| Menopause status [message #3799] |
Fri, 13 February 2015 18:57  |
AConklin
Messages: 9
Registered: November 2014
Location: Los Angeles
|
Member |
|
|
Hello,
I am trying to ascertain women's menopause status to generate a new variable to include in a multilevel model. I wanted to create a binary variable using v226 (months since last period), but some filter checks with other demographic variables raise some questions about v226...
For example, a tabulation of women coded 994 ("in menopause") shows ages from 15 to 49 which seems biologically implausible. Also, a tabulation of v226 by currently pregnant shows that 45 women are "in menopause" and "currently pregnant". Similarly among those who "never menstruated" (i.e 996), a cross tab with currently pregnant shows 57 were in the "1" category (and ages ranged from 13 to 49).
The DHS guide to statistics and recode manual indicate that the base for menopause is women aged 30-49 years, and that menopause status is determined by women who had their last period over 6 months prior to the survey, had a hysterectomy, self-reported or never mentstruated.
Should I be using a different variable/question in addition/instead to determine menopause status?
Note: my dataset is an amalgamation of 63 countries between 1999-2014 containing nearly 1.7m obs total.
Many thanks in advance for your help/clarification,
A Conklin
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
