Early education

Longitudinal follow-up of high-quality early education programs for children in economically disadvantaged families – including the Abecedarian Project, High/Scope Perry Preschool Project, and the Chicago Parent Child centers – have illustrated the potential to promote economic security and interrupt the intergenerational transmission of health and economic disparities (Campbell et al., Reference Campbell, Pungello, Burchinal, Kainz, Pan, Wasik and Ramey2012; Reynolds & Temple, Reference Reynolds and Temple1998; Reynolds, Temple, & Ou, Reference Reynolds, Temple and Ou2010). The Abecedarian Project – a center-based year-round program for almost exclusively African American children living in poverty that began in the 1970s (Campbell, Helms, Sparling, & Ramey, Reference Campbell, Helms, Sparling, Ramey, Barnett and Boocock1998) – has conducted the most extensive health assessments and found that participants assigned to the intervention condition displayed better cardiometabolic health profiles in their mid-30s relative to controls (Campbell et al., Reference Campbell, Conti, Heckman, Moon, Pinto, Pungello and Pan2014). New evidence from the High/Scope Perry Preschool intervention, also a center-based program for African American children in the 1960s, demonstrates a positive impact not only on participants, but also on their siblings and children, with the offspring of those in the treatment condition having fewer school suspensions, less criminal behavior, and greater educational attainment and employment relative to the children of participants in the control condition (Heckman & Karapakula, Reference Heckman and Karapakula2019). Moreover, a re-analysis of this intervention study to explore possible heterogeneity based on the child's socioeconomic position found that the treatment effects on both noncognitive and cognitive skills were larger and more persistent for the children from the most disadvantaged families relative to those from more advantaged families (Xie, Near, Xu, & Song, Reference Xie, Near, Xu and Song2020).

The positive outcomes from these early trials provide compelling evidence in support of investment in high-quality center-based programs as an intervention to address racial and ethnic disparities (Karoly, Kilburn, & Cannon, Reference Karoly, Kilburn and Cannon2006; Shonkoff & Fisher, Reference Shonkoff and Fisher2013; Yoshikawa, Reference Yoshikawa1995). The sample sizes are small, however (n = 111 and n = 123, respectively), and questions persist about variation in impacts across programs, outcomes, and children. Given the dramatic changes in the social context for young children since these two programs began half a century ago, these programs serve as a historical “proof of concept” (Shonkoff & Fisher, Reference Shonkoff and Fisher2013). New research is needed to accurately describe: (a) the return on investment for intensive, high-quality center-based programs; (b) features of programs and participants that lead to the best results (i.e., the content of the intervention; the location, required staff training, characteristics of families served; and the timing, intensity, and duration of the program); and (c) the underlying processes contributing to program impacts over the long run.

The Chicago Parent Child centers are unique in providing longitudinal follow-up of a large public program, showing benefits in adolescence (Reynolds & Temple, Reference Reynolds and Temple1998) and adulthood (Reynolds et al., Reference Reynolds, Temple and Ou2010). A new generation of randomized clinical trial (RCT) studies that build programming onto existing large public programs, such as Educare (Yazejian et al., Reference Yazejian, Bryant, Hans, Horm, St. Clair, File and Burchinal2017), Research-Based and Developmentally Informed-Parent home visiting program (Bierman, Welsh, Heinrichs, & Nix, Reference Bierman, Welsh, Heinrichs and Nix2018), and ParentCorps (Brotman et al., Reference Brotman, Dawson-McClure, Kamboukos, Huang, Calzada, Goldfeld and Petkova2016), find positive effects for outcomes for children, and some that extend into mid-childhood (e.g., school readiness, academic performance, mental health, and need for services into mid-childhood). It will be critical to examine the long-run effectiveness and cost–benefit analyses for this new set of programs.

Economic support for families

There is growing consensus for a causal impact of parental income for nearly all child and youth outcomes. For a thorough review of the impact of family economic security policies on child health, we refer you to comprehensive reports on this topic (Dreyer, Reference Dreyer2019; Duncan, Magnuson, & Votruba-Drzal, Reference Duncan, Magnuson and Votruba-Drzal2017; National Academies of Sciences, Engineering, and Medicine, 2019a). Researchers have used a variety of study designs to demonstrate the benefits of interventions to promote economic security for young and expectant parents. For example, researchers have conducted natural experiments using variation in state policy, demonstrating that income-related policies (e.g., federal tax reforms, minimum wage laws) can positively impact maternal and child health outcomes. As an example, one study shows that the Earned Income Tax Credit (EITC) improves infant health outcomes, with effects stronger for Black mothers than White mothers (Hoynes, Miller, & Simon, Reference Hoynes, Miller and Simon2015). Similarly, laws to increase the minimum wage above the federal level lead to a decline in low birth weight and post-neonatal mortality (Komro, Livingston, Markowitz, & Wagenaar, Reference Komro, Livingston, Markowitz and Wagenaar2016).

