A Systematic Review of Economic and Concrete Support to Prevent Child Maltreatment

Abstract

The association between poverty and maltreatment is well established, and numerous studies show the detrimental effects of economic and material hardship on maltreatment, child welfare involvement, and family well-being. In turn, an increasing body of literature suggests the value of programs and services that provide economic and concrete support in stabilizing and supporting families and, ultimately, reducing occurrences of maltreatment. In this article, we examine the effectiveness of economic and concrete support in reducing primary outcomes of child maltreatment and child welfare involvement and in improving secondary outcomes of child and adult well-being through a systematic review of the literature. We review studies published between 1990 and 2023 through searches of scholarly databases, child welfare websites, and evidence-based clearinghouses. We used a two-stage review process to identify a final sample of articles for data extraction. Twenty-four studies with low risk of bias were included in the synthesis. Regardless of delivery mechanism, we found consistent, favorable evidence of the impact of economic and concrete support as a prevention strategy to reduce child maltreatment and child welfare entry and to a lesser extent, improvements in child and adult well-being across these studies. Implications for policy, practice, and racial equity are discussed.

1. Introduction

All families must have access to resources to meet their basic needs, including food, shelter, and health care, as well as services and support to address more complex needs [1]. Such support is particularly critical during times of economic instability. Yet, for families that face economic and material hardship and instability (for example, having insufficient resources to meet basic needs, experiencing economic shocks, and experiencing housing instability), this support is not adequately or equitably addressed by a siloed social service system. Research shows geography, immigration status, family structure, and race and ethnicity are all related to inequitable access to public benefit programs [2]. In turn, families in financial distress are too often reported to child protective services when services are unavailable or insufficient to alleviate economic and material hardships before child welfare intervention becomes necessary. Importantly, families themselves are calling for policy and practice that recognize the importance of economic and concrete support before challenges become crises that put them at risk of child welfare system involvement. The recent Children’s Trust Fund Alliance Birth Parent Network and Casey Family Programs Parent Advisory Committee together created a framework for Building a 21st Century Community-based Approach to Strengthening Families that clarifies, from the perspective of parents with lived expertise, the need at both the macro policy and family service levels for economic and concrete support [3].

How best to address economic and material hardship may depend on unique family circumstances, needs, and strengths. As such, economic and concrete support services may be particularly effective when implemented by a community-based organization, a state or local public agency, and/or agency partners, with intentional flexibility in the type, amount, frequency, and disbursement mechanism of the support and in how families use the funded service [4]. In this article, we describe findings from a systematic review of evidence on the effectiveness of economic and concrete support services, delivered with flexibility across multiple delivery mechanisms and either independently or within broader programs, as a prevention strategy for the reduction of child maltreatment and related child welfare and prevention-focused well-being outcomes.

More than 50% of all Americans and 84% of all Black Americans are estimated to experience poverty by age 65 [5]. Moreover, nearly half of American families with young children will experience poverty before their child enters kindergarten [6]. Moreover, economic insecurity is dynamic, as families frequently move in and out of poverty over time [5]. Sudden life events such as death, illness, or job loss, as well as broader economic downturns or global pandemics, often precipitate periods of economic insecurity. Research indicates that nearly half of those who fall into poverty manage to escape it within a year, but more than half of those who previously escaped poverty will return to it within five years [7]. Furthermore, many low-income families that are above the federal poverty line encounter significant material hardships such as challenges in paying for housing, utilities, food, or medical care. Over 60% of these families report experiencing a financial shock in the previous year [8].

Economic insecurity and material hardship, whether chronic or intermittent, not only detrimentally affect health and well-being [9] but also contribute to referrals to child protective services (CPS) and involvement with the child welfare system [10,11]. The connection between poverty and child welfare involvement is firmly established [12], with recent evidence indicating a heightened county-level correlation between child poverty rates and CPS reporting in recent years [13]. Extensive research underscores that economic insecurity, including factors such as reductions in Temporary Assistance for Needy Families (TANF) benefits [14] and loss of income from cash assistance [15] as well as material hardship like inadequate childcare [16,17] and insecure housing [18,19], significantly increase the risk of involvement with child protective services. Income volatility further increases the likelihood of CPS investigations among low-income families, irrespective of their overall income levels [20], and families experiencing multiple material hardships are four times more likely to undergo CPS investigations and seven times more likely to face physical abuse investigations [11].

Such research highlights the importance of addressing families’ economic and material needs to prevent child welfare system involvement. For example, a growing body of research suggests that economic support programs for working parents, including Temporary Aid to Needy Families (TANF), the Supplemental Nutrition Assistance Program (SNAP), child care subsidies, and the Earned Income Tax Credit (EITC), are associated with reductions in child maltreatment and/or child welfare involvement [21,22,23,24,25,26,27,28]. In addition to studies of individual economic support programs, a recent scoping review examined the collective literature on these four economic support programs and found evidence of preventive benefits against child maltreatment [29]. Similarly, a recent analysis simulating the effects of increased household income under three anti-poverty economic support program packages found potential reductions of 11 to 20% in CPS investigations annually [30]. Furthermore, the literature shows that addressing economic and material hardships through cash transfers, including those without work requirements, can prevent maltreatment. For example, a study of the Alaska Permanent Fund Dividend (PFD), which provides unconditional cash payments to all residents, estimates that an additional USD 1000 unconditional cash payment to families in the early months of a child’s life reduces the likelihood of referrals by age 3 for child neglect by 10 percent and physical abuse by 30 percent [31].

