Geo-visualizing Diet, Anthropometric and Clinical Indicators for Children in India

The nutritional status of children in India fares much worse in global comparisons. Using the disaggregated data from the fourth National Family Health Survey (NFHS 2015-16), we present a dashboard and atlas for 31 nutritional indicators that include diet, anthropometry, clinical and service utilization measures of child nutrition for the districts of India. The geo-visualizations are presented with a motivation to help various stakeholders prioritise indicators and districts for interventions. 

Notwithstanding the value of utilising the summary data for select indicators made available by the fifth NFHS (2019-20) for 17 states and 5 Union Territories, it is important to note that a truly all-India picture covering all districts of India will not be available at least until the later part of 2021. Even as we await the availability of disaggregated data for a full range of indicators, there remains much to be investigated and learned from a more detailed examination of the fourth round of the NFHS.

Visualization

Authors: Akhil Kumar, Weixing Zhang, S V Subramanian. Dashboard on Geo-visualising Diet, Anthropometric and Clinical Indicators for Children in India. December 2020, Cambridge MA, Harvard Center for Population and Development Studies.

NOTE: The legend and rank show the Prevalence-Headcount Metric (PHM) which was calculated by adding the normalized values of prevalence and headcount together. Blank areas with unavailable data.

Methodology & Indicators

PHM Methodology

We estimated the burden for each of the nutritional deprivation indicators along two dimensions of Prevalence (P) and Headcount (H), and combined them to derive a Prevalence-Headcount Metric (PHM).

Prevalence

The metric P was calculated as children with nutritional deprivation (q) divided by the eligible sample of children (n) in the district (j) and expressed in percentage as:

Pj = (qj / nj) × 100

The P metric quantifies the risk of a child experiencing nutritional deprivation in a district. For example, in the Kupwara district of Jammu and Kashmir, 137 (q) out of 435 (n) sample of eligible children were stunted, translating into a district prevalence of 31.5%; in other words, one out of every three children is at risk of being stunted. Thus, the P metric helps identify districts where the future risk of a specific nutritional deprivation should be reduced.

However, the P metric does not contain any information on the absolute number of children who are at risk because it does not take into account the total population of children in a district. For example, consider the districts of Hyderabad (16.7%) and South Garo Hills (16.6%); both have the same P metric but differ with regards to the under-five population of 398,513 and 23,953, respectively, translating to different levels of absolute burden.

Headcount

The metric H is given as the product of P and the total eligible population N for each district.

Hj = Pj × Nj

Returning to the above example, the number (H) of stunted children is substantially large in Hyderabad (66,553) than South Garo Hills (3,970) despite both districts having the same prevalence because the total population burden by nutrition deprivation in Hyderabad is substantially larger.

Prevalence-Headcount Metric

We developed a combined Prevalence-Headcount metric (PHM) that takes into account the features of both the risk (P) as well as the headcount (H) to provide a comprehensive picture of the burden of nutritional deprivation in a district. We computed the PHM using the following steps. We exemplify these steps using the district of Kupwara, Jammu and Kashmir for the nutritional deprivation indicator of stunting.

P = Prevalence; H = Headcount;

PHM = Prevalence-headcount metric;

j = District;

q = Number of children with nutritional deprivation within the eligible sample;

n = Eligible sample; N = Eligible population;

norm = Normalized; max = district with the Maximum value;

min = district with the Minimum value;

STEP 1: Calculating Prevalence

Formula: Pj = (qj / nj) x 100

Examples: qj = 137

                    nj = 435

                    Pj = (137 / 435) x 100 = 32%

STEP 2: Calculating Headcount

Formula: Hj = Pj x Nj

Examples: Pj = 32%

                     Nj = 166791

                     Hj = (32 / 100) x 166791 = 53373

STEP 3: Normalizing the Prevalence

Formula: Pjnorm = (Pj - P(min)) / (P(max) - P(min))

Examples: Pj = 32%

                     P(max) = 65%

                     P(min) = 13%

                     Pjnorm = (32 - 13) / (65 - 13) = 0.365

STEP 4: Normalizing the Headcount

Formula: Hjnorm = (Hj - H(min)) / (H(max) - H(min))

