Early Childhood Investments to Improve the Mental Health of Disadvantaged Youth

In this issue of Pediatrics, Goldfeld et al¹  used hypothetical modeling to examine how addressing 2 key factors in early childhood, parental mental health and preschool attendance, may reduce disparities in mental health experienced by disadvantaged children later in childhood. Using a longitudinal data set of Australian children, they demonstrated that children experiencing socioeconomic disadvantage by 1 year of age experienced mental health symptoms at age 10 years nearly twice as often (32.8% vs 18.7%, a statistically significant difference after adjusting for confounders) as their nondisadvantaged peers. Disadvantaged children were more likely to have a parent experiencing psychological distress (14.8 vs 8.5%) and less likely to attend preschool (60.9 vs 69.3%). Their model showed that closing the gap in parental mental health leads to a greater reduction in child mental health compared with closing the gap on preschool attendance, though neither reached statistical significance. The authors also conducted a “maximum benefit scenario” in which all disadvantaged children attended preschool and also had parents with low psychological distress. In the maximum benefit model, a statistically significant absolute risk reduction of 1.8% was demonstrated with eliminating parental psychological distress among disadvantaged children. Though not statistically significant, addressing both parental mental health and preschool attendance resulted in the highest risk reduction. These findings illustrate the importance of future research to confirm, or challenge, these results and to evaluate the effects of other combined interventions on children’s mental health.

Given the growing mental health crisis,^2,3  models such as these are needed to evaluate the substantial investments required to successfully address this crisis. Modeling does have limitations. Changes to exposure variables are hypothetical, which can limit real-world applicability. In this study, key variables were measured at 1 time point only, such as socioeconomic status (at 1 year), parental mental health (at 4–5 years), and child mental health (at 10 years). Socioeconomic status and parental mental health are not necessarily static over time, so using comprehensive, longitudinal data to assess effects at multiple time points may provide a more robust, and accurate, assessment of outcomes. Evidence also suggests that the chronicity of parental mental health distress,⁴  in addition to the cumulative nature of other adverse childhood experiences,^5,6  further increases the likelihood of child mental health problems.

Sociodemographic differences between Australia and the United States may also affect generalizability. In the US, preschool is 1 component of early childhood education, along with child care, prekindergarten, and kindergarten. The US system includes private programs and state-funded prekindergarten programs (in 45 states⁷ ), as well as the federally funded Head Start Program for children living at or below the federal poverty level. Despite universal eligibility, Head Start serves less than half of children living in poverty.⁸  In the United States, around 60% of children ages 3 and 4 years participated in state-funded preschool before the pandemic, and enrollment decreased by 25% during the pandemic.⁹  Participation, resources, and quality varies significantly by state.¹⁰  This study measured preschool attendance as a dichotomized variable and at 1 time point, yet outcomes may differ depending on participation, quality,^5,11–14  and amount of time spent in preschool.¹⁵ 

Interestingly, even in their maximum benefit model, disparities in child mental health persisted. This highlights a critical point: these interventions alone are not sufficient, especially for those who are marginalized and socioeconomically disadvantaged. Moreover, these 2 interventions live in systems that were stressed before the COVID-19 pandemic, and now, as a result of the pandemic, significantly affected.^9,16  As the authors note, cross-sectoral, multilayered interventions will be needed to meaningfully reduce mental health inequities. Increasingly, health systems and businesses are investing in multifaceted, community-based interventions that go beyond direct service provision and instead target social determinants of health.^17–20  Given the significant investments needed, modeling may be particularly valuable in assisting with cost-minimization and resource management, which must be considered with cross-collaborative and complex systems of care.

In addition to community-based intervention, societal- and individual-level interventions will be needed to reduce disparities in child mental health. At the societal level, reducing child poverty has a potentially significant impact; possible solutions can be found in the National Academy of Sciences’ Roadmap to Reducing Child Poverty, which outlined policy recommendations and expansions of public health insurance for pregnant women, infants, and children.²¹  At the individual level, the American Academy of Pediatrics has highlighted the role of safe, stable, nurturing relationships in buffering toxic stress and promoting resiliency.⁶  Perhaps modeling can help determine the effect of these varied approaches. In the words of Frederick Douglass, “it is easier to build strong children than to repair broken men.” Being able to better quantify the cost and impact of societal and individual interventions, both separately and in combinations of multiple interventions as modeled in this study, may help us build a better, and more equitable, society.