Cohort Profile Update: The Young Lives study

Cohort profile: the Adverse Childhood Experiences cohort of the Malawi Longitudinal Study of Families and Health

PURPOSE

The Adverse Childhood Experiences (ACE) cohort of the Malawi Longitudinal Study of Families and Health (MLSFH-ACE) is a study of adolescents surveyed during 2017-2021. It provides an important opportunity to examine the longitudinal impact of ACEs on health and development across the early life course. The MLSFH-ACE cohort provides rich data on adolescents, their children and adult caregivers in a low-income, high-HIV-prevalence context in sub-Saharan Africa (SSA).

PARTICIPANTS

The MLSFH-ACE cohort is a population-based study of adolescents living in three districts in rural Malawi. Wave 1 enrolment took place in 2017-2018 and included 2061 adolescents aged 10-16 years and 1438 caregivers. Wave 2 took place in 2021 and included data on 1878 adolescents and 208 offspring. Survey instruments captured ACEs during childhood and adolescence, HIV-related behavioural risk, mental and physical health, cognitive development and education, intimate partner violence (IPV), marriage and aspirations, early transitions to adulthood and protective factors. Biological indicators included HIV, herpes simplex virus and anthropometric measurements.

FINDINGS TO DATE

Key findings include a high prevalence of ACEs among adolescents in Malawi, a low incidence of HIV and positive associations between ACE scores and composite HIV risk scores. There were also strong associations between ACEs and both IPV victimisation and perpetration.

FUTURE PLANS

MLSFH-ACE data will be publicly released and will provide a wealth of information on ACEs and adolescent outcomes in low-income, HIV-endemic SSA contexts. Future expansions of the cohort are planned to capture data during early adulthood.

Longitudinal patterns of natural hazard exposures and anxiety and depression symptoms among young adults in four low- and middle-income countries

We estimated the effect of community-level natural hazard exposure during prior developmental stages on later anxiety and depression symptoms among young adults and potential differences stratified by gender. We analyzed longitudinal data (2002-2020) on 5585 young adults between 19 and 26 years in Ethiopia, India, Peru, and Vietnam. A binary question identified community-level exposure, and psychometrically validated scales measured recent anxiety and depression symptoms. Young adults with three exposure histories ("time point 1," "time point 2," and "both time points") were contrasted with their unexposed peers. We applied a longitudinal targeted minimum loss-based estimator with an ensemble of machine learning algorithms for estimation. Young adults living in exposed communities did not exhibit substantially different anxiety or depression symptoms from their unexposed peers, except for young women in Ethiopia who exhibited less anxiety symptoms (average causal effect [ACE] estimate = - 8.86 [95% CI: - 17.04, - 0.68] anxiety score). In this study, singular and repeated natural hazard exposures generally were not associated with later anxiety and depression symptoms. Further examination is needed to understand how distal natural hazard exposures affect lifelong mental health, which aspects of natural hazards are most salient, how disaster relief may modify symptoms, and gendered, age-specific, and contextual differences.

Social Protection and Foundational Cognitive Skills during Adolescence: Evidence from a Large Public Works Program

Many low- and middle-income countries have introduced public works programs (PWPs) to fight poverty. This paper provides the first evidence that children from families who benefit from PWPs show increased foundational cognitive skills. The results, based on unique tablet-based data collected as part of a long-standing longitudinal survey, show positive associations between participation in the Productive Safety Net Programme (PSNP) in Ethiopia during childhood with long-term memory and implicit learning, and suggestive evidence for working memory. These associations appear to be strongest for children whose households were still PSNP participants in the year of data collection. Evidence suggests that the association with implicit learning may be operating partially through children's time reallocation away from unpaid labor responsibilities, while the association with long-term memory may in part be due to the program's success in remediating nutritional deficits caused by early-life rainfall shocks.

Does early nutrition predict cognitive skills during later childhood? Evidence from two developing countries

The existing evidence linking early undernutrition to educational outcomes in developing countries is largely focused on assessing its impacts on grade attainment and achievement test scores, with limited evidence on the foundational cognitive skills required to perform well at school. We use unique data collected in Ethiopia and Peru as part of the Young Lives Study to investigate the relationship between early undernutrition and four foundational cognitive skills measured later in childhood, the first two of which measure executive functioning: working memory, inhibitory control, long-term memory, and implicit learning. We exploit the rich longitudinal data available to control for potential confounders at the household and individual level and for time-invariant community characteristics. We also take advantage of the availability of data for paired-siblings to obtain household fixed-effects estimates. In the latter specification, we find robust evidence that stunting at ~ age 5 is negatively related with executive functions measured years later, predicting reductions in working memory and inhibitory control by 12.6% and 5.8% of a standard deviation. Although the main cohort of Young Lives was around 12 years old when executive functions were measured, complementary results and analysis of the data available for the younger siblings suggest that the impact of stunting on executive functions-specifically, on working memory-starts at an earlier age. Our results shed light on the mechanisms that explain the relationship between early nutrition and school achievement tests suggesting that good nutrition is an important determinant of children's learning capacities.