In addition to economic support, the provision of concrete support may also be an effective mechanism to reduce child maltreatment and child welfare system involvement. For example, the long-term benefits of high-quality child care on participating children’s safety and well-being—including decreased risk of maltreatment; increased educational attainment, income, and employment; reduced delinquent behavior; and improved social and emotional functioning—are well documented [32,33,34]. Concrete support provided to families via child welfare services is also effective. Families with screened-in maltreatment reports that were randomly assigned to a differential response track and received more concrete support experienced a lower risk of subsequent screened-in maltreatment reports and lower risk of subsequent child removals (over a period of 8 to 9 years) than families that were randomly assigned to the traditional track and received less concrete support [35]. Further, the authors found a significant relationship between income and receipt of materials services: among low-income families receiving differential response, receiving concrete supports specifically was associated with a lower likelihood of experiencing a subsequent maltreatment report. Similarly, family preservation programs with concrete support are also associated with reductions in subsequent child maltreatment reports and foster care placement. A study of a home-based family preservation program found that when home visitors give discretionary funds (up to USD 600) to help families with concrete emergency insufficiencies—such as utility payments, food, clothing for children, or transportation assistance—these families are less likely than families receiving no concrete supports to experience a subsequent child maltreatment report [36]. Increasing concrete support provided by the home visitor was also associated with greater program engagement by families, an increase in client satisfaction with services, greater perceived cultural sensitivity of services, and increased attainment of service plan goals.

The body of research on economic and concrete support as a child maltreatment and child welfare involvement preventive strategy has been growing for over 45 years [37,38] with evidence existing across different study designs and across program or service delivery mechanisms (e.g., cash, housing, child care, healthcare, food assistance). In turn, this suggests that economic and concrete support is an independent and portable evidence-based service that can be delivered directly to families through a variety of means [39], including cash transfers, income and nutrition supports, child care subsidies, vouchers, or gift cards, all centered on the provision of cash or cash equivalent as the evidence-based service, as well as concrete support in the form of goods, services, or flexible funds to meet a specific need such as diapers, food, clothing, appliances or their repair, furniture, cars or their repair, housing or rental vouchers and assistance, utility assistance, and housecleaning. Economic and concrete support is a distinct service and strengths-based approach to keep children safe and help families thrive through addressing families’ basic needs, particularly during times of financial hardship and instability [4].

As noted above, the relationship between economic and concrete support services and child maltreatment or child welfare involvement has been examined across multiple studies, and findings have been extensively synthesized [37]. This synthesis includes studies of individual economic and concrete support services such as cash transfers, specific public economic support programs, or use of flexible funds to address material needs, as well as scoping reviews that consider a collection of public economic support programs [29]. Addressing economic and material hardship is also increasingly prioritized across the social service sector, within philanthropy, and among families. However, moving research findings into a comprehensive and actionable practice that can meet the varied and complex needs of families requires a methodologically rigorous review of the causal evidence around the range of economic and concrete support services available. To provide this, our study aims to build off an extensive narrative review of the literature [37,38] compiled by the authors and summarize findings from a more rigorous systematic review of the causal relationship between economic and concrete support, regardless of delivery mechanism, and outcomes targeted specifically toward the prevention of child maltreatment and child welfare involvement. As such, our systematic review was designed to answer the following research question: What is the causal evidence for the effectiveness of economic and concrete support on child maltreatment and child welfare outcomes, and secondly on child and adult well-being?

2. Materials and Methods

To guide our review, we pre-identified clear inclusion and exclusion criteria following a PICOTS (population, intervention, comparator, outcomes, timing, and setting) framework as outlined in

Table 1. PICOTS framework for eligibility criteria.

PICOTS Element	Eligibility Criteria
Population	Child/youth (Age 18 or under), family, caregiver
Intervention	Concrete supports; economic supports
Comparator	Control or services as usual as assigned through a randomized controlled trial (RCT) or a quasi-experimental design (QED) with either (a) a matched comparison group or (b) natural experimental conditions
Outcome(s)	Outcomes designated as target domains by the Title IV-E Prevention Services Clearinghouse, including child safety/maltreatment, child welfare involvement, child welfare priorities (permanency, reunification, stability, maltreatment recurrence), and child and adult well-being
Timing	Literature published in or after 1990
Setting	Within or beyond child welfare/system of care; in the U.S. or in non-U.S. countries in the OECD

. Specifically, our search was limited to studies in which the population comprised children/youth or families/caregivers with a child(ren) under age 18. Eligible interventions included any in which an economic and/or concrete support—broadly defined—was specifically examined. We elected to exclude conditional cash transfer programs due to the requirements placed on participants to receive the economic support. Effectiveness was established by comparing an intervention condition to a control or services as usual condition in which the intervention was not provided; we did not include active controls as a comparator. Related to our definition of comparator, we note that we predetermined study designs for which intervention and comparison groups were identified. Specifically, we included only studies that tested main effects of economic and concrete support using a randomized controlled trial (RCT) or a quasi-experimental design (QED) with either a matched comparison group or a natural experimental approach.

To be eligible, studies had to assess the impact of economic and concrete support on the primary outcomes of child maltreatment, child welfare involvement, or child welfare-focused priorities (permanency, reunification, stability, and maltreatment recurrence). We also included studies that assessed impact on secondary outcomes of child well-being and adult well-being, as outlined by the Title IV-E Prevention Services Clearinghouse (“Clearinghouse”). We intentionally did not include length of sustained effect as a criterion. The Clearinghouse assigns a rating to programs and services based on sustained effects of at least 6 months post-intervention. We chose not to require a sustained effect for two reasons: First, we anticipated that some of the economic and concrete support services available to families would not have a defined end; rather, they could be continuously received (e.g., Medicaid). Second, because the literature suggests that families experience economic instability and material hardship due to economic shocks, for our purposes, we viewed even short-term impacts of economic and concrete support as notable. We included studies published in or after 1990, that were conducted within or beyond child welfare/systems of care, and that were implemented in the United States or in countries included in the Organisation for Economic Co-operation and Development (OECD). As with other systematic reviews focused on economic or concrete services [40], we limited study settings to countries in the OECD as these countries reflect economic conditions similar to the United States. The findings are reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines [41].

