Disrupting the Family Stress-Proximal Process: A Scoping Review of Interventions for Children with Incarcerated Parents

Abstract

The United States’ overreliance on incarceration has resulted in the imprisonment of millions of individuals – the majority of whom are parents of minor children. While mass incarceration has failed to effectively reduce crime or increase safety, it has dramatically harmed children and families in the United States. In turn, a wealth of research confirms the negative social, emotional, and psychological impacts of parental incarceration on children and the disproportionate impact on Black and Hispanic families and families living in poverty. As activists work towards dismantling this discriminatory and overly punitive system, it is also necessary to support children and adolescents currently impacted by parental incarceration. Using the Family Stress-Proximal Process (FSPP) model (Arditti, 2016) as a frame, the current paper critically reviews the literature on interventions to support children with incarcerated parents (CIP). The use of the FSPP frame highlights that while most intervention research has focused on promoting parenting skills of incarcerated parents and improving visit experiences, there is a dearth of research on interventions that 1) support at-home caregivers, 2) provide developmentally-targeted and -appropriate services and 3) acknowledge and counteract systems of inequality like structural racism and poverty that cause and exacerbate incarceration-related stress. These findings support a research agenda that prioritizes interventions framed around the intersectional identities of CIP and the intersecting systems that impact their lives.

Introduction

The United States has a higher rate of incarceration than any other country in the world (Sawyer & Wagner, 2020). Nearly 2.2 million people are held in prisons and jails in the U.S. on any given day, and over 1.6 million of these individuals are parents of minor children (U.S. Department of Justice, 2018). Pursuantly, over 5 million children in the United States have had a parent in jail or prison at some point in their lives (Annie E. Casey Foundation, 2016). Researchers have not found that higher incarceration rates across the U.S. lead to lower rates of violent crime or safer societies (Roeder et al., 2015; Steman, 2017). Comparatively, it is well-documented that high incarceration rates have deleterious effects for communities and families of incarcerated individuals (Hagan & Dinovitzer, 1999). Parental incarceration is associated with myriad negative social, emotional, physical, and psychological outcomes for children and adolescents (Lee et al., 2013; Miller, 2006) and children with incarcerated parents are more likely to experience additional stressful and potentially traumatic events during their childhoods (Casey et al., 2015).

Additionally, systems of mass incarceration and policing interact with other systems of oppression in the United States, like racism and capitalism, to disproportionately target economically marginalized families and families of color, particularly Black and Hispanic families (Alexander, 2010; Crenshaw, 2011; Garland et al., 2008; Rabuy & Kopf, 2015). Abolitionist activists and scholars have called for the dismantling of this discriminatory and destructive system for years (Davis, 2003; Gilmore, 2007; Washington, 2018), but as organizers work towards reform, it is necessary for researchers and clinicians to support children and families impacted by the detrimental effects of the system of mass incarceration.

The current study seeks to critically review the literature to-date on interventions and practices to support the wellbeing of children with incarcerated parents (CIP). Grounded in Joyce Arditti’s (2016) Family Stress-Proximal Process (FSPP) model, which takes a developmental ecological approach to examining the experience of parental incarceration, this review examines how interventions have sought to promote child adjustment in the context of parental incarceration. First, we define the FSPP model and note its specific contributions as a framework for critiquing interventions for CIP. Next, we describe our search strategy, in line with best practices for scoping reviews from the Joanna Briggs Institute (Peters et al., 2015). We present results of our scoping review in three sections focusing on: 1) targets of intervention, 2) developmental considerations, and 3) the integration of sociodemographic and contextual factors. Finally, we argue that this synthetic review highlights a critical need for an intersectional and ecological approach to intervention research with CIP.

Family Stress-Proximal Process Model

In Joyce Arditti’s (2016) Family Stress-Proximal Process (FSPP) Model, parental incarceration is conceptualized as a continuous stressor that directly impacts child adjustment as well as indirectly influences adjustment through proximal processes. Parental incarceration shares similarities with other types of potentially traumatic experiences, including forced separation and witnessing serious threat to a caregiver. However, the complex, ongoing, and stigmatizing nature of parental incarceration uniquely positions it to impact multiple aspects of children’s lives (Phillips & Gates, 2011). Extant research highlights the importance of accounting for stigmatizing social structures when providing support and intervention for individuals who have experienced other types of stigmatized losses, like death by suicide or homicide (Armour, 2007). Subsequent changes to family structure, potentially stressful experiences throughout the criminal justice process and during visits, and stress related to ambiguous understanding of where a parent has gone can act to exacerbate the initial trauma of separation and complicate children’s experiences of coping and grieving (Bocknek et al., 2009). The current review focuses on interventions explicitly for CIP, as opposed to more general trauma or grief treatments that could be used with CIP, to examine interventions that target this specific complex, continuous, and potentially stigmatizing stressor and its sequelae.

Proximal Processes as Points of Intervention

Arditti (2016) proposes three primary proximal processes that influence children’s adjustment: at-home caregiver parenting, contact and visiting experiences with the incarcerated parent, and child coping and social support. These proximal processes, which are also potential intervention targets, may serve to increase stress or provide protection to CIP.

Parental incarceration often has immediate impacts on children’s caregiving situation. Differences are noted between maternal and paternal incarceration, with most children of incarcerated fathers cared for primarily by mothers, and most children of incarcerated mothers cared for by other non-paternal family members, with grandmothers being most common (Glaze & Maruschak, 2010). At-home caregivers can face myriad parenting stressors including financial stress, experiences of stigma, navigation of their own emotional reactions to the parent’s incarceration, and supporting children’s emotional and behavioral reactions, as well as age-specific stressors for caregiving grandparents like health problems (Hairston, 2009), all with implications for child well-being. Critically, positive relationships with at-home caregivers have been shown to be protective for CIP (Sukhramani et al., 2018). Thus, interventions that (1) promote at-home caregiver wellbeing, (2) cultivate parenting skills, and (3) work to reduce at-home caregiver stress, may be well-poised to improve child outcomes. Further, at-home caregivers can act as gatekeepers to contact and visiting with incarcerated parents (Arditti et al., 2005).

Contact with incarcerated parents, through phone and video calls, letters, or in-person visits, varies widely from family to family and is influenced by both structural factors (e.g., jail location, financial resources) and individual and family factors (e.g., caregiver opinions, child age; Poehlman et al., 2010; Arditti et al., 2005). Findings on benefits and stressors related to incarcerated parent-child contact have been mixed, with some studies suggesting increased contact associated with better academic and emotional outcomes and others finding associations with insecure attachments and attention problems (Poehlmann et al., 2010). Policies regarding visiting have also changed in recent years, and thus more research is needed to address these changes (Poehlmann et al., 2019).

Finally, children’s adjustment can be bolstered by individual factors, including coping skills and social supports, which are critical processes linked to children’s resilient responses to stress (Compas et al., 2001; Dumont & Provost, 1999). Research suggests that CIP recruit from a wide range of coping skills (Thulstrup & Karlsson, 2017) and report receiving social support from many sources (Nesmith & Ruhland, 2008); perceptions of social support are negatively associated with internalizing and externalizing behavior problems in CIP (Hagen & Myers, 2003).

However, many CIP also report experiencing social isolation, stigma, and potentially maladaptive coping strategies, like avoidance and minimization (Nesmith & Ruhland, 2008; Thulstrup & Karlsson, 2017). The majority of this research has been qualitative, limiting causal conclusions. However, qualitative researchers have observed withdrawn behaviors, angry outbursts, and difficulties talking about emotions related to denial, minimization, or limited access to social supports (Bocknek et al., 2009; Johnson & Easterling, 2015). These findings, in tandem with results linking social supports to better behavioral health outcomes (Hagen & Myers, 2003), support child interventions that promote healthy coping and help children recruit and develop positive relationships with social supports.

