Home Nutrition, Vitamins, Minerals and Herbs Child Nnutrition ENSO impacts child undernutrition in the global tropics

ENSO impacts child undernutrition in the global tropics

ENSO impacts child undernutrition in the global tropics

Estimating the global child nutrition effects of ENSO

We capture ENSO variation (Fig. 1a) with the widely used NINO3.4 index of equatorial Pacific sea surface temperature27,28,29, which spans 5°N–5°S, 170°W–120°W. Children’s weight-for-age z-scores (WAZ) at the time of survey (Fig. 1b) are calculated using the National Center for Health Statistics/Centers for Disease Control and Prevention/World Health Organization (NCHS/CDC/WHO) International Reference Standard30 intended to provide a single measure of child nutritional outcomes comparable across ages and sexes. We first identify all countries with local climates teleconnected to ENSO (Fig. 1c) for which DHS anthropometric data exist. This yields a sample of 1.3 million children aged 0–4 years interviewed in 186 household surveys between 1986 and 2018. The sample includes 51 countries containing 38% of the world’s population and 48% of the world’s under-5 population as of 2018. We assign treatment (i.e., the ENSO state when the child was surveyed) annually by tropical year, accounting for typical annual timing in ENSO state change8,29, by calculating the mean NINO3.4 Sea Surface Temperature (SST) value between May and December of a given year. We assign that to all children interviewed by DHS during that period, as well as all children interviewed during the following year’s January–April months, i.e., before the following year’s “spring barrier” (see “Methods”).

Fig. 1: Defining the sample of teleconnected children.

a ENSO time series. El Niño or La Niña highlighted in red or blue, respectively. El Niño and La Niña states are defined as follows: when the maximum of a 3-month rolling mean of monthly Oceanic Niño Index (ONI) values is >0.5 °C (Niño-like) or <−0.5 °C (Niña-like) compared to a moving reference climatology following NOAA CPC guidelines38. b Weight-for-age z-score distribution over time in teleconnected countries (n = 1,253,176 children from 51 surveys). Box plots indicate median (middle white line), 25th, 75th percentile (box), and 5th and 95th percentile (whiskers) as well as outliers (single points). Country composition within each year is different, as a rotating sample of countries is surveyed in each year under the DHS program. c Pixel-level monthly correlation of surface temperature (1980–2010) from the UDEL climate dataset and 2-month lag of NINO3.4 Sea Surface Temperature (SST) in teleconnected locations. Teleconnections are defined as pixels where the local temperature shows ≥3 statistically significant months of correlation with the second month lag of NINO3.4 SSTs. Country boundaries indicate sample countries (those having at least 50% of the population living in locations where local temperature is significantly correlated with the second month lag (t – 2) of the NINO3.4 SST index for at least 3 months of the year and with at least two Demographic and Health Surveys measuring anthropometrics). d Pixel-level monthly correlation of precipitation (1980–2010) and 2-month lag NINO3.4 SST. There is substantial heterogeneity in how precipitation is affected by ENSO, with areas of both positive and negative correlation. Country boundaries again show sample countries.

While a warmer ENSO leads to higher temperature throughout the tropics, shifts in precipitation patterns lead to some areas getting wetter than normal while others get drier. We account for potential differences in the effects of ENSO by estimating separate responses in subnational regions where precipitation is positively correlated to warmer ENSO (Fig. 1d) vs. negatively correlated. Since only 6.4% of our sample lives in regions where warmer ENSO leads to clear wet anomalies, we largely focus our discussion on results for the majority of the sample.

The empirical distribution of WAZ is significantly and substantially different (p < 0.001) between El Niño and La Niña years, even in the absence of controls (Fig. 2a). A key aspect of our research design, however, rests on exploiting the temporal variability of the ENSO cycle. While ENSO follows a variety of non-random patterns, such as the general progression from El-Niño state to La Niña state, the timing of event occurrence is sufficiently stochastic that even state of the art models have limited prediction skill beyond 6 months into the future31. We thus use variation in ENSO anomalies—measured as a deviation from long-run average conditions—in order to statistically isolate the effect of variation in ENSO state on child malnutrition. Following standard practice in the climate impacts literature32, we purge the estimates of potentially confounding average differences33,34 between countries and within them based on rural and urban areas using fixed effects/indicator variables for spatial location, detrend the data by major world regions using an annual trend, remove monthly seasonality by major world regions using month fixed effects, and include country-specific controls for mother’s age at child’s birth and total years of mother’s education. Our results correspond to comparing children surveyed at different times in the same country but under different ENSO conditions, while appropriately detrending the data and controlling for average health differences across countries and regions.

