Severe acute maternal morbidity and maternal death audit--a rapid diagnostic tool for evaluating maternal care.

OBJECTIVE To analyse severe acute maternal morbidity (SAMM) and maternal mortality in the Pretoria region over a 2-year period (2000-2001). SETTING Public hospitals in the Pretoria region, South Africa, serving a mainly indigent urban population. METHODS A descriptive study was performed whereby women with SAMM and maternal deaths were identified at daily audit meetings and an audit form was completed for all cases fulfilling the definition of SAMM ('near miss') and for all maternal deaths. RESULTS The number of maternal deaths declined slightly but not significantly from 18 deaths in 2000 to 16 in 2001. This represents a change in the maternal mortality ratio (MMR) from 130/100,000 live births in 2000 to a MMR of 100/100,000 live births in 2001. However, when data for women with SAMM and maternal deaths were combined, there was a significant increase in major maternal morbidity from 90 cases (SAMM and maternal death rate 649/100,000 live births) in 2000 to 142 cases (SAMM and maternal death rate 889/100,000 live births) in 2001 (p = 0.006). This increase was due to a significant increase in severe maternal morbidity related to abortions and obstetric haemorrhages. CONCLUSION Analysis of maternal deaths only in the Pretoria region failed to identify abortions and haemorrhages as major maternal care problems. When data for women with SAMM were combined with data for maternal deaths, however, these problems were clearly identified, and remedial action could be taken. Including SAMM in maternal death audits increases the rapidity with which health system problems can be identified.