental rights, the foster care worker is responsible for care management prior to that point. Thus it seems reasonable that protective service and foster care records would contain information that clearly illustrate the path to 87 termination and the interventions attempted prior to the legal petition. The author felt empathy for these mothers, and recognized the magnitude of what this loss would feel like. She was upset for the mothers whose rights had been terminated. This was not because the termination might be without cause, but because there was not enough information in the file to determine whether termination was warranted. Reviewing these files methodically, similar to an “audit”, did not clearly paint a picture of the established and violated standards, indicators that should be clear when permanent loss of one’s children is the outcome. The pathway to this outcome is not explicitly and clearly documented in their case records. Termination of parental rights should be a last resort outcome. The attempted reunification processes and parental interventions that occur prior to termination should be thoroughly documented. While individual case details may vary, some level of standardization may be a way to identify parental strengths and opportunities for intervention, as well as to ensure that parents have every opportunity to reunify with their children. Additionally, although CW professionals determine whether remaining in the home is or is not safe for the child, it is not clear that placement in foster care provides better care than placement with the mother, as both TC A and TC D were listed on allegations that occurred while in foster care. Although neither was listed as the victim, they were present in the home or environment where the allegation occurred. Conclusions Due to the small sample size and the nature of case study research, results from this study cannot be generalized. Instead, the current research provides suggested direction to be explored with future research. Incarcerated mothers experience multiple risk factors, making it difficult to determine which specific risks influence CW outcomes. Although there are unique characteristics associated with each of the four mother-child dyads described in this study, 88 ultimately there are many similarities and common themes found in the struggles of these families. Future research should explore risk as it relates to the ability of a woman to navigate both the CJ and CW systems simultaneously. In addition, system-level variables of the CW system including details about individual FC placements, as well as individual variables and the role of the worker in this process, deserve further consideration. Within the CW system including both PS and FC interaction, there are additional variables that may influence outcomes. Future research designed to study CJ and CW intersection should also account for possible variances related to difference among the children themselves which could contribute to movement within the system. In addition to the individual and cumulative risk factor trends, the data revealed important complexities about the maternal child relationship. Just as the details of each of these four cases are unique, so are the relationships between each mother and target child. Therefore, CW caseworkers and CJ officials should be cautious not to assume that involvement in criminal activity is representative of a woman’s role as a mother. Nor should workers assume that perpetration of abuse or neglect is an indicator that a mother does not care about her child. Obtaining a detailed history about the maternal relationship could provide workers with foundational understanding of the woman, and potentially provide knowledge to use as motivation for change. As documented within this study, CJ involvement, substance abuse, and even formal termination of parental rights do not correlate with how a mother feels about her children. A mother’s inability to take daily care of her children due to substance abuse, poverty, or CJ involvement does not mean she does not care for them. It is important for practitioners to note this as they work with women in the CJ system and with their children. 89 CHAPTER FIVE: CONCLUSION Overview Prior research affirms the intersection of women in the criminal justice (CJ) system and their children in the child welfare (CW) system. However, the research remains in elementary stages (Beckerman, 1998; Daillaire, 2007;Johnson & Waldfogel, 2002; Lee, Genty, & Laver, 2010; Phillips & Zhao, 2010; Seymour, 1998). This dissertation explores the relationship between maternal CJ involvement and simultaneous involvement of her children in the CW system. Specifically, the guiding research question of this multiple manuscript dissertation asks: What are the maternal risk factors and the role of cumulative maternal risk on the CW outcomes of children born during maternal incarceration? Chapter Two provides a detailed literature review of the intersection of maternal CJ and CW overlap. It explores the CJ continuum from arrest, to sentencing of probation or confinement, and finally, to parole and re-entry. Chapter Three applies a cumulative risk framework to a sample of incarcerated women and their child who was born during incarceration as means to explore the relationship between maternal risk factors and child protective service involvement. Chapter Four utilizes an in-depth case study approach, combing multiple study and state level data sets to explore the pathways into foster care, and the experiences and outcomes once in the CW system. This final and concluding chapter connects overall findings and discusses implications for social work policy, practice, and future research. Major Findings This dissertation explores a small, but risk-laden population of women who give birth while incarcerated and who experience eventual CW involvement. Consistent with the literature, 90 even within a high- risk sample, those with the greatest cumulative risk were found to be more likely to experience negative CW outcomes. While these women experience many similar risk factors, their individual stories and the pathways that lead to incarceration and CW involvement are unique. Their feelings about motherhood, the hope that accompanies pregnancy and the birth of a baby is evident. Despite histories laden with sexual assault, substance abuse, CJ, and CW involvement, as one mother stated, “I have a chance to make it different for this child.” Despite the expressed desire to change, women face many barriers to reintegration that are further complicated by the role of parenting. Prior to her arrest, a woman may struggle to support herself and her children (Owen, 1998), and women frequently return to the environment from which they were arrested. Many women in the CJ system lack a high school education (Glaze & Maruschak, 2008), and for those who do have an education criminal history may make women unemployable or force them to work in low wage jobs. Collateral consequences may ban receipt of welfare