Perinatal death surveillance and response to improve survival of babies

The mortality audit (or review) process is an established tool to assess the events around a death. Applying an audit cycle can highlight breakdowns from local to national levels and ultimately improve civil registration and vital statistic (CRVS) systems and quality of care. Maternal death surveillance and response (MDSR) is a form of this strategy that has been used by many countries[1].

Less information, however, has been captured and assessed on stillbirths and neonatal deaths[2]. In 2014, 51 priority countries reported having a policy on maternal death notification, and only 17 countries had a policy for reporting and reviewing stillbirths and neonatal deaths[3].

According to the International Classification of Diseases (ICD)-10, the ‘perinatal period’ refersExpert Opinion_Perinatal death surveillance and response_box 1 to antepartum (before birth) and intrapartum (during birth) stillbirths, and early neonatal deaths (see table for WHO definition). However, in line with the WHO guide – Making Every Baby Count: Audit and review of stillbirths and neonatal deaths – the term ‘perinatal’ will also refer to
late neonatal deaths (occurring eight to 28 days after birth) as deaths during this period may be influenced by circumstances from the perinatal stage[4].

 THE SILENT TRAGEDY OF PERINATAL DEATHS: THE FACTS

Every year, more than five million babies die worldwide in the last few weeks of pregnancy, during labour or soon after delivery[5], shattering the hopes and dreams of parents for a healthy baby.

  • About 2.7 millExpert Opinion_Perinatal death surveillance and response_WHO infographicion newborn babies die in their first month of life each year[6]
  • Another 2.6 million babies die every year before birth in the last trimester of pregnancy[7]
  • Of these, 1.3 million stillbirths take place during labour (intrapartum stillbirths)[8]
  • The majority of stillbirths, especially intrapartum, are preventable[9]
  • Three quarters of all neonatal deaths are preventable[10]

Most stillbirths happen in countries where fewer women can access high quality care before and around the time of birth. With 40 million women giving birth at home each year without a skilled birth attendant, a focus on improving the quality of intrapartum care can help prevent stillbirths and about half of all maternal deaths[11].

8432697316_2e2509ae61_z

Moreover, many stillbirths and newborn deaths go unrecorded. Most stillbirths and half of all newborn deaths do not receive a birth certificate and are not registered[12].

To improve the survival of babies, we need to know more about how many stillbirths and newborn deaths taExpert Opinion_Perinatal death surveillance and response_Quote box 1ke place, where and why they occur, what are the contributing factors and what can be done to prevent similar events. This information can help identify the avoidable factors, and improve service provision and overall quality of care through health systems to prevent babies dying from similar causes in the future.

HOW AUDITS CAN REDUCE PERINATAL DEATHS

Reported country experiences of implementation suggest perinatal death audits may be a useful tool for reducing stillbirths and neonatal deaths in facilities, and can improve quality of care, as long as the audit loop can be closed[13].

The ability to respond effectively to recommendations identified through audits is pivotal to reducing mortality. When successful, audits can result in a 30% reduction in perinatal deaths[14].  This reduction will help countries meet target 3.2 of the third Sustainable Development Goal (to reduce neonatal mortality to at least 12 per 1,000 live births by 2030)[15], and Goals 1 and 2 of the Every Newborn Action Plan (to meet the target of ten or less newborn deaths per 1,000 live births and ten or less stillbirths per 1,000 total births by 2035)[16].

How can national, sub-national or facility committees make sure that perinatal death reviews are detailed enough? And how can they be supported to develop and address modifiable factors? The available literature is sparse on how best to optimise quality-of-care audits as a tool to reduce perinatal mortality in a low- and middle-income setting and further investigation is needed[17].

We asked five experts in maternal, newborn and child health to share their knowledge and experiences in perinatal death surveillance and response (PDSR) around the world.

EXPERT PROFILES

Dr Animesh Biswas, Senior Scientist at the Centre for Injury Prevention and Research, Bangladesh, supports the implementation of maternal and perinatal death surveillance and response (MPDSR) in country. He also conducts research on maternal and neonatal review systems (e.g. social autopsy) and recently received his doctorate degree in this very topic.

Ms Kate Kerber is the Senior Technical Specialist with Save the Children’s Save Newborn Lives programme based in Canada. She has worked both in South Africa and Uganda on mortality audit programmes and is currently co-leading a multi-country evaluation of MPDSR implementation. Ms Kerber is a contributor to the WHO Making Every Baby Count: audit and review of stillbirth and neonatal death guidance.

Dr Natasha Rhoda is a neonatologist in South Africa. She is an attending clinician at the Perinatal Mortality and Morbidity Meetings (PNMM) at the Groote Schuur Hospital and is the chairperson of the National Perinatal Morbidity and Mortality Committee. Previously, she was responsible for ensuring that the Perinatal Problem Identification Program (PPIP) was functional.

Dr Nathalie Roos, Program Officer at the Maternal, Newborn, Child and Adolescent Health Department at the WHO, is a gynaecologist and obstetrician providing technical guidance and implementing support to MDSR at the global level. Dr Roos is a contributor to three recent WHO publications to strengthen data on stillbirths, and maternal and neonatal deaths.

Dr Tunde Segun is a public health physician and has over twenty years of experience in maternal and newborn health.  He is the Country Director for the MamaYe-Evidence for Action programme in Nigeria and is supporting country efforts to develop a MPDSR system.