Child-centered Approach

Arditti’s (2016) FSPP model takes a developmental, child-centered approach by centering child adjustment as the primary outcome of interest. Researchers have noted the lack of child outcome measures in previous literature reviews on interventions for CIP (Johnston, 2012). Multiple reviews have examined effects of parenting programs on parenting outcomes, but without data on child outcomes, it is impossible to know whether changes in parenting skills and knowledge are effectively meeting the goal of improving parent-child interactions and ultimately child adjustment (Armstrong et al., 2018; Troy et al., 2018).

Contextual Variables and Social Inequality

Arditti (2016) notes that “imprisonment is both a result of and contributor to social inequality” (p. 68). The United States carceral system disproportionately targets individuals experiencing other types of marginalization, including racism, anti-Blackness, sexism, poverty, substance use and addiction, intergenerational incarceration, and trauma. Further, incarceration can increase experiences of oppression, acting as a barrier to employment, economic mobility, and social supports, as well as additional experiences of trauma for both the incarcerated person and their family (DeVeaux, 2013; Hagan & Dinovitzer, 1999). Interventions seeking to support CIP that fail to acknowledge and actively work against these systems of social inequality risk perpetuating harm to children and families (Heberle et al., 2020). The current paper examines whether the reviewed interventions incorporated an understanding of social positioning variables into their work with CIP and how this acknowledgement of broader social inequality impacted intervention choices.

Methods

Review Focus and Systematic Search

The current review focused on providing an overview of research on interventions for children with incarcerated parents, with particular focus on identifying available research on interventions targeting the different proximal processes identified by Arditti (2016), examining how research has been conducted regarding developmental, sociodemographic, and contextual considerations, and identifying knowledge gaps in the current body of literature. Due to this focus on broadly identifying available research and analyzing how this research has been conducted, the authors chose to utilize scoping review methodology (Munn et al., 2018).

In line with best practices for scoping reviews (Peters et al., 2015), a systematic search for studies was conducted in August 2022. The search terms “correctional institution”, or prison, or jail, or incarc*, or criminal, AND child*, or youth, or adolesc*, or kid, or teen, AND parent*, or mother*, or father*, or guardian, or caregiver, AND treatment, or intervention, or therapy, or program were entered into PsycINFO. PsycINFO was chosen for this project because of the explicit grounding of the current review in developmental psychology theories. No date or language restrictions were used. To minimize publication bias, dissertations were included. Reference lists of existing reviews, meta-analyses, and theoretical papers were examined (Armstrong et al., 2018; Goshin & Byrne, 2009; Johnston, 2012; Kjellstrand, 2017; Troy et al., 2018).

Eligibility Screening

Studies had to meet four inclusion criteria in order to be eligible for the current review.

1. The study was based in the United States.

2. The study included children of incarcerated parents, caregivers of CIP, or incarcerated parents themselves. Children were defined as minors from birth to age 18.

3. The study included an intervention.

4. The study included child outcome measures.

The systematic search identified 1,361 articles for screening (see Figure 1). The first phase of eligibility screening, excluding 1,234 articles, excluded reports if the title and/or abstract indicated that inclusion criteria were not met. When unclear, articles were retained for full-text review. The second phase involved full-text review. When reviews, meta-analyses, and theoretical papers were identified in this phase, their reference lists were reviewed for articles not already captured by the search. This resulted in six additional articles, yielding 133 full texts reviewed. During full text review, 109 articles were excluded for the following reasons: (a) based outside of the United States (n=14), (b) majority of participants were not CIP, incarcerated parents, or caregivers (n=12), (c) no intervention was studied (n=17), (d) no child outcome measures were included (n=41), (e) case study (n=6), (f) review, meta-analysis, or theoretical paper (n=16), and (g) unable to access paper, although attempts were made to contact the authors (n=3). In total, 24 articles met full inclusion criteria and were retained for this review. This included three doctoral dissertations and 21 published articles.

Figure 1.

Flow Diagram of Eligibility Screening

Results

Results are organized into three sections. The first section, guided by Arditti’s proximal processes (2016), explores how reviewed interventions align with the proximal process of at-home caregiver parenting, contact and visiting with incarcerated parents, and child coping and social support. Zero papers targeted exclusively at-home caregiver parenting. As such, a review of at-home caregiver parenting interventions is not included in the Results section, but critical analysis of this large gap in the literature is provided in the Discussion. Twelve papers targeted parenting, contact and visiting (Table 1). This section on parenting, contact, and visiting has been further divided into summaries of: 1) residential parenting programs, 2) parenting education programs, and 3) extended visiting programs. Six papers targeted child coping and social support (Table 2); this section is divided into summaries of: 1) mentoring programs and 2) group interventions. Finally, five papers targeted multiple areas (Table 3), and these are summarized together. Tables include additional methodological details on studies of interest, including age range, study design, and demographics.

Table 1.

Interventions Targeting Parenting, Contact, and Visiting Experiences

Auth or (Year)	Prison or Jail	Incarcer ated Parent	Sample Size	Child Age	Child Gender	Parent Racial/Ethnic Identity	SES	Intervention	Child Outcomes	Type of Study
Barr et al. (2014)	Prison (Juvenile detention center)	Teen fathers	41	1–15 months (M=6.88)	63% female	71% Hispanic, 17% Black, 12% MR	--	Baby Elmo Program	Observational – IGDI-IPCI	QE (individual linear growth modeling)
Barr et al. (2011)	Prison (Juvenile detention center)	Teen fathers	20	6–36 months (M=16.10)	40% female	75% Hispanic, 20% AA, 5% MR	--	Baby Elmo Program	Observational – Child subscales of Emotional Responsiveness observation	QE (individual linear growth modeling)
Byrne et al. (2010)	Prison	Mothers	30	2–16 months	--	67% “ethnic minorities”	--	Prison nursery program	Observational – SSP	QE (natural groups)
Cassidy et al. (2010)	Jail Diversion	Mothers	20	Birth-12 months	--	65% AA, 33% Caucasian, 2% Hispanic	45% HS or GED	Tamar’s Children residential jail diversion program	Observational – SSP	QE (pre/post; existing norms)
Condon (2017)	Prison	Mothers	17	Birth-29 months	--	41% White, 24% MR, 18% Latinx, 12% Native American, 5% AA	29% HS or GED	Residential parenting program	Infant observations; Mother, caregiver, and staff interviews	Qualitative
Eddy et al (2013)	Prison	Mothers & Fathers	359	1–15.6 years (M=8)		59% White, 13% AA, 11% multiracial, 8% Native American, and 8% Latino	31% HS or GED	Parenting Inside Out-Parent Management Program	Parent interviews – positive parentchild interaction	RCT
Gat (2000) a	Prison	Mothers	25	--	--	--	32% HS diploma	Mother/Offspring Life Development Program	Parent report - IPPA	QE (pre/post)
Goshin et al., (2014)	Prison	Mothers	111	Birth-18 months at intervention; 3–5 years at followup	44% female	53% Black, 24% White, 23% Hispanic	63% HS or GED, over 80% received public assistance	Prison nursery program	Parent report – CBCL	QE (comparison group; existing norms)
Harris & Landreth (1998)	Jail	Mothers	22	3–10 years (M=5)	47% female	50% Caucasian, 41% AA, 5% Hispanic, 5% Native American	Ed. range 9th grade - 1 year college, 55% incomes less than $20,000/ year	Filial Therapy	Parent report – FPC	QE (pre/post; comparison group)
Harrison (1997)	Prison	Fathers	30	8–17 years	--	--	--	Parenting and Behavior - Management Training	Child report – SelfPerception Profile	QE (pre/post; comparison group)
Landreth & Lobaugh (1998)	Prison	Fathers	32	4–9 years (M=5.94)	59% female	52% Caucasian, 30% Hispanic, 18% AA	31% not completed HS, 37% HS, 32% college	Filial Therapy	Parent report – FPC; Child report – JPPSST	QE (pre/post; comparison group)
Schubert et al. (2016)	Prison	Mothers	43	4.1–17.4 years (M=10.2)	50% female	75% Caucasian, 12.5% AA, 8.3% Native American, 4.2% MR	62.5% less than $25,000/ year; 58.3% some college	Extended Visiting Program	Interviews with mothers and at-home caregivers	Qualitative
Snyder et al. (2002)	Prison	Mothers	58	1–16 years	39% female	69% AA, 31% White	--	MotherChild Visitation Program	Interviews with mothers	Qualitative