including cash assistance or food stamps (Allard, 2002). This combined lack of resources put women at risk of poverty and material deprivation, which may in turn put children at risk for actual or perceived neglect. The findings from this study support the need for greater cross system collaboration that includes gender-responsive programming and is considerate of the emotional view of motherhood, as the lack of physical custody or absence regular contact may not be reflective of how a women views her child or thinks of herself as a mother. Due to strict policies and requirements, navigation of either the CJ or CW system independently is a complicated task. Simultaneous navigation of these systems is likely even more difficult. Within each system there are multiple levels of involvement (ex. arrest through parole, foster care through termination), and families can intersect these systems in various ways (ex. arrest and protective services, foster 91 care and incarceration). These multiple levels result in a lack of knowledge and/or experience with the other system, leaving the mother/infant dyads without the support needed to successfully navigate either. These findings provide implications for individual and systemic social work at both the policy and clinical levels. Practice There is a need for effective, evidence-based interventions (DeHart & Altschuler, 2009; Miller, 2006; Johnson, 2006) which address the complex needs and historical trauma commonly experienced by women in the criminal justice and child welfare systems. Miller (2006) describes this need for comprehensive programming, “Evidenced based interventions that employ gender specific, developmentally appropriate, and culturally sensitive treatment modalities is a practice area that calls for further development (p. 481).” Even though majority of incarcerated women are mothers, few states have programs that take into consideration this aspect of the woman’s identity and how it may relate to her criminal activity or reintegration needs. Prison nurseries are an example of this type of programming which take into account needs of both the mother and her child while preparing the mother to parent once she is released. Evaluation of a prison nursery program found high rates of secure attachment between mothers and their infants and toddlers (Byrne, Goshin, & Joestl, 2010). Follow-up concluded that 83% of children remained with their mother following the first year of re-entry. Additionally, of those that remained with their mother after the first year, 90% continued to remain with their mother after 3 years (Byrne, Goshin, & Blanchard-Lewis, 2012). Parent education has also been shown to increase child development knowledge, to improve parenting skills, to facilitate attachment and to reduce recidivism (Chappel, 2004; Thompson & Harm, 2000), however current policies do not mandate participation in such 92 courses for jailed or incarcerated parents. Only a quarter (27%) of mothers in State prison report participation in parenting or childrearing courses (Glaze & Maruschak, 2008). Given that the majority of women in prison are mothers, this may indicate an area for future growth. Given that this population is growing substantially, the demand for services is increasing as well. However, in general, current practice methods and programs are not geared specifically toward mothers with CJ involvement and their children. Not only do these families pose a challenge for caseworkers due to their multiple risk factors, caseworkers are also challenged by parallel navigation of the CJ and CW systems. These issues may make it difficult for caseworkers to successfully meet the needs of the families they work with. Incarcerated parents, caregivers, and adult children indicated that various social service agencies, including corrections department, schools, and CW system, were not attuned to their needs (Bates, Lawrence-Wills, & Hairston, 2003). As previously noted, facilitated communication between mother and child or siblings did not always occur, even when mandated by a formal reunification plan. If workers struggle to follow reunification plans, the CW system must revisit policies and explore alternative ways to facilitate communication between parents and their children. Within the CW system, data collection and quality assurance procedures should be implemented and adhered to prevent missing data, such as status plans. Correctional settings. Although some correctional facilities have implemented programs designed specifically to meet the needs of incarcerated mothers and their children, there is a need for programs targeting this group, not only for infants but for mothers and children of all ages. Despite the accessibility to family and the potential for brief interventions during visiting hours, the use of correctional facilities as an environment for human service and outreach programming 93 has not been fully utilized (Arditti, Lambert-Shute, & Joest, 2003), In addition, the research presented here supports ongoing mother/child communication as not only beneficial to the child, but as also linked to positive reunification outcomes for mothers (Houck & Loper, 2002; Poehlmann, 2005). Given these benefits to facilitating the mother/child relationship, communication must be made more accessible, whether through the use of buses to assist with transportation for visitation or by lowering costs of collect phone calls. Community settings and interagency collaboration. Not only is there is a need for prevention programs and services within correctional facilities, there is also a need at the community level as well. Incarcerated parents, their children, and caregivers have identified the need for both programs and policies targeting communities and neighborhoods as key principles for improving outcomes of children of incarcerated parents (Bates, Lawrence-Wills, & Hairston, 2003). After the passage of ASFA, Beckerman (1998) examined the effect of this policy on caseworkers and CW agencies. In addition to recommending full assessments of both the mother and the child prior to making any recommendations, the author also emphasized the need for caseworker guidelines and supports. Despite being more than a decade old, Beckerman’s assertions still hold true. There is still an absence of specific programs and policies targeted to this population. Additionally, given the many risks and the unique details of cases that overlap systems, streamlining services among agencies may be a way to concurrently meet the needs of both the mother and the child. The National Council on Crime and Delinquency (2004) supports collaboration between practitioners, researchers, and agency staff, and suggests that school be a point of contact for meeting the needs of the child. As previously discussed, arrest and incarceration usually occur after the children are removed by protective service agencies (Ehrensaft, Khashu, Ross, & Wamsley, 2003). Given the rates of substance abuse within this 94 population previously discussed, this pattern may signify increased