The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Mortality » Most recent born Vs All births in Five Years
Most recent born Vs All births in Five Years [message #22082] Fri, 29 January 2021 02:27 Go to next message
shujaat.smc@gmail.com is currently offline  shujaat.smc@gmail.com
Messages: 75
Registered: July 2020
Senior Member
Dear DHS Representative,

Kindly confirm that Neonatal mortality rates reported in Countries DHS reports including Pakistan DHS (2017-18) are estimated by "all live births" or "most recent births" women had, in the preceding five years before the conduct of the survey.

I have estimated the NMR for only the most recent birth women had in the preceding five years before the conduct of the survey (PDHS 2017-18) and that estimates are a bit lower for the provinces of Pakistan when compared to NMR reported for the provinces in PDHS (2017-18) report.

Looking forward to your response.

Best Regards

Dr. Hussain
Re: Most recent born Vs All births in Five Years [message #22085 is a reply to message #22082] Fri, 29 January 2021 07:14 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3017
Registered: February 2013
Senior Member

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

The standard rates are calculated for all births in the past five years. The NNMR, other mortality rates, and other indicators of child health are generally better for the most recent birth than they are for all births--as you have found. Shea Rutstein prepared a methodological report in 2014 describing this:

https://www.dhsprogram.com/pubs/pdf/MR14/MR14.pdf

When you look at just the most recent birth, you tend to over-represent women who had only one birth in the past five years. Outcomes are generally better for such women and births.
Re: Most recent born Vs All births in Five Years [message #22288 is a reply to message #22085] Wed, 24 February 2021 06:40 Go to previous messageGo to next message
shujaat.smc@gmail.com is currently offline  shujaat.smc@gmail.com
Messages: 75
Registered: July 2020
Senior Member
Dear Tom Pullum,

Thanks a lot for your reply and for sharing the methodological report 14.

For last-born child (the most recent born) in the preceding five years I have to make sub-pop in stata; (Pakistan 2017-18 BR-file) can I use the following variables

the variable v208==1 (Means number of live singleton births in past five years)

however, whether these singletons were most recently born? what variable I should use?

I am asking this question as I am having confusion in two variables;

1; bidx (birth column number having a range from 1-15)
2: bord (birth order number having a range from 1-5)

I am working on estimating the neonatal mortality among last (recent) born singletons in the preceding five years before the conduct of the fourth Pakistan DHS (2017-18).

How I should write the stata command for my sub-population;

subpop (v208==1 (singleton) & variable for most recent born== ?)

Kindly guide me in declaring the correct sub-population for my analysis.

Best Regards

Dr. Hussain
Re: Most recent born Vs All births in Five Years [message #22293 is a reply to message #22288] Wed, 24 February 2021 08:18 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3017
Registered: February 2013
Senior Member

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

The most recent child has bidx=1. You don't need to use v208 to identify this child. Some other categories of children are harder to identify--for example, the youngest living child who is living with the mother (b9=0). However, the most recent birth is always identified with just bidx=1. Hope this answers your question.
Re: Most recent born Vs All births in Five Years [message #22295 is a reply to message #22293] Wed, 24 February 2021 08:29 Go to previous messageGo to next message
shujaat.smc@gmail.com is currently offline  shujaat.smc@gmail.com
Messages: 75
Registered: July 2020
Senior Member
Dear Tom Pullum, Good Morning!

Thanks for clarifying my confusion. One more question;

Q; How I will contain my sub-population to those who were singleton and born in the preceding five years?


Is this stata Command Correct: tab b0 bidx if bidx==1 & v210!=1 & v208==1 to get the exact number of singleton live births in preceding five years.

I get the total number of singletons last born (recent born) live births in the preceding five years=3,738/- (Is this correct total?) Please see the stata screenshot attached.

If yes can I use my sub-pop stata command as (b0==0 & bidx==1 & v210!=1 & v208==1)

Best Regards

Dr. Hussain

[Updated on: Wed, 24 February 2021 08:54]

Report message to a moderator

Re: Most recent born Vs All births in Five Years [message #22314 is a reply to message #22295] Thu, 25 February 2021 14:14 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3017
Registered: February 2013
Senior Member

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

If you require that v208=1, then you are selecting children whose mother had only one birth in the past five years. The child with bidx=1 will be that one child. In the KR file, there won't be any other children with the same mother. That child will definitely have b0=0, that is will be a singleton. Any other value of b0 for the most recent child would imply that the mother had MORE THAN ONE birth in the past five years. That is, the requirement that b0=0 is superfluous.

I guess I don't really understand what is your intended sample. Can you answer yes or no to the following three questions? (1) Child must be a singleton? (yes/no). (2) Child must be the only child born in the past five years? (yes/no). (3) Must be the most recent birth? (yes/no). Thanks!
Re: Most recent born Vs All births in Five Years [message #22316 is a reply to message #22314] Thu, 25 February 2021 14:47 Go to previous messageGo to next message
shujaat.smc@gmail.com is currently offline  shujaat.smc@gmail.com
Messages: 75
Registered: July 2020
Senior Member
Dear Tom Pullum, Greetings!

(1) Child must be a singleton? (yes/no). Ans: YES

(2) Child must be the only child born in the past five years? (yes/no). Ans: YES

(3) Must be the most recent birth? (yes/no). Ans: YES


In other words;(Singleton births in preceding five years)

Mother had delivered the first baby= YES

Mother had a child or children and delivered a baby in past five years= YES, this baby will be taken into account.

Mother had delivered twice or thrice in the past five years= YES I will take into account only her last delivered baby.

In my understanding by this way, I am counting all the most recent singleton live births in the preceding five years.

Waiting for your reply.

Best Regards

Dr. Hussain
Re: Most recent born Vs All births in Five Years [message #22322 is a reply to message #22316] Fri, 26 February 2021 07:53 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3017
Registered: February 2013
Senior Member

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

It appears that you want the following children in the KR file: "...if v208==1 & bidx==1 & b0==0". There is some redundancy in these conditions but I recommend that you include all three. Hope your write-up will include a justification for using this subpopulation of all children born in the past five years and will describe how they may differ from other children born in the past five years. Good luck!
Previous Topic: missing values in prenatal care
Next Topic: Mothers' background information as an instrument
Goto Forum:
  


Current Time: Fri Mar 29 05:20:05 Coordinated Universal Time 2024