Note. AA = African American; CBCL = Child Behavior Checklist; FPC – Filial Problem Checklist; IGDI-IPCI = Individual Growth and Development Indicators for Infants and Children: Indicator of Parent-Child Interaction; IPPA = Inventory of Parent and Peer Attachment; JPPSST = The Joseph Preschool and Primary Self-Concept Screening Test; MR = Mixed Race; RCT = Randomized Controlled Trial; QE = Quasi-experimental; SSP = Strange Situation Procedure

^aDissertation

Table 2.

Intervention Targeting Child Coping and Social Support

Author (Year)	Prison or Jail	Incarce rated Parent	Sample Size	Child Age	Child Gender	Child Racial/Ethnic Identity	SES	Intervention	Child Outcomes	Type of Study
Bruster & Foreman (2012)	Not specified	Not specified	35	Average between 10–11 years	29.7 % female	54% AA, 24% White	--	Mentoring Children of Prisoners Program	Child and at-home caregiver surveys interviews	Descriptive
Stumpet al (2018)	Prison	Mothers and Fathers	1484	M=11.59 years	52% female	60.92 male % “racial minority” 19.43 % Hispanic	M=3.08 “1=Less than $10,000 ; 16=$20 0,000 or more”	Enhancements of the Big Brother Big Sister Program	Child report-Academic achievement, educational expectations, grades, attendance	QE
Laakso & Nygaard (2012)	Prison	Mothers and Fathers	22	9–16 years (M=10.72)	60% female	33.3% White, 28.6% MR, 23.8% AA, 4.8% Latin@, 4.8% Asian American, 4.8% PI	54.5% HS or GED	CHIPP Mentoring	Intervie ws with children, at-home caregivers, and mentors	Qualitative
Sagar (2012)a	Prison	60% Father, 30% Mother, 10% Both	10	6–12 years (M=9.20)	40% female	80% AA, 20% Caucasian	Recruited from agencies serving low-income families	Group Intervention	Child report – CDI, CMAS, EFIC; Caregiver report CBCL	QE (pre/post)
Shlafer & Poehlmann (2010)	Not specified	Mothers and Fathers	57	4–15 years	60% female	93% “from minority racial or ethnic group”	35% less than HS, 68% incomes below federal poverty line, 84% received public assistance	Mentoring Connections	Interviews with child, at-home caregiver, and mentor; Child report – IPPA Caregiver and teacher report-CBCL	Mixed Methods
Shlafer et al. (2009)	Not specified	Mothers and Fathers	57	4–15 years	60% female	49% Black, 40% MR	35% less than HS, 68% incomes below federal poverty line, 84% received public assistance	Mentoring Connections	Interviews with child, at-home caregiver, and mentor; Child report – IPPA Caregiver and teacher report-CBCL	Mixed Methods
Springer et al. (2000)	Not specified	Not specified	10	4th and 5th graders	60% female	100% Hispanic	--	Group Intervention	Child report - HSS	QE (pre/post; comparison group)

Note. AA = African American; CBCL = Child Behavior Checklist; CDI = Child Depression Inventory; CHIPP = Children of Incarcerated Parents Program; CMAS = Children’s Manifest Anxiety Scale; EFIC = Enright Forgiveness Inventory for Children; HSS = Hare Self-Esteem Scale; IPPA = Inventory of Parent and Peer Attachment; MR = Mixed Race; PI = Pacific Islander; QE = Quasi-experimental

^aDissertation

Table 3.

Interventions Targeting Multiple Processes

Author (Year)	Prison or Jail	Incarcerated Parent	Sample Size	Child Age	Child Gender	Racial/Ethnic Identity	SES	Intervention	Child Outcomes	Type of Study
Block & Potthast (1998)	Prison	Mothers	16 mothers, 13 daughters	Not reported – but must be 717 years to participate	100 % female	Parents: 90% AA, 10% not reported	--	Girl Scouts Beyond Bars	Interviews with at-home caregivers	Mixed Methods (Quant used descriptively)
Goepferich (2011)a	Prison	Mothers	7 mothers	Not reported – but must be 717 years to participate	100 % female	Parents: 57% White, 43% Black	--	Girl Scouts Beyond Bars	Interviews with mothers	Qualitative
Miller et al. (2013)	Jail	62% Mothers, 38% Fathers	31 at-home caregivers, 29 children	4–14 years	62% female	Children: 62% AA, 38% White	37% below federal poverty guidelines	Strengthening Families Program	At-home caregiver report POCA, BERS, SRP	QE (pre/post; 4 month followup)
Phillips & O’Brien (2012)	Both	67% Mothers, 33% Fathers	6 families	4–9 years	--	--	--	Family Therapist / Case Manager Services	Interviews with at-home caregivers	Qualitative
Poehlmann et al. (2021)	Jail	Fathers	71 families	3–8 years old	44% female	Children: 31% White, 28% AA, 28% MR, 11% Latinx	100% below federal poverty guidelines	Sesame Street	At home caregiver report—SDQ, child observations -- JPOC,	RCT

Note. AA = African American; BERS = Behavior and Emotional Rating Scale; MR = Mixed Race; PI = Pacific Islander; POCA = Parent Observation of Child Adaptation; QE = Quasi-experimental; RCT = Randomized Controlled Trial; SRP = Strength/Resilience and Parenting Scales JPOC = Jail-Prison Observation Checklist

^aDissertation

The second and third sections of the results reviews additional gaps noted in the literature The second section critically examines how researchers have taken (or failed to take) a developmental approach when tailoring interventions to children. The final, third section highlights how researchers integrate information about sociodemographic and contextual factors impacting CIP and their families. This final section is divided into summaries of how the body of research examines 1) gender, 2) racial and ethnic inequality, 3) socioeconomic status, and 4) institutional settings for interventions.

Proximal Process 1: Parenting, Contact, and Visiting

Residential Parenting Programs

Residential parenting programs (RPP), often referred to as prison nurseries, allow mothers and infants to remain together for children’s early years. Researchers hypothesized that infants who remained with mothers for the beginning of life would be better able to form secure attachments (Bowlby, 1982), creating a foundation for well-being and healthy adjustment. Four papers described impacts of RPPs (Byrne et al. 2010; Cassidy et al., 2010; Condon, 2017; Goshin et al., 2014) where infants were allowed to stay with their mothers for varying lengths of time from 12–29 months. One program (Cassidy et al., 2010) used a jail diversion model in which women who were pregnant, substance-abusing, and had been convicted of non-violent offenses were eligible to participate instead of serving traditional jail time. The other three programs all took place in longer-term prison facilities. All four RPPs included parenting education and/or therapeutic support for mothers parenting while incarcerated.

