Impact of COVID-19 on child development, academic achievement, and behavioral development in Rural Bangladesh

• In Event: The impact of COVID on child development and learning loss

Abstract

Background: The COVID-19 pandemic represents an unprecedented effect on the health and livelihoods of rural families across South Asia. School closures in Bangladesh were some of the longest in the world, lasting 543 days. During this time the Ministry of Education delivered school materials through TV, radio, zoom, Facebook, and texts, but these methods of learning were insufficient for many low-income rural households without access to consistent internet or electricity. The aim of this research is to quantify the effects of the COVID-19 pandemic and associated shut-downs on child development and academic achievement amongst primary school-aged children in Rural Bangladesh.

Methods: Participants are children in the WASH-Benefits Bangladesh (WASH-B) study cohort (Tofail et al. 2018). Approximately half of the children in the 7-year child development follow-up of the WASH-B study were assessed prior to the COVID pandemic (September 2019 to March 2020; “pre-covid”), and approximately half were assessed during COVID related school closures (October 2020 to February 2021, “during covid”). The during covid sample experienced an average of 9 months (SD: 1.3) of covid-related school closures prior to assessment. Primary outcomes are Cognitive development measured by the Full Scale IQ of the Wechsler Preschool and Primary Scale of Intelligence (WPPSI-IV), School Achievement in Reading, Literacy, and Numeracy with a culturally adapted tool based on the principals of Wide Range Achievement Test, and socio-behavioral development measured with the Strengths and Difficulties Questionnaire. For each outcome adjusted mean differences will be determined comparing the pre-covid and during-covid samples using generalized linear models accounting for the block-level clustering of the original study design, study arm, child age, and theory-driven confounders as covariates. Analyses will be conducted both within the whole sample, and within the sample with an age overlap.

Results: Children assessed pre-covid (n=2007) were younger on average (mean age: 80.6 months, SD: 2.14) than children assessed during covid (n=1828, mean age: 86.6 months, SD: 1.51), though there is overlap in age for around 25% of the total sample. Children assessed at the pre-covid time point were slightly better off at baseline (collected in utero) on some sociodemographic characteristic compared to those assessed during covid (table 1). The proportion of children measured in each intervention arm was the same across groups. While analyses are ongoing, unadjusted means for outcomes are presented in the pre-covid and during covid group (table 2). Children assessed at the pre-covid time point had higher scores on the Full Scale IQ, reading, and writing scores (table 2). Adjusted analyses accounting for theorized confounders and child age, and analyses will be presented for all outcomes in both the full and reduced-age samples.

Discussion: Documenting the impact of the COVID-19 pandemic on child development is an important step to understand the implications of school closures in this setting. The consequences of school closures on cognitive and behavioral development and school achievement should be considered in future policy decisions. These results may also highlight the areas where the greatest deficits have occurred to target interventions.

Authors

• Helen Osborne Pitchik, University of California - Berkeley
  Presenting Author

• Lia C.H. Fernald, University of California - Berkeley

• Md. Mahbubur Rahman, International Center for Diarrhoeal Disease Research, Bangladesh

• Alaka Holla, World Bank - District of Columbia

• Laura Becerra Luna

• Stephen P. Luby, Stanford University

• Fahmida Tofail, International Center for Diarrhoeal Disease Research, Bangladesh