Authors' response

The issues raised by Gupta et al1 have been stated in our published article2. Data from NFHS series showed that there was no consistent decline in prevalence of anaemia in pregnancy between 1998 and 2015 at State or national level. Data from DLHS 2, 4 and AHS CAB showed decline in prevalence and severity of anaemia in all States except Uttarakhand (where AHS CAB survey was not done in all districts)2. Factors associated with reduction in anaemia in different States were not explored in the published article2.

Global and Indian studies have shown that Hemo Cue is not an accurate method for Hb estimation. Use of this convenient but not accurate method could have been the reason for the reported lower levels of anaemia, as well as lack of consistent decline in prevalence of anaemia in NFHS series. Cyanmethaemoglobin method is the gold standard method for Hb estimation; DLHS 2, 4 and AHS CAB used this method and showed that in 2002-2004 prevalence of anaemia was high and in 2014-2015 there was a reduction in prevalence of anaemia in pregnancy.

Data on prevalence of anaemia in pregnancy from NFHS series (NFHS 2, 3 and 4) were compared (not combined) with DLHS 2, 4, and AHS CAB. A national team of experts designed the sampling frames for all these massive surveys, so that State and national level estimates of parameters including prevalence of anaemia could be made.