Findings suggest that RPPs successfully promote secure attachment. Both Byrne et al. (2010) and Cassidy et al. (2010) used the Strange Situation procedure (Ainsworth et al., 2015) to classify infants as securely or insecurely attached. The Strange Situation procedure is a validated test involving a one-time observation of specific mother-child interactions (Aisnworth et al., 2015). Given ethical constraints to randomization, these authors compared rates of secure attachment to those found in a low-risk community sample and various higher risk samples including dyads with low incomes, dyads with depressed mothers, and dyads with substance-abusing mothers. Both studies found no significant differences between rates of secure attachments in their samples and the low-risk comparison, and both found significantly higher rates of secure attachments in their samples compared to the high-risk comparisons. Condon (2017) also evaluated attachment security, using unstructured ethnographic interviews and observations instead of the Strange Situation procedure. She found that the majority of infants had a secure attachment with at least one person, however, 25% of the infants in her study had formed a secure attachment with a caregiver in the prison who was not their biological mother and who would not remain in their lives after release. Further, she rated 41% of mother-infant dyads as at risk for relational health problems at release.

Unlike other studies, Goshin et al. (2014) examined long-term outcomes of an RPP on preschoolers’ behavior problems, comparing the RPP sample to a sample of families from the Fragile Families and Wellbeing Study (Reichman et al., 2001) in which mothers and infants had been separated due to incarceration with no residential parenting option. The authors found that, by preschool, children from the RPP sample demonstrated significantly fewer internalizing behaviors than same-age children from Fragile Families. However, no significant differences were observed on attention deficit/hyperactivity or aggressive behaviors. This study adds an important longitudinal perspective to the impacts of RPP for infants of incarcerated mothers.

Taken together, the current body of research supports RPPs’ positive associations with secure attachment in infancy and reduced internalizing behavior problems in preschool. However, more research and a wider range of methodological choices are needed to clarify these effects. Future research, specifically mixed methods work that would combine the rigor of quantitative measures with the nuance and detail of qualitative measures, as well as more longitudinal follow-up, could help unpack the results of extant studies.

Parenting Education Programs

Seven papers examined the impacts of parenting education programs. These included evaluations of the Baby Elmo program for incarcerated teen fathers (Barr et al., 2014; Barr et al., 2011), filial therapy for mothers in jail (Harris & Landreth, 1998) and fathers in prison (Landreth & Lobaugh, 1998), the Mother/Offspring Life Development (MOLD) program for incarcerated mothers (Gat, 2001), a behavior management training program for incarcerated fathers (Harrison, 1997) and Parenting Inside Out: a parent management training for incarcerated parents (Eddy et al.. 2013). While the Baby Elmo program was directed towards fathers of 1–36 month-olds, other programs served parents with children ranging from 1–17 years old (for specific ranges see Table 1). All programs focused on enhancing parenting skills and knowledge.

The MOLD paper (Gat, 2001) was a dissertation with methodological concerns. The intervention itself was a conglomeration of parenting programs and visit changes. This “kitchen sink” approach made it impossible to discern what actual intervention the participant mothers had received. The comparison group for this study was a group of mothers who had participated in MOLD years before, which confounded the comparative conclusions that could be drawn, and the child outcome measure was a parent report of adolescent-mother attachment, which is typically completed by the adolescent themselves. Gat (2001) saw no significant differences in attachment, but given the methodological concerns, results are uninterpretable.

While the seven parenting education studies were designed to examine outcomes related to parenting, they all included at least one child outcome measure. The Baby Elmo studies (Barr et al., 2014; Barr et al., 2011) observed significant improvements in the quality and communication of infant-father interactions with incarcerated teen fathers. The Parenting Inside Out randomized control trial (Eddy et al 2013) found an increase in parent-reported positive parent-child interaction for parents assigned to the intervention. The filial therapy studies (Harris & Landreth, 1998; Landreth & Lobaugh, 1998) found significantly decreased parent-reported child problem behaviors and Landreth and Lobaugh (1998) also saw increased child-reported positive self-concept. Conversely, Harrison (1997) found no improvements to child-reported positive self-perception. While fathers in the program developed improved attitudes about their children, children themselves did not seem to benefit. Critically, unlike parents in the studies where benefits were observed, the fathers participating in Harrison’s (1997) behavior management training did not interact with their children as part of the program, suggesting that the combination of parent education with opportunities for practice and interaction may drive improvements in child behavior and self-concept. More research is needed to parse out the distinct impacts of educational groups and opportunities for practice and interaction; however, programs able to include both may be best poised to improve child outcomes.

Extended Visiting Programs

The final two papers examined changes to visit experiences (Schubert et al., 2016; Snyder et al., 2002). Both programs occurred in women’s prisons and thus were aimed at incarcerated mothers and their children. These programs sought to counteract negative aspects of prison visits by creating more child-friendly visiting environments, extending visiting times, allowing physical contact and hugging, and providing structured activities like crafts and games. The authors suggested that improving visits could improve mother-child relationships and reduce stress related to the prison visit experience.

Qualitative interviews with mothers from both studies were suggestive of benefits from the extended and expanded visiting programs. Mothers stated that the program helped them build better relationships with their children and maintain these relationships. At-home caregivers interviewed by Schubert and colleagues (2016) had more mixed feelings. One caregiver stated, “[The children] really need those visits. It’s like therapy. They were so insecure when I got them, but now they’re strong and happy because they get to see her” (p. 225). Conversely, others noted, “A few days after the visits, his anxiety level soared. He got all stirred up” (p. 225), and, “as much as she looks forward to seeing mom, there was always a major behavioral issue before going, which was not fun for anyone” (p. 226). Inconsistent reporting between mothers and some at-home caregivers highlights the critical role of at-home parents in interventions. Additionally, given thematic qualitative findings, outcomes related to children’s relationships with incarcerated parents, self-confidence, and positive emotionality may be prudent outcomes to measure in future visit program research.

Proximal Process 2: Child Coping and Social Support

Mentoring Programs

Four papers examined the effects of mentoring programs on CIP. Three of these provided CIP-specific mentoring within the broader Big Brothers, Big Sisters program, (Laakso & Nygaard, 2012; Shlafer & Poelmann, 2010; Shlafer et al., 2009; Stump et al 2018) and one was part of a youth shelter (Bruster & Foreman, 2012). All programs cited the goal of mentoring to be providing a stable, consistent role model who could help children learn positive coping skills as well as provide children with opportunities for entertaining and educational activities. Inclusion criteria for mentoring programs were broad and included children as young as 4 years to as old as 16 years.

Qualitative results on the effectiveness of mentoring programs were positive. Children, at-home caregivers, and mentors indicated in interviews that they felt that mentoring programs had improved youth behavior, academic skills, and self-concept (Laakso & Nygaard, 2012; Shlafer et al., 2009). One child said, “I would think probably I changed a lot. Probably from my behavior. It’s better” (p. 22, Laakso & Nygaard, 2012) and a caregiver noted, “I’ve noticed an attitude change for the better” (p. 515, Shlafer et al., 2009). Similarly, while the Bruster and Foreman (2012) study was primarily descriptive and offered limited data on the effects of mentoring, their surveys found that the majority of children indicated that their mentor helped them feel good about themselves and helped them succeed in school.

