| number of pregnancies - Kenya DHS 2008 [message #1524] |
Fri, 07 March 2014 09:33  |
pkram
Messages: 3
Registered: March 2014
Location: United States
|
Member |
|
|
Hello,
I am looking to calculate the number of pregnancies to enrolled women in the 5 years preceding survey in the 2008 Kenya DHS. Please advise on where I can find the calendar information, and any tips to generate the number of pregnancies of 7+ months duration shown in Table 8.4 (Perinatal Mortality). I need the total number of pregnancies, not just the number of live births because I am working on an analysis involving stillbirths.
Thanks very much,
Pavani
|
|
|
|
| Re: number of pregnancies - Kenya DHS 2008 [message #1526 is a reply to message #1524] |
Fri, 07 March 2014 11:05  |
Liz-DHS
Messages: 1516
Registered: February 2013
|
Senior Member |
|
|
Dear User,
The recode calendar variables are in Rec82 and are called VCOL and VCAL. If you downloaded the data, you should have a map and a dictionary indicating the location of REC 82. This is part of the woman's data and is what you should be using. If you are using CSPro to do your analysis, the name of the dataset containing the woman's data is called KEIR52.zip. You can refer to The Standard Recode Guide http://dhsprogram.com/pubs/pdf/DHSG4/Recode6_DHS_22March2013 _DHSG4.pdf for additional information regarding the calendar variables and how they are used. The publication is bookmarked and under the bookmark called "Section and Variable Description - Individual" there is a bookmark called "Section W82 (REC82)- Calendar". The Guide to DHS Statistics also has a section on PERINATAL MORTALITY RATE, P96. This section has the Definition and the Calculation including the handling of missing values. In addition our website has a section on Data Variables and Definitons http://dhsprogram.com/data/Data-Variables-and-Definitions.cf m. You may also want to refer to Further Analysys #59 Contraceptive Use Dynamics http://dhsprogram.com/pubs/pdf/FA59/FA59.pdf This is an older publication but the analysis was done using calendar data.
We also have a very useful section on our website on Using Datasets for Analysis http://dhsprogram.com/data/Using-DataSets-for-Analysis.cfm. Finally, also on our website there is a section on the various types of datafiles and the naming convention. http://dhsprogram.com/data/File-Types-and-Names.cfm This will provide a clear example of the various datasets. Please feel free to post again if this does not answer your question.
|
|
|
|
|
|
|
|
| Re: number of pregnancies - Kenya DHS 2008 [message #1533 is a reply to message #1529] |
Mon, 10 March 2014 15:06 |
Liz-DHS
Messages: 1516
Registered: February 2013
|
Senior Member |
|
|
Dear Pavani,
Here is a response from one of our experts, Dr. Shea Rutstein:
"The reproductive calendar is given as a string of positions, each representing a month. The first calendar string indicates, births (B), non-live birth terminations (T), each month of pregnancy (P), and numbers and letters that represent a month of use of contraception of various methods. See the data dictionary for the codes used. The position of the month of interview in the calendar is given by V018, the date of the start of the calendar by v017, and the length of the calendar string is given by V019. To use the calendar data, string functions are required that can be repeated for each position.
"Non-live births in column 1 of the reproductive calendar are given by "T"s. Live births are given by "B"s. Count the number of "T"s and "B"s in the 60 positions prior to the position of the survey date to get the number of pregnancies in the five years preceding the survey. For still births, only count the number of "T"s that are preceded by 6 to 8 "P"s (a month where the woman was pregnant) since we count as still births all those non-live birth terminations with pregnancy durations of 7 to 9 months (the T is the last month of the pregnancy). Note that for pregnancies that began prior to the calendar, the duration of pregnancy will not be given by the number of Ps.