substance use as a coping mechanism for grief associated with the removal of the children. Knowledge of this pattern, along with interagency collaboration, would allow us to tailor an intervention and to connect women to necessary services at the time their children are removed, possibly preventing subsequent criminal behavior. Policy The lack of policy addressing children of incarcerated parents may be a reflection of the need for a broader social understanding of the collateral consequences of CJ policy and the effect is has not only on the offender, but on their family and children as well. Current drug policy continues to focus on enforcement rather than treatment, and mandatory minimums result in harsh sentences for first-time offenders. In addition, the construct of these policies disproportionately impacts women, one of the major reasons why this population is increasing so steadily. Legislative decisions that produce unintentional collateral consequences and punish women with “lifetime sentences” also create additional risk factors for the children of these women. Following conviction, women pay their dues through a sentence of probation, fines, and/or jail or prison, however collateral consequences continue to punish the woman and hinder her ability care for her children. There is need for exploration of these collateral consequences in the context of motherhood and reunification with children following confinement. No government entity is responsible for collecting data on the children women leave behind when incarcerated (Gentry, 1998; Lee, Genty, & Laver, 2005; Wright & Seymour, 2000). Some states, however, have taken the initiative and implemented policies to begin tracking this population. As of 2007, the state of New Mexico has implemented a protocol that requires 95 identification of minor children upon parental arrest (Child Protection Best Practices Bulletin, 2006). Identification is the first step to ensuring that children of incarcerated mothers are connected with the proper and necessary services. In addition, the Rhode Island Department of Corrections has recently begun collecting detailed information on family members and children as part of their intake policy (Brazzell & LaVigne, 2008); however, they are in the minority. The CW League of America (1997) indicated that only 6% of states had policies in place specific to children with an incarcerated parent. Without state leadership, the lack of policy is present at the agency level as well. In a survey of CW, law enforcement, and corrections agencies, more than 80% of those surveyed indicated that they did not have specific policies about placing children when mothers are involved in the CJ system (Smith & Elstein, 1994). Halperin and Harris (2004) refer to the absence of policy as a “policy vacuum”: “The failure of CW and corrections bureaucracies to formulate policy to address these problems results in a serious compromise of these women’s rights as parents” (p. 339). Although there may have been a renewed focus on the safety and permanency of the child established by the ASFA, the intended goals of ASFA are not being met, and children are lingering in foster care without being adopted (Lowry, 2004). In addition, as previously discussed, an unintended consequence of ASFA is a higher rate of termination of parental rights in this population. When parents are absent, efforts must be made to locate them. The same principle should apply to incarcerated parents. Within the State of Michigan’s (SOM, 2008) policy regarding “absent” parents, “Absent Parent Policy: Identifying, Locating, and Notifying Absent Parents in Child Protective Service Proceedings”, the parental rights of an absent parent are still considered. A Supreme Court Decision supporting parental rights states: A parent has an inherent right to the care, custody, and upbringing of his or her child. The Supreme Court has consistently protected these rights against state's 96 unwarranted usurpation. In Santosky v Kramer, 455 US 745 (1982), the Supreme Court declared “…the fundamental liberty interest of natural parents in the care, custody, and management of their child is protected by the 14th Amendment, and does not evaporate simply because they have not been model parents or have lost temporary custody of their child to the State. Within this same document, there is a section dedicated to reiterating that absent mothers should be located within this same framework and given the same considerations (SOM, 2008) that biological and putative fathers receive. The reason for parental absence is not mandated, and therefore, this policy and these same considerations should be applied to incarcerated mothers. Social Work Education Within social work education, there is an opportunity to instill in future social workers the important of interagency collaboration and of critical thinking and collateral consequences as they relate to policy. The social work education process should provide an overview of all social systems so that there is a baseline level of understanding prior to individuals working in the field. For students who specialize in child welfare, there should be mandatory course work that examines the overlapping population of child welfare and criminal justice populations. Reinforcing the need for better cross-system collaboration, Ross (2009), comments “child welfare agencies often have responsibility for a problem but do not have the authority or capacity to solve the problem without cooperation from other government departments”. Moreover, it is critical to train social workers to collaborate with criminal justice personnel such as judges, correctional treatment specialists, and probation/parole officers. Forensic social work classes could also be cross listed with criminal justice, psychology and other human service professionals as an avenue for social change (Springer & Roberts, 2007). To best serve vulnerable families, and to help families within either CJ or CW systems, this cooperation is key. 97 In addition, within social work education, there is an opportunity to train specialized caseworkers who have comprehensive understanding of the complexities both criminal justice and child welfare systems, and how these systems often fail to work together. In addition to specialized caseworkers, case management handbooks, and model case plans are possible methods of introducing the importance of cross-system collaboration to social work students (Beckerman, 1998). Essential to collaboration with a family is an understanding of responsibilities within respective systems (Ross, 2009). If social workers are better educated about these responsibilities, they will be better able to advocate for services for the families they work with and to hold agencies accountable for services. Limitations of the Study This study uses high-risk mother child dyads, resulting in a small sample. This small sample is prohibitive for certain analytical approaches and future research would benefit from larger sample sizes that allow for further