Maternal death inquiry and response in India - the impact of contextual factors on defining an optimal model to help meet critical maternal health policy objectives

BACKGROUND: Maternal death reviews have been utilized in several countries as a means of identifying social and health care quality issues affecting maternal survival. From 2005 to 2009, a standardized community-based maternal death inquiry and response initiative was implemented in eight Indian states with the aim of addressing critical maternal health policy objectives. However, state-specific contextual factors strongly influenced the effort's success. This paper examines the impact and implications of the contextual factors. METHODS: We identified community, public health systems and governance related contextual factors thought to affect the implementation, utilization and up-scaling of the death inquiry process. Then, according to selected indicators, we documented the contextual factors' presence and their impact on the process' success in helping meet critical maternal health policy objectives in four districts of Rajasthan, Madhya Pradesh and West Bengal. Based on this assessment, we propose an optimal model for conducting community-based maternal death inquiries in India and similar settings. RESULTS: The death inquiry process led to increases in maternal death notification and investigation whether civil society or government took charge of these tasks, stimulated sharing of the findings in multiple settings and contributed to the development of numerous evidence-based local, district and statewide maternal health interventions. NGO inputs were essential where communities, public health systems and governance were weak and boosted effectiveness in stronger settings. Public health systems participation was enabled by responsive and accountable governance. Communities participated most successfully through India's established local governance Panchayat Raj Institutions. In one instance this led to the development of a multi-faceted intervention well-integrated at multiple levels. CONCLUSIONS: The impact of several contextual factors on the death inquiry process could be discerned, and suggested an optimal implementation model. District and state government must mandate and support the process, while the district health office should provide overall coordination, manage the death inquiry data as part of its routine surveillance programme, and organize a highly participatory means, preferably within an existing structure, of sharing the findings with the community and developing evidence-based maternal health interventions. NGO assistance and the support of a development partner may be needed, particularly in locales with weaker communities, public health systems or governance.