Subject Area="Monitoring, evaluation and quality control."
Objective="Performance monitoring to improve services."
Details for Reform Option "Auditing Maternal Deaths, Tamil Nadu"
Reporting of maternal mortality did not include the direct and indirect causes of deaths. This is essential for deciding the interventions for reduction of maternal mortality and morbidity.
Systematic reporting and auditing of maternal deaths was first introduced in Tamil Nadu in 2000. All maternal deaths are reported directly within 24 hours to the Commissioner- Maternal and Child Health & Family Welfare by the field health functionaries, Anganwadi workers, Primary Health Centre (PHC) medical officers. This is followed by a detailed investigation report by an obstetrician within 15 days. The multiple reporting is rectified at the state level by the statistical staff.
A clear system has been laid down which specifies the person responsible for reporting the death (according to whether it occurs at home, in transit, in a health sub-centre, PHC, public hospital or in a private health facility or nursing home).
The Deputy Director of Health Services (DDHS) at the district level is responsible for collecting the relevant information and monitoring the reporting to the Commissioner, the Director of Public Health and the Joint Director of Health Services within 24 hours of the occurrence/ receipt of the information.
A district level maternal death investigation team has been formed to improve the quality of investigation. The team, whose obstetrician rotates on a monthly basis, is accompanied on its visit by the local PHC medical and nursing staff. The DDHS makes the logistical arrangements (fixes the dates, arranges the vehicle).
To guide its work, the team uses a “Maternal Death Investigation Case Sheet” which collects information on:
(i) the location of the death,
(ii) the economic, social and educational profile of the family,
(iii) the deceased’s obstetric history and record of antenatal, delivery and postnatal care, referral and
(iv) the circumstances of death.
The team meets the relatives of the deceased and visits the health premises where she was treated to examine case records and interview staff.
It is the obstetrician’s responsibility to analyse the direct and indirect obstetrical causes which led to death. The report is also expected to highlight any system failures.
The other members of the team examine specifically the non-medical causes of death including antenatal care, risk factors and/or complications, delay in referral or in initiation of treatment, non-availability of specialists, equipment, blood, etc.
The team feeds back their findings to all the personnel who were involved in care with suggestions as to corrective measures designed to prevent recurrence.
The findings are also placed before the maternal deaths medical audit committee on a monthly basis. This committee’s minutes are placed before the District Reproductive and Child Health (RCH) Committee chaired by the District Collector, which also receives relatives of the deceased who are invited to give their account of the events.
The minutes of both meetings are placed before the Commissioner. The findings are also fed back to the relevant First Referral Units (FRUs) and PHCs. An annual analysis of maternal deaths is carried out at district level to give insight into the corrective measures required.
A state level committee, which has responsibility for assessing the quality of maternal death investigations, visits FRUs at random and reports back to the Commissioner every month.
A quarterly meeting is held with the Joint Director of Health Services and the DDHS to discuss measures for the reduction of maternal mortality and morbidity on the basis of the district investigation reports.
No specific evaluation of the impact of maternal death audits has been carried out till now. Anecdotal evidence suggests that it is an effective tool in raising awareness of the preventable and very often non-medical causes of maternal deaths.
Now the maternal death audit has been improved further and verbal autopsy of maternal death is being conducted.
Two state level training programmes have been conducted- one for the district health managers and another for the district public health nurses to brief them about the new format.
The new format includes the community-based survey i.e. interview with the relatives of the deceased and facility-based survey.
Following the state level meeting, two district level sensitisation meetings are being held for the PHC medical officers, the obstetric specialists from the hospitals and the officers from the health-related departments.
All the maternal deaths reported from April 2004 are being investigated and the first state level meeting was held in August 2004, to review the use of the new formats.
Apart from the purchase at state level of a computer capable of reading the machine-readable forms at an approximate cost of INR 5 lakh, the main costs are staff time and stationery (paper, etc.).
The cost of sensitisation meetings conducted at the state level for the district officers, at the district level for the PHC medical officers and at the PHC level for the female health functionaries is around INR 4,00,000 (€7,154).
The entire verbal maternal death audit activity is funded by UNICEF.
Tamil Nadu from 2000.
Verbal audit initiated from April 2004.
Total twelve months, including:
Development of investigation format – discussion with expert groups: two months.
Pre-testing of forms: one month.
Printing of forms: one month.
State level workshops for the district health administrators and district public health nurses: two months.
District level workshops for the PHC medical officers: three months.
Conduct of PHC level sensitisation meetings: three months.
Focussed: The system focuses attention on maternal deaths and their relatively preventable nature, something which has frequently been ignored.
Inclusive approach: Taking a non-judgmental approach, it focuses attention at all levels, including on the bereaved families and on the types of situation and circumstances which lead to maternal death.
Fear factor: Some health staff may not be cooperative fearing they will be blamed for a death.
Possible opposition: From the investigators who may feel that it is an additional responsibility.
-- Policy clearance.
-- Administrative will.
-- Willingness of service providers to investigate.
-- Funding support.
-- Single department to monitor the collection of reports, analysis and feed back.
-- Monitoring system to check the quality of investigation.
Who needs to be consulted
Policy makers to be informed about the process. All programme officers at state level. All district health managers. Hospital administrators. Obstetric specialists of the hospitals. Professional associations. PHC medical officers and paramedical functionaries.
The system does not have any major costs attached to its implementation so its sustainability depends essentially on political will and management effectiveness.
Chances of Replication
At present (September 2004) there was no known replication of this specific system but since there are few costs involved, other than staff time, it should be easily replicable.
Maternal death is a traumatic event and studies proved that a maximum recall period is 8 years for maternal deaths. It is therefore important to conduct the investigation before memories fade and details become blurred.
Sara Joseph, Researcher, ECTA, New Delhi. September 2004.