The DHS Program User Forum: Biomarkers

Home » Topics » Biomarkers » cardiometabolic data

Show: Today's Messages :: Show Polls :: Message Navigator

cardiometabolic data [message #3212]

Fri, 07 November 2014 19:43

AConklin
Messages: 9
Registered: November 2014
Location: Los Angeles

Member

Hello DHS

Do all country surveys include a question about high blood pressure/ hypertension which would provide self-reported data? If so, then why are they not all listed under the survey characteristics section of the survey search tool?
It would be helpful to understand whether 12 countries selected by blood pressure data refer to self-reported and/or based on clinical measurement.

Also, I am interested to know the distribution of adult BMI in the 69 countries with anthropometry (presumably measured height and weight). So, is it possible to obtain from DHS the BMI variable for only those countries to run descriptives or do I need to download all 69 country datasets separately?

Thank you in advance,
A Conklin

Report message to a moderator

Re: cardiometabolic data [message #3262 is a reply to message #3212]

Thu, 13 November 2014 11:35

Liz-DHS
Messages: 1516
Registered: February 2013

Senior Member

Dear User,
I checked the standard recode dictionary for DHS6 and found only the variable M42C "During pregnancy: blood pressure taken." Any other variables having to do with hypertension have been country specific. The survey characteristics show countries where blood pressure questions were asked as well as countries where blood pressure was measured.

Regarding the BMI, I will refer your question to our data archive specialist.
Thank you!

Report message to a moderator

Re: cardiometabolic data [message #3266 is a reply to message #3262]

Thu, 13 November 2014 15:53

AConklin
Messages: 9
Registered: November 2014
Location: Los Angeles

Member

Thank you very much for the reply.
I look forward to hearing from the data archive specialist.
Kind regards,
A Conklin

Report message to a moderator

Re: cardiometabolic data [message #3269 is a reply to message #3266]

Fri, 14 November 2014 19:55

Trevor-DHS
Messages: 805
Registered: January 2013

Senior Member

Concerning the BMI for adults, you can get some information from the STATcompiler at this link: http://statcompiler.com/?share=229104AD593

These data, though, are for maternal BMI, that is restricted to women who gave birth to a child in the three or five years preceding the survey. If this doesn't suffice, you will need to refer to the individual reports or to download the datasets.

Report message to a moderator

Re: cardiometabolic data [message #3280 is a reply to message #3269]

Mon, 17 November 2014 16:58

AConklin
Messages: 9
Registered: November 2014
Location: Los Angeles

Member

Hello Trevor,
Thank you very much for the reply. I have downloaded individual country datasets as I am interested in all adults who were measured.

With respect to hypertension, I wondered if you could help elucidate why the men's recode dataset for Azerbaijan Phase 5 contains two different variables ("sm101asn" and "sm509asn") are both labeled "blood pressure measurement 1" but with different values. Should I ignore the 3 different bp measurement variables (e.g. "sm101asn", "sm509asn" and "sm839sn") since only only two (e.g. "sm509asn" and "sm839sn") were used to compute an average (e.g. "sm840sa")?

Look forward to your advice and further information,
BW,
A Conklin

Report message to a moderator

Re: cardiometabolic data [message #3281 is a reply to message #3280]

Mon, 17 November 2014 17:15

Trevor-DHS
Messages: 805
Registered: January 2013

Senior Member

The DHS approach to blood pressure measurement is to take three measurements during the survey. The first one is just to get the respondent used to the blood pressure measurement and there is often a measurement effect with a higher reading for this first one. We then take the blood pressure twice more and use the average of the two. If for some reason only one the remaining two measurements is taken then we use just that one measurement. For sm509asn, it looks like that should be labeled 2 not 1. I would recommend just using the average (e.g. sm840sa).

Report message to a moderator

Re: cardiometabolic data [message #3306 is a reply to message #3281]

Wed, 19 November 2014 16:03

AConklin
Messages: 9
Registered: November 2014
Location: Los Angeles

Member

Thank you very much, Trevor! I appreciate the advice on appropriate variables to use.

-a

Report message to a moderator

Previous Topic:	Kyrgyz DHS 2012
Next Topic:	Decision tree analysis

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Thu Jan 30 14:23:59 Coordinated Universal Time 2025