The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Nepal » Staff trained for ANC
Staff trained for ANC [message #21888] Tue, 05 January 2021 10:46 Go to next message
Messages: 21
Registered: April 2014
"I would like to create the variable "Staff trained for ANC" using Nepal Health Facility Survey-2015 (Report-Table 6.2 page 120). First I merged the provider data with facility inventory data. Then I tried using the following code but the result was not similar to the report:

gen trained_staff=0
replicate trained_staff=1 if P502_1 ==1 | P502$2==1 | P502$3==1 | P502$4==1 | (P505$1==1 | P505$5==1)
Re: Staff trained for ANC [message #21972 is a reply to message #21888] Tue, 12 January 2021 06:56 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2270
Registered: February 2013
Senior Member

Please see the following response from SPA Specialist, Rajendra Dangol:

You need to count the no. of providers with ANC in-service training in facility level, then only merge with inventory data file as necessary to get the result. Following is the logic we have used in CSPro to get this indicator. Please also see the note from table 6.2 below. I hope this helps you replicate the table in Stata.

haveanc=(Q102(5)=1 & Q1401 in 1:31,missing);
if haveanc & P500 in 1:3 then  {anc services for T6.2, T6.17}
  {for t6.2 - cntanc2}
  if count(P502000 where P502=1) or P505(1)=1 or  P505(5)=1 or P206(5)=1  then

Facility has at least one interviewed staff member providing ANC services who reports receiving in-service training in some aspect of antenatal care during the 24 months preceding the survey. The training must have involved structured sessions; it does not include individual instruction that a provider might have received during routine supervision.

Re: Staff trained for ANC [message #22549 is a reply to message #21972] Tue, 30 March 2021 12:20 Go to previous messageGo to next message
Messages: 21
Registered: April 2014
Thank you, Bridgette, for the response. Here "P502000" is not clear to me, couldn't find any variable under this name. Also, there are five (p502_1 p502_2 p502_3 p502_4 p502_5) variables in p502, do we need to use all of them? Similarly, p505 has six (p505_1 p505_2 p505_3 p505_4 p505_5 p505_6) variables. The last part cntanc2=cntanc2+1 is also unclear, why we add 1 at the end? I would appreciate it if you could help writing this in Stata.

Thank you!
Re: Staff trained for ANC [message #22589 is a reply to message #22549] Wed, 07 April 2021 15:09 Go to previous message
SaraDHS is currently offline  SaraDHS
Messages: 3
Registered: December 2020

In Stata, if you use the below code after merging the provider data with the facility inventory data, you should be able to recreate the final report values for number of providers with ANC in-service training.

gen ancfacil =0
replace ancfacil=1 if (q102_05==1 & inrange(q1401,0,31)) | q102_05==3

gen anc_staff=0
replace anc_staff=1 if (p500==1 | p500==2 | p500==3)

gen trained_staff=0 if ancfacil==1 & anc_staff==1
replace trained_staff =1 if ancfacil==1 & anc_staff==1 & (p502_1==1 | p502_2==1 | p502_3==1 | p502_4==1 | p502_5==1 | p505_1==1 | p505_5==1 | p206_5==1 | p206_6==1)

Hope that's helpful!


Sara Riese, PhD Senior Demographic and Health Researcher, DHS Program
Previous Topic: How to create a single weight when combining surveys
Goto Forum:

Current Time: Wed May 12 17:57:36 Coordinated Universal Time 2021