Scaling Early Childhood Development – what to read this month | September 2026

Scaling Early Childhood Development – what to read this month is curated by Bet Caeyers (Lead Editor, Chr. Michelsen Institute), Meghan Taylor, (Editor, Oxford Policy Management), and Daniel Munday (Editor, Oxford Policy Management).

Cuartas et al. make the point that taking parenting programmes to scale requires looking beyond caregiver behaviour to the wider conditions that enable – or constrain – positive parenting. Their ‘geo-bioecological’ perspective shows how parenting is shaped by factors ranging from poverty and parental mental health to social networks, cultural norms, conflict and climate-related shocks. Economic hardship, food insecurity and displacement, for example, can increase parental stress and harsh parenting, while social support can reduce depression and provide practical caregiving help. Parenting programmes can improve engagement, responsive caregiving and non-violent discipline, but effects sometimes fade, and many still fail to reach entire populations. The authors therefore argue for combining parenting support with wider services and policies – such as social protection, mental-health support, childcare and parental leave – that create more supportive conditions for nurturing care.

Figure 1: Visual representation of the ecology of parenting (Cuartas et al. 2026)

Turning to parenting programme delivery, Jeong et al. test an approach that combines parenting support with the reach and influence of local faith leaders. In Kenya, Moments that Matter combined monthly parenting groups and home visits delivered by community volunteers with faith leaders reinforcing nurturing-care messages. The programme improved stimulation, dietary diversity, caregiver mental health and male involvement, and reduced harsh discipline, parenting stress and intimate partner violence. Yet there was no overall impact on child development. The authors suggest that caregivers may have needed more opportunities to practise with their children, alongside coaching and feedback. Importantly, play and stimulation emerged as the main pathway linking the programme to child development. The findings remind us to focus on these core elements of parenting support, rather than assuming broader improvements in family wellbeing will translate into better child development.

Staying with parenting programme delivery, a pilot mixed-methods process evaluation by Jeong et al. shows how designing programmes around both mothers and fathers from the outset can help address the persistent challenge of engaging fathers. Familia Bora was co-designed with parents in Tanzania and combined parenting and responsive caregiving with couple relationships, gender norms, mental health and economic strengthening. Fifteen mixed-gender group sessions were delivered to 62 couples, with similar attendance among fathers (61%) and mothers (66%) and 97% retention. Parents particularly valued learning together as couples, although work remained an important barrier for fathers and some messages around discipline and gender roles initially met resistance. The next test will be whether similar engagement and delivery quality can be achieved when the programme moves from well-supported NGO facilitators to existing community health workers – an important step in understanding whether this model can work within routine systems.

Turning to the wider systems supporting child development, Onuorah et al. show how school meal programmes can pursue environmental goals alongside nutrition and social protection, while highlighting the challenge of embedding these objectives consistently across countries. Using Global Child Nutrition Foundation surveys from 2019–2024, they find widespread use of diverse plant-based foods and food-waste reduction, but less consistent adoption of local sourcing and formal environmental policies. Practices also varied substantially across regions and income groups. Programmes with environmental policies were more likely to have diverse plant-based menus and source food locally, suggesting that policy frameworks may help embed sustainability into routine delivery. The authors stress, however, that greener school meals need strong nutrition standards and monitoring so environmental goals reinforce, rather than compromise, children’s nutrition and equity.

Also on nutrition, Chowdhury et al. show how strengthening existing delivery systems can improve child nutrition at scale without introducing new nutrition interventions. Reflecting on evidence from 410 districts in India, they report that strengthening frontline workers, digital monitoring and community mobilisation increased nutrition-service coverage and improved height-for-age by 0.18 standard deviation (SD), corresponding to a 6–7 percentage-point reduction in stunting. Adding financing, infrastructure and performance incentives produced no further gains, suggesting substantial returns from strengthening basic implementation capacity before adding more complex components. Yet delivery systems alone cannot address poverty, food insecurity, sanitation, women’s empowerment and other underlying drivers of undernutrition. The authors therefore argue for stronger district-level systems embedded within coordinated, multisectoral approaches, rather than relying on standalone nutrition interventions.

Turning to childcare, Goldberg and Meghir examine how expanding provision can support both women’s economic opportunities and child development in low- and middle-income countries. Reviewing evidence across countries, they find that childcare can increase mothers’ employment, but effects depend on available jobs and social norms; where relatives provide care, it may instead free grandparents or older siblings to work or attend school. For children, what matters is the quality of childcare relative to the care it replaces: high-quality provision can improve development, while replacing better home or alternative care with poorer-quality services can be harmful. The authors therefore argue that expanding access alone is not enough: investment in childcare quality needs to accompany expansion, alongside attention to the labour market and social constraints that determine whether families can benefit.

On measurement, Fernald et al. provide practical guidance for an important challenge as early childhood development (ECD) programmes expand: how to measure child development reliably across different contexts and for different purposes. The World Bank toolkit covers 147 assessment tools for children aged 0–8 and guides users through choosing, adapting and implementing measures for population monitoring, programme evaluation and research. Importantly, the authors caution against assuming that a tool validated elsewhere will work equally well in a new setting: translation, cultural adaptation, pre-testing and piloting are needed to maintain validity. At the same time, extensive adaptation can reduce comparability across settings, creating an important trade-off. For programmes and systems seeking to track ECD more widely, the toolkit shows that measurement choices need to balance local validity, comparability, sensitivity, cost and feasibility rather than relying on a single universal measure.

Finally, Seiden et al. examine how much ECD assessments in infancy can tell us about children’s development several years later. Following 487 children in Tanzania, they find that caregiver-reported early development instrument (CREDI) scores at around 12 months significantly predicted international early learning and development assessment (IDELA) and early childhood development index (ECDI2030) scores at ages 3.5–6.5, although associations were small (around 0.1–0.2). At preschool age, the directly assessed IDELA and much shorter caregiver-reported ECDI2030 were substantially more strongly correlated (around 0.54). The findings therefore caution against treating ECD scores in infancy as strong predictors of later development, while supporting brief caregiver-reported tools for population-level monitoring.