We searched selected databases, including Academic Search Complete, ERIC, CINAHL via EBSCOhost, Sociological Abstracts, SCOPUS, Social Services Abstracts, PubMed, and Cochrane Library, as well as child welfare websites and evidence clearinghouses, including the California Evidence-Based Clearinghouse for Child Welfare (CEBC), the Child Welfare Information Gateway, and Campbell Collective, to retrieve peer-reviewed articles and relevant publications or reports based on a predetermined set of keywords, including: “concrete service” OR “concrete services” OR “concrete support” OR “concrete supports” OR “income support” OR “economic service” OR “economic services” OR “economic support” OR “food assistance” OR “housing voucher” OR “housing support” OR “tax credit” OR “rental assistance” OR “cash transfer” OR “economic intervention” OR “social welfare” OR “government benefits” OR “social safety net” OR “relief” OR “financial assistance” OR “stipend” OR “direct payment” OR “cash allowance” OR “cash grant” OR “vouchers” AND (“child welfare” OR “ well-being” OR “maltreatment” OR “abuse” OR “neglect” OR “child safety” OR “permanency” OR “stability”) AND (“evaluation” OR “trial” OR “impact” OR “outcome” OR “effectiveness” OR “efficacy” OR “RCT”). Additionally, we scanned reference lists from recent systematic reviews, where applicable, for relevant studies to augment the bibliographic search. An initial search was conducted in September 2021 and a second search, using the same search strategy, was conducted in August 2023 to update with more recent records.

We deduplicated the retrieved articles using reference management software (EndNote version 20.4.1) and entered all into a web-based tool called Rayyan for screening [42]. At least two reviewers screened each reference against inclusion/exclusion criteria. We conducted two levels of screening. At level 1, we screened the title and abstract of each reference retrieved from the literature search. Records that met inclusion criteria were promoted for second-level screening. At level 2, we reviewed the full text of each record. We resolved conflicts between reviewers at both level 1 and level 2 through a larger team discussion during which the team focused on interpretation of the eligibility criteria established by the study design. We recorded reasons for exclusion among ineligible references at both levels. Once the final sample of references was confirmed, two researchers extracted data and entered data into a spreadsheet. Information that we extracted for each reference included citation; economic or concrete support category and type; specific program, intervention, or service name; study design type and details; sample and sample size; information on baseline equivalence; data sources used; key exposure; intervention and comparison groups; outcome category and specific measure; and effect sizes.

Once data were extracted, we synthesized findings overall and by type of study design, delivery mechanism of the economic and concrete support service, and outcome category. In interpreting effectiveness of economic and concrete support services, we used conventional standards of statistical significance (p < 0.05) to indicate a positive or negative vs. null finding. Because some studies measured multiple outcomes, we synthesized findings for all eligible outcomes included in each study. In our synthesis, we did not include findings from subgroup analyses but rather focus on main effects of the economic and concrete support on the full population studied.

3. Results

3.1. Study Selection

Our search yielded 7553 records, of which we excluded 6912 during Level 1 screening. Most of the excluded records were considered ineligible because they did not include a program or service that qualified as an economic or concrete support (n = 5356). Of the remaining excluded records, common reasons for exclusion included the study not measuring the impact of the economic or concrete support, such as implementation studies; presenting a systematic, meta-analysis, literature review, or descriptive overview; and being conducted outside of the target setting or on a different population than was defined in our eligibility criteria. We promoted 641 records to Level 2 screening and read the full text of each to determine final eligibility.

With additional information available in the full text, we excluded a significant number of records at this level, mostly due to ineligible study design for assigning treatment and control conditions. Additionally, although we retrieved studies of home visiting and family preservation programs, many of which may include concrete support, we only included studies of such programs when the provision of an economic or concrete support was clearly detailed and tested. Similarly, while subsidized early childhood education (ECE) programs, such as Head Start, can be considered an economic and concrete support, we excluded studies of ECE programs in this review due to the potential for confounding bias in which the effects of an economic and concrete support element within a larger program on our outcomes of interest can be mixed with effects of other elements of the program, making it difficult to establish a clear causal link between economic and concrete support and outcomes [43].

Our result of level 1 and level 2 screening is an initial sample of 40 studies. Of the final studies, 16 had methodological issues that were significant enough to make interpretation of causal findings difficult. We identified four main risks of bias across these studies, including (1) baseline equivalence not established, (2) integrity to assignment not maintained, (3) insufficient sample size to detect hypothesized effects, and (4) significant sample attrition (generally more than 20%). We highlight these methodological considerations because they are critical to our definition of “comparator” and the associated methods to define and compare treatment conditions. For example, to be a valid comparator, the comparison group must be equivalent at baseline with the intervention group. Notably, these factors are considered in study design standards by the Clearinghouse. Although we cannot confirm that all studies included in our final sample would be rated strongly by the Clearinghouse, for our review, we followed key Clearinghouse design standards around type of study, use of an equivalent comparison group, and outcomes assessed. Given the high risk of bias in these 16 studies, we excluded them from our synthesis and analysis of findings and, rather, focused the presentation on economic and concrete support impact from the set of studies for which risk of bias was determined to be low (n = 24). Studies in the final sample for which findings were analyzed and synthesized in the next section as well as those excluded from analysis are presented in Appendix A and Appendix B, respectively. Screening and eligibility determination is summarized in Figure 1.

3.2. Study Characteristics of Final Analytic Sample

• Population. All studies included in this review focused on services provided to populations and/or samples of families with children.
• Intervention. A range of economic and concrete support services were represented in the final sample (see

  Table 2. Interventions examined in analytic sample studies.