Quantitative results were more mixed. Shlafer and Poehlmann (2010) focused primarily on children’s relationships with their incarcerated parents and caregivers unrelated to mentoring, but included data on behavior problems before and after six months of mentoring. The sample was too small to conduct statistical testing, due to the high dropout rate of one third of the participants, however, descriptively, authors found that the percentage of children exhibiting borderline or clinically significant externalizing problems remained consistent throughout mentoring and the percentage with internalizing problems increased. Another analysis of these data found that more frequent contact with mentors, reported at a single time point, was associated with fewer internalizing and externalizing symptoms at that time, though longitudinal pre-post testing did not find any significant change in problem behaviors (Shlafer et al., 2009). Stump et al (2018) conducted secondary analyses on mentoring program enhancements such as having specific goals for mentees, and found a significant increase in children’s educational expectations.

Given the finding about frequency of contact with mentors, and outlining specific goals for mentees, it is possible that various moderators impact the effectiveness of mentoring, such as amount of time spent with and intentionality of mentors, that may have been masked by the conglomeration of data. Additionally, while qualitative data on mentoring is overwhelmingly positive, further quantitative research is needed. Specifically, studies with larger sample sizes should investigate the effects of mentoring on problem behaviors, but also, qualitative data suggest additional targets – specifically, academic performance, positive self-concept and confidence.

Group Interventions

Two reports examined the effects of therapeutic group interventions (Sagar, 2011; Springer et al., 2000), highlighting that group interventions can work to reduce stigma and expand social support networks to include other children who have also experienced the incarceration of a parent. Groups differed in theoretical underpinnings. Springer and colleagues’ (2000) school-based group with fourth and fifth grade students was described as strengths-based, goal-focused, and interactional, using the interactions as a microcosm for broader difficulties children experienced and an opportunity to practice prosocial interaction. Sagar’s (2011) clinic-based groups with 6-to-12-year-olds focused on bibliotherapy with the Bible, art therapy, and cognitive-behavioral exercises. Results of these two studies were not promising. Springer et al.’s (2000) students improved from pre to posttest on a measure of self-esteem, but there was no significant difference between the wait-list comparison and experimental group, suggesting self-esteem growth may have been normative and unrelated to the intervention. Sagar (2011) found no pre to posttest changes on caregivers’ reports of children’s behavior problems or children’s reports of their own depression, anxiety, or ability to forgive.

Importantly, we cannot extrapolate about the effectiveness of all therapeutic group interventions from the results of these two small studies. First, both included only ten participants each, limiting generalizability and power. Second, though Springer et al. (2000) included thick description of each session, neither included measures of adherence. Third, both studies have specific and perhaps inappropriately selected outcomes measures. Sager (2011) explained that all of their outcome measures fell in the average range before intervention. We would not expect the intervention to significantly improve depression, anxiety, or other outcomes that are already in a non-clinical, healthy range. Springer et al. (2000) exclusively measured self-esteem, however in another paper that the same authors published describing lessons they learned from developing this intervention (Springer et al., 1999), they noted that the goals that individual children set often had less to do with self-esteem or their feelings about their incarcerated parent, and more to do with behavior and relationships. Children wanted to improve classroom behavior, get in fewer fights at school, or develop better relationships with their siblings and peers, suggesting these may be promising targets for future group interventions for CIP. Similarly, both programs cited increases in social support and feelings of solidarity as goals of group intervention work, however neither measured these outcomes.

More empirical research, with outcomes aligned to therapeutic goals and theoretical frameworks, is needed before any broad claims about the effectiveness of therapeutic group interventions can be made. Youth coping and social support are identified as key moderators in the FSPP model. However, the 6 articles including youth coping and social support did not find empirical support for their moderating role in stress.

Multiple Target Areas

Five papers examined interventions that integrated multiple target areas (Block & Potthast, 1998; Goepferich, 2011; Miller et al., 2013; Phillips & O’Brien, 2012; Poehlmann er al., 2021). Specifically, all of these programs included interventions for incarcerated parents or at-home caregivers and CIP, as well as opportunities for group work with youth and adults together. Two qualitative studies evaluated the Girl Scouts Beyond Bars (GSBB) program, providing extended visits in the form of a Girl Scout troop meeting for mothers and their 7–17-year-old daughters at the prison, as well as separate mother-only meetings to focus on planning and parenting skills, and daughter-only meetings outside of the prison to focus on youth development and social support (Block & Potthast, 1998; Goepferich, 2011). One paper examined the Strengthening Families Program (SFP), developed for families with substance-abusing parents and adapted for families with incarcerated parents (Miller et al., 2013). Through SFP, at-home caregivers and 4–14-year-old CIP attended separate group sessions and merged family sessions. This allowed for youth and their caregivers to work on family communication skills and individual problem-solving skills. Another paper (Phillips & O’Brien, 2012) evaluated an individualized service program where a family therapist worked as a case manager and service provider, referring families with 4–9-year-old children to the appropriate services or providing interventions for children, parents, at-home caregivers, or full family units as needed. Poehlmann et al (2021) evaluated the efficacy of the Sesame Street educational group in a randomized control trial with 3–8-year-old CIP and their at-home caregivers. Caregivers and children in the educational group received developmentally appropriate information about the father’s incarceration.

Qualitative reports from incarcerated parents and at-home caregivers indicated support for integrated programs. Regarding GSBB, at-home caregivers highlighted that the program improved communication between mothers and daughters and had a number of positive outcomes for daughters’ behaviors and emotionality including decreased fighting and talking back to teachers, improved academic grades, increased self-esteem, and less frequent feelings of sadness, anger, and worry (Block & Potthast, 1998). Incarcerated mothers noted the importance of social support from other girls in GSBB. At-home caregivers in the family therapist referral program also reported that participation improved their own ability to provide structure and emotional support and had helped children identify feelings and concerns about their parents’ absences and communicate about these feelings (Phillips & O’Brien, 2012).

Miller and colleagues (2013) and Poehlmann et al (2021) included quantitative information about the impact of their programs on youth. In Miller and colleagues’ (2013) SFP, they found greater effects on at-home caregiver parenting outcomes than child outcomes, but nonetheless reported decreases in youth overt aggression and criminal behavior from pretest to 4-month follow up. Additionally, family strength and family organization and cohesion increased while family conflict decreased. No changes were seen in children’s internalizing symptoms, concentration, or family involvement. In Poehlmann et al (2021)’s Sesame Street educational group, children’s reactions to visits with their parent in jail were measured by the Jail-Prison Observation Checklist (JPOC). The JPOC is an observational rating scale used by trained researchers to observe children during visits. Children in the intervention group who were told developmentally appropriate explanations about their incarcerated parent were observed to have more positive affect during visits. Caregivers in the educational group reported a positive change in how they talked to their children about incarceration. Although not studied directly, interventions to help demystify parental incarceration through educational materials may lessen the impacts of ambiguous loss associated with parental separation due to incarceration (Bokneck et al 2009; Poehlman et al 2021). More research is needed to directly assess how educational materials could buffer experiences of ambiguous loss for children.

While available research supports all of the integrated interventions, programs like GSBB and Phillips and O’Brien’s (2012) family therapist case manager program that have relied exclusively on qualitative data would benefit from quantitative follow-up to empirically verify parent and caregiver reports of outcomes. Confirming measurable improvements to parent-child communication as well as child behavior, emotionality, academic success, and social support would add power to their conclusions of effectiveness.