The results from studies of state-level policies described above are consistent with evidence from the Great Smoky Mountains study, a natural experiment that took place when a subset of Native American children (ages 9–13 at baseline) already enrolled in a longitudinal study received an additional $4000 in annual household income when a casino opened in their community. Assessed four years after income supplementation began, children in households that received supplemental income saw improvements in conduct and oppositional defiant disorder symptoms, becoming similar to the level observed among counterparts who were never poor (Costello, Compton, Keeler, & Angold, Reference Costello, Compton, Keeler and Angold2003). Longer-term follow-up revealed that the beneficial impact continued into adulthood, with benefits observed for educational attainment and minor criminal offenses, eliminating the Native American-White gap for both of these outcomes (Akee, Copeland, Keeler, Angold, & Costello, Reference Akee, Copeland, Keeler, Angold and Costello2010). In addition, subgroup analyses show stronger effects for the younger subset of the cohort. A new RCT is underway to test the causal impact of monthly unconditional cash transfers on child development from birth to three years (Noble, Reference Noble2017; Rojas et al., Reference Rojas, Yoshikawa, Gennetian, Lemus Rangel, Melvin, Noble and Magunson2020), which will expand our understanding of income supplementation in the earliest years of life. The importance of this line of intervention research is reinforced by observational cohort studies which show that income in the prenatal period and early infancy may be a sensitive periods for the effect of socioeconomic disadvantage on adult body mass index (BMI) (Gilman et al., Reference Gilman, Huang, Jimenez, Agha, Chu, Eaton and Loucks2018; Ziol-Guest, Duncan, & Kalil, Reference Ziol-Guest, Duncan and Kalil2009) and inflammation (Slopen et al., Reference Slopen, Loucks, Appleton, Kawachi, Kubzansky, Non and Gilman2015).

Federal investments directed towards taxes and benefits for families determine the child poverty rate, and therefore present an opportunity to foster resilience. For example, a UNICEF report making cross-national comparisons based on 2014 data shows dramatic variation in the reduction in the child poverty rate before and after taxes and transfers; while the child poverty rate is minimally changed in some countries (e.g., less than 10% reduction in Mexico, Romania, and Israel), taxes and transfers reduce the child poverty rate by approximately two-thirds in Finland, Iceland, and Norway. Beyond income and tax and benefit policies, other work-place policies related to paid family and medical leave that vary by race and ethnicity (Bartel, Kim, Nam, Rossin-Slater, Ruhm, & Waldfogel, Reference Bartel, Kim, Nam, Rossin-Slater, Ruhm and Waldfogel2019) could be important in addressing child health disparities. For example, access to paid leave is associated with children's well-child visits, access and use of healthcare and dental services, and fewer ER visits (Asfaw & Colopy, Reference Asfaw and Colopy2017; Clemans-Cope, Perry, Kenney, Pelletier, & Pantell, Reference Clemans-Cope, Perry, Kenney, Pelletier and Pantell2008; DeRigne, Stoddard-Dare, & Quinn, Reference DeRigne, Stoddard-Dare and Quinn2016; Shepherd-Banigan, Bell, Basu, Booth-LaForce, & Harris, Reference Shepherd-Banigan, Bell, Basu, Booth-LaForce and Harris2017).

