The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Mortality » Maternal Mortality
Re: Maternal Mortality [message #19752 is a reply to message #9403] Sun, 09 August 2020 00:02 Go to previous messageGo to next message
placidplus is currently offline  placidplus
Messages: 31
Registered: August 2020
Member
Hello Trevor,
Thanks for your quality contribution is the estimation of maternal mortality rates. Please I want to find out how derived column F. Was it from column C. I tried computing column F from C and it does not match. For instance, (11/102)* 100 does not equal 19.6%. I will appreciate
your help. I was trying to replicate this for the Nigeria situation. I have just attached the rerun script and output for your guide. This was from the script you shared. Thank you.

Kind Regards,
Placid.
Re: Maternal Mortality [message #19758 is a reply to message #19752] Mon, 10 August 2020 09:49 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 774
Registered: January 2013
Senior Member
I think you are referring to the Excel spreadsheet linked to message 9302. Column F is simply the proportion of all respondents in each age group from the IR dataset. Just use tab v013 to get the distribution.
Re: Maternal Mortality [message #19764 is a reply to message #19758] Tue, 11 August 2020 19:22 Go to previous messageGo to next message
placidplus is currently offline  placidplus
Messages: 31
Registered: August 2020
Member
Hello Trevor,
Thanks for the guide in estimating maternal mortality rates/ratio. I think I am almost there. The attached file is the improvement I made concerning your earlier correction. I am wondering how you computed the "Unweighted Ns(Maternal Deaths & Exposure". On the table at the left. I got my exposure by Exposure (years) in the excel by running the following script

* EXPOSURE NUMBER OF YEARS BY AGE GROUP
tab agegrp [iw=expo*wt/12] if mm1 == 2

I also got deaths as shown on the excel by running:
* Deaths from table 15.4 - restict to maternal deaths
tab agegrp [iw=deaths*wt ] if mm1 == 2 & mm9 >= 2 & mm9 <= 6

I am actually trying to arrive at how you got your 107.
Based on the excel sheet, would you also say that the estimated mortality rate is 357? I was thinking that the value should be a ratio(fraction) instead of an absolute value as big as 357. Though, based on your values I got 355.
Please I will also want to find out how you got your TFR (7 years). Finally, I will need an assistance on how to get 95% CI for the maternal mortality rate/ration estimated value.
Re: Maternal Mortality [message #19765 is a reply to message #19764] Wed, 12 August 2020 11:03 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 774
Registered: January 2013
Senior Member
Hi Placid

First I should note that DHS no longer calls this the maternal mortality ratio, but rather the pregnancy-related mortality ratio. In newer surveys we calculate both the pregnancy-related mortality ratio, and an updated calculation of the maternal mortality ratio. Please see the Guide to DHS Statistics chapter 16 for more information about the calculation of both the pregnancy-related mortality ratio and the maternal mortality ratio. In particular, the calculation of the numerator differs between the two:
Numerators:
1) Pregnancy-related mortality: Number of female siblings of respondents who died during pregnancy, delivery or within two months of delivery (mm1 = 2 & mm2 = 0 & mm9 in 2:6) in the period 0-6 years prior to the interview by five-year age group at time of death (this is what you have currently)
2) Maternal mortality: Number of female siblings of respondents who died during pregnancy, delivery or within 42 days of delivery (mm1 = 2 & mm2 = 0 & mm9 in 2:6 & mm12 in 100:141,198,199), excluding due to accidents or violence (mm16 ≠ 1 & mm16 ≠ 2), in the period 0-6 years prior to the interview by five-year age group at time of death.

Older surveys do not include variables mm12 and mm16 and it is only possible to calculate the pregnancy-related mortality ratio.

For more information on the difference between pregnancy-related mortality and maternal mortality, see the Youtube video

To answer your questions, you get the unweighted numbers by using the same commands, but without the sample weight (wt), as follows:

* Deaths from table 15.4 - restrict to pregnancy-related deaths
tab agegrp [iw=deaths] if mm1 == 2 & mm9 >= 2 & mm9 <= 6
* Exposure from table 15.4 (expo is in months, division by 12 to give years) - restrict to women only
tab agegrp [iw=expo/12] if mm1 == 2

For the pregnancy-related (or the maternal) mortality ratio, this is presented per 100,000 live births, so in the example here, the 357 means 357 women per 100,000 live births. So actually a relatively small (but important) number.

