The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Ethiopia » children mortality rate for Ethiopia's KR file
children mortality rate for Ethiopia's KR file [message #21895] Wed, 06 January 2021 12:29 Go to next message
id709nvz is currently offline  id709nvz
Messages: 29
Registered: March 2019
Member
Dear senior members,
I want to calculate children mortality rate in KR file for Ethiopia DHS 2005, 2011, 2016
However, age at death (months) for DHS 2005 is not available. How can I construct this in Stata, please?

I saw the codes in Github, and answers in the forums.

Kind regards
Zecharias
Re: children mortality rate for Ethiopia's KR file [message #21908 is a reply to message #21895] Thu, 07 January 2021 10:04 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2253
Registered: February 2013
Senior Member

Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

First, it is very important that you use the BR file for the under-five rates, NOT the KR file. The rates include exposure (or risk) from children who were born more than 5 years ago. The b variable that you want, b7, IS in ETBR51FL.dta. It is also in ETKR51FL.dta. I don't know why it would not show up. Perhaps you have modified the file somehow?

Re: children mortality rate for Ethiopia's KR file [message #21913 is a reply to message #21908] Thu, 07 January 2021 12:44 Go to previous messageGo to next message
id709nvz is currently offline  id709nvz
Messages: 29
Registered: March 2019
Member
Dear,
Thank you. You're right. I realize that I have dropped dead children in the first place since I wanted to use only alive children.
The reason I am using KR rather than BR is that I want to analyse children under five whose mother is also in the household
Re: children mortality rate for Ethiopia's KR file [message #21917 is a reply to message #21913] Thu, 07 January 2021 13:45 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2253
Registered: February 2013
Senior Member

Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

The variable b9 in the birth history tells you whether the mother and child are in the same household. That variable is in both the KR and BR files. I will repeat that with just the KR file you cannot calculate the under-five death rates. You can tell whether a child born in the past five years has died, and if so, at what age, but that's not sufficient for calculating the under-five death rates for the five years before the survey.

Re: children mortality rate for Ethiopia's KR file [message #21918 is a reply to message #21917] Thu, 07 January 2021 14:48 Go to previous messageGo to next message
id709nvz is currently offline  id709nvz
Messages: 29
Registered: March 2019
Member
Dear,
Thanks for your reply. The issue is the following.
I presented a summary statistics of alive children and dead children by the intervention regions of a policy X.
But my supervisory asked me to report also the mortality rate. Generally, my research question is on under-five children.
I am thinking if I use BR, the calculation will also consider those above age 5 isn't it?

Kind regards
Re: children mortality rate for Ethiopia's KR file [message #21934 is a reply to message #21918] Fri, 08 January 2021 10:29 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2253
Registered: February 2013
Senior Member

Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

For your purpose, and in order to satisfy your supervisor, I suggest that you just work with the KR file and calculate the proportion of children born in the past five years who died in the first month of age (b7=0), the first year of age (b7=0 through 11), or the first five years of age (you can get that just from b5=0). These proportions must not be interpreted as rates (this is important!) but they will be good indicators of mortality within subgroups.
Re: children mortality rate for Ethiopia's KR file [message #21936 is a reply to message #21934] Fri, 08 January 2021 12:09 Go to previous messageGo to next message
id709nvz is currently offline  id709nvz
Messages: 29
Registered: March 2019
Member
Dear,
Thank you can you leave a few lines of Stata command to achieve that?

By the way, I am surprised by the fact that when I append DHS 2005, 2011 and 2016 the variable b5, b7 etc have zero observations, but observation exists in each round.
I think that happens because of the process I used to merger height/weight variables for 2005 (I dropped cases b16==0|b16==.).

Thank you in advance.
Re: children mortality rate for Ethiopia's KR file [message #21960 is a reply to message #21936] Mon, 11 January 2021 08:18 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2253
Registered: February 2013
Senior Member

Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

If you limit the files to surviving children, then b5 will equal "1" for all cases and b7 will be NA (".). You reduced the files to children who survived AND were living in the same household as the mother. To analyze mortality you must go back and reconstruct the files, retaining all children.
Previous Topic: Replicating table 8.4: ETHIOPIA DHS 2000
Next Topic: Iron supplementation
Goto Forum:
  


Current Time: Wed Apr 21 12:57:32 Coordinated Universal Time 2021