from EVIDENCE to POLICY Learning what works for better programs and policies July 2021 INDIA: Can we make parenting programs more cost-effective? EARLY CHILDHOOD DEVELOPMENT In the first years of life, all children need healthy food, a clean from low-income settings has found that encouraging parents to environment, and stimulation to thrive and reach their full devel- play and interact more with their children can improve children’s opmental potential. However, poverty prevents millions of young brain development, with impacts that can last into adulthood. De- children in low- and middle-income countries from receiving ad- livering these parenting programs at scale and in a cost-effective equate nutrition and stimulation. As a result, many disadvantaged manner, however, has been a challenge, in part because some of the most successful programs have been delivered through inten- sive and relatively costly home-based programs. The World Bank’s Strategic Impact Evaluation Fund support- ed a randomized evaluation in Odisha, India that tested different ways of delivering weekly parenting sessions for two years. Chil- dren and their mothers received either nutritional education, nu- tritional education and individual home-based child development sessions, or nutritional education and group-based child develop- ment sessions, depending on which community they lived in. The evaluation found that compared to receiving no interven- tion, the individual home-based sessions and group sessions both improved children’s language and cognition, and they were equally effective. The group sessions were much cheaper to implement, however, costing $38 per child per year, which was roughly a quar- ter of the cost of home visits. Most impacts occurred in the first year and were sustained in the second year. The nutrition educa- tion had no impact on developmental outcomes, suggesting that knowledge may not be an important barrier to better nutrition in this context. These findings suggest mother-child group sessions, Photo: John Isaac / World Bank when implemented well, offer a cost-effective alternative to home children’s brain development lags behind that of their well-off visits as a strategy to improve young children’s cognitive develop- peers, which can have lifelong consequences. Previous research ment in low-income settings. This policy note is based on “Group Sessions or Home Visits for Early Childhood Development in India: A Cluster RCT,” Grantham-McGregor S, Adya A, Attanasio O, et al. Pediatrics. (2020). Context India has experienced steady economic growth in recent decades, study in Jamaica that found long-term benefits of early childhood and many social and health indicators have greatly improved stimulation that lasted into adulthood. This parenting model is along the way, including indicators of early childhood develop- now known as Reach Up and Learn. In a previous evaluation of ment, but there is still a long way to go in addressing gaps in early Reach Up and Learn in India, researchers found that the program childhood development. UNICEF estimates that 43 percent of improved children’s development, but it was labor-intensive and children under age five in India are at risk of not fulfilling their costly. Together with Pratham, researchers decided to investigate developmental potential. Odisha, a state in eastern India on the whether adapting the home-visiting program into group sessions Bay of Bengal, has experienced steady growth, but remains one of could deliver similar impacts at lower cost. The evaluation took the country’s poorest and most populated states. place in in three districts of Odisha state: Cuttack, Salepur, and Pratham Education Foundation, the implementing partner Bolangir. in this study, is a large and established non-governmental orga- 39% of children are stunted nization in India. In the context of this study, Pratham imple- Poverty in Odisha state... 44% lack drinking water on premises mented a program, which like many others around the world, 30% of adults are illiterate is based on the curriculum used in a foundational experimental Source: World Bank, 2016 Evaluation Researchers conducted a randomized controlled trial to evaluate The third group received nutrition education and weekly group the impact of nutrition education, home-based visits, and moth- sessions where facilitators taught an adapted version of the cur- er-child group sessions on five main outcomes: child cognition, riculum used in the individual sessions to groups of 7-8 children language, motor development, physical growth, and morbidity. and their mothers. During the sessions, all mothers and children A total of 192 villages (clusters) were randomly assigned to one performed the same activities at the start, such as free play, sing- of four groups –– three treatment groups and a control group. ing, review of the previous week’s activities, and child-rearing dis- Each experimental group had 48 villages and approximately 360 cussions. The groups were then divided in half based on children’s children. ages for age-specific play activities. The first treatment group received nutritional education only. The sessions all ran weekly for two years, starting in December The nutrition education program focused on improving the qual- 2015, and were led by female facilitators from within the local ity of children’s diets and basic hygienic practices in households communities, nearly all of whom had secondary school education through games, stories, and cooking demonstrations. or higher (40 percent had bachelor’s degrees). Home visits lasted The second group received both nutritional education and about one hour, group meetings lasted 90 minutes, and nutri- home-based individual early childhood development sessions. tional education visits were 40-minutes long. The facilitators came to the house once a week for one-hour Researchers first collected data at baseline, before the pro- sessions, where they showed mothers how to interact with and grams began. At that time, they assessed children’s development respond to their children in ways likely to promote their child’s using an adapted version of the Ages and Stages Questionnaire development. They demonstrated play activities and encouraged (ASQ-3). At 12 and 24 months after baseline, caregivers brought mothers to participate by using toys made from locally available children to testing centers, where testers with tertiary education or materials and books. Mothers received play materials to use at equivalent experience working with children measured children’s home that would be exchanged each week with every new set of cognitive, language, and motor development using the Bayley activities demonstrated during the session. Scales of Infant and