statistical testing including regression and survival analyses. This study explores one subgroup of mother-child dyads dually involved with criminal justice and child welfare, those who were born during maternal incarceration. There is a need for future research which continues to explore specific points of intersection of the criminal justice and child welfare continuums. Results of the current study may not generalize to all children of incarcerated mothers or to all mother-child dyads dually involved with criminal justice and child welfare systems. Recommendations for Future Research Focusing on Children of Criminally Involved Mothers 98 Currently, the majority of available research focuses on parental incarceration lumping together children of incarcerated mothers and fathers. This research provides general information about this population of vulnerable children, and these children may have many of the same characteristics and experiences. However, the differences between these populations, as discussed within this dissertation, create a need for separate bodies of research. One of the main reasons for this is the parent-child relationship, both before and after confinement. As illustrated, the relationships between children and their incarcerated parent differ based on the gender of the parent, specifically regarding direct care-giving of the child before and after confinement. Identifying and Tracking Efforts Future research studies need to be intentionally designed in a way to determine whether identified outcomes are a product of the various contextual risk factors, or occur as a result of the incarceration itself. In other words, better efforts need to be made to isolate criminality/ incarceration as variables. This research could be done using a matched, high-risk comparison group as a way to attempt to isolate incarceration as the cause. In addition, there is also a need for longitudinal research with this population including ongoing both data collection that occurs at multiple time points and the use of longitudinal administrative data. Rigorous longitudinal research with mother-child dyads would help to better understand patterns or cycles of criminal behavior and CW involvement. Longitudinal research may help to disentangle the cumulative risk factors and help identify what specifically service providers should target with their interventions. As previously mentioned, no agency or governmental body is responsible for routinely tracking the children of incarcerated parents; therefore we can only estimate the number of children affected by maternal incarceration. Lack of tracking or data collection also means lack 99 of a national database, which precludes ongoing analysis of this group and associated trends. A lack of standardized methods means that we can only estimate the number of children with an incarcerated mother. Given the stigma associated with criminal behavior and the general fear of CW involvement, it is likely that many women do not disclose information about their children. Without the true measure of how many children fall into this group, it is more difficult to design tailored interventions, when the specifics of the target population are not known. To fully know the extent of this issue, the responsibility of tracking children impacted by maternal criminality and/or incarceration must be assigned to a singular agency. This would also allow for identification of those involved in the CW system. One such local tracking mechanism is the Reentry Mapping Network (RMN), which uses the address of the offender at the time of their arrest and self-reported data on number of children, to create a neighborhood map that illustrates the geographic distribution of children with incarcerated parents (Brazzell & LaVigne, 2008). In addition to tracking children to ensure an accurate count, just as this project did, tracking should be a priority of social service and corrections agencies, as it highlights the area that would most benefit from additional programming aimed at children of incarcerated parents, or support services aimed at their caregivers. 100 REFERENCES 101 REFERENCES Adoption Assistance and Child Welfare Act (AACWA) of 1980. (1979). H.R. 3434--96th Congress. Adoption and Safe Families Act (ASFA) of 1997. (1997). H.R. 867--105th Congress Allard, P. (2002). Life sentences: Denying welfare benefits to women convicted of drug offenses. Washington, DC: The Sentencing Project. Allen, R. L. (1974). Reluctant reformers: Racism and social reform movements in the United States. Washington, DC: Howard University. Amnesty International.( 2005). Women in custody. Retrieved from www.amnestyusa.org/pdf/custodyissues.pdf. Appel, A.E., & Holden, G.W. (1998). The co-occurrence of spouse and physical child abuse: A review and appraisal. Journal of Family Psychology, 12(4), 578-599. Appleyard, K., Egeland, B., Dulmen, M. H., & Alan Sroufe, L. (2005). When more is not better: The role of cumulative risk in child behavior outcomes. Journal of Child Psychology and Psychiatry, 46(3), 235-245. Arditti, J. A., Lambert‐Shute, J., & Joest, K. (2003). Saturday morning at the jail: Implications of incarceration for families and children. Family Relations, 52(3), 195-204. Atzaba‐Poria, N., Pike, A., & Deater‐Deckard, K. (2004). Do risk factors for problem behaviour act in a cumulative manner? An examination of ethnic minority and majority children through an ecological perspective. Journal of Child Psychology and Psychiatry, 45(4), 707-718. Bates, R., Lawrence-Wills, S., & Hairston, C. F. (2003). Children and families of incarcerated parents: A view from the ground. Research Brief on Children, Families and the Criminal Justice System. Chicago, IL: University of Illinois at Chicago. Baxter, P. & Jack, S. (2008). Qualitative case study methodology: Study design implementation for novice researchers. The Qualitative Report, 13(4), 544-559. Beckerman, A. (1998). Charting a Course: Meeting the Challenge of Permanency Planning for Children with Incarcerated Mothers. Child Welfare, 77(5), 513-29. 