The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Service Provision Assessment (SPA) » Medicine Availability for routine antenatal care in Bangladesh (2014 SPA Bangladesh )
Medicine Availability for routine antenatal care in Bangladesh [message #21691] Tue, 08 December 2020 23:20 Go to next message
bakerchowdhury
Messages: 23
Registered: April 2014
Member
Dear DHS Experts and SPA User,
I am analyzing the Bangladesh 2014 SPA survey. It appears there are two data set for each item (Facility, Provider, Staff Listing). For example, Facility has "Facility Recode" and "Facility Raw" files, what is the key difference between these two files? Which one should be used for analysis?

In my analysis, I am looking at the Availability of medicines for routine antenatal care, which corresponds to table 6.4 in the report. In the report table 6.4, it shows a total of 1,508 (121-rural, 1387 urban) facilities offering ANC.

When I run the frequencies, it appears 1493 (urban-403, rural-1090) facilities offer the ANC service.

My Stata code is
Use "BDFC7AFLSP.DTA"
keep if q102_05==1 // this keeps facilities has ANC service available (from questionnaire report page 202)
tab ftype

*
urban/rural | Freq. Percent Cum.
------------+-----------------------------------
urban | 445 27.88 27.88
rural | 1,151 72.12 100.00
------------+-----------------------------------
Total | 1,596 100.00
*
For the columns "Iron tablets" "Folic acid tablets" "Combined iron and folic acid" the urban-rural percentages are way off. For example, iron tablets are available 92.1% and 96% in urban and rural areas, respectively in the report. However, I'm getting very low frequencies. I am using the variables related to question 1422(01, 02,03) on page 204 of the report.

Stata Codes:
tab q1422_1 ftype //which gives 73.94 % available

According to the questionnaire, it looks like I am looking the right variable, but for some reason I am not getting the frequencies as of the table. Do I need to consider anything else? I would greatly appreciate any help on this matter.

Many thanks
Re: Medicine Availability for routine antenatal care in Bangladesh [message #21729 is a reply to message #21691] Mon, 14 December 2020 10:17 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 SPA Team Member, Rajendra Dangol:

You can use either one of the datasets for analysis. The recode files are standardized across SPA surveys and preferred for comparative studies. It appears as if you are using the raw dataset, which is perfectly fine - it's easy to match the variable names in the raw data with the question numbers in the questionnaires.

I recommend you reference the Word document which is within the zip files you downloaded. This document provides guidance on how to apply the weights, along with some other basic instructions.

It seems you didn't apply the weights while doing your analysis. Please use the proper weight (FACWT/1000000) for inventory and (PROVWT/1000000) for provider level analysis.

Please refer to table 3.1 in the report, we have 1,493 cases unweighted and 1,508 cases weighted for ANC availability.

Hope this helps.
Re: Medicine Availability for routine antenatal care in Bangladesh [message #21806 is a reply to message #21691] Wed, 23 December 2020 02:49 Go to previous messageGo to next message
bakerchowdhury
Messages: 23
Registered: April 2014
Member
Thank you for the response. Yes. I am using the correct weight variable now. However, I am unable to match some of the tabulations.

Could you please help in identifying the variables to match the below columns?

"Percentage of facilities offering ANC where ANC services are offered on indicated days^1" in table 6.1 (Report Page-83)
"Urine protein^2" in table 6.3 (Report Page-86)
"Combined iron and folic acid" in table 6.4 9Report Page-87)

Thank you!
Re: Medicine Availability for routine antenatal care in Bangladesh [message #21816 is a reply to message #21806] Wed, 23 December 2020 07:01 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 SPA Team Member, Rajendra Dangol:

Please use the following definitions for your queries:
"Percentage of facilities offering ANC where ANC services are offered on indicated days^1" in table 6.1 (Report Page-83)
Please use Q1401 and recode as 1-19 = 1 (Provides, but not everyday) and 20-31 = 2 (Provides everyday);

"Urine protein^2" in table 6.3 (Report Page-86)
if Q837B(1)=1 or Q1406(2)=1 or Q1420(4)=1

"Combined iron and folic acid" in table 6.4 (Report Page-87)
if (Q1422(3)=1 or (Q210=1 & Q906(4)=1)) for Combined iron and folic acid
Re: Medicine Availability for routine antenatal care in Bangladesh [message #21887 is a reply to message #21816] Tue, 05 January 2021 10:29 Go to previous messageGo to next message
bakerchowdhury
Messages: 23
Registered: April 2014
Member
Hello,

I am trying to replicate Table 8.2 Bangladesh Health Facility Survey, 2017 (page 152), the percentage for injectable insulin in the report is 29.2%. However, I am getting 11.9. I am not sure if I am using the appropriate filter or variable. I have tried in the following way in Stata:

use "BDFC7IFLSP.DTA"
gen wgt=facwt/1000000
keep if q2301==3
tab q903_18,m
gen insuline=q903_18
recode insuline 1=1 2/5=0
tab insuline [iweight=wgt]

Thank you,
Re: Medicine Availability for routine antenatal care in Bangladesh [message #21905 is a reply to message #21887] Thu, 07 January 2021 08:07 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 SPA Team Member, Rajendra Dangol:

The specification we used for this variable is Q903(15)=GLUCOSE INJECTABLE SOLUTION
However, you used Q903(18)=INSULIN INJECTIONS [ANTI DIABETIC].
Please use Q903(15) instead of Q903(18).
Re: Medicine Availability for routine antenatal care in Bangladesh [message #21919 is a reply to message #21905] Thu, 07 January 2021 21:13 Go to previous message
bakerchowdhury
Messages: 23
Registered: April 2014
Member
Thank you for your response. It looks like the Q903(15)=GLUCOSE INJECTABLE SOLUTION which I was able to replicate with report percentage (19.2).
However, I am unable to match the 7th column "Injectable Insulin" percentage (29.2) with the report (table 8.2 page 152). Please help in identifying the "Injectable Insulin" variable in the data set.

Thank you for your attention.
Previous Topic: DR Congo 2017-18 Sick Child Observation dataset error on child weighed
Next Topic: Bangladesh Health Facility Survey 2017
Goto Forum:
  


Current Time: Fri Mar 29 04:48:21 Coordinated Universal Time 2024