The DHS Program User Forum: Nutrition and Anthropometry » Children age 6-59 months who are anaemic (<11.0 g/dl)

Home » Topics » Nutrition and Anthropometry » Children age 6-59 months who are anaemic (<11.0 g/dl) (Discrepancy in Calculating Childhood Anemia Prevalence from NFHS-5 Dataset)

Show: Today's Messages :: Show Polls :: Message Navigator

Children age 6-59 months who are anaemic (<11.0 g/dl) [message #30519]

Fri, 13 December 2024 12:55

NIkita@123
Messages: 2
Registered: December 2024

Member

I am trying to calculate the prevalence of anemia among children aged 659 months using the NFHS-5 dataset, but my results are not aligning with the NFHS-5 report. Below are the steps and issues encountered:

Current Approach
Dataset and Variables Used:
PR file of NFHS-5.

Code Used:

stata code
drop any_anem
gen any_anem = 0 if hc55 == 0 & hv103 == 1 & (hc1 > 5 & hc1 < 60) & hv042 == 1
replace any_anem = 1 if hc57 < 4 & any_anem != .
mean any_anem [iw=wt] if hv024 == 9, over(shdist)

Indicator Definition:
Children aged 659 months who are anemic (<11.0 g/dl).

Issue:
The calculated prevalence is not matching the reported figures in the NFHS-5 report.

alternatively if we could use the hc56 variable, which contains hemoglobin level in g/dl with 1 implied decimal place, codes for coverting the same to anemia

[Updated on: Fri, 13 December 2024 12:56]

Report message to a moderator

Re: Children age 6-59 months who are anaemic (<11.0 g/dl) [message #30542 is a reply to message #30519]

Tue, 17 December 2024 13:19

Janet-DHS
Messages: 938
Registered: April 2022

Senior Member

Following is a response from DHS staff member, Tom Pullum:

Please say specifically which table you are trying to match--or which indicator in STATcompiler--and the specific numerical value.

Report message to a moderator

Re: Children age 6-59 months who are anaemic (<11.0 g/dl) [message #30545 is a reply to message #30542]

Wed, 18 December 2024 01:02

NIkita@123
Messages: 2
Registered: December 2024

Member

indicator in STAT compiler
Indicator: Children with any anemia
Measure Percent
Definition Percentage of children under age 5 classified as having any anemia
Denominator Children age 6-59 months
Group Child Nutrition: Prevalence of anemia in children
Type Indicator
Decimals 1
Indicator ID CN_ANMC_C_ANY

I wanted to match the india (Prevalence: 67.1 n:152,752 as reported in NFHS5 report), region-wise prevalence of any anemia among children and the district factsheet for Sambhal (Uttar Pradesh), which reports a prevalence rate of 69.8%.

Codes used in STATA:
*Using PR file
drop any_anem
gen any_anem=0 if hc55==0 & hv103==1 & (hc1>5 & hc1<60) & hv042==1
replace any_anem=1 if hc57<4 & any_anem!=.
ta any_anem
ta any_anem [iw=wt] //India
mean any_anem [iw=wt] if hv024==9 , over(shdist) //not matching with report

***********
*Using KR file
drop any_anem
gen any_anem=0 if hw55==0 & b19>5
replace any_anem=1 if hw57<4 & any_anem!=.
ta any_anem
ta any_anem [iw=wt] //India

Report message to a moderator

Re: Children age 6-59 months who are anaemic (<11.0 g/dl) [message #30563 is a reply to message #30545]

Tue, 24 December 2024 14:46

Janet-DHS
Messages: 938
Registered: April 2022

Senior Member

Following is a response from DHS staff member, Tom Pullum:

Your code looks fine to me. I did some checking and basically I match you--but I do not match the report.

The KR file has 179,262 (weighted) children age 6-59 months with a valid hemoglobin measurement. Table 10.12 says there are 152,752 such children. I don't know how to get to 152,752.

Perhaps another user, or someone from IIPS, can help.

Report message to a moderator

Previous Topic:	Mozambique MAD indicator etc..
Next Topic:	Merge BR & HW data

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Tue Nov 4 04:09:59 Coordinated Universal Time 2025