Re: Maternal Mortality [message #19767 is a reply to message #19765] Wed, 12 August 2020 19:58 Go to previous messageGo to next message
placidplus is currently offline  placidplus
Messages: 31
Registered: August 2020
Member
Thank you very much Trevor! This is useful. You hard work in providing attention globally is appreciated.
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20451 is a reply to message #19765] Wed, 04 November 2020 22:51 Go to previous messageGo to next message
placidplus is currently offline  placidplus
Messages: 31
Registered: August 2020
Member
Dear Courtney,
Thanks for making the scripts for the Chapter 16 Stata Scripts available. I am working on estimating MMR & PRMRatio for Nigeria and specific regions of the country. After executing the "AM_gfr.do", The "AM_rates.do" ran into issues for me at the point of running the "main_adult_mm_mortality" program. Please find attached a snapshot of the error and the overall copy of the executed script. Please, I am seeking your assistance to fix this challenge. Thank you very much for your enormous sacrifices.

Kind Regards,
Placid Ugoagwu.
  • Attachment: MMortalR.log
    (Size: 94.95KB, Downloaded 229 times)
  • Attachment: MMR_Error.docx
    (Size: 174.89KB, Downloaded 232 times)
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20452 is a reply to message #20451] Wed, 04 November 2020 23:49 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 774
Registered: January 2013
Senior Member
Hi Placid
I'm not getting the error that you are seeing when I run the code. There is a lot going on in this code, so it is difficult to highlight where the problem is. I'm fairly sure the error lies in AM_gfr.do, but because there are a lot of 'program's in this file and they run quietly it does not highlight where the problem is. I suggest modifying the code so that when it is calling a program call it noisily so that you can identify exactly where the error is coming from. For example, in AM_gfr.do on lines 333 and 342, make these
noisily setup_gfr, and
noisily make_exposure_and_births
and similarly in lines 194 and 195, change the "quietly" to "noisily".
Hopefully these will help to identify exactly which command is failing.
Also, which version of Stata are you using? It is possible that there is a Stata command that doesn't work in earlier versions of Stata.
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20453 is a reply to message #20452] Thu, 05 November 2020 00:05 Go to previous messageGo to next message
placidplus is currently offline  placidplus
Messages: 31
Registered: August 2020
Member
Thanks for your response, Trevor. The AM_gfr.do actually ran without any errors. I noticed the error at the point of running the program referred to as "main_adult_mm_mortality" in the AM_rates.do file. This is almost at the end of the "AM_rates.do" file program script. I will try out what you just advised and see if it helps. Thank you!
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20454 is a reply to message #20453] Thu, 05 November 2020 00:08 Go to previous messageGo to next message
placidplus is currently offline  placidplus
Messages: 31
Registered: August 2020
Member
Sorry that I forgot to respond to your question about the version. I am using version Stata version 12.1
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20455 is a reply to message #20452] Thu, 05 November 2020 00:51 Go to previous messageGo to next message
placidplus is currently offline  placidplus
Messages: 31
Registered: August 2020
Member
Hello Trevor,
After adding noisily to the specific program lines of "AM_gfr.do" as advised. AM_gfr.do was executed without any errors. But after running the AM_rates.do program. I had issues executing program line 1346 of the AM_rates.do file. The error says "you are using old merge syntax; see [D] merge for new syntax)" and then "unrecognized command: - invalid command name". The earlier snapshot that I sent shows the specific point of execution when the error prompts out. I am re-attaching the file again. The attached log has the run so far attained before the error. Please look at the end of the log file. I am using Stata version 12.1. Thank you so much.

Kind Regards,
Placid.
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20458 is a reply to message #20455] Thu, 05 November 2020 10:48 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 774
Registered: January 2013
Senior Member
Hi Placid
AM_gfr.do appears to run ok because all it is doing is defining up the 'program's that are being called by AM_rates.do. When AM_rates.do is running it calls the programs that are defined in AM_gfr.do, and it is n the code defined in AM_gfr.do that the error appears. The message about using the old merge syntax is not important - it is using older syntax, but that part still works properly, as you can see by the tab _merge output that follows. We need to find the actual command that it is failing on, which is buried in the code.
Line 1346 is main_adult_mm_mortality, which calls several other programs, and it is in these programs, many of which are in AM_gfr.do that the problem occurs. For example, you can see in your output that temp1.dta is saved. This happens in line 330 of AM_gfr.do. After this the programs setup_GFR, make_exposure_and_births, and calc_rates are run. Setup_gfr is fairly simple and seems to have run. I think the issue is in the make_exposure_and_births program. Can you try commenting out the recast command on line 202? I think it is possible that the recast command does not exist in your version of Stata, and it also looks like it is not needed. Please send me your output after trying this. Good luck.

Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20459 is a reply to message #20458] Thu, 05 November 2020 13:02 Go to previous messageGo to next message
placidplus is currently offline  placidplus
Messages: 31
Registered: August 2020
Member
Hello Trevor,
Thanks for following up on this for me. I commented out the "recast" command on line 202 as advised. Please find attached the log file. Thank you.
  • Attachment: MMortalR.log
    (Size: 96.93KB, Downloaded 228 times)
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20460 is a reply to message #20459] Thu, 05 November 2020 13:27 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 774
Registered: January 2013
Senior Member
Hi Placid
Can you send your AM_gfr.do file. I now think there is an extraneous character or a character missing in the code somewhere around lines 200-220.
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20461 is a reply to message #20460] Thu, 05 November 2020 14:30 Go to previous messageGo to next message
placidplus is currently offline  placidplus
Messages: 31
Registered: August 2020
Member
Okay. Please find attached both files. Thank you for your help.
  • Attachment: AM_gfr.do
    (Size: 10.04KB, Downloaded 225 times)
  • Attachment: AM_rates.do
    (Size: 46.10KB, Downloaded 223 times)
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20462 is a reply to message #20461] Thu, 05 November 2020 14:50 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 774
Registered: January 2013
Senior Member
Hi Placid
It is exactly as I thought. You code has:
	*--------------------------------------------------------------------------
	*We now have a woman-level file that includes each woman's exposure and births 
	*in each of the age intervals and in the specified time interval.
	
	*It will help to know the woman-years of exposure used for the GFR; call it 
	*yexp_GFR to distinguish from the person-years of exposure for the mortality rates
	--------------------------------------------------------------------------*/
however the version on github has
	/*--------------------------------------------------------------------------
	We now have a woman-level file that includes each woman's exposure and births 
	in each of the age intervals and in the specified time interval.
	
	It will help to know the woman-years of exposure used for the GFR; call it 
	yexp_GFR to distinguish from the person-years of exposure for the mortality rates
	--------------------------------------------------------------------------*/
In the version on github the comment starts with the /* in the first line and ends with */ on the last line.
In your version the first of these has been changed to just a * (meaning a single line comment, rather than a continuous comment). This means that in the last line the dash (-) at the beginning of the line is being treated as a command, and it doesn't recognize it.
Just add the slash (/) at the beginning of the first line of the comment and it should fix the problem.
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20463 is a reply to message #20462] Thu, 05 November 2020 15:43 Go to previous messageGo to next message
placidplus is currently offline  placidplus
Messages: 31
Registered: August 2020
Member
Thank you very much and I am sorry for putting you through the trouble. I am sincerely grateful.

Best Regards,
Placid.
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20464 is a reply to message #20462] Thu, 05 November 2020 16:31 Go to previous messageGo to next message
placidplus is currently offline  placidplus
Messages: 31
Registered: August 2020
Member
Hello Trevor,
Please one last question on the MMR. I am trying to make sense of the estimates I just generated. The DHS reports MMR as 512 per 100,000 live births(Nigeria). I have just generated the attached results for Nigeria in Excel. If you look at the "Maternal Mortality Ratio" sheet tab (as attached), you will find "0.002551416" under the MMR column. When you look at the "Mortality summary" summary sheet, you will find a column referred to as "unwtd_prdeaths" with an estimate of"511" which seems close to this estimate that was reported per 100,000 live births? "511" seems to me as pregnancy-related deaths. I actually need to report this as well as its 95% confidence intervals as done on page 374 of the DHS report. Thank you for your help.

Kind Regards,
Placid.
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20465 is a reply to message #20462] Thu, 05 November 2020 23:07 Go to previous messageGo to next message
placidplus is currently offline  placidplus
Messages: 31
Registered: August 2020
Member
I figured out the estimates and the intervals. Thank you Trevor.
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20467 is a reply to message #20465] Fri, 06 November 2020 13:54 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 774
Registered: January 2013
Senior Member
Hi Placid

There is an error in the calculation of the maternal mortality rates and ratios. This may be only an issue for Nigeria 2018, but might also affect other surveys. In AM_RATES.do, the following line should change from:
replace  mdeaths_in_`li'=1 if deaths_in_`li'==1 & mm9>=2 & mm9<=5 & mm16==0
to:
replace  mdeaths_in_`li'=1 if deaths_in_`li'==1 & mm9>=2 & mm9<=5 & mm16!=1 & mm16!=2
This is because the question on whether the death was due to accidents or violence are skipped in the Nigeria survey if the death was during delivery. The change above thus only excludes deaths due to accidents or violence, while the prior version was effectively excluding all deaths during delivery (as mm16 was missing, and thus not 0).