Examples: Hj = 53373

                     H(max) = 460209

                     H(min) = 346

                     Hjnorm = (53373 - 346) / (460209 - 346) = 0.115

STEP 5: Calculating the PHM

Formula: PHMj = (Pjnorm + Hjnorm)/2

Examples: Pjnorm = 0.365

                     Hjnorm = 0.115

                     PHMj = (0.365 + 0.115) / 2 = 0.24

 

Types of Nutritional Deficit

Indicator

Age Group

Definition

Diet and Anthroprometric Failure

6-23 months

Children with both dietary as well as anthropometric failures

Diet Failure Only

6-23 months

Children with at least one or more dietary failures but no anthropometric failure

Anthropometric Failure Only

6-23 months

Children with at least one or more anthropometric failures but no dietary failure

Dietary Measures

Indicator

Age Group

Definition

Inadequate Diet

6-23 months

Children who did not receive minimum acceptable diet

Inadequate Diet Diversity

6-23 months

Children who did not receive minimum dietary diversity

No Solid/Semi-Solid Food

6-23 months

Children who did not consume solid or semi solid food in the day or night preceding the interview

No Dairy

6-23 months

Children who did not consume milk and milk products in the day or night preceding the interview

No Nuts/Legumes

6-23 months

Children who did not consume nuts and legumes in the day or night preceding the interview

No Grains/Roots/Tubers

6-23 months

Children who did not consume grains in the day or night preceding the interview

No Eggs

6-23 months

Children who did not consume eggs in the day or night preceding the interview

No Flesh Foods

6-23 months

Children who did not consume fish, chicken, meat in the day or night preceding the interview

No Vit-A Rich Fruits/Vegetables

6-23 months

Children who did not consume Vit-A rich fruits and vegetables in the day or night preceding the interview

No Other Fruits/Vegetables

6-23 months

Children who did not consume other fruits and Vegetables in the day or night preceding the interview

Anthropometric/Clinical Measures

Indicator

Age Group

Definition

Stunting or Underweight or Wasting

0-59 months

Children who are either stunted, wasted or underweight

Stunting & Underweight & Wasting

0-59 months

Children who are stunted, underweight and wasted

Stunting & Underweight

0-59 months

Children who are stunted and underweight but not wasted

Underweight & Wasting

0-59 months

Children who are underweight and wasted but not stunted

Stunting

0-59 months

Children who are stunted (short height-to-age)

Severe Stunting

0-59 months

Children who are severly stunted (short height-to-age)

Underweight

0-59 months

Children who are underweight (low weight-to-age)

Severe Underweight

0-59 months

Children who are severely underweight (low weight-to-age)

Wasting

0-59 months

Children who are wasted (low weight-to-height)

Severe Wasting

0-59 months

Children who are severely wasted (low weight-to-height)

Anemia

6-59 months

Children with hemoglobin level less than 11.0 g/dL

Severe Anemia

6-59 months

Children with hemoglobin level less than 7.0 g/dL

Low Birth Weight

0-59 months

Children with written record of birthweight less than 2.5 kg

Breastfeeding Practices

Indicator

Age Group

Definition

No Early Breastfeeding

0-12 months

Children who were not breastfed within 1 hour of birth

No Exclusive Breastfeeding

0-12 months

Children who were not exclusively breastfed

Service Utilization

Indicator

Age Group

Definition

No Hot Cooked Meal

> 36 months

Children who did not receive supplemnetary nutrition under ICDS

No Take Home Ration

6-36 months

Children who did not receive supplemnetary nutrition under ICDS

No Vit-A Supplementation

6-59 months

Children who did not receive Vit-A dose in the six months preceding the survey

Publications

Gausman J, Perkins JM, Lee H-Y, et al. Ecological and social patterns of child dietary diversity in India: a population-based study. Nutrition. 2018;53 :77-84.Abstract