|
|
|
|
| Re: number of pregnancies - Kenya DHS 2008 [message #1534 is a reply to message #1533] |
Mon, 10 March 2014 15:26 |
 |
user-rhs
Messages: 132
Registered: December 2013
|
Senior Member |
|
|
Pavani,
From the recode manual for DHS5 (which is what was used for the 08/09 Kenya DHS):
Column 1 description (VCAL_1):
Records each of the births and pregnancies during the calendar period, as well as each
episode of contraceptive use and non-use. The following codes are used in column 1:
0 Non-use of contraception
1 Pill
2 IUD
3 Injections
4 Diaphragm
5 Condom
6 Female sterilization
7 Male sterilization
8 Periodic abstinence/rhythm
9 Withdrawal
L Lactational amenorrhea
C Female condom
F Foam and Jelly
α Country-specific method 1
ß Country-specific method 2
τ Country-specific method 3
? Unknown method/missing data
P Pregnancy
T Terminated pregnancy/non-live birth
B Birth
W Other traditional methods
N Norplant
A Abstinence
So I think what you are interested in would be counting the number of "P"s within the string (divide by 9 months or how long pregnancies last), or rather, the first instance of "P" after a series of non-"P" characters. Unfortunately, there is no differentiation between abortion or stillbirth, but I suppose you can always go back to the data and see whether the woman ever had any stillbirths, and match the terminations with stillbirths in order to separate abortions from stillbirths.
Link: https://www.dhsprogram.com/pubs/pdf/DHSG4/Recode5DHS_23Augus t2012.pdf, VCAL coding is on page 72 of document (p. 76 of PDF file)
HTH,
RHS
[Updated on: Mon, 10 March 2014 15:30] Report message to a moderator |
|
|
|
|
|
| Re: Alignment of vcal_1 and vcal_2 codes in the data [message #8337 is a reply to message #1534] |
Mon, 12 October 2015 04:20 |
Kisaakye
Messages: 15
Registered: August 2015
|
Member |
|
|
Hello DHS experts,
On reading the calendar data, i understand the vcal_1 codes should align well with vcal_2 data codes. I am specifically talking about the gaps between the vcal_2 to match with vcal_1. However, it appears that the months between two discontinuations is not related to the duration of use/non-use of a contraceptive method (the misalignment) in the data. For now i am using the Uganda DHS 2011 but this problem appears to run through other DHS where vcal_1 and vcal_2 data were collected. When you browse the data, all the first codes for discontinuation are placed in the first column from right to left. For example, for a woman who stopped using a method after 7 months, should have no code for vcal_2 in the first 6 months from right and it should appear in the seventh month as you move to the left and so on. However, this code appears in the first column of Vcla_2. Clearly, the gaps are not matching with the vcal_1 data.
I have gone through a thread of reproductive calendar on this forum but have not found a solution.
Hoping for a response
Thanks
Peter
|
|
|
|
| Re: Alignment of vcal_1 and vcal_2 codes in the data [message #8363 is a reply to message #8337] |
Thu, 15 October 2015 15:15 |
Liz-DHS
Messages: 1516
Registered: February 2013
|
Senior Member |
|
|
Dear User,
Here is a response from Senior Data Processing expert, Mr. Guillermo Rojas,
Quote:To identify discontinuations in column 2, users need to start from the last row (position 80) in column 1 and keep looking up until they found that a method changed. In the same position in column 2 of the last month of use there should be a code indicating the reason for discontinuation. To illustrate the problem let me put columns 1 and 2 aligned on top of each other.
LLBPPPPPPPP000LLLLLLLLLLLLLLLLLLLLLLLLLLBPPPPPPPP0099LLLLLLLLLLLLLLBPPP
2 2 8
Quote: In the example above the respondent started using method L in position 76 (5th month from the bottom of the calendar) in column 1. The respondent used the method for 14 months and changed to method 9, she gave code 8 as the reason for discontinuation. In position 63 (which corresponds to the last month of use of method L), the discontinuation code is recorded in column 2. The respondent then used method 9 for two months and discontinued for reason 2, and so on.
Users need to be aware that chronologically the first month of the calendar correspond to the last position in column 1. Or to put in another way, position 80 of column 1 correspond to the first month where calendar data is collected. That first position correspond to the CMC code given by variable V017.
If you have additional questions, please feel free to post again. Thank you!
|
|
|
|
| Re: Alignment of vcal_1 and vcal_2 codes in the data [message #8369 is a reply to message #8363] |
Fri, 16 October 2015 03:08 |
Kisaakye
Messages: 15
Registered: August 2015
|
Member |
|
|
Thank you Mr. Guillermo Rojas for this answer.
Yes, we understand how to read the calendar data and the explanation you give here makes sense to us. However it appears to us that the data is misaligned. The codes in vcal_2 do not appear to be in the right positions they should be. I am attaching a screen grab from the 2011 UDHS to make my point clearer. Please prove to us that the vcal_2 codes are in the right positions.
Thanks
Peter
|
|
|
|
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