Fig. 2: Negative effects of ENSO on child weight-for-age.

a Distribution of detrended weight-for-age z-scores (WAZ) during years classified as El Niño (red) and La Niña (blue) according to NOAA definition using NINO3.4 SSTs, with means of each distribution shown. b Epanechnikov kernel-weighted local polynomial (bandwidth 0.7) estimate of Table 1. Model 1 showing conditional association of WAZ with ENSO, differentiating countries where precipitation is negatively correlated with ENSO for >3 months in the year over >50% of country area (green) and where it is positively correlated (orange). 95% confidence intervals are shown for the estimated curves. Controls include fixed effects (indicators) for each country; country-specific mother’s age at child’s birth, total years of mother’s education, and rural vs. urban indicator; as well as UNICEF world region-specific linear trends in survey year and fixed effects for the month of interview. Histograms represent the number of observations in each precipitation correlation subsample. c Effects of ENSO on WAZ within each decade (blue) and UNICEF world region (green) in the sample (n = 1,253,176 children from 51 surveys), estimated using only locations with non-positive precipitation teleconnections. Dots signify point estimates, bars signify 95% confidence intervals, and gray shaded region and dashed line show main effect from Table 1. The 95% confidence intervals for each decade are: 1980s: [−0.070, 0.011], 1990s: [−0.086, −0.002], 2000s: [−0.063, 0.026], 2010s: [−0.047, 0.021]; and for each region are: Latin America [−0.087, −0.026], Sub-Saharan Africa [−0.050, 0.003], South Asia [−0.086, 0.006].

ENSO’s effects on contemporaneous nutrition

We estimate that a 1 °C increase in the ENSO index is associated with 0.03σ (p = 0.02) average decrease in WAZ after detrending the data and controlling for location-specific unobservable confounders and mother characteristics (Table 1). We allow the relationship between ENSO and WAZ to vary flexibly (Fig. 2b) and find that the negative association remains across the distribution of ENSO values. The result is substantively similar across a broad range of model specifications (Supplementary Table 1) and across other outcomes reflecting recent nutrition, including weight-for-height and body mass index (BMI; −0.04σ/°C and p < 0.01 for both measures). Using WHO z-score classification thresholds, warmer ENSO increases the prevalence of underweight (below −2σ in weight-for-age) significantly by 0.6 percentage points per 1 °C (p < 0.05). We find that the risk of wasting (below −2σ in weight-for-height) is similarly positive but not significant (0.3 p.p./°C, p = 0.21), consistent with higher measurement error in height measurements due to the difficulty of measuring child height/length compared to weight35,36, which decreases the precision of our estimates. All of these patterns are reversed in the minority of places in our sample (6.4%) where warmer ENSO is correlated to wet anomalies. The heterogeneity in results across regions of wet and dry anomalies points toward the importance of agriculture in mediating the ENSO–nutrition link, though others (e.g., conflict) cannot be ruled out.

Table 1 Anthropometric effects of ENSO.

Comparing the 2015 El Niño to large-scale nutrition interventions

The several degree variation in ENSO cycle implies that it is a meaningful source of variation in population nutrition in the tropics. To give context to the size of these effects, we provide illustrative order-of-magnitude calculations of the scale of public health interventions needed to offset undernutrition on the scale we estimate was caused by the 2015 El Niño, using published effect sizes of nutritional interventions37. According to our results, the 1.92 °C increase in the detrended mean NINO3.4 index during the 2015 El Niño event38, one of the largest on record, likely caused average WAZ in the representative child of our sample countries to decrease by 0.078σ based on the average treatment effect estimated in Supplementary Table 5 column 1. The human scale of this impact is large given that the under-5 population in our sample countries was 311 million in 2015. By calculating the effect size of the 2015 El Niño summed over all children and dividing by the mean effect size for each nutrition intervention, Fig. 3 shows that offsetting the effects of the 2015 El Niño would require approximately 134 million children receiving multiple micronutrient supplementation (confidence interval (CI) 75–193 million) or 72 million (CI 33–105 million) receiving provision of complementary foods or 72 million (CI 26–118) receiving nutrition education. The effect of the 2015 El Niño is also equivalent to the WAZ reduction from moving 46 million children from urban to rural areas, based on our model results. Using the same 1.92 °C increase and the coefficient in Supplementary Table 5 column 4, the 2015 El Niño increased risk of being below the WHO threshold for underweight by 1.9 percentage points, i.e., an increase of nearly a tenth of the current population rate of 24%. This corresponds to an additional 5.9 million children being driven into underweight status.

Fig. 3: Interventions required to offset 2015 El Niño.

