The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Child Health » Child growth
Re: Child growth [message #8433 is a reply to message #6942] Tue, 27 October 2015 13:30 Go to previous messageGo to previous message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 746
Registered: January 2013
Senior Member
I'm going to give you this in two parts as this is how it is calculated:
1) First using the birth history (Births' recode dataset - BR) calculate the early neonatal mortality.
2) Second using the calendar data from the women's individual recode dataset (IR) calculate the stillbirths.

For the first part you can use the code below:
* Early neonatal mortality
* use Births' Recode
use "BDBR61FL.dta", clear

* create a child Alive or Died variable using the b5 variable
gen alive=b5
lab def alive 0 "Died" 1 "Alive"
lab val alive alive
lab var alive "Alive or Died by the time of survey"

*** Age at death using variables b6 and b5
gen age_death=.
replace age_death = 0  if  b6<=106 & b5==0
replace age_death = 1  if  b6>=107 & b6<=130 & b5==0
replace age_death = 2  if (b6> 130 & b6<=999) | b5==1
lab def age_death 0 "Early Neonatal Death" 1 "Late Neonatal Death" 2 "Survived Neonatal Period"
lab val age_death age_death
lab var age_death "Neonatal Mortality Status"

gen wt=v005/1000000
svyset v021 [pw=wt], strata(v023) 

svy: tab age_death if v008-b3<60, per count form(%7.3g) 
svy: tab age_death if v008-b3<60, per col form(%7.3g)

For the second part, open the individual recode dataset (BDIR61FL.dta), and then use the code posted here in this message. In this code, you can ignore births and nlbirths, and just use stillbirths.

These are the two parts that go into table 8.5. The first part will give you 220 early neonatal deaths out of 8789 total births in the period 0-59 months preceding the survey. The second part will give you 232 total stillbirths (223 women with 1 and 4.56 with 2. 223*1 + 4.56*2 = 232).

The perinatal mortality rate is then calculated as (220 + 232) / (8789 + 232) = 0.050 (or 50 per 1000).

In the above code, the first part counts twins, but if you only want pregnancies and not births of 7+ months then you can look at the code for births in the second part, where twins are ignored.

I hope this helps.
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic: Concentration Index
Next Topic: Birth interval estimation
Goto Forum:

Current Time: Mon May 17 09:19:22 Coordinated Universal Time 2021