Having worked in West Africa I was interested in Lelia Duley's (1992) paper and would like to make two comments. Firstly, Ms. Duley is obviously aware of the difficulties in obtaining accurate data on maternal mortality but it continues to worry me that ‘pockets’ of extremely high maternal mortality exist unbeknown to the WHO database. For example Ghana's MMR is quoted as 670/100 000 whereas it was 1700/100 000 in the hospital in which I worked (Fiander 1991). The worst maternal mortality scenarios affect the least privileged and the most isolated of the world's women and their plight deserves documentation. In this respect could not the WHO set up a ‘notification procedure’ by which health workers could report their local experience?
Secondly, it is my observation that the problem of hypertensive disease varies between geographical areas within a country such as Ghana. For example, hyertensive disease was an important cause of maternal mortality in the more southern tropical rain forest regions but I rarely saw pregnancy-related hypertension and never saw eclampsia in the northern savannah regions. Obviously environmental, ethnic or dietary factors must be operating either by preventing or facilitating hypertensive disease. These require further investigation and evaluation as they may have important implications for the prevention of maternal death.
As Ms Duley concludes ‘how to reduce deaths associated with hypertensive disorders of pregnancy will not become clear until there is better evidence than is currently available.’