Toddler Development. Other enumerators measured children’s growth at home in all data collection rounds line by maternal reports by using the Strengths and Difficulties using World Health Organization (WHO) guidelines and stan- Questionnaire. Secondary outcomes, such as home environment dard tools for measuring height and weight. Researchers assessed and parents’ knowledge of infant development, were measured in children’s wellness at midline and endline through mothers’ re- all rounds in homes. All surveys and questionnaires were trans- ports on occurrences of diarrhea, fever, and cough in the previous lated into the local language, Odiya, and extensively piloted before two weeks using WHO definitions. the evaluation began. Children’s socioemotional development was assessed at end- Findings After one year, children in communities offered group There are a few possible reasons why this approach to improv- or individual home-based sessions had better cogni- ing children’s nutrition didn’t work. For one, the nutrition tive development than children in the control group. counselling may have occurred too late: stunting (being too short for one’s age) may have been better addressed through Both home-based and group sessions led to improvements in earlier intervention when children were younger. Second, children’s cognition, amounting to 0.313 standard deviations household food security was low in this population –– 15 per- (SD) and 0.298 SD, respectively. Children in the communities cent of households reported at least one household member offered group sessions also had better language skills on aver- skipping meals in the past week – so nutritional supplemen- age (a 0.313 SD improvement), while those in the home-based tation may have been required. A further problem may have sessions didn’t see significant impacts at the one-year mark. No been poor sanitation, with less than half of households owning statistically significant changes were observed in children’s motor skills or physical growth relative to the control group. At the two-year mark, children in both the group ses- sions and home-based intervention had better lan- guage and cognitive development than children in the control group. Significant impacts on cognition persisted and remained stable from year 1 to year 2, as did the effects on language for chil- dren in the group-based sessions. After two years, children in the group offered home-based sessions showed a 0.23 SD lan- Photo: Simone D. McCourtie / World Bank guage advantage over the control group that was marginally toilets. Such inadequate sanitation adds to risks of diarrhea, significant. intestinal disease, and parasites, preventing absorption of nu- trients and contributing to malnutrition and poor growth. The nutritional education intervention had little to no impact on developmental outcomes. Beyond language and cognition, researchers found limited impacts in other domains of child Researchers had hypothesized that the nutrition intervention development. may lead to better nutrition and improve child growth, but there was no impact on children’s height or weight (for their Researchers also looked at a range of other outcomes that age). The approach also didn’t lead to any improvements in might be affected by the programs, but found little beyond other health indicators like diarrhea or coughing. the main outcomes. For example, no statistically or economi- cally meaningful impacts were observed on children’s motor Group sessions may have worked partially through skills. Group sessions led to reductions in reported fever after peer modeling or by promoting cultural acceptance of two years, but had no other effects on reported morbidity. The the practices that were taught. home-based intervention had a marginal impact on prosocial skills but there weren’t any impacts on child behavior (internal- Overall attendance was higher in the home-visit group than the izing and externalizing problems) or on parents’ knowledge of group session group (75 percent compared with 51 percent), but child development. impacts were the same. Because impacts are estimated based on families who were offered the sessions in the experiment, not just The group sessions were much more cost-effective those who actually received the sessions, the identical impacts than the home visits. with differential attendance suggests that impacts of the group sessions for participants was much higher than for participants The costs of training, materials, salaries, and overhead of the of the home-based intervention. The research team speculates home visiting program led to a cost of $135 per child per year that this is because mothers and children in the group sessions when each home visitor made 15 weekly visits. Group sessions learned skills through observing others and that socially isolated cost $38 per child per year when each facilitator ran eight groups mothers may have received support they needed by interacting per week with eight children per group. Given the impacts were with peers. They also hypothesize that group sessions aided cul- the same across the two program variants, the group session tural acceptance of the child stimulation and rearing practices model ended up being much more cost-effective. that the program was trying to promote. Conclusion This was the first study to compare home-based and group- The results also suggest that the nutrition education model based parenting sessions. Group sessions cost only 28 percent tested in this study isn’t effective, and other approaches, pos- of what home-based visits cost in this context, with equiva- sibly supplementation – which has been effective in other con- lent average effectiveness among those offered the programs texts – or improvements in sanitation may be necessary to im- (and larger impacts for those who actually participated in the prove children’s physical growth. programs). The Strategic Impact Evaluation Fund, part of the World Bank Group, supports and disseminates research evaluating the impact of development projects to help allevi- ate poverty. The goal is to collect and build empirical evidence that can help governments and development organizations design and implement the most appropriate and effective policies for better educational, health, and job opportunities for people in low and middle income countries. For more information about who we are and what we do, go to: http://www.worldbank.org/sief. The Evidence to Policy note series is produced by SIEF with generous support from the British government’s Foreign, Commonwealth and Devel- opment Office and the London-based Children’s Investment Fund Foundation (CIFF). THE WORLD BANK, STRATEGIC IMPACT EVALUATION FUND 1818 H STREET, NW, WASHINGTON, DC 20433