This will now give you the 512 estimate for maternal mortality which will match the results on page 374 of the report and in table 14.4 of page 377.
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20469 is a reply to message #20467] Fri, 06 November 2020 17:05 Go to previous messageGo to next message
placidplus is currently offline  placidplus
Messages: 31
Registered: August 2020
Member
Oh, Great Thanks to you Trevor. The results are coming out accurately. I can now use the template for some other regions of the country. I am grateful for your interest and patience in this. I am happy this has fixed.


Kind Regards,
Placid.
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20471 is a reply to message #20467] Sat, 07 November 2020 14:49 Go to previous messageGo to next message
placidplus is currently offline  placidplus
Messages: 31
Registered: August 2020
Member
Hello Trevor,
The whole process of generating the MMR produces a dataset referred to as the "micro.dta". Something I really appreciate on this dataset is that the mm9_**(sibling's death and pregnancy multiple variables) has now been reduced to one single unifying variable referred to as mm9. Same as "sex of child" in that event of pregnancy. Please how can I reduce the "micro.dta" dataset to an individual level dataset just like the IR data and maintain the single variables in "micro.dta"? This is so that I can use it to fit a regression model. I would then later generate a binary variable for maternal deaths from this dataset and fit a categorical model. I will appreciate your response.

Best Regards,
Placid.
Re: MMR & PRMRatio (pregnancy related mortality ratio) [message #20478 is a reply to message #20471] Mon, 09 November 2020 07:41 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2587
Registered: February 2013
Senior Member


Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

The "micro.dta" file has one record for each sibling in the sibling histories. It's analogous to the BR file, which has one record for each child in the birth histories. It's built into the program mainly for data quality checking and simple tabulations. If you want to use it for data analysis, you need to take into account that there tends to be some rounding (and probably displacement) in ages of siblings, years since death, and age at death (for those who died). There may be some omission, especially for siblings who were much older than the respondent and died when young. And characteristics of the respondent, such as level of education and use of health services, cannot safely be attributed to the sibling. These are the reasons why DHS doesn't recommend estimating adult and maternal mortality for earlier time periods and subpopulations. Even so, you may find the file useful. You can add the respondent's IR variables onto micro.dta using v001 v002 v003 and a normal merge command.

Re: Maternal Mortality [message #23270 is a reply to message #11291] Fri, 13 August 2021 10:37 Go to previous messageGo to next message
Peter is currently offline  Peter
Messages: 1
Registered: August 2021
Member
Hi Trevor-DHS,
thanks for your contribution on this topic, it's really of great help to my study of maternal mortality. But i still do understand somethings, from the spss syntax the part where it
*Set up the period length and related variables.

is it the same has this?

* Define the period to use, the upper limit (months preceding the interview), and the lower limit.
define periodlen ().
* To use 5 years durations.
compute period = 5.
compute kmax = 1.
compute kmin = 12*period.
compute totexp = kmin-kmax+1.
!enddefine.

And secondly, my research is on some of the factors affecting maternal mortality in Nigeria 2013 and 2018. please how will merge the data?
thanks!
Re: Maternal Mortality [message #23281 is a reply to message #23270] Mon, 16 August 2021 09:52 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 774
Registered: January 2013
Senior Member
Yes, setting the code given is setting the period (5 years) and the minimum and maximum months preceding the survey (kmax and kmin will be 1 and 60), and the total exposure in that period will be 60 months.

On your second question this depends on what the variables are that you are interested in. If they are variables that already exist in the individual recode (IR) file then you just need to carry them through in your program, by including those variables on the /keep parameter of the varstocases command. You shouldn't need to merge files if the variables of interest are already in the IR file.
Re: Maternal Mortality [message #25103 is a reply to message #8851] Wed, 31 August 2022 13:07 Go to previous messageGo to next message
rhopper1 is currently offline  rhopper1
Messages: 4
Registered: August 2022
Member
Dear DHS team

I am working on a project in Sierra Leone and I am trying to calculate adult mortality and maternal mortality rates.

Using the DHS 2019 data for sierra leone I have used the same formula listed by Trevor (and used by many others successfully). I have, unfortunately, not been able to generate the same results as are presented in the DHS 2019 report. The numbers for both mortality and exposure are very different to those in the DHS 2019 publication.

I have made changes to the weight as suggested (gen wt = (v005/1000000)*(awfactt/100)) but this has not made a difference.

