The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Dataset use in Stata » treatment seeking for ARI: Bangladesh DHS 2014
treatment seeking for ARI: Bangladesh DHS 2014 [message #9477] Thu, 31 March 2016 22:29 Go to next message
zobidaislam is currently offline  zobidaislam
Messages: 6
Registered: March 2016
Location: Dhaka
Member
Hello,

I tried to estimate the proportion of children with ARI that sought treatment from a health facility or provider (according to the table 10.8, page-147 in Bangladesh DHS 2014) using the following stata code, but my estimate is different from the table. My estimate is 44.47, but the table states 42.0. Could you kindly tell me whats wrong with my code?

** I used kids record (BDKR4JFL)

* generate sampling weight
g wgt=v005/1000000

*symptoms of ARI (%)
gen ari=0 if b5==1 // b5: child is alive
replace ari=1 if b5==1 & h31b==1 & (h31c==1 | h31c==3)
tab ari [iw=wgt]

*suspected ARI taken to health facility
gen treat= 0 if ari==1
foreach x in a b c d e f i j l p s t u v w z {
replace treat=1 if ari==1 & h32`x'==1
}

tab treat if b5==1 [iw=wgt]

Thank you.
Zobi

[Updated on: Thu, 31 March 2016 22:32]

Report message to a moderator

Re: treatment seeking for ARI: Bangladesh DHS 2014 [message #9540 is a reply to message #9477] Mon, 11 April 2016 09:43 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3016
Registered: February 2013
Senior Member
Following is a response from DHS Stata Specialist, Shireen Assaf:

Your code is correct except for the inclusion of some of the treatment variables particularly h32z.

If you use the following code below you will match the report:

***********
use "BDKR70FL.DTA" , clear

* generate sampling weight
g wgt=v005/1000000

*symptoms of ARI (%)
gen ari=0 if b5==1 // b5: child is alive
replace ari=1 if b5==1 & h31b==1 & (h31c==1 | h31c==3)
tab ari [iw=wgt]

* treatment for ARI
gen treat= 0 if ari==1
foreach x in a b c d e f g h i j l n s t u v w {
replace treat=1 if ari==1 & h32`x'==1
}

ta treat [iw=wgt]

------------

As described in the DHS recode manual, h32z is described as:
"Whether the child was taken to a medical facility for treatment of the fever and/or cough. This usually includes being taken to all Public Sector facilities and all Medical Private Sector
facilities except for Pharmacy. This variable is a summary of these preceding variables as is used in the final reports."

http://www.dhsprogram.com/publications/publication-dhsg4-dhs -questionnaires-and-manuals.cfm

So h32z should not be included when you generate the treat variable.
Previous Topic: Stata Help
Next Topic: Is there a way to add household member's information to birth's recode file?
Goto Forum:
  


Current Time: Thu Mar 28 06:48:02 Coordinated Universal Time 2024