The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Reproductive Health » Place of delivery questions -- Mozambique and Tanzania (Place of delivery questions -- Mozambique and Tanzania)
Place of delivery questions -- Mozambique and Tanzania [message #19924] Tue, 01 September 2020 17:33 Go to next message
asheffel is currently offline  asheffel
Messages: 4
Registered: July 2017
Member
I'm working on a multi-country analysis that utilizes place of delivery (variable m15) and have found small issues in two datasets that I could use help clarifying.

- In the Mozambique 2011 DHS, is there a reason why the place of birth was imputed for more than a third of respondents? Can you provide any information on how this data was imputed?
- In the Tanzania 2004 DHS, some of the response option labels are repeated while there are no "private" facilities. I am assuming some of the duplicates are supposed to say "private". For example, referal/ spec. hospital and district hospital are both listed twice. Is there a way to get clarification on these value set labels?

Many thanks for any information you can provide!
Ashley
Re: Place of delivery questions -- Mozambique and Tanzania [message #19941 is a reply to message #19924] Fri, 04 September 2020 10:17 Go to previous messageGo to next message
Shireen-DHS is currently offline  Shireen-DHS
Messages: 140
Registered: August 2020
Location: USA
Senior Member
Hello Ashley,

I am not sure that values were imputed for Mozambique as you say. I used our standard code which is also available on our DHS Code Share Library on GitHub (https://github.com/DHSProgram/DHS-Indicators-Stata) and get a match with the final report (Table 9.5). If you would like to check this code, please go to the GitHub link and see the RH_DEL.do file in Chapter 9. Please read the readme file before running this code, you would need to run the RHmain.do file. However, the code I provide below will give you a match for place of delivery for the 2011 Mozambique DHS.

The same code can be used for Tanzania 2004 survey, however the standard code groups religious institutions with private, you can use the code to separate that category out if you like. The repeated values you mention are the difference between public and private. Values between 20/29 are reserved for public institutions and values between 30/39 are reserved for private or other non-government institutions.

Hope this helps.

Best,
Shireen Assaf
The DHS Program


*Standard DHS Stata code for place of delivery
*open BR file for the survey

gen period = 60
gen age = v008 - b3

//Place of delivery - by place type
recode m15 (20/29 = 1 "Health facility - public") (30/39 = 2 "Health facility - private") (10/19 = 3 "Home")(40/98 = 4 "Other") (99=9 "Missing"), gen(rh_del_pltype)
replace rh_del_pltype = . if age>=period
label var rh_del_pltype "Live births by type of health facility"

gen wt=v005/1000000
ta rh_del_pltype [iw=wt]



Re: Place of delivery questions -- Mozambique and Tanzania [message #19944 is a reply to message #19941] Fri, 04 September 2020 12:09 Go to previous messageGo to next message
asheffel is currently offline  asheffel
Messages: 4
Registered: July 2017
Member
Hi Shireen,

Many thanks for your quick reply!

We are working with the recode data files provided by DHS and would like to write our own do-file to classify place of care as per the needs of our analysis (particularly because we would like to disaggregate by facility type and managing authority). As such, we need to clarify the labels in the recode datasets for Mozambique and Tanzania.

For Tanzania 2004, it is still not clear exactly what the delineation is between private and religious facilities. Are response codes 21/27 public, 31/33 private, and 34/37 religious?

For Mozambique 2011, when we tabulate m15 the response options are as follows:

Respondent's home
Other home
Other home (imputed)
Government hospital
General/provincial hospital
Rural hospital
Health center
Other public sector
Private hospital/clinic
Private doctor
Other private sector
Health facility (imputed)
Other
99

While you can certainly collapse these categories to create the variable in your standard code, it doesn't explain the presence of the response options "Other home (imputed)" and "Health facility (imputed)". We are particularly concerned about what "Other home (imputed)" means as this is the response for 34% of the sample. We would like to know what these "imputed" data mean so that we can make a decision on whether or not to include this dataset in our analysis. Any additional information on this would be very much appreciated.

Best,
Ashley
Re: Place of delivery questions -- Mozambique and Tanzania [message #19945 is a reply to message #19944] Fri, 04 September 2020 12:17 Go to previous message
Shireen-DHS is currently offline  Shireen-DHS
Messages: 140
Registered: August 2020
Location: USA
Senior Member
Hello Ashley,

Yes for Tanzania the categories would be 21/27 public, 31/33 private, and 34/37 religious.

For Mozambique, I am not able to answer how this was imputed for home. I will try to find someone that can.

Best,
Shireen
Previous Topic: Skilled Birth Attendant Cambodia DHS 2014
Next Topic: Creating variable of first prenatal care visit
Goto Forum:
  


Current Time: Thu Mar 28 19:56:40 Coordinated Universal Time 2024