102 Begle, A.M., Dumas, J.E., & Hanson, R.F. (2010). Predicting child abuse potential: An empirical investigation of two theoretical frameworks. Journal of Clinical Child & Adolescent Psychiatry, 39(2), 208-219. Berger, L. M. (2004). Income, family structure, and child maltreatment risk. Children and Youth Services Review, 26(8), 725-748. Berrick, J.D., Choi, Y., D’Andrade, A., & Frame, L. (2008). Reasonable efforts? Implementation of the reunification bypass provision of the ASFA. Child Welfare, 87 (3), 163-182. Black, M. C., Basile, K. C., Breiding, M. J., Smith, S. G., Walters, M. L., Merrick, M. T., ... & November, M. R. S. (2011). National intimate partner and sexual violence survey. Atlanta, GA: Centers for Disease Control and Prevention (CDC). Bloom, B., & Steinhart, D. (1993). Why Punish the Children?: A Reapraisal of the Children of Incarcerated Mothers in America. National Council on Crime and Delinquincy. Bocknek, E.L., Sanderson, J., & Britner, P.A. (2009). Ambiguous loss and posttraumatic stress in school-age children of prisoners. Journal of Child & Family Studies, 18, 323333. Brazzell, D. & LaVigne, N. (2008). Mapping community data on children of prisoners: Strategies and insights. Washington DC: The Urban Institute. Brown, M., & Bloom, B. (2009). Reentry and renegotiating motherhood maternal identity and success on parole. Crime & Delinquency, 55(2), 313-336. Browne, A., & Finkelhor, D. (1986). Impact of child sexual abuse: A review of the research. Psychological Bulletin, 99(1), 66. Bush-Baskette, S.R. (2000). The war on drugs and the incarceration of mothers. Journal of Drug Issues, 30 (4), 919-928. Bureau of Justice Statistics. (2012). Frequently Asked Questions. Washington, DC: USDOJ, Bureau of Justice Statistics. Byrne, M. W., Goshin, L. S., & Joestl, S. S. (2010). Intergenerational transmission of attachment for infants raised in a prison nursery. Attachment & human development, 12(4), 375-393. Chaffin, M. & Bard, D. (2006). Impact of intervention surveillance bias on analyses of child welfare report outcomes. Child Maltreatment, 11(4), 301-312. Chappel, C.A. (2004). Post-secondary correctional education and recidivism. A meta-analysis of research conducted 1990-1999. The Journal of Correctional Education, 55, 148-169. 103 Chesney-Lind, M., & Bloom, B. (1997). Feminist criminology: Thinking about women and crime. In B. MacLean & D. Milovanovic (Eds.), Thinking critically about crime (pp. 54-65). Vancouver: Collective Press. Chesney-Lind, M., and Pasko, L. (2004). The female offender: Girls, women, and crime. Thousand Oaks, CA: Sage. Chibnall, S., Dutch, N., Jones-Harden, B., Brown, A., & Gourdine, R. (2003). Children of color in the child welfare system: Perspectives from the child welfare community. Washington, DC: U.S. Department of Health and Human Services, Children's Bureau. Child Welfare Information Gateway. (2012). Child Maltreatment 2010: Summary of key findings. Washington, DC: U.S. Department of Health and Human Services, Children’s Bureau. Child Welfare League of America. (1998). State agency survey on children with incarcerated parents. Washington, DC: Child Welfare League of America Press. Clarke J.G., Hebert M.R., Rosengard C., Rose, J.S., DaSilva, K.M., & Stein M.D. (2006). Reproductive health care and family planning needs among incarcerated women. American Journal of Public Health, 96:834–839. Connell, C.M., Bergeron, N., Katz, K.H., Saunders, L., & Tebes, J.K. (2007). Re-referral to child protective services: The influence of child, family, and case characteristics on risk status. Child Abuse & Neglect, 31:573–588. Conron, K.J., Beardslee, W., Koenen, K.C., Buka, S.L., & Gortmaker, S.L. (2009). A longitudinal study of maternal depression and child maltreatment in a national sample of families investigated by child protective services. Archives of Pediatric and Adolescent Medicine, 163, 922-930. Covington, S., & Bloom, B. (2003). Gendered justice: Women in the criminal justice system. In B. Bloom, (Ed.), Gendered justice: Addressing female offenders. Durham, NC: Carolina Academic Press. Courtney, M. E. (1995). Reentry to foster care of children returned to their families. The Social Service Review, 226-241. Courtney, M., McMurtry, S., & Zinn, A. (2004). Housing problems experienced by recipients of child welfare services. Child Welfare, 83, 389-392. Covington, S. (2003). A Woman s Journey Home: Challenges for Female Offenders. In J.Travis & M. Waul, (Eds.), Prisoners once removed: The impact of incarceration and reentry on children, families, and communities, (pp. 67-103). Washington, D.C.: Urban Institute Press. Craig, C. D., & Sprang, G. (2007). Trauma exposure and child abuse potential: Investigating the 104 cycle of violence. American journal of orthopsychiatry, 77(2), 296-305. Craig, S. C. (2009). A historical review of mother and child programs for incarcerated women. The Prison Journal, 89(1 suppl), 35S-53S. Creswell, J. W. (2007). Qualitative inquiry and research method: Choosing among five approaches (2nd. ed.). Thousand Oaks, CA: Sage. Dalley, L. P. (2002). Policy implications relating to inmate mothers and their children: Will the past be prologue?. The Prison Journal, 82(2), 234-268. Dallaire, D.H., & Wilson, L.C. (2009). The relation of exposure to criminal activity, arrest, and sentencing to children’s maladjustment. Journal of Child and Family Studies, 19(41), 404-418. D'Andrade, A. C., & Nguyen, H. (2013). The relationship between use of specific services, parental problems, and reunification with children placed in foster care. Journal of Public Child Welfare.(In press). Deater-Deckard, K., Dodge, K. A., Bates, J. E., & Pettit, G. S. (1998). Multiple risk factors in the development of externalizing behavior problems: Group and individual differences. Development and Psychopathology, 10(03), 469-493. DeBellis, M.D., Broussard, E.R., Herring, D.J., Wexler, S., Moritz, G., & Benitez, J.G. (2001). Psychiatric co-morbidity in caregivers and children involved in maltreatment: A pilot research study with policy implications. Child Abuse & Neglect, 25(7), 923-944. DeHart, D. D., & Altshuler, S. J. (2009). Violence exposure among children of incarcerated mothers. Child and Adolescent Social Work Journal, 26, 467-479. Del Vecchio, T., Erlanger, A. C. E., & Slep, A. M. S. (2013). Twelve theories of child abuse. Handbook of Family Theories: A Content-Based Approach, 208. Dhillon, A. (2005). Keeping families together and safe: A primer on the child protectionhousing connection. Washington, DC: CHILD WELFARE League of America. Dressel, P.L. & Barnhill, S.K. (1994). Reframing gerontological thought and practice: The case of grandmothers with daughters in prison. The Gerontologist, 34(5), 685-691. Edwards, J. L. P. (2003). Judicial oversight of parental visitation in family reunification cases. Juvenile and Family Court Journal, 54(3), 1-24. Ehrensaft, M., Khashu, A., Ross, T., & Wamsley, M. (2003). Patterns of criminal conviction and incarceration among mothers of children in foster care in New York City. New York: Vera Institute of Justice and Administration for Children’s Services. 