Finally, there is an opportunity to address racial and ethnic disparities in early child development via workforce training for parents of young children. A recent line of research is now underway to advance on the hypothesized limitations of workforce training initiatives for low-income parents that were tested in the 1980s and 1990s (e.g., Project Redirection, New Chance) that are summarized as “largely ineffective” (p. 434) (Chase-Lansdale et al., Reference Chase-Lansdale, Sabol, Sommer, Chor, Cooperman, Brooks-Gunn and Morris2019). These new programs incorporate a “two generational approach” and merge education and training programs for parents with high-quality early education for children (i.e., sectors that typically exist in silos). For example, CareerAdvance, developed by the Community Action Project of Tulsa, is a partnership between Head Start and community college partners that aligns child and adult services and provides additional on-site services at the Head Start centers (e.g., coaching support, peer meetings, in-kind assistance for transportation, wraparound child care on top of Head Start services, etc.) (Chase-Lansdale et al., Reference Chase-Lansdale, Sommer, Sabol, Chor, Brooks-Gunn, Yoshikawa and King2017; Chase-Lansdale et al., Reference Chase-Lansdale, Sabol, Sommer, Chor, Cooperman, Brooks-Gunn and Morris2019; Sabol et al., Reference Sabol, Sommer, Chase-Lansdale, Brooks-Gunn, Yoshikawa, King and Ross2015; Sommer et al., Reference Sommer, Sabol, Chor, Schneider, Chase-Lansdale, Brooks-Gunn and Yoshikawa2018). The training program is tailored to the local economy and focuses on health care careers, which constitute a growth area for the region. At one year of follow-up of 287 low-income, racially diverse families, participants in CareerAdvance had a higher rate of certification and employment in the healthcare sector, and reported higher levels of optimism, self-efficacy, and career identity, relative to a matched comparison group (Chase-Lansdale et al., Reference Chase-Lansdale, Sabol, Sommer, Chor, Cooperman, Brooks-Gunn and Morris2019). Of note, the program did not have an effect on short-term income, employment, or reports of material hardship or stress, even though participants in the intervention condition had additional demands of training, along with family responsibilities and employment. In future studies, it will be important to evaluate long-term impacts on income and employment, as well as outcomes for children.

Food security

Food insecurity is approximately twice as common in Hispanic, non-Hispanic Black, and American Indian and Alaska Native households relative to non-Hispanic White households (Coleman-Jensen, Rabbitt, Gregory, & Singh, Reference Coleman-Jensen, Rabbitt, Gregory and Singh2019; Jernigan, Huyser, Valdes, & Simonds, Reference Jernigan, Huyser, Valdes and Simonds2017), which has implications for disparities in birth outcomes and subsequent child development (American Academy of Pediatrics, 2015; Gross & Mendelsohn, Reference Gross and Mendelsohn2019). Solutions to address food insecurity include making it easier for families to access federal food and nutrition programs, and strengthening these programs. Quasi-experimental studies exploiting county- and state-level variation and change over time provide support for the benefits of programs such as Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) and Supplemental Nutrition Assistance Program (SNAP) for improving food security (Mabli & Worthington, Reference Mabli and Worthington2014), birth outcomes (Almond, Hoynes, & Schanzenbach, Reference Almond, Hoynes and Schanzenbach2011; Hamad, Collin, Baer, & Jelliffe-Pawlowski, Reference Hamad, Collin, Baer and Jelliffe-Pawlowski2019), and children's health (East, Reference East2020). Furthermore, longitudinal quasi-experimental evidence from the Panel Study of Income Dynamics (PSID) shows that children of parents with lower education who received food stamps in utero through 4 years of age had reduced incidence of metabolic syndrome, relative to those who did not (Hoynes, Schanzenbach, & Almond, Reference Hoynes, Schanzenbach and Almond2016). Among women only, access to food stamps during the in utero or early childhood period was associated with increased high school graduation rates and lower use of safety net programs. Consistent with earlier research using the Panel Study of Income Dynamics showing that income support during this early period has a stronger impact on long-term health relative to later in childhood (Ziol-Guest et al., Reference Ziol-Guest, Duncan and Kalil2009), this study found that the positive outcomes are specific to this early period.

National data from 2018 show that just over half (56%) of families reporting food insecurity participated in one of the three largest federal food and nutrition assistance programs (i.e., Supplemental Nutrition Assistance Program, Special Supplemental Nutrition Program for Women, Infants, and Children, or the National School Lunch Program) in the 30 days leading up to the survey (Coleman-Jensen et al., Reference Coleman-Jensen, Rabbitt, Gregory and Singh2019). Beyond improving ease of enrollment and support for these government programs, nongovernmental programs may be particularly important to address food insecurity among immigrant families, as current immigration policies discourage eligible immigrant families, fearing consequences at the time of application for permanent residency, from enrolling in benefit programs (Bernstein, McTarnaghan, & Gonzalez, Reference Bernstein, McTarnaghan and Gonzalez2019; Laird, Santelli, Waldfogel, & Wimer, Reference Laird, Santelli, Waldfogel and Wimer2019; Sommers et al., Reference Sommers, Allen, Bhanja, Blendon, Orav and Epstein2020).