  Economic and Concrete Support Service	Number of Studies	Study Design
  Income support (cash/cash equivalent:)	14
  Earned Income Tax Credit (EITC)	6	Natural experiments
  Child Tax Credit (CTC)	2	Natural experiments
  Direct cash transfers/guaranteed income	3	RCTs
  Child support income	1	RCT
  Child Development Accounts (CDAs)	2	RCTs
  Support for basic needs:	10
  Healthcare support (Medicaid)	2	Natural experiments
  Housing support	2	RCTs
  Food assistance	4	Mixed: 2 RCTs; 1 matched comparison, 1 natural experiment
  Other concrete support	2	Mixed: 1 RCT; 1 matched comparison
  Total	24	11 RCTs 11 natural experiments 2 QEDs w/matched comparison

  ). Just under two-thirds of the final analytic sample studies (n = 14) examined programs or services that directly address financial needs and supplement income via cash or cash equivalent, while the remaining 10 studies focused on programs or services that address basic material needs, such as healthcare, housing, and food.
• Comparator. Within the sample, all comparators reflected services as usual or no/nominal exposure to the treatment condition. Comparison and treatment conditions were formed through 11 RCTs, 11 QEDs from natural experiments, and 2 from QEDs using matched comparison methods.
• Outcomes. Studies examined the full range of primary and secondary outcomes identified in the eligible list. Child safety and maltreatment was typically defined through maltreatment report or investigation rates and child welfare involvement through removals from home (foster care entry). Only one study examined child welfare priority outcomes of reunification and maltreatment recurrence. Specific child and adult well-being outcomes varied considerably across studies, including physical and emotional health outcomes; mental health; substance use; parenting skills, competencies, or practices; income volatility; economic or housing instability; food insecurity; child behavioral issues or delinquency; and child cognition.
• Timing. All studies were published after 1990, with most (n = 20) published after 2015.
• Setting. Four of the studies were limited to a child welfare system setting, with families selected from a population of child welfare-involved families. The remaining 20 studies were set in non-child-welfare-specific settings. Though international studies were screened, due to other eligibility criteria, all included studies were conducted on services delivered in the United States with the exception of one study in Wales.

3.3. Overall Effectiveness on Primary and Secondary Outcomes

Results support the hypothesis that providing economic and concrete support, regardless of delivery mechanism, has a causal effect on primary outcomes of child maltreatment, child welfare involvement, and child welfare-related priorities, as well as secondary outcomes of child and adult well-being. Overall, 20 of the 24 studies include at least one finding of a statistically significant causal relationship between the economic and concrete support and improvements in the targeted outcomes, using a 5% threshold as the criteria for assessing significance, and two additional studies found at least one statistically significant finding at a 10% threshold. No studies reported a statistically significant unfavorable finding, meaning the economic or concrete support tested caused worse outcomes among the treated group versus the comparison group.

Table 3. Number of studies with favorable or null effects ¹.

Type of Outcome	Studies Including Outcome	Favorable	No Effect
Primary outcomes:
Child safety and maltreatment	5	4	1
Removal from home	3	3	0
Child welfare priority outcomes	3	3	0
Child well-being:
Cognitive functioning	3	2	1
Behavioral/emotional functioning	4	1	3
Physical development and health	4	3	1
Substance use or misuse	1	1	0
Delinquent behavior	0	n/a	n/a
Adult well-being:
Parenting practices/skills	3	3	0
Mental health	7	4	3
Physical health	3	2	1
Substance use or misuse	2	0	2
Criminal behavior	0	n/a	n/a
Family functioning	1	1	0
Economic and housing stability	4	3	1
Total studies included	24

¹ Some studies have more than one outcome. In this table, studies are only listed as having a null finding if there were no significant effects on any of the respective outcomes measured per category. Studies with at least one favorable finding per outcome may also have null findings for other outcomes or within the same outcome category. A not applicable (n/a) entry in the table reflects that the outcome was not measured and therefore, reporting whether favorable or null is not applicable.

presents evidence by outcome, including the number of studies that examined each, the number of studies that found at least one favorable and statistically significant (p < 0.05) effect on the outcomes tested, and the number that found no significant effects (all null findings, p > 0.05) on the outcomes in each category. As shown, for most of the outcome categories, studies generally found significant, favorable effects. This is clearest for primary outcomes related to child maltreatment, child welfare involvement, and child welfare priority outcomes. Only one study found no effect on the outcome (child safety measured through abusive head trauma admissions), though it is notable that the finding neared significance at p = 0.08 [41]. Findings were less robust for child and adult well-being. While most studies that examined child cognitive functioning, physical development and health, or substance use found significant, favorable effects, most of the studies on behavioral or emotional functioning found no effects of the economic and concrete support service. Evidence for adult well-being was even more mixed. Economic and concrete support services generally showed favorable effects on parenting practices or skills, family functioning, physical health, and economic or housing stability across studies, but not mental or emotional health or substance use.

3.4. Effects by Economic and Concrete Support Service

The finding that every type of support examined some impact on outcomes confirms our hypothesis that economic and concrete support, regardless of delivery mechanism, is an evidence-based service that can be provided independently or within broader programs. Though it was beyond the scope of this review to test the relative size or strength of effectiveness of each of economic or concrete support services delivery mechanism, in this section, we summarize findings for each type in greater detail.

