Implementation of maternal and perinatal death surveillance and response (MPDSR) in humanitarian settings: insights and experiences of humanitarian health practitioners and global technical expert meeting attendees

Abstract

Background Maternal and perinatal death surveillance and response (MPDSR) is a system of identifying, analysing and learning lessons from such deaths in order to respond and prevent future deaths, and has been recommended by WHO and implemented in many low-and-middle income settings in recent years. However, there is limited documentation of experience with MPDSR in humanitarian settings. A meeting on MPDSR in humanitarian settings was convened by WHO, UNICEF, CDC and Save the Children, UNFPA and UNHCR on 17th–18th October 2019, informed by semi-structured interviews with a range of professionals, including expert attendees. Consultation findings Interviewees revealed significant obstacles to full implementation of the MPDSR process in humanitarian settings. Many obstacles were familiar to low resource settings in general but were amplified in the context of a humanitarian crisis, such as overburdened services, disincentives to reporting, accountability gaps, a blame approach, and politicisation of mortality. Factors more unique to humanitarian contexts included concerns about health worker security and moral distress. There are varying levels of institutionalisation and implementation capacity for MPDSR within humanitarian organisations. It is suggested that if poorly implemented, particularly with a punitive or blame approach, MPDSR may be counterproductive. Nevertheless, successes in MPDSR were described whereby the process led to concrete actions to prevent deaths, and where death reviews have led to improved understanding of complex and rectifiable contextual factors leading to deaths in humanitarian settings. Conclusions Despite the challenges, examples exist where the lessons learnt from MPDSR processes have led to improved access and quality of care in humanitarian contexts, including successful advocacy. An adapted approach is required to ensure feasibility, with varying implementation being possible in different phases of crises. There is a need for guidance on MPDSR in humanitarian contexts, and for greater documentation and learning from experiences.

Maternal and perinatal death surveillance and response: a systematic review of qualitative studies

“… 51 , 73 Third, an extreme example given was refusal of admission to referral facilities of women who seemed likely to die, possibly to avoid damaging mortality statistics. 76 Fourth, serious repercussions were reported for a woman who had complained that a midwife had treated her harshly; the midwife recognized herself in the audit session and complained to the woman’s parents. 75 …”

“…Fear of blame usually resulted from insufficient confidentiality or anonymity, and the death review process not being separated from disciplinary procedures. 76 Telling participants that the process was blame-free was insufficient to allay fears when senior managers were present who would also be in charge of disciplinary procedures 53 , 76 or when litigation against health workers was increasing. 73 …”

“…Under-reporting of deaths was often due to fear of blame or other negative consequences, such as reduced funding, 21 , 53 , 73 , 76 but also resulted from social stigma, 33 cultural beliefs, non-mandatory reporting 53 and political pressure. 51 , 72 , 73 …”

“…Inaccurate and/or incomplete information undermines the review process. Although poor record-keeping was common, 42 , 53 several reports noted deliberate falsification of records 25 , 57 , 70 , 73 or misclassification of deaths 70 , 76 to avoid blame or reputational damage. Sometimes staff did not collect the information because they simply did not have time 45 or the correct forms, 60 or did not understand the purpose of maternal and perinatal death surveillance and response.…”

“…Furthermore, implementation rarely happened if: responsibility for implementation was unclear; 44 the individuals responsible for implementation were not involved in the review; 21 , 38 , 54 , 60 recommendations were not fed back to those responsible for implementation; 30 , 44 implementers avoided taking responsibility; 40 , 43 or no mechanism was in place to follow up on implementation. 76 , 77 Insufficient resources also prevented implementation. 25 , 36 , 48 , 72 …”