The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Mortality » Infant and child mortality
Infant and child mortality [message #3085] Mon, 13 October 2014 14:38 Go to next message
jacobm is currently offline  jacobm
Messages: 4
Registered: October 2014
Location: Polokwane
Member
Good evening

I am using the Angola Malaria Indicator Survey (AMIS) 2011 to study socio-economic factors contributing to infant and child mortality in Angola. I need to find out the following:
1. Can this survey (AMIS 2011) accurately measure infant and child mortality for the country or is it (its sample) only focusing on malaria-related cases?
2. Secondly, the infant and child mortality levels (per 1000) from my descriptive results are different from the ones in the AMIS 2011 report. What could be causing this?

Regards,
Jacob


JacobM
Wits University
Re: Infant and child mortality [message #3091 is a reply to message #3085] Mon, 13 October 2014 18:28 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 787
Registered: January 2013
Senior Member
Hi Jacob,

1) The sample does not focus only on malaria-related cases, and so it is possible to calculate all-cause mortality rates, however the survey did not focus on the collection of data for child mortality.
2) The calculation of the mortality estimates in DHS surveys is described in the Guide to DHS Statistics ( http://www.dhsprogram.com/publications/publication-dhsg1-dhs -questionnaires-and-manuals.cfm) pages 90-94, so you can see if your approach differs from the DHS standard approach.
Re: Infant and child mortality [message #3099 is a reply to message #3091] Wed, 15 October 2014 10:25 Go to previous messageGo to next message
jacobm is currently offline  jacobm
Messages: 4
Registered: October 2014
Location: Polokwane
Member
Good day

Thank you very much for the information and response.

When I did the frequencies for variable 'child', I get ages from 0-4 years for current at of child and 1 - 59 months for age at death - months imputed. Can't these then be used/grouped into infants (0 years - current age and 1-11 age at death - months imputed) and children (1-4 years-current age and 12-59 months - age at death- months imputed)?

The above is the process I have followed to formulate and compute infant and child mortality using this data-set and ended up with very high infant mortality levels (e.g 192 per 1000) and very low child mortality levels (e.g 25 per 1000). The question raised was that 'what caused the sharp instant increase in death just from 12 months and higher in Angola?'. Again the AMIS 2011 report had presented both infant and child mortality levels.

Regards,
Jacob


JacobM
Wits University
Re: Infant and child mortality [message #3102 is a reply to message #3085] Wed, 15 October 2014 14:01 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 787
Registered: January 2013
Senior Member
Hi Jacob,

I'm not sure what variable you are referring to as there is no variable 'child', but I'm guessing that you are using the KR file (e.g. AOKR61FL), and you are trying to calculate simple ratios. This unfortunately will give you completely biased estimates. You need to use the births recode (BR) dataset as you need to include children who were born before the last 5 years, but were exposed to the risk of death within the last 5 years, and some of whom may have died in the last five years. Mortality rates are actually probabilities, and are typically calculated using life table approaches. Simple ratios as you have done won't given you good estimates. Please see the explanation in the Guide to DHS Statistics.

When I look at the data presented in the AMIS 2011 report I do not see a "sharp instant increase in death just from 12 months and higher". The table on page 38 shows a decrease in mortality rates in the last 5 years.
Re: Infant and child mortality [message #3588 is a reply to message #3102] Wed, 14 January 2015 00:59 Go to previous messageGo to next message
jacobm is currently offline  jacobm
Messages: 4
Registered: October 2014
Location: Polokwane
Member
Morning

I am requesting a step-by-step STATA Do-file on how to compute infant and child mortality rates.

I have went through the discussions/explanations made on the DHS forum but (being a beginner on analysis) I am still struggling to use the information successfully. I have also familiarised myself with the procedures as stated in the Guide to DHS Statistics (pp 90-94).

Please email the do file directly to my email address: jacobm@statssa.gov.za



JacobM
Wits University
Re: Infant and child mortality [message #3676 is a reply to message #3085] Sun, 25 January 2015 23:03 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 787
Registered: January 2013
Senior Member
We don't have Stata do files for the calculation of infant and child mortality currently available for distribution.
Re: Infant and child mortality [message #3703 is a reply to message #3676] Thu, 29 January 2015 09:16 Go to previous messageGo to next message
user_ha is currently offline  user_ha
Messages: 4
Registered: January 2015
Member
Could you at least let us know whether it is possible to reproduce the DHS estimates with Stata's ltable command (as recommended by the World bank Institute)? I try to estimate the yearly U5MR & IMR for Ghana (1980-1990) using the ltbale command, but I haven't succeeded yet to come up with the same figures/estimates. Thanks in advance!
Re: Infant and child mortality [message #3762 is a reply to message #3085] Fri, 06 February 2015 08:57 Go to previous messageGo to next message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 787
Registered: January 2013
Senior Member
I have never tried with ltable, but I suspect that the calculation approach is slightly different, so it is likely that you will get slightly different (but not meaningfully different results). Can you send us a link to where the World Bank Institute suggests using ltable?
Re: Infant and child mortality [message #5571 is a reply to message #3762] Tue, 09 June 2015 13:54 Go to previous message
busbyj2 is currently offline  busbyj2
Messages: 3
Registered: June 2015
Location: Austin Texas
Member
Page 32
http://siteresources.worldbank.org/INTPAH/Resources/Publicat ions/459843-1195594469249/HealthEquityCh3.pdf
Previous Topic: mother's age at birth
Next Topic: multilevel analysis
Goto Forum:
  


Current Time: Thu Mar 28 09:22:49 Coordinated Universal Time 2024