Developmental Considerations

Overall, the body of literature on interventions for CIP would benefit from more explicit developmental framing. Taking a developmental, child-centered approach, it is important to look explicitly at the age groups served by these interventions and how interventions were adapted to children’s ages. Studies with children from birth to 17 years-old were included in this review and most papers included very wide age ranges of child participants. The exception to this was studies of interventions specifically targeting parents of infants and toddlers, which included age-specific programming to support this group (Barr et al., 2014; Barr et al., 2011; Byrne et al., 2010; Cassidy et al., 2010; Condon, 2017; Goshin et al., 2014). These studies had the smallest ranges of ages and focused on parenting behaviors specific to parenting an infant or toddler. A few studies targeted primarily preschool and elementary-aged children (Harris & Landreth, 1998; Landreth & Lobaugh, 1998; Springer et al., 2000; Phillips & O’Brien, 2012), however, this still represented a large age range from three to eleven years old. While Phillips and O’Brien (2012) provided one-on-one services, which were able to be adjusted based on child age, and Springer and colleagues (2000) included only fourth and fifth grade students, Harris and Landreth (1998) and Landreth and Lobaugh (1998) were limited in their ability to provide developmentally targeted parenting interventions to groups with such wide age ranges.

Most other programs were available to a very large age range of youth. Some, like mentoring programs, that provided one-on-one services were able to differentiate treatment, but others seemed to be providing a one-size-fits-all treatment to an overly wide age range of children (e.g., Sagar, 2012). Even for programs differentiating treatment, large age ranges made it difficult to appropriately evaluate effects. Some included measures that could only be used with a portion of their participants, which further limited sample size and findings (e.g. Shlafer et al., 2009; Shlafer & Poehlmann, 2010). Further, age-specific differences may be masked by large age ranges and studies that use age as a control variable rather than a variable of interest. Most notably, zero studies purposely targeted teens, except to include them as teen parents (Barr et al., 2014; Barr et al., 2011), leaving a gap in specific services for adolescents who may be experiencing parental incarceration differently than younger children. Finally, as indicated in Tables 1 and 3, three studies failed to describe the ages of youth participants entirely, which suggests a concerning lack of regard for the ways developmental differences may impact effectiveness (Block & Potthast, 1998; Gat, 2000; Goepferich, 2011).

Qualitatively, only studies of mentoring programs included children’s voices (Bruster & Foreman, 2012; Laakso & Nygaard, 2012; Shlafer & Poehlmann, 2010; Shlafer et al., 2009). All other qualitative studies relied on parent or caregiver interviews. Youth are experts on their own experiences and provide subjective information unavailable through communicating exclusively with adults (Grover, 2004). Including youth voices will increase the breadth and depth of understanding about interventions for CIP and how they are perceived by youth themselves.

Sociodemographic and Contextual Considerations

Arditti’s (2016) model highlights the ways contextual variables as well as social positioning variables like race, gender, and socioeconomic status impact families’ experiences of incarceration and its sequelae. Racism, sexism, and capitalism as well as other institutional and governmental barriers can increase the likelihood of incarceration and exacerbate the negative impacts of incarceration on family systems. These systems of oppression are also directly associated with many of the outcomes of interest in the reviewed studies; for example, experiences of racism and poverty are associated with worse mental health outcomes including depression, anxiety, behavior problems, and reduced self-esteem and self-worth (Priest et al., 2013; Yoshikawa et al., 2012). Further, contextual variables relating to institutions themselves, like prisons and jails, as well as schools and clinics serving CIPs, control the services they are able to provide. When working to support children impacted by parental incarceration, it is necessary to consider sociodemographic and contextual factors. Below, we discuss how the papers included for review incorporated considerations about gender, race and ethnicity, socioeconomic status, and institutional settings into interventions.

Gender

Gender of both parents and youth, but particularly parents, impacted what services were available and how services were conceptualized. Due to gendered separation of carceral facilities, all programs that took place in a prison or jail were exclusively for incarcerated mothers or fathers, not both. Conversely, interventions focused on youth coping that took place in schools and clinics did not differentiate by gender of incarcerated parents.

All interventions that focused on a single-gender parent highlighted the differences between maternal and paternal incarceration, including differences in location and caregiving expectations. Programs for incarcerated mothers (e.g. Schubert et al., 2016) noted that because there are fewer women’s prisons, families tend to have to travel far distances to visit, which can impede the ability of families to participate in interventions that include visits. For this reason, some programs included free transportation (Block & Potthast, 1998; Snyder et al., 2002) and one program allowed overnight visits (Gat, 2000). Additionally, programs for mothers (e.g. Goepferich, 2011) emphasized that societal expectations for mothers to be primary caregivers meant that many mothers would be immediately returning to the role of primary caregiver after release. Programs cited this as an important reason for prioritizing parenting education and support programs for mothers (e.g. Cassidy et al., 2010; Harris & Landreth, 1997). Conversely, programs for fathers often highlighted that the lack of societal expectation for paternal caregiving meant that there was a gap in programming for incarcerated fathers (e.g. Barr et al., 2011; Harrison, 1997). These programs sought to reverse sexist ideas that fathers are uninvolved in children’s development by promoting father-child relationships.

Regarding child gender, single-gendered parenting programs were open to parents of children of all genders, with the exception of Girl Scouts Beyond Bars, which was a program exclusively for mothers and daughters. Interestingly, while participation of all children was allowed in other parenting programs, many did not collect data on child gender (Byrne et al., 2010; Cassidy et al., 2010; Condon, 2017; Gat, 2000, Harrison, 1997). Again, as with child age, this demonstrates a disregard for the ways youth gender may impact interventions for CIP.

Unlike most parenting programs, mentoring programs often centered child gender identity and tried to match children with same-gender mentors. Laakso and Nygaard (2012) noted that gender matches were aligned with most families’ gendered requests for “big sister[s]” for daughters and “male role model[s]” for sons (p. 18), although Bruster and Foreman (2012) highlighted that finding male mentors had been difficult. Zero articles discussed how transgender parents or children might be supported by programs for CIP. While many programs theoretically could be utilized by trans parents or youth, the lack of acknowledgement of trans people in families impacted by parental incarceration perpetuates the erasure of trans experiences and suggests that programs are not considering how trans individuals could access these resources or how discrimination against trans people impacts the experiences of familial incarceration.

Findings from the current review suggest that many programs for CIP have considered how gender may impact the needs and experiences of incarcerated parents and their children. Some programs have been developed for exclusively single-gender groups or matched same-gendered mentors to help address gender-related needs. However, broader gender-related systemic issues like locations of women’s prisons, societal norms regarding caregiving roles, and discrimination against trans people in the carceral system have not been fully addressed.

Racial and Ethnic Inequality

Despite the fact that structural racism in the United States has caused the disproportional impact of incarceration on families of color and particularly Black families, discussion of this contextual information or how interventions for CIP could best serve parents and children of color was remarkably absent from the literature. Only two papers included explicit reference to disproportionate incarceration rates by race (Bruster & Foreman, 2012; Goepferich, 2011) and only Goepferich (2011) noted that these disproportionalities are rooted in legislation targeting communities of color. Lack of acknowledgement of the ways that racism has led to the disproportionate incarceration of Black parents and other parents of color suggests that, from conceptualization, the interventions included in this review were not designed with a critical consciousness towards race and racism at a systems level.

Although lacking a systems-level analysis, two papers did actively incorporate racial and ethnic identity at the individual level into the conceptualization of their interventions. Because their sample was predominantly African American, Miller and colleagues (2013) chose an intervention that had previously been successful for African American parents and the authors noted that cultural considerations were made when choosing food and prizes for their participants. Springer and colleagues (2000) emphasized how they considered family relationships, cultural centrality, and communication styles of Hispanic children when considering how to lead the group intervention. While these two programs demonstrate an acknowledgement of the role of culture in intervention effectiveness and satisfaction, neither expands these cultural considerations beyond the bounds of the intervention content. Cultural considerations at the individual and content level are an important step, however, programs that acknowledge race and ethnicity without acknowledging systems of racism and ethnocentrism in the United States risk perpetuating these systems of oppression or failing to fully serve the needs of families of color.