3.4.1. Income Support—Cash or Cash Equivalent

• Earned Income Tax Credit. Six studies examined the effect of the Earned Income Tax Credit (EITC), an economic support for workers with low to moderate income, the amount of which depends on a recipient’s income and number of children. States can elect to implement state-level EITC policies as a supplement to the federal program. In terms of our primary outcomes of interest—child maltreatment, child welfare involvement, or other children welfare priority outcomes—EITC showed consistent favorable impacts in each of the four of the six studies that examined these primary outcomes. Biehl and Hill [28] found that expansion of federal EITC decreased foster care entry rates by 7.6% per year in states with a state-level EITC, relative to those without (b = −0.293, p < 0.001). Similarly, Rostad et al. [27] found a refundable EITC was associated with an 11% decrease in foster care entries compared to states without a state-level EITC (incidence rate ratio = 0.89, p < 0.05). Kovski et al. [26] found a 10-percentage-point increase in the generosity of refundable state EITC benefits was associated with 241 fewer neglect reports (p < 0.05), 220 fewer overall reports (p < 0.10), and 51 fewer substantiated reports (p < 0.10). However, generosity was not significantly associated with physical, emotional, or sexual abuse reports. Finally, Klevens et al. found that refundable EITC was associated with a marginally significant (p = 0.08) decrease of 3.1 abusive head trauma admissions per 100,000 population in children aged 2 and under [44]. EITC also led to improved child well-being, although the findings were less robust than for child maltreatment outcomes. One study found food insecurity among children decreased during the months immediately following EITC refund (b = −0.0094 per USD 1000, p = 0.052) but there was no EITC impact on the more distal outcomes of child body mass index (BMI) or behavioral issues [45]. Living in a state with refundable EITC was not associated with improved parent well-being, measured through self-reported general physical health, psychological distress, alcohol use, or smoking [46].
• Child Tax Credit. The Child Tax Credit (CTC) is another federal economic support program for families with children. The CTC was temporarily expanded during the COVID-19 pandemic to be dispersed monthly through automatically transferred advance payments to eligible families rather than through an annual tax refund. The CTC was also made fully refundable to low-income and unemployed parents. We included two eligible studies of the expanded CTC in this review, both of which examined the impact of CTC on well-being outcomes. One study found fewer depressive symptoms (b = −1.7, p < 0.01) and anxiety symptoms (b = −3.4, p < 0.01) but not mental healthcare utilization after the CTC expansion [47]. Another study found favorable effects of the expanded CTC on child well-being [48]. Specifically, the authors found a lower likelihood of injuries requiring medical attention (odds ratio = 0.58, p < 0.05) and significantly fewer behavior problems (b = −2.07, p < 0.05) among children with mothers eligible to receive a refundable CTC.
• Child Support Income. One study examined the impact of child support income on child maltreatment. The Child Support Demonstration Evaluation, implemented statewide in 1997–1998, randomly assigned all participants of the Temporary Assistance to Needy Families (TANF) program to either a full pass-through and disregard group for whom every dollar of child support paid by the nonresident father was passed on to the mother, or to a partial pass-through and disregard group (services as usual). Child support was then disregarded in determining the TANF cash benefit for mothers in the full pass-through experimental group. Cancian et al. [49] found receipt of the full pass-through resulted in a 12% reduction in screened-in maltreatment reports (odds ration = 0.88, p < 0.05).
• Cash Transfers/Guaranteed Income Support. Conditional cash transfers have been widely implemented and tested internationally with robust findings of favorable impact on a range of outcomes. Direct (unconditional) cash transfers have become more commonly implemented in recent years. This review included three studies of unconditional cash transfer or guaranteed income programs. Two studies examined the causal impact of the Baby’s First Years program, in which low-income mothers with newborns across four geographically diverse metropolitan areas in the United States were randomly assigned to receive either a large monthly cash gift (USD 333) or a nominal monthly cash gift (USD 20) during the early years of the infant’s life. One study found increased brain activity at approximately one year of age among infants of mothers who had received the large cash gift; differences between the treatment and control groups were statistically significant for most, but not all measures of brain activity, but were greatest in parts of the brain most associated with cognitive development [50]. A second study of Baby’s First Years found no statistically significant differences between mothers in the treatment and control groups in terms of self-reported substance use or expenditures on alcohol and cigarettes [51]. While this is a technically a null finding, it is notable that this indicates the large cash payments did not lead to increased spending on substances; the estimated share of the USD 313 additional cash provided to the treatment group that was spent on alcohol or tobacco was less than one percent. Our review included another program that provided cash as part of a guaranteed income pilot project. The Stockton Economic Empowerment Demonstration (SEED) provides a random sample of low-income households in Stockton, CA with a monthly unconditional cash transfer of USD 500 for two years. After one year, when compared to the control group, recipients reported lower income volatility (t = 1.76, p = 0.039), lower psychological distress (F = 4.98, p = 0.027), greater emotional health (t = 14.85, p = 0.012) and emotional well-being (t = 7.7, p = 0.022), higher energy (t = 7.3, p = 0.023), and lower physical pain (t = 7.87, p = 0.047) [52]. Due to the COVID-19 pandemic, the study suffered from attrition at the second-year follow-up that did not meet the eligibility criteria of our review. Although the follow-up showed similarly positive effects of the program, including a statistically significant impact on increased ability to pay for a USD 400 emergency, the follow-up study generally had too small of a sample to detect statistical significance.
• Child Development Accounts. Two studies examined the impact of Child Development Accounts (CDAs) on parent well-being. CDAs are an economic intervention that encourages families to accumulate assets for children’s long-term development, such as post-secondary education. Both studies focused on the SEED for Oklahoma’s Kids (SEED OK) experimental program, which randomly selected a group of infants in 2007 and randomly assigned mothers to a treatment or control condition. Those in the treatment group received an automatically opened Oklahoma 529 College Savings Plan account and received an initial USD 1000 deposit. Huang et al. found no statistically significant effect of the CDA on mothers’ depressive symptoms [53]. Similarly, Nam et al. did not find the CDA reduced parenting stress or parenting practices, with the exception of reducing the frequency of screaming at the child (b = −0.06, p < 0.05) [54]. Because this was only one small effect among a number of null effects, we took a conservative approach and considered this study to show all null findings in the overall summary noted in Table 3.

