| Suggested changes to presentation of child mortality indicators in DHS reports [message #17093] |
Fri, 15 March 2019 12:54 |
UN IGME
Messages: 1
Registered: March 2019
Location: New York, USA
|
Member |
|
|
The following suggestions are being submitted on behalf of the United Nations Inter-agency Group for Child Mortality Estimation (UN IGME) in light of the ongoing review period on DHS-8 questionnaires.
These suggestions pertain to the way child mortality is presented in the DHS reports. The UN IGME would first suggest reporting standard errors for the child mortality indicators wherever possible. UN IGME would also suggest to avoid combining child mortality indicator values from different surveys in the report for a specific survey, and rather to display only the trend in retrospective mortality rates referring to the 5-year periods before the survey derived from the survey at hand.
Reporting standard errors with the child mortality indicators would better inform the public health community and users of these DHS indicators as to the quality and precision of the child mortality estimate. These standard errors also inform the global modelling and monitoring of child mortality at the country level. Furthermore, the calculation of the standard errors should consider the survey design in terms of the stratification.
Presenting values from different surveys linked together with a trend line (see figure 8.1 p. 146 of Pakistan DHS 2017-18 report, https://dhsprogram.com/pubs/pdf/FR354/FR354.pdf for example) should be avoided whenever possible as it can obscure varying data quality issues with different surveys, ignores the other retrospective values calculated from the birth histories greater than 5 years prior to the survey, and suggests the most recent estimate referring to the 5-year period before the survey is the true value for the child mortality indicator for the survey year. Presenting the child mortality estimates derived from the survey at hand could avoid such issues.
Thank you for your consideration of these suggestions.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