While not explicitly discussing the role of race in their interventions, multiple mentoring programs shared information about mentor and child racial identity, which highlighted the racial mismatch between participants, who were predominantly children of color, and mentors, who were predominantly White (Laakso & Nygaard, 2012; Shlafer et al., 2009). Neither paper discussed how this might impact mentor/mentee relationships. Similarly, some authors used race and ethnicity as a control variable (Goshin et al., 2014; Barr et al., 2014), but did not discuss it further. Three papers did not identify the race or ethnicity of their participants (Gat, 2000; Harrison, 1997; Phillips & O’Brien) and two conglomerated all non-White identities and exclusively reported the percentage of “ethnic minorities” (Byrne et al., 2010; Shlafer & Poehlmann, 2010), further perpetuating the erasure or dismissal of the role of racial and ethnic identity on intervention effectiveness through colorblindness.

Socioeconomic status

Socioeconomic status (SES) and economic marginalization impact access to and resources available for interventions for CIP. Data on SES of participants, including reference to either pre-incarceration income or educational attainment, was missing in half of the papers included for review (denoted with “--” on Tables 1–3). Again, failing to report social positioning variables limits not only the interpretation of findings, but also the extent to which researchers and interventionists can adequately understand the contextual experiences of incarcerated parents and their children.

Authors who did include SES information noted myriad ways that SES impacted families’ abilities to participate in interventions. Lack of access to transportation kept some families from participating (Harrison, 1997), although one program offered in-home services to counteract this barrier (Phillips & O’Brien, 2012) and some provided transportation or child care (Block & Potthast, 1998; Miller et al., 2013; Snyder et al., 2002). Frequent residential transience and changes in phone numbers made it difficult for mentors to contact families (Bruster & Foreman, 2012; Shlafer et al., 2009), and limited access to education acted as a barrier for programs that required written applications (Condon, 2017; Snyder et al., 2002).

It is well-documented that income and wealth disparities are driven by structural racism and sexism that interact with capitalism to cause the disproportionate economic marginalization of families of color and female-headed households (Claude, 1986; Wilson, 2018; Wolff, 2016). Thus, calls for increased attention to gender and race, as well as systemic issues of sexism and racism, would be incomplete without an additional call to acknowledge the intersections of racism, sexism, and economic marginalization. A focus on these intersecting systems of oppression within and surrounding the system of incarceration will appropriately contextualize the experiences of families involved in the criminal justice system and correctly situate the origins of “problems” in systems of oppression, not individuals and families. This will allow interventions for CIP to counteract intersecting racist, sexist, and economic barriers to participation and wellbeing and effectively link pre-, during, and post-incarceration supports.

Institutional Settings for Interventions

The institutional settings of interventions and their related environmental, financial, and legal contexts impacted whether and how interventions happened. Many programs took place in prisons or jails, and authors highlighted that these settings are often loud, frightening, and uninviting for children (e.g. Barr., et al., 2011; Condon, 2017; Schubert et al., 2016); therefore, their programs aimed to create more child-friendly environments for visits. However, prisons and jails are not equivalent carceral facilities, with individuals tending to spend less time in jails than prisons. Thus, programs associated with jails needed to shorten the length of interventions (Harris & Landreth, 1997) or find ways to continue contact with families after release (Cassidy et al., 2010; Miller et al, 2013). Harris and Landreth (1997) also noted that visits are often not allowed in jails, and they needed to secure special permission for mothers to have contact visits with their children.

Programs for children and at-home caregivers outside of carceral facilities took place in schools (e.g. Springer et al., 2000), individual’s homes, (e.g., Phillips & O’Brien, 2012), or community settings (e.g., Miller et al., 2013). Springer and colleagues (2000) and Miller and colleagues (2013) noted that their relationships with school personnel and community sites facilitated logistics of programs for CIP. Further, they underscored the importance of collaborating with trusted community partners. For example, Miller and colleagues (2013)’s SFP program took place at Motherly Intercession, a community non-profit devoted to helping incarcerated and formerly incarcerated mothers, and suggested that Motherly Intercession’s preexisting relationships with families and standing in the community promoted trust, attendance and participation. However, other community partnerships faced legal barriers, with Phillips and O’Brien (2012) highlighting that partner schools were unable to assist in referring participants due to concerns around confidentiality.

Similarly, legal barriers related to custody and parental rights barred some families from participation. Both authors (e.g. Snyder et al., 2002) and interviewed mothers (Goepferich, 2011) noted that many state policies legally terminate parental rights for incarcerated parents. Often parents hoped to participate in extended visiting programs or parent-child interventions, but were unable to because they could not consent to their child’s participation. While children’s wellbeing and safety should always be the priority, many of these laws terminate parental rights solely due to incarceration without examining individual circumstances (Hagar & Flagg, 2018). Termination of parental rights due to incarceration is a larger political issue, but psychologists have a role to play in accounting for child wellbeing when determining whether this is the best course of action and advocating against the process should it not be in the child and family’s best interest. Some programs, like Cassidy et al.’s (2010) jail diversion program, provided an option for mothers that increased their likelihood of retaining parental rights. Other interventionists could consider how they may amend their programming to include parental rights advocacy.

Often interrelated with legal barriers, many programs faced funding barriers. In particular, residential parenting programs experienced frequently changing legislation that impacted allocated funds and even whether the programs were allowed to continue to exist (Byrne et al., 2010; Cassidy et al., 2010; Goshin et al., 2014). Similarly, outside providers faced funding difficulties due to the fact that CIP may not meet diagnostic criteria for a specific mental health disorder, limiting insurance-based funding.

Together, incarceration-related environmental, legal, and financial barriers interacted to impede interventions for CIP. This suggests that a multi-pronged planning approach that accounts for these overlapping contextual variables and allows time for developing needed community partnerships is necessary when planning interventions to support CIP. Further, interventionists and researchers must stay apprised of current legal policy and funding decisions in order to effectively plan and execute interventions.

Discussion

The current review drew upon Arditti’s (2016) Family Stress-Proximal Process model to critically examine the available literature on interventions to promote the wellbeing and adjustment of children with incarcerated parents. While parental incarceration has been recognized as a unique stressor and a potentially traumatic event for children and adolescents (Arditti, 2012; Felitti et al., 1998), limited research has examined the literature on interventions for CIP through a lens that prioritizes child outcomes and the broader contextual systems that impact familial experiences of incarceration. Twenty-four papers met inclusion criteria and were based in the United States, studied interventions targeting CIP, their parents or caregivers, and included child outcomes. A critical review of these papers, grounded in the FSPP framework, revealed three main recommendations for future research: 1) increased support for at-home caregivers, 2) more developmentally-appropriate services, particularly for adolescents, and 3) explicit acknowledgement and incorporation of identity variables and systems influences.

Increased Support for At-Home Caregivers

Systematic examination of the targets of intervention revealed that programs focused on contact with incarcerated parents have been prioritized. These findings align with the fact that multiple previous reviews have focused exclusively on parenting interventions without examining other targets (Armstrong et al. 2018; Goshin & Byrne, 2009). While many parenting and visit-based programs have demonstrated success, it is important to think holistically about the multiple influences in the lives of CIP. In particular, at-home caregivers spend the most time with CIP, and supportive parenting from these caregivers has been shown to promote positive outcomes (Sukhramani et al., 2018). At-home caregiver parenting is noted in the Arditti (2016) FSPP model as one of the primary proximal processes impacting child adjustment during parental incarceration. However, only two reviewed studies, both part of integrated intervention models, provided services for at-home caregivers. Promoting parenting skills and reducing stress for at-home caregivers are important intervention goals for those seeking to support CIP and more research is needed on how interventions can best support these caregivers to promote positive child adjustment.