3.4.2. Non-Cash Support for Basic Needs

• Medicaid. As part of the Affordable Care Act (ACA), 24 states and the District of Columbia expanded Medicaid benefits in 2014, with another 7 states expanding the program in 2016. Brown et al. used this inconsistent adoption (and opt-out) to test the effect of the Medicaid expansion on child maltreatment [55]. The authors found Medicaid expansion resulted in 422 fewer screened-in maltreatment reports for neglect (95% CI: −753, −91; p < 0.05), but not physical abuse. Boudreaux et al. examined the long-term impacts of Medicaid exposure and found Medicaid exposure during early childhood (age 0–5) vs. no exposure in early childhood results in improved adult health (ages 25–54) based on a composite health index (b = −0.35, p < 0.01) but not economic conditions [56].
• Housing Support. Stable and safe living conditions are critical to family well-being. Two studies examined the impact of supportive housing on outcomes: First, through a collaboration between the Connecticut Department of Children and Families (DCF) and the Department of Mental Health and Addiction Services (DMHAS) on the development of a supportive housing program for women in recovery from substance use and their children. The program was then expanded to a larger set of families involved in the child welfare system. An RCT study of the program found that, compared to services as usual, children in families randomly assigned to a supportive housing treatment condition were more likely to reunify within 18 and 24 months; less likely to be removed from home within 12, 18, and 24 months (x² = 14.7, p < 0.01; x² = 21.9, p < 0.01, x² = 17.1, p < 0.01); and less likely to have a subsequent substantiated maltreatment report at 12, 18, and 24 months (x² = 5.1, p < 0.05; x² = 7.1, p < 0.01, x² = 3.0, p < 0.1) [57]. The supportive housing program had similarly significant impacts when modeling outcomes at the family level rather than child level. However, the supportive housing program did not impact time to reunification or maltreatment recurrence (including unsubstantiated reports). Another RCT study of a pilot intensive housing subsidy program for families with prior stays in family shelters found families in the program exited from shelter faster (95% CI 0.589, 0.942), spent a longer time out of shelter (95% CI 0.379, 0.880), and spent fewer days in shelter (376 vs. 449, p < 0.05) [58].
• Food and Nutrition Support. The Supplemental Nutrition Assistance Program (SNAP) and the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) provide food assistance to low-income families. SNAP provides eligible participants with an Electronic Benefits Transfer card with which funds are provided for food purchases while WIC provides a monthly dollar benefit to purchase certain types of foods as well as infant formula and other products. Wang et al. took advantage of temporal and state variations in SNAP policy rules to identify the impact of SNAP participation vs. no SNAP participation. The authors found SNAP reduced parental stress (b = −1.104, p < 0.01) yet had a negative, though non-significant impact of parental engagement with children (b = −2.27, p < 0.10) [59]. To further address food insecurity among school-aged children in low-income families, the Summer Electronic Benefits Transfer for Children (SEBTC) demonstrations provided summary food assistance in the form of electronic benefit transfer cards. An evaluation of the demonstrations randomly assigned households to a monthly USD 60-per-child benefit, a monthly USD 30-per-child benefit, or no benefit [60]. The main treatment condition (USD 60-per-child benefit) reduced prevalence of low or very low food security by one-third compared to the no benefit control group (6.1% vs. 9.1%, p < 0.001); hence, families that received the food assistance benefit were less likely to experience low or very low food security. The program reduced prevalence of food insecurity by nearly one-fifth, with less than 35% of households in the treatment group reporting food insecurity compared to 43% of households in the control group (p < 0.001). Similarly, children receiving free school meal entitlements ate a greater number of health items at breakfast (b = 0.16, p < 0.05) [61]. As further evidence of the importance in addressing families’ basic food and nutrition needs, another study found a significant and positive impact of in utero exposure to WIC on short-term cognitive skill development before age 2 (b = 0.0625, p = 0.03) and school-age cognitive skills, specifically letter/word recognition and broad reading scores (b = 0.256, p < 0.05; b = 0.260, p < 0.05) [62]. The treatment did not impact the applied problems cognitive skills outcome.
• Other Concrete Support. In addition to the cash, cash equivalent, or subsidies for basic needs, the provision of other concrete support is hypothesized to reduce child maltreatment and improve family well-being. Child welfare leaders across the country indicate that economic and concrete support is important in a comprehensive prevention strategy, as described in a recent Chapin Hall national landscape survey [63], and multiple states have identified resourcing families through economic and concrete support as a strategy for preventing child abuse and neglect and foster care entry and have specifically emphasized the need for such support as an evidence-based service in their Family First plans and in their use of flexible funds in child welfare services. Two studies examined the impact of concrete supports on outcomes. Wimer et al. examined Room to Grow, a program for parents and children ages 0–3, and found the program resulted in more toys and related developmental goods and books in the home (b = 0.39, p < 0.001; b = 0.51, p < 0.001, respectively) [64]. The program also increased parental sense of competence (b = 0.32, p < 0.001) and reduced aggravation in parenting (b = 0.28, p < 0.01). Effects on chaos in the home, level of cognitively stimulating activities in the home, maternal depressive symptoms, perceived stress, and financial worries were marginal or insignificant. Looking specifically at families involved with an open child welfare case, another study found that families who received concrete support through discretionary funds up to USD 600 as part of a home-based program reported greater program satisfaction (b = 0.058, p = 0.011), program engagement (b = 0.209, p = 0.004), and perceived cultural competence of services (b = 0.079, p = 0.018) but not program retention (b = 0.143, p = 0.149) or parenting stress (b = 3.67, p = 0.249) [36]. Receipt of concrete support did have a significant impact on child welfare re-entry. Specifically, receiving concrete support vs. not receiving such support was associated with a nearly 17% reduction in the odds of re-entry (hazard ratio = 0.83, p = 0.03).

4. Discussion

The findings in this systematic review highlight the importance of resourcing families as a child welfare prevention and intervention strategy and, to a lesser extent, a means of improving child and family well-being. Using a systematic and rigorous process, our results confirm the evidence, accumulated over time and across methodologies, on the effectiveness of economic and concrete support as a prevention tool. Importantly, this review confirms there are multiple types of economic and concrete support services that reduce maltreatment incidents and removals from home and improve child welfare priority outcomes, including maltreatment recurrence and reunification. Of the 24 studies, 8 examined these primary outcomes using both RCTs and natural experiments and with a focus on different types of economic and concrete supports as the intervention, including the EITC, child support income, Medicaid, supportive housing, and other concrete support. All except one study found at least one significant effect and, in most cases, significant effects on multiple primary outcomes. When families receive economic and concrete support, whether through cash or income support, healthcare or housing support, or other concrete support services, child maltreatment is reduced.