As a direction for future research, behavioral interventions related to policy utilization is an under-studied yet potentially useful area of research for developmentally informed intervention science (Yoshikawa, Whipps, & Rojas, Reference Yoshikawa, Whipps and Rojas2017). For example, policy interventions related to utilization of income support, federal food and nutrition assistance programs, health care policies, or workplace policies require individual behaviors for access, and navigation of these supports and interactions with service providers can be challenging for parents. Intervention studies related to policy utilization, informed by a developmental understanding of the cognitive processes of specific subgroups of parents (e.g., young parents vs. grandparent caregivers; native-born vs. immigrant families, etc.), could improve access to such programs and ultimately result in improved support for low-income families (Yoshikawa et al., Reference Yoshikawa, Whipps and Rojas2017).

Primary health care

There are a number of approaches to promote the well-being of young children of color using interventions embedded into health care practices. A commonly discussed example is Nurse Family Partnership (NFP), a home-visiting program that begins in the prenatal period and extends until the child is 2 years old (Karoly et al., Reference Karoly, Kilburn and Cannon2006). In three separate RCTs with varied populations and locations, Nurse Family Partnership has shown positive effects for child well-being, including fewer emergency department visits and adverse birth outcomes, better language development, and lower health care encounters for injury and substantiated abuse or neglect. CenteringPregnancy, a group prenatal care program, is another health care-centered intervention that has been extensively studied, and has documented the potential to improve birth outcomes for minority and Medicaid-eligible women (Abshire, McDowell, Crockett, & Fleischer, Reference Abshire, McDowell, Crockett and Fleischer2019; Crockett et al., Reference Crockett, Heberlein, Smith, Ozluk, Covington-Kolb and Willis2019).

There is strong interest in screening for the social determinants of child health in the pediatric setting, to connect individuals to needed social resources (Garg, Marino, Vikani, & Solomon, Reference Garg, Marino, Vikani and Solomon2012; Gottlieb et al., Reference Gottlieb, Adler, Wing, Velazquez, Keeton, Romero and Hessler2020) or legal assistance (Beck et al., Reference Beck, Klein, Schaffzin, Tallent, Gillam and Kahn2012; Klein et al., Reference Klein, Beck, Henize, Parrish, Fink and Kahn2013). These efforts can have a positive impact on families and children (Beck et al., Reference Beck, Cohen, Colvin, Fichtenberg, Fleegler, Garg and Kahn2018; Fierman et al., Reference Fierman, Beck, Chung, Tschudy, Coker, Mistry and Federico2016). For example, a cluster-RCT of a screening and referral program (Well Child Care, Evaluation, Community Resources, Advocacy, Referral, Education, [WE CARE]) of mothers with infants in eight urban community health centers (n = 336) found that screening and referral during well-child visits can lead to positive changes 12 months later across a range of outcomes related to social services, child care needs, and economic security (Garg, Toy, Tripodis, Silverstein, & Freeman, Reference Garg, Toy, Tripodis, Silverstein and Freeman2015). Another example, Healthy Steps for Young Children (Valado, Tracey, Goldfinger, & Briggs, Reference Valado, Tracey, Goldfinger and Briggs2019), is a clinic-based program that combines social needs screenings with child development specialists during well-child visits, developmental and family checkups, home visiting, and various parent supports. Experimental and quasi-experimental studies have demonstrated improvements in quality of health care and parenting practices, including reading, communication, and discipline (Johnston, Huebner, Anderson, Tyll, & Thompson, Reference Johnston, Huebner, Anderson, Tyll and Thompson2006; Johnston, Huebner, Tyll, Barlow, & Thompson, Reference Johnston, Huebner, Tyll, Barlow and Thompson2004; Minkovitz et al., Reference Minkovitz, Strobino, Mistry, Scharfstein, Grason, Hou and Guyer2007).

Medical Legal Partnership (Murphy, Lawton, & Sandel, Reference Murphy, Lawton and Sandel2015; Tobin-Tyler & Teitelbaum, Reference Tobin-Tyler and Teitelbaum2019) is an example of a type of program that families may be connected to as the result of social screening. Medical legal partnerships embed legal expertise into the health care team to address unmet legal needs that could impact child health, such as substandard housing, denial of government supports or educational needs, family violence, and barriers to health insurance or health care. Studies have shown that medical legal partnerships can result in improvements in preventive care for infants (Sege et al., Reference Sege, Preer, Morton, Cabral, Morakinyo, Lee and Kaplan-Sanoff2015) and parenting stress (Rosen Valverde, Backstrand, Hills, & Tanuos, Reference Rosen Valverde, Backstrand, Hills and Tanuos2019), as well as economic hardships (e.g., debt relief [Teufel et al., Reference Teufel, Werner, Goffinet, Thorne, Brown and Gettinger2012], retroactive receipt of benefits [Klein et al., Reference Klein, Beck, Henize, Parrish, Fink and Kahn2013], avoidance of utility shutoffs [Taylor et al., Reference Taylor, Bernstein, Carroll, Oquendo, Peyton and Pachter2015]).

