AFRO-NETS> Maternal Deaths Disproportionately High... (5)

[Ehi Uche Igumbor <ehiuche@yahoo.com>]

Maternal Deaths Disproportionately High... (5)
----------------------------------------------
 
Dear All,
 
REPORT ON MATERNAL MORTALITY: ISSUES FOR HEALTH MEASUREMENTS
 
The recently reported estimates of maternal mortality are cer-
tainly startling in revelations (of numbers) as they are ground-
breaking in methodological approach. It seems to me that a sig-
nificant improvement has been made in the current report compared 
to previous reports of 1990 and 1995.
 
Building on the methods in previous reports on "dual strategy", 
viz. adjusting nationally-derived data and creating model-based 
estimates, the current work has also taken into account deaths 
due to HIV/AIDS and adjusted values accordingly. This is a major 
strength of the method used as it is axiomatic that HIV/AIDS 
would have exaggerated the estimates. The values therefore are 
delineated to ONLY maternal deaths based on the ICD-nomenclature.
 
However, certain gaps are observable that may impact on the accu-
racy of the measurements. A practical limitation was the unavail-
ability of data in most countries. "The methods at arriving at 
the final values for each country varied according to data avail-
ability and type". With the dearth of basic data in countries of 
sub-Saharan Africa, one can only imagine how much this could have 
meant strong assumption implications for most of the countries.
 
A review of the various methods used in measuring maternal mor-
tality highlights some key issues in use of vital registration, 
household surveys, sisterhood approaches, verbal autopsies, cen-
sus and RAMOS as means of measurement. The inadequacies are noted 
in the report with a caution that values should not be considered 
absolute but rather as estimates.
 
If this brilliant epidemiological work is to lend its methods to 
national and sub-national health measurements, there will be need 
to critically re-visit the varied approaches used and the impli-
cations of assumptions in the statistical model.
 
The "Next Step" aptly identifies improving epidemiological data 
as a priority. Health measurements in sub-Sahara Africa must 
arise to this awakening not only in terms of measuring Maternal 
Mortalities but indeed in all health estimations.
 
ON FINDINGS OF THE ESTIMATES
 
It is interesting, but there was surmounting evidence of im-
proved/improving maternal child health in some quarters with de-
creased maternal mortality estimates in the 1980s through to 
early 1990s (MCH reports). So where are these gains all gone to? 
This is particularly critical when it is noted that the report 
has adjusted values to exclude HIV/AIDS deaths. What then is hap-
pening? Why are values for sub-Saharan Africa for 2000 still re-
markably high?
 
Clearly, maternal child health should be given priority and bet-
ter understanding of "why maternal mortalities remain high..... 
how can we reduce it??" is needed.
 
A forum like ours could get the ball rolling in drawing up an-
swers to these crucial questions.
 
Ehi Uche Igumbor
South Africa
mailto:ehiuche@yahoo.com

--
To send a message to AFRO-NETS, write to: afro-nets@healthnet.org
To subscribe or unsubscribe, write to: majordomo@healthnet.org
in the body of the message type: subscribe afro-nets OR unsubscribe afro-nets
To contact a person, send a message to: afro-nets-help@healthnet.org
Information and archives: http://www.afronets.org