Can someone please assist? I really would appreciate some help with this!

Thanks

Robbie
Re: Maternal Mortality [message #25106 is a reply to message #25103] Wed, 31 August 2022 15:14 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2587
Registered: February 2013
Senior Member

Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

The programs to produce adult and pregnancy-related and maternal mortality rates are in chapter 16 of our GitHub site (https://github.com/DHSProgram).

The adjustment for awfactt is only needed for surveys limited to ever-married women. The Sierra Leone survey was not an EMW survey.

Let us know if you have difficulties If you use the GitHub programs, you should carefully read the introduction.
Re: Maternal Mortality [message #25109 is a reply to message #25106] Thu, 01 September 2022 05:14 Go to previous messageGo to next message
rhopper1 is currently offline  rhopper1
Messages: 4
Registered: August 2022
Member
Dear Tom

Thanks so much for your reply. I used the GitHub site chapter 16 and the code runs fine until the end when I go to execute it, when it says...

//Execute the program
. main_adult_mm_mortality
C:\Users\robbi\OneDrive\Consultancy Work\WB HCR Sierra Leone\Data\DHS
file \`lfn_IR'.dta not found
r(601);

This code is returned even if I extract the DHS_IR dataset from the my desktop and begin the program with use "C:\Users\robbi\Desktop\DHS_2019_IR_ORDERED.dta"... then the output above is still returned exactly the same, despite me using a different folder.


. //Execute the program
. main_adult_mm_mortality
C:\Users\robbi\OneDrive\Consultancy Work\WB HCR Sierra Leone\Data\DHS
file \`lfn_IR'.dta not found
r(601);



Am I missing something obvious here?

Thanks

Robbie
Re: Maternal Mortality [message #25111 is a reply to message #25109] Thu, 01 September 2022 08:45 Go to previous messageGo to next message
Shireen-DHS is currently offline  Shireen-DHS
Messages: 128
Registered: August 2020
Location: USA
Senior Member
Hello,

To execute any code from the DHS Program GitHub, please check the main file and change to your paths and the survey you are working with in the main file. This is indicate in the readme file for the repository and also there are notes in the main files.

After you change the paths to your own and the survey name in the main file, you should run the main file not the separate do files.

Thank you.

Best,
Shireen Assaf
The DHS Program

Re: Maternal Mortality [message #25163 is a reply to message #25111] Thu, 08 September 2022 09:47 Go to previous messageGo to next message
rhopper1 is currently offline  rhopper1
Messages: 4
Registered: August 2022
Member
Dear DHS team

Thanks so much for your assistance so far. I have read the blanked-out text in the do files and the read me more thoroughly and with some persistence managed to generate the information I needed. Apologies for not realising all the information I needed was there in the first place.

One thing I have not been able to do, or understand, is toi work out if the probabilities of child/infant mortality generated by the do file on child mortality controls for any other variables. I understand logistic regressions are being run but I do not believe that any other variables are controlled for. I was wondering if this is possible, and/or it is possible to produce interaction terms using the do file so I can see the impact of say... a postnatal check within 2 months (m70) depending on who delivers the check (m72).

Apologies for the potential obvious nature of this questions. I hope to hear from you soon.

I look forward to hearing from you

Robbie
Re: Maternal Mortality [message #25164 is a reply to message #25163] Thu, 08 September 2022 11:21 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 2587
Registered: February 2013
Senior Member

Following is a response from DHS staff member, Tom Pullum:

All of the demographic rates in DHS reports and on STATcompiler--fertility, under-5 mortality, adult and maternal mortality--are calculated for aggregates of women, or children, or siblings, and for exposure + events that occur in an interval of time, such as the last 3 years or the last 5 years or the last 7 years. There are some differences but basically they are based on a pooling of exposure or risk, and events, for an aggregate. The calculations can be repeated within aggregates or subpopulations such as the wealth quintiles, but cannot control for other potential covariates unless you go down to smaller and smaller subpopulations.

The GitHub programs to calculate the rates do include logit or poisson regressions with no covariates but basically as a shortcut to calculate proportions or means and to get standard errors.

If you want to do multivariate (multivariable?) analysis with controls, interactions, etc., you need to shift to a statistical framework that gives results in terms of coefficients rather than rates. You can do logit regressions that adjust for exposure, e.g. with hazard models, or poisson regressions, etc., depending on the outcome. Bridging demographic and statistical approaches is a methodological challenge.
Previous Topic: Under five mortality
Goto Forum:
  


Current Time: Tue Sep 27 23:31:37 Coordinated Universal Time 2022