At present, routine screening for a range of social needs, which presents both logistical and ethical challenges, is rare in pediatric offices in the United States (Garg et al., Reference Garg, Cull, Olson, Boyd, Federico, Dreyer and Racine2019). Numerous questions remain that call for careful study, including (a) the contents of a screening measure; (b) who should administer the screening; (c) how to effectively develop trust between parents and providers to enable parents to be honest about needs; (d) how frequently screenings should be administered; and (e) how both pediatrics practices and social policies must be restructured to ensure that screening generates connections to effective community-based services and state and federal safety-net programs (Brundage, Reference Brundage2019; Garg, Homer, & Dworkin, Reference Garg, Homer and Dworkin2019).

Place-based interventions

Structural inequities in contexts, such as neighborhoods and schools, affect health, educational, and economic disparities, and shape day-to-day interactions and the supports and stressors in daily living. As a result, many children of color live in neighborhood contexts that fail to provide the same opportunities for healthy development relative to White children. The Child Opportunity Index, a composite index of 29 indicators measured at the census tract level, combines information on neighborhood resources and conditions that promote optimal child development (e.g., early childhood education and schools, air and water quality, parks and playgrounds, safe housing, economic characteristics, etc.) (Acevedo-Garcia et al., Reference Acevedo-Garcia, Noelke, McArdle, Sofer, Huntington, Hardy and Reece2020). According to this metric, in the 100 largest metropolitan areas in the United States in 2015, one in five White and Asian children live in communities characterized as high or very high opportunity, while two-thirds of African American children and more than half of Latino and Native American children live in low- or very low-opportunity neighborhoods (i.e., 67%, 58%, and 53%, respectively) (Acevedo-Garcia et al., Reference Acevedo-Garcia, Noelke, McArdle, Sofer, Huntington, Hardy and Reece2020). Research shows that the Child Opportunity Index is related to child health outcomes and acute care use (Beck et al., Reference Beck, Huang, Wheeler, Lawson, Kahn and Riley2017; Kersten et al., Reference Kersten, Adler, Gottlieb, Jutte, Robinson, Roundfield and LeWinn2018); therefore, these disparities in neighborhood opportunity by race and ethnicity illustrate the importance of place-based interventions as a strategy to improve life opportunities for children of color.

Comprehensive community development programs, sometimes referred to as “place-based initiatives” or “structural interventions” aim to make changes to multiple levels of the social and environmental context via cross-sectoral and sustainable collaborations and stake-holder engagement and shared decision-making (Brown et al., Reference Brown, Ma, Miranda, Eng, Castille, Brockie and Trinh-Shevrin2019; Raphael, Reference Raphael2018). Place-based interventions emerged in response to clear evidence on the importance of community context for health and inequities (Dankwa-Mullan & Pérez-Stable, Reference Dankwa-Mullan and Pérez-Stable2016). There are a number of developmentally informed place-based initiatives that are designed to promote healthy child development, with initiatives tailored to the specific setting and needs of the community (Dean, Seymour, & Rider, Reference Dean, Seymour, Rider, Shapiro and Harper Browne2016; Inkelas, Bowie, & Guirguis, Reference Inkelas, Bowie and Guirguis2017; Rider, Winters, Dean, & Seymour, Reference Rider, Winters, Dean and Seymour2014; Schuchter & Jutte, Reference Schuchter and Jutte2014; Tough, Reference Tough2009).

