India is one of the world's hotspots for undernourished children, both in terms of prevalence and absolute numbers. Successive rounds of National Family Health Surveys (NFHS) have revealed that the progress observed since the early 1990s is far from what is expected when compared to the country's economic growth.
The Lancet study reports an astonishing number (i.e. 20%, 18.2%, 10.6%, 17.2%, 17.8%, respectively in 1992-93, 1998-99, 2005-06, 2015-16, 2019-21) of infants and young children with “zero food” consumption in last 24 hours prior to the survey date of successive National Family Health Surveys. The study also reports an increase in “zero-food” intake in India during 2015-16 (17.2%) to 2019-21 (17.8%). They have also reported a rise in “zero-food” intake in several states. The increase reported for states such as Chhattisgarh and Uttar Pradesh have certainly pushed a panic button.
Considering the serious shock waves these estimates have created among the Indian policymakers and popular media, we have evaluated the accuracy of the estimates reported by the Lancet study in this commentary note. Although we acknowledge that India’s rank (107 out of 121 countries) in the Global Hunger Index is not encouraging, the estimates reported by the Lancet study with a newly derived concept of "zero food" intake is completely misleading and technically incorrect. Below, we have systematically highlighted where they might have gone wrong.
Despite the authors’ recognition of the contribution of breastmilk to the overall calorific requirement of children in different age brackets (page 3, line no. 6 to 24), question arises as to why do the authors consider breastmilk as having "zero calorific” significance in their definition of “zero-food” intake, even for infants?
Further, to validate their estimates and find the true magnitude of the problem, we have replicated the authors' methodology to define “zero food” and re-produced the estimates for the latest two rounds of NFHS (2015-16 and 2019-21). The latest two rounds are better comparable in terms of their sampling design and the questions used for estimating “zero food” by the Lancet study.
However, our sample size is 92 less and 218 more for 2015-16 and 2019-21, respectively. Unlike what was stated in the Lancet study, for the latest two rounds of NFHS, there is no need to make any re-adjustment of boundaries, and we assume that this could be the one reason why there is a minor difference in the sample sizes for our estimates compared to them.
Using a similar definition of "zero food", comparison of our estimates with the Lancet study suggests a slight difference which can be attributed to sample size variations, as pointed above. However, the state-level patterns align with the Lancet study estimates.
After adjusting to breastfeeding (Figure 1), we found that the percentage of children without any food (and also not breastfed) is much less (1.1% in 2015-16 and 1.3% in 2019-21) compared to what is reported in the Lancet study (17.2% in 2015-16 and 17.8% in 2019-21). Similar differences are also observed across the states.
However, we do agree with the authors that even after adjusting for breastfeeding, both at the all-India level and also across the considerable number of states “zero food” intake have increased. Given that overall levels of zero food intake including breastfed, is very small, it is difficult to predict why these children are not given anything in the past 24 hours prior to the survey.
Apart from the assumption that these families do not have anything to feed their children, the other reason could be that these children are sufficiently fed through breastmilk. However, this can’t be verified accurately using limited information given in the NFHS.
However, we do agree with the authors that even after adjusting for breastfeeding, both at the all-India level and also across the considerable number of states “zero food” intake have increased. Given that overall levels of zero food intake including breastfed, is very small, it is difficult to predict why these children are not given anything in the past 24 hours prior to the survey.
Apart from the assumption that these families do not have anything to feed their children, the other reason could be that these children are sufficiently fed through breastmilk. However, this can’t be verified accurately using limited information given in the NFHS.
In conclusion, we posit that the concept of "zero food" intake invented by the Lancet study is misleading to the readers and especially for the policymakers. Our main contention with Lancet study' methodology of defining "zero food” is exclusion of breast milk out of the definition of “food”.
If mothers opt for breastfeeding over complementary feeding, it can also be a conscious decision that they have taken based on their capacity to produce to sufficient breast milk for their babies, especially those with younger children i.e. 6 to 11 months.
---
*Srinivas Goli is Associate Professor, International Institute for Population Sciences (IIPS), Mumbai; Shalem Balla is Project Officer, IIPS Mumbai; Harchand Ram is Research Analyst, IIPS Mumbai
If mothers opt for breastfeeding over complementary feeding, it can also be a conscious decision that they have taken based on their capacity to produce to sufficient breast milk for their babies, especially those with younger children i.e. 6 to 11 months.
Assessing the age-wise “zero food” intake results presented by the Lancet study in Table 6 and the age-wise contribution of children to "zero food" intake shown in Figure 2 in this note suggest that a major share (about 65%) of children with “zero food” intake belongs to 6-11 months who have a greater chance to be breastfed and accomplish calorific sufficiency.
Further, our estimates in Figure 3 show nearly 93% of children under 6 to 23 months were being currently breastfed at the time of the survey and this number is even higher for younger children i.e. 6 to 11 months. Therefore, in the absence of breastfeeding from the definition of “food”, the newly designed concept of “zero food" intake artificially inflates the numbers and creates unwarranted alarm among readers and policy makers.---
*Srinivas Goli is Associate Professor, International Institute for Population Sciences (IIPS), Mumbai; Shalem Balla is Project Officer, IIPS Mumbai; Harchand Ram is Research Analyst, IIPS Mumbai
Comments
Further they "suggest that a major share (about 65%) of children with “zero food” intake belongs to 6-11 months who have a greater chance to be breastfed and accomplish calorific sufficiency." Is UNICEF changing its recommendation on the need for complementary feeding at teh age of 6 months? Is it being suggested that caloric sufficiency is the only goal of IYCF? I am baffled by this blog from these authors.
Don't read between the lines. The full sentence is "Apart from the assumption that these families do not have anything to feed their children, the other reason could be that these children are sufficiently fed through breastmilk. However, this can’t be verified accurately using limited information given in the NFHS". We are trying to tell that the 1.3% of the children who are not eating anything in past 24 hours can belong to any of the above cited categories which can't be verified with the limited information. We are not saying that they have caloric sufficiency through anyone means.
Moreover, none of us belong to UNICEF, rather two of our RAs working in a Project funded by UNICEF at the IIPS, Mumbai. These opinions are independent views of the authors. Nothing to do with UNICEF mandate.