105 Fedock, G., Fries, L., & Kubiak, S. P. (2013). Service Needs for Incarcerated Adults: Exploring Gender Differences. Journal of Offender Rehabilitation, 52(7), 493-508. Ferszt, G. G., & Clarke, J. G. (2012). Health care of pregnant women in US State prisons. Journal of Health Care for the Poor and Underserved, 23(2), 557-569. Festinger, T. (1996). Going home and returning to foster care. Children and Youth Services Review, 18(4), 383-402. Foster, H., & Hagan, J. (2009). The mass incarceration of parents in America: Issues of race/ethnicity, collateral damage to children, and prisoner rentry. The Annals of the American Academy of Political and Social Science, 623, 179-194. Gabel, K., & Johnston, D. (Eds.). (1995). Children of incarcerated parents. Lexington Books. Gaudin Jr, J. M., & Sutphen, R. (1993). Foster care vs. extended family care for children of incarcerated mothers. Journal of Offender Rehabilitation, 19(3-4), 129-147. Genty, P. M. (1998). Permanency planning in the context of parental incarceration: Legal issues and recommendations. Child Welfare, 77(5), 543-59. Glass, J.C., & Honeycutt, T.L. (2002). Grandparents raising grandchildren: The courts, custody, and educational implications. Educational Gerontology, 28(3), 237-251. Glaze, L.E. & Bonczar, T.P. (2009). Probation and parole in the United States, 2008 (NCJ 228230). Washington, D.C.: U.S. Department of Justice, Bureau of Justice Statistics. Glaze, L. E. & Maruschak, L. M. (2008). Parents in prison and their minor children (NCJ 222984). Washington, D.C.: U.S. Department of Justice, Bureau of Justice Statistics. Goodman-Brown, T. B., Edelstein, R. S., Goodman, G. S., Jones, D. P., & Gordon, D. S. (2003). Why children tell: A model of children’s disclosure of sexual abuse. Child Abuse & Neglect, 27(5), 525-540. Greenfield, L., & Snell, T. (1999). Bureau of Justice Statistics Special Report: Women offenders. Washington, DC: US Department of Justice. Grella, C.E., & Greenwell, L. (2006). Correlates of parental status and attitudes toward parenting among substance abusing women offenders. The Prison Journal, 86, 89-113. Griffin, M. L., & Armstrong, G. S. (2003). The effect of local life circumstances on female probationers' offending. Justice Quarterly, 20(2), 213-239. Hairston, C.F. (2007). Focus on children with incarcerated parents: An overview of the research literature. Annie E. Casey Foundation: Baltimore, MD. 106 Hall, L. A., Sachs, B., & Rayens, M. K. (1998). Mothers' potential for child abuse: The roles of childhood abuse and social resources. Nursing Research, 47(2), 87-95. Halperin, R., & Harris, J.L. (2004). Parental rights of incarcerated mothers with children in foster care: A policy vacuum. Feminist Studies, 30 (2), 339-352. Hanlon, T. E., O'Grady, K. E., Bennett-Sears, T., & Callaman, J. M. (2005). Incarcerated drugabusing mothers: Their characteristics and vulnerability. The American Journal of Drug and Alcohol Abuse, 31(1), 59-77 Harm, N. J., & Phillips, S. D. (2000). You can't go home again: Women and criminal recidivism. Journal of Offender rehabilitation, 32(3), 3-21. Harrison, P. M., & Beck, A. J. (2006). Prison and jail inmates at midyear 2005. Bureau of Justice Statistics. NCJ, 213133. Hayward, R. A., & DePanfilis, D. (2007). Foster children with an incarcerated parent: Predictors of reunification. Children and Youth Services Review, 29(10), 1320-1334. Heffernan, E. (1993). Alderson: The early years. In American Correctional Association (Ed.), Female offenders: Meeting needs of a neglected population. Baltimore: United Book Press. Hill, C. E., Thompson, B. J., & Williams, E. N. (1997). A guide to conducting consensual qualitative research. The counseling psychologist, 25(4), 517-572. Hotelling, B. A. (2008). Perinatal needs of pregnant, incarcerated women. The Journal of perinatal education, 17(2), 37. Houck, K. D., & Loper, A. B. (2002). The relationship of parenting stress to adjustment among mothers in prison. American Journal of Orthopsychiatry, 72(4), 548-558. Johnson, E.I., & Waldfogel, J. (2002a). Children of incarcerated parents: Cumulative risk and children’s living arrangements. Chicago, IL: Northwestern University/University of Chicago Joint Center for Poverty Research. Johnson, E. I., & Waldfogel, J. (2002b). Parental incarceration: Recent trends and implications for child welfare. Social Service Review, 76(3), 460-479. Jones, L. (2005). The prevalence and characteristics of substance abusers in a child protective service sample. Journal of Social Work Practice in Addictions, 4(2), 33-50. Kendler, K. S., Bulik, C. M., Silberg, J., Hettema, J. M., Myers, J., & Prescott, C. A. (2000). Childhood sexual abuse and adult psychiatric and substance use disorders in women: An epidemiological and cotwin control analysis. Archives of General Psychiatry, 57(10), 953. 107 Knight, M., & Plugge, E. (2005). Risk factors for adverse perinatal outcomes in imprisoned pregnant women: a systematic review. BMC Public Health, 5(1), 111. Kohl, P., Edleson, J.L., English, D.J., & Barth, R.P. (2005). Domestic violence and pathways into child welfare services: Findings from the national survey of child and adolescent well-being. Child and Youth Services Review, 27(11), 1167-1182. Kruttschnitt, C., & Gartner, R. (2003) Women’s imprisonment. In M. Tonry (Ed.), Crime and justice: A review of research, Vol. 3, (pp. 1-82). University of Chicago Press. Kubiak, S. P., & Arfken, C. L. (2007). Beyond gender responsivity: Considering differences among community dwelling women involved in the criminal justice system and those involved in treatment. Women & Criminal Justice, 17(2-3), 75-94. Kubiak,, S.P. Kasiborski, N., Karim, N., & Schmittel, E. (2012). Does subsequent criminal justice involvement predict foster care and termination of parental rights for children born to incarcerated women? Social Work in Public Health, 27, 129-47. Kubiak, S.P., Kasiborski, N., & Schmittel, E. (2010). Assessing long-term outcomes of an intervention designed for pregnant incarcerated women. Research on Social Work Practice, 20(5), 528-535. Kubiak, S.P., Siefert, K., & Boyd, C.J. (2004). Empowerment and public policy: An exploration of the implications of section 115 of the personal responsibility and work opportunity act. Journal of Community Psychology, 32(2), 127-142. Kurshan, N. (1992). Women & imprisonment in the United States: History & current reality. Baltimore: Monkeywrench Press. Lapidus, L., Luthra, N., Verma, A., Small, D., Allard, P., & Levingston, K. (2005). Caught in the net: The impact of drug policies on women and families. New York: American Civil Liberties Union, the Brennan Center, and Break the Chains. Larrieu, J. A., Heller, S. S., Smyke, A. T., & Zeanah, C. H. (2008). Predictors of permanent loss of custody for mothers of infants and toddlers in foster care. Infant Mental Health Journal, 29(1), 48-60. LaVigne, N.G., Davies, E., & Brazzell, D. (2008). Broken bonds: Understanding and addressing the needs of children with incarcerated parents. Washington, DC: The Urban Institute. Lee, A.F., Genty, P.M., &Laver, M. (2005) The impact of the Adoption and Safe Families Act on children of incarcerated parents. Washington, DC: Child Welfare League of America. 108 Lee, B. J., & George, R. M. (1999). Poverty, early childbearing, and child maltreatment: A multinomial analysis. Children and Youth Services Review, 21(9/10), 755-780. Leeb, R. T. (2008). Child maltreatment surveillance: Uniform definitions for public health and recommended data elements. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Lochner, L., & Moretti, E. (2004). The effect of education on crime: Evidence from prison inmates, arrests, and self-reports. American Economic Review, 94.155-189 Love, M. C., Roberts, J., & Klingele, C. (2013). Collateral Consequences of Criminal Convictions: Law, Policy and Practice. Minnesota: Thomson West Lowry, M. (2004). Putting teeth into ASFA: The need for statutory minimum standards. Children and Youth Services Review, 26(11), 1021-1031. MacDonald, H. (1994). The ideology of family preservation. Public Interest, 115, 45–61. MacKenzie, M. J., Kotch, J. B., & Lee, L. C. (2011). Toward a cumulative ecological risk model for the etiology of child maltreatment. Children and Youth Services Review, 33(9), 16381647. Magruder, J., & Shaw, T. V. (2008). Children ever in care: An examination of cumulative disproportionality. Child Welfare, 87(2), 169-188. Martin, M. (1997). Connected mothers: A follow-up study of incarcerated women and their children. Women & Criminal Justice, 8(4), 1-23. Maruschak, L. M., & Bonczar, T.P. (2013). Probation and parole in the United States, 2012. Washington, DC: Bureau of Justice Statistics, US Department of Justice. Mastony, C. (2010). Childbirth in chains. Chicago Tribune, 1. Maurer, M., Potler, C., and Wolf, R. (1999). Gender and justice: Women, drugs, and sentencing policy. Washington DC: The Sentencing Project. Maurer, M. (2013). The changing racial dynamics of women’s incarceration. Washington DC: The Sentencing Project. Meyer, A.S., McWey, L.M., McKendrick, W., & Henderson, T.L. (2010). Substance using parents, foster care, and termination of parental rights: The importance of risk factors for legal outcomes. Children and Youth Services Review, 32, 639-649. Miller, K. (2006). The impact of parental incarceration on children: An emerging need for effective interventions. Child and Adolescent Social Work Journal, 23(4), 472-486. 109 Milner, J. S., Thomsen, C. J., Crouch, J. L., Rabenhorst, M. M., Martens, P. M., Dyslin, C. W., ... & Merrill, L. L. (2010). Do trauma symptoms mediate the relationship between childhood physical abuse and adult child abuse risk? Child Abuse & Neglect, 34(5), 332344. Moe, A. M., & Ferraro, K. J. (2007). Criminalized mothers: The value and devaluation of parenthood from behind bars. Women & Therapy, 29(3-4), 135-164. Moses, M. (2006). Correlating incarcerated mothers, foster care, and mother-child reunification. Corrections Today, 68(6), p.98. Mumola, C. J. (2000). Incarcerated Parents and Their Children (NCJ 182335). Washington, DC: Bureau of Justice Statistics Special Report. Murphey, D. A., & Braner, M. (2000). Linking child maltreatment retrospectively to birth and home visit records: An initial examination. Child Welfare Journal, 79, 711728. Myers, B. J., Smarsh, T. M., Amlund-Hagen, K., & Kennon, S. (1999). Children of incarcerated mothers. Journal of Child and Family Studies, 8(1), 11-25. Nair, P., Schuler, M. E., Black, M. M., Kettinger, L., & Harrington, D. (2003). Cumulative environmental risk in substance abusing women: Early intervention, parenting stress, child abuse potential and child development. Child Abuse & Neglect, 27(9), 997-1017. National Council on Crime & Delinquency, (2004). Evaluation of the project development of National Institute of Corrections/Child Welfare League of America’s planning and intervention sites funded to address the needs of children of incarcerated parents. Oakland, CA: National Council on Crime and Delinquency. National Inventory of Collateral Consequences of Conviction. (2013). Retrieved from http://www.abacollateralconsequences.org. National Women’s Law Center. (2010). Mothers behind bars.Washington, DC: National Women’s Law Center. Needell, B., & Barth, R. P. (1998). Infants entering foster care compared to other infants using birth status indicators. Child Abuse & Neglect, 22(12), 1179-1187. Neville, K. R. (2010). Forgotten children: Law enforcement agencies, child protective services, and children of arrested parents in Michigan. McNair Scholars Research Journal, 2(1), 13. Nieto, M. (2002). In danger of falling through the cracks: Children of arrested parents. Sacramento, CA: California Research Bureau, 3–36. 110 Nikulina, V., Widom, C. S., & Czaja, S. (2011). The role of childhood neglect and childhood poverty in predicting mental health, academic achievement and crime in adulthood. American Journal of Community Psychology, 48(3-4), 309-321. Osborne, J. W., & Overbay, A. (2004). The power of outliers (and why researchers should always check for them). Practical assessment, research & evaluation, 9(6), 1-12. Owen, B. (1998). In the mix: Struggle and survival in a women’s prison. Albany, NY: State University of New York Press. Patton, M.Q. (2002). Qualitative research and evaluation methods. Thousand Oaks, CA: Sage. Paxton, C., & Waldfogel, J. (1999). Welfare reform, family resources, and child maltreatment. In B. Meyer & G. Duncan (Eds.), The incentives of government programs and the wellbeing of families. Chicago: Joint Center for Poverty Research. Pecora, P. J., Kessler, R. C., Williams, J., O'Brien, K., Downs, A. C., English, D., et al. (2005). Improving family foster care: Findings from the Northwest Foster Care Alumni Study. Seattle, WA: Casey Family Programs. Petersilia, J. (1985). Probation and felony offenders. Fed. Probation, 49, 4. Phillips, (2012). Video visits for children whose parents are incarcerated: In whose best interest? Washington, DC: The Sentencing Project. Phillips, S. D., Burns, B. J., Wagner, H. R., & Barth, R.P. (2004). Parental arrest and children involved with child welfare services agencies. American Journal of Orthopsychiatry, 74(2), 174-186. Phillips, S. D., & Dettlaff, A. J. (2009). More than parents in prison: The broader overlap between the criminal justice and child welfare systems. Journal of Public Child Welfare, 3(1), 3-22. Phillips, S. D., Dettlaff, A. J., & Baldwin, M. J. (2010). An exploratory study of the range of implications of families' criminal justice system involvement in child welfare cases. Children and Youth Services Review, 32(4), 544-550. Phillips, S. D., Erkanli, A., Costello, E. J., & Angold, A. (2006). Differences among children whose mothers have been in contact with the criminal justice system. Drugs, Women, and Justice, 17, 43-61. Phillips, S. D., Leathers, S. J., & Erkanli, A. (2009). Children of probationers in the child welfare system and their families. Journal of Child & Family Studies, 18, 183-191. Poehlmann, J. (2005). Incarcerated mothers' contact with children, perceived family relationships, and depressive symptoms. Journal of Family Psychology, 19(3), 350. 