The Harlem Children's Zone (HCZ) is an example of a multisectoral community intervention that began in the 1990s, designed to interrupt intergenerational cycles of poverty by providing wide-ranging supports to children and their families (Tough, Reference Tough2009). The initiative is designed to comprehensively address threats to healthy child development, including issues related to housing, education, and pervasive drug use, crime, and chronic health problems in the community (Nicholas et al., Reference Nicholas, Jean-Louis, Ortiz, Northridge, Shoemaker, Vaughan and Hutchinson2005; Northridge et al., Reference Northridge, Scott, Swaner, Northridge, Jean-Louis, Klihr-Beall and Caraballo2009; Spielman et al., Reference Spielman, Golembeski, Northridge, Vaughan, Swaner, Jean-Louis and Cushman2006; Taylor, Schorr, Wilkins, & Smith, Reference Taylor, Schorr, Wilkins and Smith2018). Various aspects of the initiative have been evaluated over time. For example, the educational programs have shown positive impacts for (a) kindergarten readiness (Harlem Children's Zone, 2020), (b) reducing racial gaps in academic achievement (Dobbie & Fryer, Reference Dobbie and Fryer2011, Reference Dobbie and Fryer2015), and (c) reducing incarceration among males, and teenage pregnancy among females (Dobbie & Fryer, Reference Dobbie and Fryer2015). HCZ was used as a model by the US Department of Education to design the Promise Neighborhoods program which was established in 2010. This initiative provides competitive grants to support coordination between the education sector and family and community services in now more than 80 communities around the United States (Promise Neighborhood Institute, 2020).

There is increasing recognition that efforts to create communities of opportunity and to achieve population-level impacts are most likely to be effective when they generate systemic and multisectoral changes (National Academies of Sciences, Engineering, and Medicine, 2019b). Taylor and colleagues (Reference Taylor, Schorr, Wilkins and Smith2018) outline features of systemic approaches within place-base initiatives such as HCZ and Promise Neighborhoods that can provide a starting point for effective systems changes, including: leveraging infrastructure (i.e., connecting existing infrastructure to make improvements to the overall system and individual components), community orientation (i.e., responding to community-identified priorities), skills and resources to handle complexity (i.e., structures to enable partners to collaborate and align efforts in a strategic manner), commitment to rigor (i.e., investment in skilled personnel for data collection and rigorous analysis), and use of multiple methods and data for evaluation and continuous improvement. It is important to note that while place-based initiatives hold great potential for sustainable population-level impact, there are many challenges, as they require substantial time and resources, and effort and strategy to build trust, accountability, and shared goals across partners and stakeholders (Raphael, Reference Raphael2018; Taylor et al., Reference Taylor, Schorr, Wilkins and Smith2018).

Community empowerment

Community empowerment, defined as “processes of interaction between individuals and organizations to enhance community living, thereby effecting changes in the large social system” (p. 810) (Cyril, Smith, & Renzaho, Reference Cyril, Smith and Renzaho2015), is a community-level trait that may promote resilience for children and families experiencing discrimination. For example, a seminal study by Chandler and Lalonde examined youth suicide rates in relation to a measure of community empowerment and continuity across 196 First Nations communities in British Columbia, Canada (1987–1992) (Chandler & Lalonde, Reference Chandler and Lalonde1998). The measure of community empowerment and continuity – designed to reflect the degree to which a community had taken steps to preserve and continue cultural practices and use legal and political strategies for advocacy and empowerment (i.e., pursuit of land claims, right to self-governance, control over education, police, fire, health services, and the presence of cultural facilities) – displayed an inverse dose–response relationship with youth suicides. Of note, all six of the indicators of community empowerment were protective against youth suicides, and these results were replicated in a later study with subsequent years of data (i.e., 1993–2000) (Chandler & Lalonde,, Reference Chandler, Lalonde, Kirmayer and Valaskakis2008).

Research on empowerment as a community-level trait is consistent with (a) a small literature on the positive effects of fostering racial/ethnic pride for adolescent health behaviors (DiClemente et al., Reference DiClemente, Wingood, Harrington, Lang, Davies, Hook and Robillard2004; Enriquez, Kelly, Cheng, Hunter, & Mendez, Reference Enriquez, Kelly, Cheng, Hunter and Mendez2012), and (b) studies of sociopolitical control, which refers to individual beliefs about one's skills and abilities to create change within social and political systems (Peterson, Peterson, Agre, Christens, & Morton, Reference Peterson, Peterson, Agre, Christens and Morton2011; Zimmerman & Zahniser, Reference Zimmerman and Zahniser1991). The research on individual perceptions of sociopolitical control finds that higher perceptions of sociopolitical control are associated with better youth outcomes such as community and school participation and drug use (Peterson et al., Reference Peterson, Peterson, Agre, Christens and Morton2011). Accordingly, this work illustrates the potential for political movements that promote (a) justice within governmental processes and (b) the allocation of power and resources within communities most affected by health disparities, as a means to achieve health equity. These topics represent potentially useful directions for future research on health disparities in children and adolescents.