Development and Evaluation of Developmentally Appropriate Services for Children

While an ecological frame that accounts for the impacts of incarcerated parents and at-home caregivers is important for holistic consideration of the impacts of parental incarceration on children, it is equally important to provide direct intervention to support children themselves. This review found that only four mentoring programs and two therapeutic group interventions had been empirically studied to examine child outcomes. However, mental health practitioners know that many CIP are receiving direct services. The National Child Traumatic Stress Network (NCTSN) suggests Bounce Back and the Trauma and Grief Component Therapy for Adolescents (TGCT-A) as potential group interventions for children and adolescents, respectively, who have experienced separation from a parent due to incarceration (NCTSN, n.d.). Evaluative studies examining the use of these general trauma and grief interventions for CIP is necessary to determine effectiveness with this unique population. Similarly, some case studies have examined the use of individual therapies like child-centered play therapy (Brown & Gibbons, 2018) and trauma-focused cognitive behavioral therapy (TF-CBT; Morgan-Mullane, 2017) with CIP, however they were not included in the current review due to their descriptive nature. Future empirical research on direct individual and group therapies for CIP will help better inform practitioners looking to provide therapeutic support for these children and their families.

Additionally, it is important to acknowledge that parental incarceration can impact children and adolescents differently at different stages in development. Life course theory (Elder Jr, 1996) highlights that significant events can have differing impacts depending on when in the course of life they fall. With this in mind, it is clear that interventions for CIP must explicitly consider the developmental needs of children of different ages, including choosing developmentally appropriate measures for the assessment of program efficacy among younger children. Overall, as child participants aged, program models demonstrated less developmental specificity and, of note, no interventions specifically targeted adolescents. Adolescence is a unique period of development where children demonstrate both developmental strengths, but also specific needs and sensitivities, particularly regarding identity development (Meeus, 2011) and social interaction (Blakemore, 2008). Group interventions, like TGCT-A cited above (NCTSN, n.d.) may be particularly salient for this age group as peer relationships become priorities. Overall, the body of literature on interventions for CIP could benefit from examination of developmental differences and how best to target intervention based on developmental needs and strengths.

Explicit Incorporation of Sociodemographic and Systems-Level Influences

In addition to acknowledging multiple aspects of family systems and developmental differences, it is necessary for research on interventions for CIP to utilize an intersectional frame analyzing the role identities and social systems play in families’ experiences with the United States’ carceral system. Colorblind approaches, and identity-blind approaches in general, impede the recognition of individual and group strengths or protective factors and situate “problems” within individuals or families rather than systems of oppression (Neville et al., 2016; Heberle et al., 2020). While a few papers in the current review incorporated aspects of racial and ethnic culture into their intervention work, this is the bare minimum. On an individual level, acknowledgement of racial and ethnic identity variables will allow interventionists to prioritize incorporating racial-ethnic protective factors and cultural asset-based frameworks into their programming, which can capitalize on cultural strengths (Gaylord-Harden et al., 2012; Jones & Neblett, 2016).

Further, intersectional theory urges researchers, interventionists, and politicians to attend to the ways in which people’s identities intersect and position them within systems of oppression and power (Crenshaw, 1989; Overstreet et al., 2020). Only a few papers in the current review cited how the intersecting impacts of systems like sexism and capitalism (Snyder et al., 2002) as well as racism (Goepferich, 2011) catalyze and exacerbate the impacts of incarceration and promote social inequality. To address the impact of systems level racism, researchers may include measures of racism-based stress and experiences of racism within the carceral system to appropriately contextualize families’ experiences and examine moderators of interventions’ success (Carter & Pieterse, 2020). Additionally, interventionists should seek to develop programs that directly target the harm caused by racist systems, as well as the intersections of racist systems and other systems of oppression (Bryant-Davis, 2007), to legitimize the experiences of many families in the carceral system. Direct acknowledgement of both participant identity variables and related systems of oppression (e.g. SES and economic marginalization, gender and sexism or anti-trans oppression) throughout background information, methodological choices, and discussions of findings, as well as the integration of these concepts into the development of intervention programs themselves, is necessary for the creation of programs and research that appropriately incorporate systems-level considerations. Future research cannot be blind to the intersecting effects of these systems and must consider how interventions can actively counteract discrimination and promote antiracist and equitable policies (Graham & Harris, 2012).

Limitations

The review was conceptualized from a developmental psychology discipline, informing the use of PsycINFO as the search engine of choice. It is possible that this choice meant that interventions from other disciplines, like criminology, were not captured in the systematic search. Further, the inclusion criteria of this review may have excluded interventions that take a more systems-level approach to directly challenge the oppressive system of incarceration. Programs that help families avoid incarceration, like drug court interventions (Gifford et al., 2016; Logan et al., 2001) were not examined here, but can be utilized as models of decarceration to change the institution of mass incarceration itself. Inclusion criteria purposefully limited reviewed studies to those occurring in the United States, chosen given its egregiously disproportionate use of incarceration as punishment compared to other nations. However, parental incarceration occurs throughout the world, and interventions for CIP in different countries may target different aspects of families’ experiences. Finally, inclusion criteria around outcome measures excluded case studies and descriptions of interventions. Many of these excluded papers (e.g. Lopez & Burt, 2013; Morgan-Mullane, 2017; Quintanilla, 2010) presented interventions based on 1-on-1 or group therapy models, which were conspicuously minimal in the present review. Future researchers may seek to broaden the scope of the current review to include additional search engines, countries, and types of studies. However, the specificity of the current review allows for distinct, empirically-supported, actionable suggestions for social scientists in the United States seeking to support CIP and their families. Further, our recommendations align with similar calls from the Administration for Children and Families (McCormick et al., 2021) to address gaps in services that attend to at-home caregivers, developmental level, and intersectional identity and systems factors, which adds powers to the validity of the current findings. Aside from the specific calls for improvement noted in this review – attention to home-based, developmental and social context – the overall quality and quantity of research outputs that focus on actual outcomes on CIP remains so varied in approach, and so mixed in quality, that it suggests a research field still early in its development. As rigorous, well-conducted empirical work in this field continues to accumulate, systematic reviews that include analyses of biases in the existing literature, as well as meta-analytic approaches, can continue to guide the field forward to build an evidence base of support for CIP and their families – but first, we need more high quality research.

Conclusion

In conclusion, results from this critical review support a research agenda that prioritizes a developmental ecological frame and acknowledges intersectional identity and intersecting systems of oppression and power. Specifically, future research on interventions for CIP should 1) examine supports for at-home caregivers, 2) develop and evaluate developmentally appropriate direct services, particularly for adolescents, and 3) explicitly acknowledge and incorporate identity variables and systems-level influences, like racism and capitalism, that compound the stress of parental incarceration. It is necessary to think about context – family context, life course context, and systemic context – in order to adequately address the needs of CIP and promote wellbeing.

Highlights.

• Critically reviews the literature on interventions to support children with incarcerated parents.

• Twenty-four studies reporting on child outcomes in the United States were reviewed.

• Most research has focused on promoting parenting skills and improving visits.

• Little research on programs supporting at-home caregivers.

• Services are typically not developmentally specified and do not acknowledge systems of inequality.