The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Bangladesh » Replicating BDHS 2011 Table 15.5.1 & 15.5.2 Table on Diabetes prevalence
Replicating BDHS 2011 Table 15.5.1 & 15.5.2 Table on Diabetes prevalence [message #28320] Wed, 13 December 2023 08:55 Go to next message
Delulu is currently offline  Delulu
Messages: 2
Registered: September 2023
Member
Hello,

I am trying to replicate tables Table 15.5.1 & 15.5.2 on Diabetes prevalence among man and women. I am finding it difficult to arrive at the same prevalence values. When I tabulate variable sh284a which is plasma blood glucose (mmol/l) it shows that everyone has Elevated FPG. Am I doing something wrong here?
Please can anyone help me replicate these tables and include Stata commands where possible?

Thank you so much.
Re: Replicating BDHS 2011 Table 15.5.1 & 15.5.2 Table on Diabetes prevalence [message #28325 is a reply to message #28320] Thu, 14 December 2023 07:37 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3087
Registered: February 2013
Senior Member

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

Most of the biometric variables, including this one (sh284a), have two aspects of their coding that you must take into account. The first is that there are codes for "refused", etc., that must be bypassed in any analysis. The second is that the measurements have been multiplied by some power of 10 to get rid of a decimal point.

If you enter "describe sh284a" you will find that the label is SH284A. Then enter "label list SH284A" to identify the special codes. Here they are:

. label list SH284A
SH284A:
994 not present
995 refused
996 other

Then enter either "histogram sh284a if sh284a<994" or "summarize sh284a if sh284a<994" to find out what the multiplier is. The latter command shows that the (unweighted) mean is 55 and the range is 16 to 271. The cutoff values in table 15.5 are 3.9, 6.0, etc. Clearly sh284a is 10 times the scale of the glucose indicator in the table. Therefore you need to construct a new variable as follows: "gen glucose=sh284a/10 if sh284a<994". Please try again to construct the tables using "glucose".



Re: Replicating BDHS 2011 Table 15.5.1 & 15.5.2 Table on Diabetes prevalence [message #28339 is a reply to message #28325] Sat, 16 December 2023 07:54 Go to previous message
Delulu is currently offline  Delulu
Messages: 2
Registered: September 2023
Member
Thank you so much. It worked!!
Previous Topic: Determining child's birth order
Next Topic: Merging DHS household Survey data with SPA data
Goto Forum:
  


Current Time: Wed Jun 19 19:58:28 Coordinated Universal Time 2024