Research

Abstract

Dams are a key and costly investment in agricultural productivity. Despite controversies over environmental degradation and community displacement, dams continue to be built with the expectation that they will generate benefits for society. This paper explores an understudied link between irrigation dams and children's health in India. Using two recent rounds of the National Family Health Survey, a global dam database, a high-resolution river and river basin data, a comprehensive agricultural panel survey, and numerous remote-sensed data, I find that irrigation dams constructed within river basins increase neonatal mortality incidences by approximately 7.4 percent, while no changes are observed in downstream areas. I show that these results can be linked to dam-induced changes in agrichemical exposure, highlighting the role of agricultural pollutants in increasing child health risks. Given the established link between infant health and later-life productivity and well-being, this paper highlights that the long-term benefits of irrigation dams may be greatly overestimated and there is need to better manage the unintended consequences of these irrigation dams in India.

Abstract
Seasonal changes in food availability and disease incidence put pressure on children’s health in Sub-Sahara Africa. Using year-round survey data from Senegal, we examine how seasonality in key health inputs (dietary diversity, diarrhea, and fever) helps predict seasonality in children’s health (weight-for-height z-score). We first parameterize seasonal variation in health and health inputs using second-order trigonometric polynomials, then decompose the seasonal curve of children’s health into component parts explained by seasonality in each health input. We find that lagged seasonality in disease incidence predicts seasonality in child health, while seasonality in dietary diversity does not – likely because diets are poor in Senegal even during the most food-plentiful part of the year. We also observe noticeable heterogeneity in the way these health inputs predict children’s health across different wealth levels and regions. 

Abstract

Background: Child wasting is known to exhibit seasonal patterns but few studies have examined how seasonality of wasting affects the tracking of wasting trends across multi-year periods.

Objective: We aimed to explore the seasonality of wasting in Senegal relative to changes over time and examine implications for tracking.

Methods: The average prevalence of child wasting (weight-for-height z-score < -2) and 95% confidence intervals were calculated by month and year from the continuous Demographic and Health Surveys (DHS) from 2012-2019. Peak and low wasting season estimates were defined as the 3 highest and lowest months of average wasting prevalence. A seasonally-adjusted annual estimate was generated using month-fixed effects linear regression, and compared with unadjusted estimates.

Results: Nationally, wasting fluctuated from 2013-2019, with the lowest annual prevalence of 6.0% (95% CI, 5.1 – 7.0%) recorded in 2014 and the highest in 2017 at 9.0% (95% CI, 8.3% - 9.7%).  Pooled across years, the peak wasting season prevalence was 8.7% (95% CI, 8.2% - 9.3%), and low wasting season prevalence was 6.3% (95% CI, 5.5% - 7.2%).  Considerable fluctuations in annual wasting prevalence were noted when data were exclusively sampled from either the peak wasting or low wasting seasons, but season-adjusted estimates did not differ notably from raw wasting prevalence estimates.

Conclusions: Seasonal fluctuations in the prevalence of wasting can be large enough to bias the interpretation of multi-year trends in situations where surveys are not conducted in the same season. Efforts should be made to conduct national surveys at the same time of year or adjust for seasonality when possible and tracking systems should report dates of data collection to facilitate interpretation of annual trends.

Description

Over the past few decades, Nepal has made significant progress in reforestation by transferring the management of state-owned forests to local communities. However, during the same period, both the frequency and intensity of forests fires have increased, largely due to drier and hotter dry-season and a rise in negligent and accidental fires. This project aims to assess the impacts of these forests fires, particularly on infants' health, due to their vulnerability to air pollutants. Using remote sensed measures of forest fire, air pollutant levels, and vegetation biomass, coupled with a detailed census of Community Forest User Groups and a nationally representative household survey, I measure the health costs of these forests fires and examine whether community forests mitigate or exacerbate these effects. 

In India, state election (Vidhan Sabha) results significantly impact constituencies, as state governments hold substantial power over policies and expenditures. These elections occur at the assembly level, a smaller administrative unit than districts. "Low-caste parties," or political parties focused on representing marginalized groups such as Scheduled Castes (SC), Scheduled Tribes (ST), and Other Backward Classes (OBC), can run for office. Using a fuzzy regression discontinuity design and assuming close election results are quasi-random, we examine the welfare impacts when low-caste parties win elections, particularly on health outcomes, which have historically been worse for these marginalized households. We also test whether these effects are lasting or depend on subsequent elections.