Millions of children who must be targeted with specific interventions in order to reverse the effects on malnourishment caused by the 2015 El Niño. Effect sizes calculated using treatment effects in Bhutta et al.37. Bars represent the central estimate, with whiskers representing the 95% confidence interval of these estimates. See supplementary information for details of calculation.

Robustness and implications of ENSO impacts

Our main result is consistent across alternative specifications, observation weighting, and ENSO variable definitions, as well as across age categories within the sample (Supplementary Tables 14). Results are robust to a variety of model specifications controlling for plausible observable and unobservable factors (Supplementary Table 1). While our main results weigh observations so that interpretations are the effect of ENSO on a child in the average country, we also calculate effect on the average child in all sample countries in order to estimate global effects (Fig. 3), and we show that results are consistent if no observation weights are used (Supplementary Table 2). Alternative indicators for ENSO state yield similar results (Supplementary Table 3), with positive deviations from the mean ENSO state decreasing anthropometric z-scores and negative deviations from the mean ENSO state (La Niña events) reducing undernutrition (opposite patterns occur in the few places where precipitation increases with ENSO SST). Coefficients for alternate definitions of ENSO are not statistically distinguishable from our main effect. Supplementary Table 4 estimates the effects of ENSO allowing for different effects by child age categories of 0–5, 6–11, 12–23, 24–35, and 36–59 months. With few exceptions, coefficients are consistent in sign and magnitude across age groups for each outcome variable and with the corresponding coefficient in Table 1.

For the purposes of population-wide attribution statistics, we also calculate the average effect of warmer ENSO in the average country, without separating the sample by whether warm ENSO leads to dry or wet anomalies (Supplementary Table 5). The average effect across the sample suggests that warmer ENSO leads to a 0.04σ/°C reduction in weight-for-age (p = 0.02), and a 1 percentage point increase in prevalence of underweight (p < 0.01). We also test the use of alternative detrending of the data with decade fixed effects to guard against the possibility that results are an artifact of detrending specification (Supplementary Table 6). The lagged effects of ENSO (Supplementary Table 7) indicates no persistent effect of ENSO on child weight-for-age, weight-for-height, or BMI, except in the subsample with positively correlated rainfall. This is consistent with child weight recovering quickly once nutrition returns to adequate levels. On the other hand, child stunting remains affected years after negative shocks from ENSO (Supplementary Table 8), consistent with height being slower to respond to health shocks than weight20 and with the first 2 years of life being the riskiest period for growth faltering due to scarring25,39.

Supplementary Tables 9 and 10 show that results are robust to alternative definitions of teleconnection. Supplementary Table 9 extends the main sample to include countries that are teleconnected with NINO3.4 in terms of precipitation but not temperature, while Supplementary Table 10 restricts the sample to include only those countries that have both a significant teleconnection with ENSO via temperatures and precipitation. Supplementary Figs. 2 and 3 and Supplementary Tables 12 and 13 explore whether the ENSO state might affect the timing of DHS surveys within the year and therefore spuriously lead to changes in child anthropometrics due to seasonality. The timing of DHS surveys does not vary as a result of ENSO state (Supplementary Fig. 2 and Supplementary Table 12). ENSO’s effect is evident regardless of what time of year the child was surveyed (Supplementary Fig. 3), indicating that effects are not limited to a specific part of the growing or post-harvest season.

Supplementary Table 14 varies the standard error adjustments for serial and spatial autocorrelation and shows that results remain unchanged. Supplementary Table 15 shows that employing logistic regressions for the dichotomous outcome variables results in odd ratios with the same qualitative interpretation as the corresponding linear probability models in Supplementary Table 1. Finally, teleconnected areas remain unchanged across different NINO SST indices (Supplementary Fig. 1) and results hold under a placebo randomization test (Supplementary Fig. 4).

While our estimates of ENSO’s effect on child undernutrition are robust, there are nonetheless limitations imposed by both the nature of the data and structure of this research design. DHS data only selectively report migration, making it difficult to deal with any possible migration into or out of the sample that might occur in response to the ENSO cycle. Sufficiently severe ENSO events may also differentially influence the likelihood of being in sample, both at local scale, where, e.g., worse-impacted children may be less likely to end up surveyed due to mortality or illness, and at larger scales, where events such as civil conflict that are known to respond to ENSO8 may plausibly inhibit the DHS’s ability to gather data or ensure data quality. While these aspects of sample selection may lead to unavoidable biases in our results, missing more vulnerable populations would likely bias us away from finding an effect of ENSO on health. Moreover, the consistency of the result across specifications and subsamples suggests that the influence of these limitations on the overall result is likely small.

Source link


Please enter your comment!
Please enter your name here

For security, use of Google's reCAPTCHA service is required which is subject to the Google Privacy Policy and Terms of Use.

I agree to these terms.

en English