1 June 2012

Health Baseline within Different IAs - Presentation 1

Health Baseline for IA in Resource Poor Settings

  • It's about data and collection and there is a time component
  • Heath status and determinants of a given population wrt time, place and context.

Baseline as a snapshot


But what is it for, what is it supporting, what are we going to do with it?


Understand how the project might affect a population's health status.



Example: Malaria

  • WHO estimate 655,000 malaria deaths in 2010 but another agency says over 1 million
  • 3 important issues - need to know number of deaths, within what size of population and what time period
  • Challenge that using secondary data which can lead to very different estimates of baseline health status
  • Health clinic data means nothing without knowing what population is using the clinic, how are people accessing the service, why are they accessing, etc.
  • Often many disease cases aren't recorded in clinics or elsewhere.
  • Lots of challenges using secondary data.

Therefore, need to be very clear about what we are reporting, is the data really measuring what we think we are measuring, is the whole population captured, are all cases of disease reported.


Are we using primary data or secondary sources. What are the weaknesses in the sources and data.


Doing a baseline should help to develop health indicators e.g. Anaemia, pregnant women have a different range of normal values than non-pregnant women, similarly with differents age groups of children.


What should we report?

For Dengue should we use annual incident rate or annual case fatality e.g annual incidence may be low but case fatality may be high, I.e. few people get it but those who do more likely to die of it.



  • Describe health status
  • Estimate prevalence and distribution of main health outcomes and determinants
  • Contribute to understanding how project might affect population Wellbeing and equity
  • Use information to define an dmonitor recommendations and management plan
  • Need to be careful of using secondary data to infer meaningful conclusions
  • If local data is not available then the data used is not a baseline but perhaps should call it community profiling










No comments:

Post a Comment