Economic and concrete support had less robust effects on adult well-being measures of mental health and substance use. Mental health and substance use challenges are common among families who experience substantiated CPS investigations, and having sufficient income to meet basic needs may provide an important buffer [1]. However, economic and concrete support may not be solely effective in reducing substance use or improving mental health, but rather may be one part of a larger set of services needed to address the complex needs associated with addiction or mental health issues. It was beyond the scope of this systematic review to confirm why the economic and concrete supports tested largely did not directly impact adult mental health or substance use. Additional literature review or testing in this area is recommended, particularly given that economic and concrete support may be indirectly related to improvements in these outcomes by improving participants’ engagement and retention in mental health or substance abuse treatment. Families must be able to access and continue to engage with evidence-based prevention services designed to prevent or treat mental health or substance use challenges to benefit from these services. Multiple studies have shown a relationship between economic and concrete support and participation or retention in substance use treatment and/or mental health programs. For example, Marsh et al. compared participation and outcomes for among mothers receiving substance use treatment as usual to mothers in an enhanced substance use treatment program that provided transportation to and from treatment, as well as outreach and child care to mothers [65]. They found access to these concrete supports through the enhanced program increased use of other social services, such as health care, parenting classes, family counseling, education/job training, domestic violence counseling, and housing and legal assistance, which in turn led to significantly less drug use. In another study of a substance treatment program for mothers involved with the child welfare system, those who enrolled their children in the day care program that was offered were three times more likely to complete treatment [66]. Similarly, a demonstration study of a residential substance abuse treatment program highlights the importance of concrete support as a protective factor, particularly for mothers with co-occurring mental health disorders and who are at risk of child welfare involvement. Mothers who participated in an enhanced intervention that included concrete support in time of need, including safe housing, access to health care and public benefits, employment assistance, and child care, remained in treatment for an average of 206 days compared to only 128 days for mothers who did not participate in the enhanced intervention [67]. Hence, concrete support may have helped to address mothers’ immediate needs, which allowed for participation in more in-depth substance abuse and mental health services.

The consistency with which economic and concrete support impacted child maltreatment and child welfare outcomes in this rigorous systematic review points to the utility of meeting families’ economic and material needs as a child maltreatment prevention and intervention strategy. Providing such support is also responsive to child welfare practitioners and families. Based in part on the growing body of literature, child welfare leaders across the country indicate that economic and concrete support is important in a comprehensive prevention strategy and that cross-sector collaboration is critical for provision, as detailed in a recent Chapin Hall and APSHA national landscape survey [63]. Additionally, multiple jurisdictions have identified resourcing families through economic and concrete support as a strategy for preventing child abuse and neglect and foster care entry and have specifically emphasized the need for such support as an evidence-based service in their Family First plans and in their use of flexible funds. As noted previously, families also are calling for the provision of economic and concrete support [3]. The findings of our systematic review provide rigorous evidence to support both practitioners and those with lived experience in their efforts to change system responses to families in need of economic or concrete support.

While the findings from the studies in this review show the effectiveness and flexibility of economic and concrete support as a prevention and intervention strategy, there are some caveats worth noting. First, our search was extensive yet resulted in what is likely a conservative final sample of studies due to strict eligibility criteria that we employed. Specifically, we excluded studies of programs that included economic or concrete support services but did not separately detail or test those components. This resulted in the exclusion of most studies of home visiting interventions and family preservation or differential response programs, many of which intentionally include economic and concrete support as a component, and which are significantly associated with less child maltreatment or child welfare involvement. Similarly, due to the potential for confounding bias, we excluded studies of programs or services that, in and of themselves, could be considered an economic or concrete support for families, such as subsidized early childhood education programs and child care. Therefore, due to our methodological decisions, we note that our review did not include all the potential economic and concrete support programs and services that may be effective in preventing child maltreatment or improving family well-being. Future reviews including these study designs would provide further information on the effectiveness of macroeconomic supports and home visiting or other child welfare-focused interventions that intentionally include concrete support.

Second, although nearly all studies reviewed found at least one favorable effect on the targeted outcomes and no studies found unfavorable effects, many reported mixed results, with some favorable and some null findings depending on the specific outcome. This is the case for multiple studies of the same program that assess different outcomes. For example, the Baby’s First Years program showed early impacts on infant brain activity [50] but no significant effect on substance use [51]. In using the outcome domains prioritized by Title IV-E Prevention Services Clearinghouse, our review included studies that measured impact on multiple indicators of child maltreatment and other child welfare-focused outcomes as well as a broad range of child and adult well-being outcomes. With so many eligible outcomes included in our review, it is perhaps not surprising that the economic and concrete support services assessed had no statistically significant effect on some of these outcomes. Most of the studies that reported null findings were those aiming to impact child or adult well-being. As noted by multiple authors of these studies, it may be that the amount or extent of economic or concrete support provided was insufficient to adequately address complex issues related to physical or emotional well-being. Economic and concrete support, however, had a more consistently favorable impact on child maltreatment- and child welfare-related outcomes.

Finally, we limited our review of findings to those reflecting main effects of economic and concrete support services on outcomes on full populations. However, we note that multiple studies reviewed, both those that were ultimately included and those excluded, presented subgroup analyses of economic and concrete support effectiveness. For example, neither study of Child Development Accounts included in our review found a significant main effect on outcomes. However, Huang et al. also examined the impact of CDAs on children of married vs. unmarried mothers [53]. While they found no main effect (impact on children pooled across both married and unmarried mothers), the authors found a statistically significant and large impact of CDAs on social–emotional development in children of unmarried mothers such that CDAs eliminated the disparity in social–emotional development between children of single mothers and those of married mothers. As another example, McGinty et al. (not included in final sample presented here) found Medicaid reduced maltreatment rates when examining expansion of the program in different age groups of children [68]. Other studies identified during our review showed statistically significant differences in the impact of economic and concrete support services on targeted outcomes by gender, age, and race/ethnicity.

