[afro-nets] Trends in maternal mortality: 1990 to 2008 - Estimates developed by WHO, UNICEF, UNFPA and The World Bank (2)

["Ikuo Takizawa." <Takizawa.Ikuo@jica.go.jp>]

Owing  to the UN Millennium Summit, health indicators such as maternal mortality  and  child  mortality  are  now  center of attention.  As a public  health practitioner, I appreciate the commitment of the global leaders, including our own (linked below), in mobilizing resources for the  health  of  mothers  and  children, even though we need continued advocacy and monitoring.
http://www.un.org/en/mdg/summit2010/debate/JP_en.pdf

While  we  are all concerned about the numbers to tell us where we are in  the  achievement  of  MDGs, it seems we have two different sets of trend  estimate for both maternal mortality and child mortality, i.e., the   one   by   inter-agency  group  of  international  organizations and another   set by  Institute  for  Health  Metrics  and  Evaluation (IHME), the University of Washington.

It  is  a good news for the world that both estimates are in agreement in  important aspects.  Both indicate that there is an acceleration of child  mortality  decline  in  2000s  compared to the previous decade. Both  suggest that maternal mortality is likely to be in decline since 1990   unlike   previous  estimates.  Sub-Saharan  Africa  is  not  an exception.

However,   when  you  look  at  the  trend  estimates  for  individual countries, there are several important disagreements.  Some examples;

-  Countdown  to 2010 Group classifies countries which achieved annual reduction  of  4%  or more in under five mortality rate since 1990 (or countries  with  the  latest  under five mortality of less than 40 per 1,000 live births) as "On Track" for achieving MDG4.  If you apply the criteria  to the  inter-agency  group  estimate, Eritrea, Malawi, Cape Verde,  Botswana,  Mauritius  and  the Seychelles are classified as On Track in Sub-Saharan Africa, as of 2008.  On the other hand, if we use the IHME  estimate, Liberia, Madagascar, Mauritius and Cape Verde will be  the ones On Track (the estimate for the Seychelles is not provided in IHME estimate), as of 2010.

   -  Annual  rate  of reduction in maternal mortality of 5.5% or more is needed  to achieve MDG5. According to the inter-agency group estimate, Equatorial  Guinea  and Eritrea are the only two countries on track in Sub-Saharan  Africa,  while  it  is Cape Verde only if we use the IHME estimate, both as of 2008.

I  fully understand estimating mortality trend at country level is not an  easy task and both groups are trying hard to come up with the best possible  estimate.   We  should  not interpret these numbers with too rigid  precision  because  of  statistical  errors and wide intervals, especially  for  MMR.   I also appreciate that having different expert groups working   on   the  same  numbers  will  contribute  to  better understanding of the issue and refinement of the methodology.

However,  most  of  the  public  health  practitioners  like mysel are "information  takers"  rather  than  "information makers".  We need to rely on the works of those experts who are dedicated to the production of these  numbers.   Having  two  sets of "best estimate" for the same numbers leads us into confusion.

How  are the information makers going to deal with this situation?  Do we  expect  to  have  another  set  of  "best" estimate in near future consolidating the two?

Ikuo TAKIZAWA, MSc.
Regional Project Formulation Advisor for Health
Regional Strategy Unit for Africa (Nairobi)
Japan International Cooperation Agency (JICA)
Tel: +254-20-2724877
Fax: +254-20-2720190
mailto:Takizawa.Ikuo@jica.go.jp