The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Mortality » survival analysis of neonatal mortality (i want to do servival analisis of neonaltal mortality )
survival analysis of neonatal mortality [message #30070] Thu, 19 September 2024 07:01 Go to next message
kidus28sewagegn@gmail.com is currently offline  kidus28sewagegn@gmail.com
Messages: 3
Registered: September 2024
Member
hello I was trying to calculate the age of neonates in days but all I found was calculating age in month i.e. v008-b3 but I cant calculate the neonatal age in day if any body who knows how to calculate please help me thank you
Re: survival analysis of neonatal mortality [message #30079 is a reply to message #30070] Fri, 20 September 2024 10:48 Go to previous messageGo to next message
Janet-DHS is currently offline  Janet-DHS
Messages: 878
Registered: April 2022
Senior Member
Thank you for submitting your question. Could you please provide some more detailed information so we can better advise you?
Can you let us know:
• Which survey you are using (Include country name and year)?
• Which data files you are referring to?
• Which software you are using (Stata, SPSS, R, etc.)?
If you are trying to match a Table in a final report, please also indicate which table and which estimate you are trying to match.
Re: survival analysis of neonatal mortality [message #30088 is a reply to message #30079] Sat, 21 September 2024 02:29 Go to previous messageGo to next message
kidus28sewagegn@gmail.com is currently offline  kidus28sewagegn@gmail.com
Messages: 3
Registered: September 2024
Member
Dear janat-dhs

I am asking for help with the DHS dataset for Sub-Saharan Africa, specifically for the period from 2010 to 2022.

In the course of my analysis, I discovered that the datasets from 2015 to recent years have a 'century day code', also known as CDC, but for the years 2010-2015, the datasets only provide a 'century month code', also known as CMC.

Although the 'century day code' provides more precise information to calculate the age in days, I am having difficulty deriving the age in days for the period 2010-2015 due to the absence of this precise information.

For instance, when using the women's dataset from the 2011 Ethiopian DHS (ETIR61FL.DTA), I encountered the same problem. The 'CMC' variable only provides information on the century and month (0-12), which restricts the calculation of age to an approximate number of months.

I kindly seek your advice or suggestion on possible ways to work around this limitation, as using alternative Sub-Saharan African DHS datasets yielded a similar outcome.

Thank you for taking the time to read and respond to my query.

Best regards,
kidus

Re: survival analysis of neonatal mortality [message #30118 is a reply to message #30088] Fri, 27 September 2024 11:32 Go to previous message
Janet-DHS is currently offline  Janet-DHS
Messages: 878
Registered: April 2022
Senior Member
Following is a response from DHS staff member, Tom Pullum:

The century day code only appears for b18, in the KR and BR files, for surveys for which day of birth was collected in the birth history. Day of birth (within the month of birth) appears as b17. The cdc code is elapsed days since some reference date, which I believe is January 1, 1960 for the algorithm DHS uses.

If you are studying neonatal mortality, I don't think you need to use that code at all. You just need age at death, in days, for deaths in the first 28 days after birth. That variable is b6, for codes in which the first digit is 1 (signifying that age at death is given in days). Days 1-28 corresponds with completed days 0-27, given by b6 values from 100 through 127.
Previous Topic: Discrete Time Vs Continous Time Survival Model
Next Topic: Case-crossover analysis with DHS
Goto Forum:
  


Current Time: Mon Nov 4 20:01:42 Coordinated Universal Time 2024