Considering the unique needs and strengths of families, subgroup analyses are likely important in identifying services that are culturally responsive and effective for different populations (e.g., by age, gender, gender identity, marital status). Subgroup analyses considering race and ethnicity are particularly important. To prevent child maltreatment and foster care placements will require the delivery of services that address deeply rooted systemic factors to reduce racial and ethnic disparities and promote equity for all families. This includes focusing attention on underserved populations and increasing culturally specific evidence-based interventions and services. Subgroup analyses are, therefore, important in detecting differential effects, which could lead to a greater understanding of what services are most effective, for whom.

Optimizing the flexibility of economic and concrete support services to meet the economic and material hardship facing families and, in turn, preventing the unnecessary use of child protective services and foster care, requires transformation—both within child welfare and across the social and human services sector—and collaboration. Multiple systems play a role in supporting families, as evidenced in this review, albeit insufficiently and without the coordination necessary to address what Rollins and colleagues describe as a design flaw in the child protective system that leaves far too many families “in an abyss between health, economic, caregiving, and human service systems, where supports are either insufficient or unattainable, and family needs swell—only to find themselves in contact with child welfare, a system not designed to help them” [1] (p. 1). As seen in our systematic review, multiple social and human service systems provide economic or concrete support to families in need—through federal or state tax credits, investment accounts, cash transfer or guaranteed income programs, food assistance programs, housing supports, the Medicaid system, and the child welfare system—and all are shown here to be effective in either reducing maltreatment and child welfare involvement or strengthening family well-being. Moving even further toward prevention of child maltreatment will require these systems to collaborate. Rollins et al. provide an organizing framework for an integrated and holistic family and child well-being system, outlining the shared responsibility and accountability needed across systems [1]. Equally important is power sharing between agencies, communities, and families in decision-making and development of relevant programs, practices, and policies. Many agencies currently engage families and communities via formal platforms to obtain feedback on recommended decisions and newly developed initiatives. However, to truly ensure power sharing and to place equity at the center of community partnerships requires public systems to use strategies designed to collaboratively develop solutions, break down racist and stigmatizing narratives about Black and Brown communities, refocus on relationships and trust, and bridge the gaps between what individuals and families say they need and what systems deliver [69].

Although our review did not examine the effectiveness of economic and concrete support for different racial groups, we know that racial disparities occur at nearly every major decision-making point along the child welfare continuum, stemming from structural and institutional racism historically within the child welfare system and as a part of society at large [70]. For example, 53% of all Black children are estimated to be subject to at least one investigation by age 18 [71], and Black, American Indian, and Alaska Native children are at greater risk than other children of substantiated maltreatment and placement in out-of-home care [72]. At the same time, racism and socioeconomic bias in broader policymaking result in insufficient access and opportunity for families of color to meet their basic needs and to address economic and material hardship. According to U.S. Census figures, Hispanic, Black, American Indian, and Alaska Native individuals are overrepresented in the poverty population, with American Indian and Alaska Native individuals representing a share of the poverty population that is twice as large as their share of the total population [73]. Families of color are more likely to face economic and material hardship due to longstanding systemic inequities and structural racism [74].

Hence, evidence-based economic and concrete support services can begin to address longstanding fiscal inequities and restricted access to resources in communities of color [75], which serve as drivers of disparity and disproportionality in child welfare. For example, in an analysis simulating the effects of increased income under anti-poverty packages (including income support programs like the EITC) [30] found the potential for an 11 to 20% annual reduction in CPS investigations (affecting 386,000 to 669,000 children) and possible reductions of between 19 and 29% for Black children. However, while this idealized simulation found potential for reductions in racial disproportionality, an analysis by Puls et al. using state data found that state public benefit spending was associated with reduced CPS investigations for White children, but not for Black or Hispanic children [76]. If benefit programs add relative advantages for White children compared with Black and Hispanic children, this will contribute to further racial and ethnic disparities in CPS investigations [76]. Our systematic review illustrates the effectiveness of economic and concrete support in addressing families’ needs and preventing child maltreatment. However, the recent Puls et al. study reinforces the importance of intentionality in distributing such support so as not to create even greater racial disparities and calls for a close examination of how public benefit programs and other economic and concrete support services are implemented. To ensure that policy is equitable in practice requires implementation research to understand if and why there are racial and ethnic differences in the distribution of economic and concrete support.

5. Conclusions

The results synthesized here reflect the utility of economic and concrete support services as a prevention strategy. Given the strict eligibility criteria we used to identify “treatment” and “control” groups and associated study designs, this should be viewed as a conservative review and does not reflect the much larger body of literature that shows a preponderance of evidence of the relationship between economic and concrete support and outcomes of interest. For example, we did not include any correlational studies showing that increasing economic or concrete support significantly predicts better outcomes for families. Additionally, because we included a number of eligible outcomes of interest, it is not surprising that most studies resulted in at least some null findings. This was particularly true for measures that reflected child or adult wellbeing, given the wide range of constructs that these could include, for example, health, substance use, behavior or behavioral health, parenting skills, or economic stability in the home. Measures of child maltreatment or child welfare involvement were less varied and the impact of economic and concrete support services more robust. It is notable, however, that none of the studies examined found evidence of economic and concrete support services having an unfavorable effect on outcomes. Further, our results should be viewed as conservative given only main effects were included; we did not include significant findings on the impact of economic and concrete support services for different subgroups or those presented as moderating effects. Additional reviews are needed to identify if there is a differential impact of economic and concrete support for subgroups and the level to which these supports make a difference for families facing certain challenges or in specific contexts. Overall, the evidence of the effectiveness of economic and concrete support synthesized through our systematic review points toward policy and practice that encourages preventive strategies and interventions focused on resourcing families to alleviate economic and material hardship in addition to addressing safety. Both must be addressed through a coordinated and integrated response, built upon collective responsibility and accountability, to maximize the positive impact of public resources for families.