111 Poehlmann, J., Dallaire, D., Loper, A. B., & Shear, L. D. (2010). Children's contact with their incarcerated parents: Research findings and recommendations. American Psychologist, 65(6), 575. Richie, B. E. (2001). Challenges incarcerated women face as they return to their communities: Findings from life history interviews. Crime & Delinquency, 47(3), 368-389. Renzetti, C. (2001). “One strike and you’re out”: Implications of a federal crime control policy for battered women. Violence Against Women, 7, 285. Ross, T., Khashu, A., & Wamsley, M. (2004). Hard data on hard times: An empirical analysis of maternal incarceration, foster care, and visitation. New York, NY: Vera Institute of Justice. Sabol, W. J., & Couture, H. (2008). Prison inmates at midyear 2007 (NCJ 221944). Washington, DC: Bureau of Justice Statistics. Sameroff, A. J., Bartko, W. T., Baldwin, A., Baldwin, C., & Seifer, R. (1998). Family and social influences on the development of child competence. In C. Feiring & M. Lewis, Families, Risk, and Competence, (pp. 161-185). Mahwah, NJ: Erlbaum. Sameroff, A. J., Seifer, R., Baldwin, A., & Baldwin, C. (1993). Stability of intelligence from preschool to adolescence: The influence of social and family risk factors. Child Development, 64(1), 80-97. Sameroff, A. J., Seifer, R., Barocas, R., Zax, M., & Greenspan, S. (1987). Intelligence quotient scores of 4-year-old children: Social environmental risk factors. Pediatrics, 79, 343-350. Sameroff, A. J., Seifer, R., & Zax, M. (1982). Early development of children at risk for emotional disorder. Monographs of the Society for Research in Child Development, 47 (Serial No. 199). Sameroff, A., Seifer, R., Zax, M., & Barocas, R. (1987). Early indicators of developmental risk: Rochester Longitudinal Study. Schizophrenia Bulletin, 13(3), 383. Schirmer, S., Nellis, A., & Mauer, M. (2009). Incarcerated parents and their children: trends 1991-2007. Washington, DC: The Sentencing Project. Schmittel, E., Kubiak, S. P., & Kasiborski, N. (October 2010). Foster care experiences of children born to incarcerated women. Oral paper presented at the 2010 Council on Social Work Education Annual Program Meeting, Portland, OR. Schofield, G., & Beek, M. (2005). Providing a secure base: Parenting children in long-term foster family care. Attachment & Human Development, 7(1), 3-25. Sedlak, A. J., & Broadhurst, D. D. (1996). Third national incidence study on child abuse and 112 neglect. Washington, DC: U.S. Department of Health and Human Services. Sedlak, A., Mettenburg, J., Basena, M., Petta, I., McPherson, K., Greene, A., & Li, S. (2010). Fourth national incidence study of child abuse and neglect (NIS-4): Report to Congress. Washington D.C. U.S. Department of Health and Human Services. Sedlak, A., & Shultz, D. (2005). Racial differences in protective services investigations of abused and neglected children. In Derezotes et al (Eds.) Race matters in child welfare: The overrepresentation of African American children in the system (pp.77-118). Washington DC: Child Welfare League of America. Shin, S. H., & Lee, S. (2011). Caregiver Social Support for Children Involved With Child Welfare: Correlates and Racial/Ethnic Variations. Journal of Public Child Welfare, 5(4), 349-368. Sidebotham, P., & Heron, J. (2006). Child maltreatment in the “children of the nineties”: A cohort study of risk factors. Child Abuse & Neglect, 30(5), 497-522. Siefert, K., & Pimlott, S. (2001). Improving pregnancy outcome during imprisonment: A model residential care program. Social Work, 46(2), 125-134. Simms, M.C., Fortuny, K., & Henderson, E. (2009). Racial and ethnic disparities among lowincome families. Washington, DC: The Urban Institute. Smith, B.E., & Elstein,S.G. (1994). Children on Hold: Improving the Response to Children Whose Parents Are Arrested and Incarcerated. Washington, D.C.: American Bar Association, Center for Children and the Law. Smith, D. K., Johnson, A. B., , Pears, K. C., Fisher, P. A., & DeGarmo, D. S. (2007). Child maltreatment and foster care: Unpacking the effects of prenatal and postnatal parental substance use. Child Maltreatment, 12,150–160. Stake, R. E. (1995). The art of case study research. Thousand Oaks, CA: Sage. State of Michigan. (2008). Michigan absent parent protocol: Identifying, locating, and notifying absent parents in child protective proceedings. Lansing, MI: State Court Administrative Offices. Stein, T. J. (2003). The adoption and safe families act: How congress overlooks available data and ignores systemic obstacles in its pursuit of political goals. Children and Youth Services Review, 25(9), p. 669-682. Sussman, D. (2008). Bound by Injustice: Challenging the Use of Shackles on Incarcerated Pregnant Women. Cardozo Journal of Law & Gender, 15, 477. Swann, C. A. and Sylvester, M. S. (2006). The foster care crisis: What caused caseloads to 113 grow? Demography, 43, 309-335. Swift, K. J. (1995). Manufacturing “bad mothers”: A critical perspective on child neglect. Toronto: University of Toronto Press. Tajima, E. A. (2000). The relative importance of wife abuse as a risk factor for violence against children. Child Abuse & Neglect, 24(11), 1383–1398. Taylor, S. D. (1996). Women offenders and reentry issues. Journal of Psychoactive Drugs, 28, 85-93. Thompson, R. (2006). Exploring the link between maternal history of childhood victimization and child risk of maltreatment. Journal of Trauma Practice, 5(2), 57-72 U.S. Bureau of Labor Statistics. (2013). Earnings and unemployment rates by educational attainment. Washington, DC: Office of Occupational Statistics and Employment Programs. . U.S. Department of Housing and Urban Development. (2000). Multi-family tenant characteristics system data base. Washington, DC: Government Printing Office. U.S. Government Accountability Office. (2005). GAO-05-238 denial of federal benefits. Washington, DC: Government Printing Office. U.S. Government Accountability Office (2007). African American Children in Foster Care: Additional HHS Assistance Needed to Help States Reduce the Proportion in Care. Washington, DC: Government Printing Office. Villanueva,C.K., Frome, S.B, & Lerner, G. (2009). Mothers, infants and imprisonment: A national look at prison nurseries and community based alternatives. New York: Women’s Prison Association. West, H. C., & Sabol, W. J. (2009). Prison inmates at midyear 2008—statistical tables. Bureau of justice statistics special report. Washington, DC: Government Printing Office. West, H. C., Sabol, W. J., & Greenman, S. J. (2010). Prisoners in 2009 (NCJ 231675). Washington, DC: Bureau of Justice Statistics, US Department of Justice. Wilsnack, S. C., Vogeltanz, N. D., Klassen, A. D., & Harris, T. R. (1997). Childhood sexual abuse and women's substance abuse: National survey findings. Journal of Studies on Alcohol and Drugs, 58(3), 264. Wright, L., & Seymour, C. B. (2000). Working with children and families separated by incarceration. Child Welfare League of America Press. Wooldredge, J. D., & Masters, K. (1993). Confronting problems faced by pregnant 114 inmates in state prisons. Crime & Delinquency, 39(2), 195-203. Wulczyn, F. (2004). Family reunification. The Future of Children,14 (1), 95-113. Wu, S. S., Ma, C. X., Carter, R. L., Ariet, M., Feaver, E. A., Resnick, M. B. & Roth, J. (2004). Risk factors for infant maltreatment: A population-based study. Child Abuse & Neglect, 28:1253-1264. Yin, R. K. (2003). Case study research: Design and methods (3rd ed.). Thousand Oaks, CA: Sage. Young, N. K, Boles, S. M.,& Otero, C. (2007). Parent substance use disorders and child maltreatment: Overlaps, gaps, and opportunities. Child Maltreatment, 12(2), 137-149. Zemans, E. S., & Smith, M. N. (1946). War activities of women and girls in state correctional institutions 1943-1944. Chicago: Central Howard Association. Zlotnick, C., Tam, T. & Bradley, K. (2007). Impact of adult trauma on homeless mothers. Community Mental Health Journal, 43, 13-32. 115