Response to the study by V. Kumaravel et al., titled "Are the current Indian growth charts really representative?".

Sir,

We read the article entitled “Are the current Indian growth charts really representative? Analysis of anthropometric assessment of school children in a South Indian district”, with interest.[1] Authors objective was to assess the anthropometry of 5-18-year-old school children, estimate the prevalence of childhood thinness, overweight and obesity and to assess how their population compares with that of Agarwal et al. Authors have used Agarwal's growth charts which have been recommended by the IAP for classifying children together with the IOTF standards.[2]

While the dataset is large and interesting, it is quite surprising that authors have analysed both socio-economic classes together for most results. Also, in our opinion, to judge how the authors’ population fared in relation to Agarwal et al.'s and IOTF data, it would have been more appropriate to derive Z scores based on the two datasets.

Kumaravel et al., have mentioned that in the Khadilkar and Marwaha data, the height, weight, and BMI cut-offs are higher.[3] Our study (Khadilkar et al., 2009) was conducted as India is still in a phase of nutrition transition and it is suggested that in developing countries, growth references should be regularly updated.[3] Also, there are doubts as to whether two-decade-old data is representative of the growth of present day Indian children. To correct for the higher weights in Khadilkar et al.'s study, prescriptive BMI charts have been published for Indian children.[4] In this study, we have defined BMI cut offs for obesity and overweight in such a way that they were linked to adult 23 and 28 equivalent cut-offs which are the recommended Asian cut-offs for defining overweight and obesity by WHO. These cut offs have been derived keeping metabolic risks in mind.[4] These charts and cut-offs are designed to be applicable irrespective of the socioeconomic class.

Thus, when comparing with current growth charts, it would have been appropriate to refer to this paper as well as Cole's extended BMI for Asian children.[5] Without these comparisons, the authors’ conclusion of obesity being under diagnosed when using Indian charts is, in our opinion, not on firm ground.

It would thus be interesting to see how the data by Kumaravel et al., performs when compared with prescriptive BMI charts appropriate for present day Indian children.[4]