Effects of early-life poverty on health and human capital in children and adolescents: analyses of national surveys and birth cohort studies in LMICs

The survival and nutrition of children and, to a lesser extent, adolescents have improved substantially in the past two decades. Improvements have been linked to the delivery of effective biomedical, behavioural, and environmental interventions; however, large disparities exist between and within countries. Using data from 95 national surveys in low-income and middle-income countries (LMICs), we analyse how strongly the health, nutrition, and cognitive development of children and adolescents are related to early-life poverty. Additionally, using data from six large, long-running birth cohorts in LMICs, we show how early-life poverty can have a lasting effect on health and human capital throughout the life course. We emphasise the importance of implementing multisectoral anti-poverty policies and programmes to complement specific health and nutrition interventions delivered at an individual level, particularly at a time when COVID-19 continues to disrupt economic, health, and educational gains achieved in the recent past.

Lifetime impact of being underweight or overweight/obese during childhood in Vietnam

BACKGROUND

There is limited evidence about lifetime burden of child malnutrition. This study aimed to estimate the lifetime impact of being underweight or overweight/obese during childhood in Vietnam.

METHODS

We developed a life table model in combination with a Markov model for Vietnamese children aged 5-19 years and simulated until they reached 75 years of age or died using published data. The starting year was 2019 and the model estimated number of deaths, years of life lived and quality-adjusted life years (QALY) with an annual discount rate of 3%. We performed scenario, one-way, and probabilistic sensitivity analyses to assess the impact of uncertainties in input parameters.

RESULTS

The model estimated 9.68 million deaths (6.44 million men and 3.24 million women), 622 million years of life lived (317 million men and 305 million women), and 601 million QALYs (308 million men and 293 million women). Scenario analyses showed that the reduction in either underweight or overweight/obesity alone, and reduction in both underweight and overweight/obesity resulted in fewer deaths, more years of life lived and more QALYs gained. In the scenario where everyone was a healthy weight, the model estimated 577,267 fewer deaths (6.0% less), 2 million more years of life lived (0.3% more), and 3 million QALYs gained (0.6% more) over base-case results which represents current situation in Vietnam.

CONCLUSIONS

Our results suggest that addressing underweight and overweight/obesity will contribute to reducing deaths and increasing years of life lived and QALYs. Policies and interventions in alignment with Sustainable Development Goals to address underweight and overweight/obesity are necessary to achieve health for all.

Development of an online food frequency questionnaire and estimation of misreporting of energy intake during the COVID-19 pandemic among young adults in Peru

Background

The Young Lives longitudinal study switched to remote data collection methods including the adaptation of dietary intake assessment to online modes due to the physical contact restrictions imposed by the COVID-19 pandemic. This study aimed to describe the adaptation process and validation of an online quantitative food frequency questionnaire (FFQ) for Peruvian young adults.

Methods

A previously validated face-to-face FFQ for the adult Peruvian population was adapted to be administered through an online self-administered questionnaire using a multi-stage process. Questionnaire development was informed by experts' opinions and pilot surveys. FFQ validity was assessed by estimating misreporting of energy intake (EI) using the McCrory method, and the FFQ reliability with Cronbach alpha. Logistic regressions were used to examine associations of misreporting with sociodemographic, body mass index (BMI), and physical activity covariates.

Results

The FFQ was completed by 426 Peruvian young adults from urban and rural areas, among whom 31% were classified as misreporters, with most of them (16.2%) overreporting daily EI. Men had a lower risk of under-reporting and a higher risk of over-reporting (OR = 0.28 and 1.89). Participants without a higher education degree had a lower risk of under-reporting and a higher risk of over-reporting (OR = 2.18 and 0.36, respectively). No major difference in misreporting was found across age groups, areas, studying as the main activity, being physically active or sedentary, or BMI. Results showed good internal reliability for the overall FFQ (Cronbach alpha = 0.82).

Conclusion

Misreporting of EI was mostly explained by education level and sex across participants. Other sociodemographic characteristics, physical activity, sedentary behavior, and BMI did not explain the differences in EI misreporting. The adapted online FFQ proved to be reliable and valid for assessing dietary intakes among Peruvian young adults during the COVID pandemic. Further studies should aim at using and validating innovative dietary intake data collection methods, such as those described, for informing public health policies targeting malnutrition in different contexts after the COVID-19 pandemic.