Dietary diversity (DD) measures dietary variation in children. Factors
at the child, community, and state levels may be associated with poor
child nutritional outcomes. However, few studies have examined the role
of macro-level factors on child DD. This study seeks to 1) describe the
distribution of child DD in India, 2) examine the variation in DD attributable
to the child, community and state levels, and 3) explore the relationship
between community socioeconomic context and child DD. Using nationally
representative data from children aged 6-23 months in India, multilevel
models were used to determine the associations between child DD and
individual- and community-level factors. There was substantial variation in
child DD score across demographic and socioeconomic characteristics. In an
age and sex-only adjusted regression model, the largest portion of variation
in child DD was attributable to the child level (75%) while the portions of
variance attributable to the community-level and state level were similar to
each other (15% and 11%). Including individual-level socioeconomic factors
explained 35.6 percent of the total variation attributed to child DD at the
community level and 24.8 percent of the total variation attributed to child
DD at the state level. Finally, measures of community disadvantage were
associated with child DD in when added to the fully adjusted model. This
study suggests that both individual and contextual factors are associated with
child DD. These results suggest that a population-based approach combined
with a targeted intervention for at-risk children may be needed to improve
child DD in India.

Rajpal S, Joe W, Subramanian SV. Living on the edge? Sensitivity of child undernutrition prevalence to bodyweight shocks in the context of the 2020 national lockdown strategy in India. Journal of Global Health Science. 2020;2.Abstract

The National Family Health Survey (NFHS) 2015–16, finds that every second
child in India suffers from at least one form of nutrition failure. Dichotomised
indicators of underweight and wasting based on z-score cut-off does not
provide any information regarding those children who are clustered around
the threshold and are at an elevated risk of undernutrition through any minor
weight-loss. This paper aims to estimate the effect of bodyweight shocks on
net increments in the prevalence of child underweight and wasting among
the poorest households in India. We used cross-sectional information from
NFHS 2015–16 to estimate possible increase in the prevalence of child
underweight and wasting as a result of reduction in their bodyweight. The
shocks are presumed to range from a minimum of 0.5% to a maximum
5% reduction in the bodyweight for every child from the poorest 20%
households. Various raw weight measures scenarios were developed and
transformed into age- specific z-scores using World Health Organization child
growth standards. Nutritional status of children is sensitive to smallest of the
shocks to bodyweight. In fact, a reduction of 0.5 and 1 percent in weight can
lead to substantial increase in underweight and wasting prevalence. Under a
scenario of bodyweight shock of 0.5 percent, the prevalence of underweight
and wasting will increase by 1.42 and 1.36 percentage points, respectively.
These estimates get translated into 410,413 and 392,886 additional cases
of underweight and wasting, respectively. With such high concentration of
children around the undernutrition threshold, any minor shock to nutritional
health of the children can have major implications. In the current scenario of
national lockdown and restrictions due to coronavirus disease 2019 pandemic,
it is critical to ensure an uninterrupted supply of nutritious meals and food
supplements to the poor children while arresting the infection spread.

Subramanyam M. Why childhood under-nutrition persists in India and how to intervene. The Indian Express. 2012.Abstract

Recently, Prime Minister Manmohan Singh released a survey on child under-nutrition in rural India in 2010-11 (Hunger and Malnutrition Survey,HUNGaMA). Sadly,the new data reinforced the existence of an India marked by substantially low levels of something absolutely vital for adequate human development. The survey found that 42 per cent of the under-five children were underweight and 59 per cent were stunted in the 100 focus districts. Remarkably,in six districts with the best child development index,the prevalence of underweight (33 per cent) and stunting (43 per cent) among children,while somewhat lower,was still substantially high — suggesting the endemic and persistent nature of the under-nutrition burden. Even though child under-nutrition remains very high,do the data from HUNGaMA suggest an improvement over previous assessments? Data from the district-level health survey (DLHS) of 2002-2004 provide some answers. The DLHS includes data on underweight among children under six from hundreds of districts across India. In the 100 focus districts,the prevalence of underweight appeared to have reduced 11 percentage points from 53 per cent in the DLHS to 42 per cent in the HUNGaMA Survey. A similar comparison of changes in the prevalence of stunting is not possible since DLHS did not measure the height of children. Other aspects of the results from the HUNGaMA survey reiterate older patterns. For instance,under-nutrition is inversely associated with socio-economic status; thus children from low income households or whose mother had low levels of education have higher prevalence of under-nutrition.

  •  
  • 1 of 11
  • »
More Publications