Minimizing unintended negative effects of interventions

Future research on policies to promote resilience and improve health equity will benefit from focused attention to identifying and minimizing unintended negative outcomes stemming from interventions. Unintended negative effects could occur due to: (a) the widening of inequalities, as the result of interventions that differentially impact subgroups of the population (Krieger, Reference Krieger2011; Thomson et al., Reference Thomson, Hillier-Brown, Todd, McNamara, Huijts and Bambra2018; Whitehead, Reference Whitehead2007); or (b) positive impacts of an intervention for some outcomes or subgroups simultaneous to harmful consequences for others outcomes or subgroups.

Variation in the impact of an intervention across child health outcomes and subgroups (e.g., defined by age or sex) have been observed for housing mobility interventions (Chetty, Hendren, & Katz, Reference Chetty, Hendren and Katz2016; Fauth, Leventhal, & Brooks-Gunn, Reference Fauth, Leventhal and Brooks-Gunn2007; Kessler et al., Reference Kessler, Duncan, Gennetian, Katz, Kling, Sampson and Ludwig2014) and income supplementation (Akee, Simeonova, Copeland, Angold, & Costello, Reference Akee, Simeonova, Copeland, Angold and Costello2010). For example, in an analysis of data from the Moving to Opportunities study (n = 2872 children aged 0–8 years at baseline, assessed at 13–19 years), boys whose families received vouchers to relocate from high- to low-poverty neighborhoods displayed increased depression, posttraumatic stress disorder, and conduct disorder relative to the control group, whereas girls in families that received traditional vouchers were less likely to experience major depression and conduct disorder relative to the control group (Kessler et al., Reference Kessler, Duncan, Gennetian, Katz, Kling, Sampson and Ludwig2014).

Considering the Great Smoky Mountains Study described above, one analysis reported an increase in accidental deaths during the specific months that households received the income supplement, presumably linked to increases in vehicular travel and substance use (Bruckner, Brown, & Margerison-Zilko, Reference Bruckner, Brown and Margerison-Zilko2011). In addition, the cash supplements were also associated with increased adolescent obesity among American Indian families whose incomes were low at the inception of the income supplements with no effect for those families whose income was high (Akee, Simeonova, et al., Reference Akee, Simeonova, Copeland, Angold and Costello2010). These findings suggest the potential importance of a comprehensive, multilevel strategy, that could combine income interventions with individual-level interventions that increase health knowledge and engagement that empowered individuals and families to minimize risks associated with substance use and unhealthy nutritional patterns.

Outcome specificity of resilience promoting factors

A recent phenomenon identified within resilience research, referred to as “skin deep resilience” has documented a consistent pattern where disadvantaged adolescents who display external signals of achievement and success (e.g., self-control, academic excellence) and positive mental health outcomes have poorer physical health outcomes according to biomarkers that predict risk of subsequent chronic diseases (Brody et al., Reference Brody, Yu, Chen, Miller, Kogan and Beach2013; Chen, Miller, Brody, & Lei, Reference Chen, Miller, Brody and Lei2015; Gaydosh, Schorpp, Chen, Miller, & Harris, Reference Gaydosh, Schorpp, Chen, Miller and Harris2018; Miller, Cohen, Janicki-Deverts, Brody, & Chen, Reference Miller, Cohen, Janicki-Deverts, Brody and Chen2016; Miller, Yu, Chen, & Brody, Reference Miller, Yu, Chen and Brody2015). In the first study to show this pattern, Brody and colleagues analyzed longitudinal data from a sample of 489 African American youth in the rural Southeast followed from age 11–19, with teacher ratings of academic and social competence assessed annually during the first three years of the study. Individuals with high cumulative socioeconomic-related risk and identified as “resilient” based on teacher reports had low levels of depressive symptoms and externalizing behaviors at age 19, but had the highest allostatic load scores at this time (i.e., a count of physiological indicators in the top quartile of risk, including body mass index, resting diastolic and systolic blood pressure, and overnight cortisol, epinephrine, and norepinephrine) (Brody et al., Reference Brody, Yu, Chen, Miller, Kogan and Beach2013).

Since this initial publication, this pattern of resilience showing an association with better psychosocial outcomes and worse physical health has been replicated for other outcomes among African American participants, ranging from epigenetic aging of immune cells (Miller et al., Reference Miller, Yu, Chen and Brody2015), to susceptibility to infection (Miller et al., Reference Miller, Cohen, Janicki-Deverts, Brody and Chen2016), to diabetes (Brody, Yu, Miller, & Chen, Reference Brody, Yu, Miller and Chen2016b). Importantly, at least two studies have shown that this relationship between resilience, measured in these studies using assessments of conscientiousness and “striving”, and compromised physical health does not extend to White participants (Brody, et al., Reference Brody, Yu, Miller and Chen2016b; Miller et al., Reference Miller, Cohen, Janicki-Deverts, Brody and Chen2016). The nongeneralizability of these results across racial and ethnic groups raises critical questions about the external social conditions, interpersonal interactions, and individual coping methods that lead to these race-specific findings. To understand these patterns, it may be valuable for future studies to consider the potential role of systemic inequities in community context (Williams & Collins, Reference Williams and Collins2001), personally mediated racism (Williams et al., Reference Williams, Lawrence and Davis2019), social isolation (Cole & Omari, Reference Cole and Omari2003), and “John Henryism” (i.e., a stress-related coping method that involves sustained high effort to the point of adverse physiological consequences) (Bennett et al., Reference Bennett, Merritt, Sollers Iii, Edwards, Whitfield, Brandon and Tucker2004; James, Reference James1994).

This line of research has raised significant questions on the outcome specificity related to resilience, and indicates a need for research on: (a) the unique challenges of social disadvantage and upward mobility that are distinct for racial minority children and adolescents, to inform interventions; (b) supportive strategies to protect the physical health of high-achieving minority students; (c) the longer-term consequences of these patterns observed in young adulthood (i.e., it is unknown if the benefits of higher education over the long run are most consequential for long-term risk of age-related diseases); and (d) the interactive relationship between resilience-promoting factors at multiple levels, including political and economic forces. In future research, it will be valuable to document how carefully designed programs and interventions have the potential to work synergistically with broad social policies to protect the long-term health of at-risk youth in the United States.

Increasing public support to address racial and ethnic health inequities

Although substantial progress across disciplines has advanced knowledge regarding policies and interventions that can foster resilience among children who face intersecting systems of disadvantage, bridging the gap from research to practice or policy is a challenge (Yoshikawa et al., Reference Yoshikawa, Whipps and Rojas2017), and will require increasing public support to address racial and ethnic inequities (Williams & Cooper, Reference Williams and Cooper2019, Reference Williams and Cooper2020). To do this, there is a need to elevate awareness of racial and ethnic inequalities in economic and health outcomes, which are under-recognized by the general public in the United States (Benz, Espinosa, Welsh, & Fontes, Reference Benz, Espinosa, Welsh and Fontes2011; Kraus, Onyeador, Daumeyer, Rucker, & Richeson, Reference Kraus, Onyeador, Daumeyer, Rucker and Richeson2019; Niederdeppe, Bigman, Gonzales, & Gollust, Reference Niederdeppe, Bigman, Gonzales and Gollust2013). Connected to this is increased understanding of and compassion towards the diverse circumstances of children in the United States could play an important role in building public support and political will to mitigate inequities (Gollust, Nagler, & Fowler, Reference Gollust, Nagler and Fowler2020; Williams & Cooper, Reference Williams and Cooper2019), and research shows that empathy can be developed (Weisz & Zaki, Reference Weisz and Zaki2017; Zaki, Reference Zaki2019). For example, a randomized field experiment set in Miami, Florida found that a single 10-minute conversation with a door-to-door canvasser designed to encouraged empathy led to a change in attitudes towards transgendered individuals and increased support for a discrimination law, and these changes were sustained for at least three months (Broockman & Kalla, Reference Broockman and Kalla2016). As another example, in a randomized trial with middle-school math teachers (n = 31 teachers, n = 1682 students), an empathy-focused intervention designed to address implicit bias reduced suspension rates from 9.6% to 4.8% over the course of one year (Okonofua, Paunesku, & Walton, Reference Okonofua, Paunesku and Walton2016). Similarly, in a randomized trial of 34 White female pre-service teachers, an empathy-focused intervention to improve implicit bias (as measured by the Implicit Association Test) led to a decline in implicit bias towards Black individuals (Whitford & Emerson, Reference Whitford and Emerson2019); however, future research is needed to establish how this type of change in implicit bias leads to a meaningful decline in discriminatory behavior towards children, or to increased support for progressive policies.