QUESTIONS AND ANSWERS

Q: What are your thoughts on investing in a perinatal death surveillance and response (PDSR) system, particularly where MDSR already exists but is not fully functional?

Kerber: At the facility level, ideally, perinatal death review should take place aExpert Opinion_Perinatal death surveillance and response_Quote box 2nywhere that maternal death reviews are happening. A full death review isn’t as feasible at the community level given the additional number of stillbirths and neonatal deaths compared to maternal deaths. All systems should be looking to at least count stillbirths and newborn deaths, and collect basic information on each of these.

Roos: Yes – as the mother and baby share the same periods of risk, there are lessons to be learnt from perinatal deaths as well as maternal deaths to save the lives of both mothers and their babies. The information should therefore not be separated.

Segun: I agree; perinatal deaths should be included in the review process. That makes sense. However, the challenges (in incorporating a perinatal component when healthcare providers are struggling with maternal death reviews) are in the implementation. We need to ask:  are there enough human resources to review the deaths? If we are reviewing all maternal deaths, can we also review all perinatal deaths knowing that the numbers are higher?

Rhoda: The 18-years of experience of MDSR in South Africa has given the participating clinical staff (doctors and nurses) insight into the importance of reviewing maternal deaths with the main focus on prevention and improvement of the quality of care provided. On the back of this, we have extended this review system to the perinatal population, starting with the review of all neonatal deaths. For us, it is the correct time to invest in perinatal deaths. We have the data, [we have identified the] avoidable causes and now people who understand a MDSR process can see the benefit of including perinatal deaths.

Q: How can countries implement PDSR mechanisms? What recommendations would you suggest?

Roos: The existing platform or processes for MDSR should be used when collecting information and reviewing perinatal deaths with the idea of building on what is already in place rather than creating parallel systems. “Think big, start small and grow steadily”. When the system is working well then you could add on perinatal deaths.

Kerber: I don’t know if the perceived dichotomy between maternal and perinatal systems exists as much as a dichotomy between facility and community. It is much more difficult to count every death that occurs in the community. With very few exceptions, those involved in maternal death review should be the same as those involved in perinatal death review.

The WHO application of ICD-10 to deaths during the perinatal period: ICD-PM publication highlights the inseparable link between mother and baby by providing a classification system that enables a maternal condition to be added to a baby’s cause of death. CommExpert Opinion_Perinatal death surveillance and response_Quote box 3unication amongst health professionals is key, especially across disciplines, and the maternal and perinatal death surveillance and response (MPDSR) system can really facilitate that. It’s important, however, to make sure that blame is not shifted to other cadres and that safeguards are put in place to maintain an atmosphere of learning and improvement rather than punishment.

Rhoda: [When we first starting reviewing perinatal deaths], our health staff was numbed by the large numbers of stillbirths and neonatal deaths to the extent that they had come to ignore them. However with carefully crafted newborn messages and strategies, we have been able to build a solid platform to address perinatal deaths.

One of the biggest obstacles in developing PNMM in South Africa was working with the management staff – partly because they were not clinically proficient and partly because they saw this as a possible intrusion on their work. Once we engaged with them and discussed why this was important, the PNMM functioned better. It is also important that a qualified and senior clinician is present at the PNMM to ensure that sound clinical knowledge is used to make decisions and the causes of perinatal deaths are coded accurately.

Q: What are the options or possible approaches to introducing MPDSR to avoid ‘dilution’ of effect with MDSR efforts? Please provide examples or recommendations.

Segun: One option is to approach [MPDSR] in a phased manner i.e. start with MDSR before proceeding to MPDSR. This was the approach in Nigeria – MDRs were first established before integrating the “P”.

Rhoda: In retrospect, the running of MDSR [in South Africa] has provided a strong foundation on which to build perinatal death surveillance. We have already had ten years of recording data on the perinatal deaths in the PPIP and have published nine [national] Saving Babies reports based on the data. These were all necessary steps which have allowed perinatal death surveillance nationally. Our next step is to ensure that facilities not only report, but act more efficiently on the avoidable factors.

Roos: One recommendation [to help strengthen CRVS] is that all hospitals should collect data on all births and deaths, which is the minimum set of perinatal indicators.

Kerber: Yes – I agree – improving care at the time of birth affects both mother and baby. Improving the civil registration and vital statistics system and data collection processes improves data across the health system so there are a lot more synergies than competitions. However, I think it is wise not to review the factors for each death (whether stillbirth, neonatal or maternal) in settings where MPDSR is just getting started – rather select particular cases for a more-in-depth review. Otherwise the process can be very time consuming in places where the mortality burden is still quite high.

Q: What can be done to tackle the extra workload that perinatal death audits may add to already over-stretched health workers?

Expert Opinion_Perinatal death surveillance and response_Quote box 4Roos: As noted, perinatal deaths are much higher in number than maternal deaths, which can be a challenge in settings where health workers are already overstretched. The new WHO guide for conducting audit and review of stillbirths and neonatal deaths, suggests a way forward to review a sample of all perinatal deaths. Even though it may not be possible to review each death, one case can provide learnings to prevent similar events from happening in the future.

Kerber: As quality of care improves, fewer complications and deaths occur and work-related stress may actually decrease, if not the actual number of deliveries and workload. I also think good leaders can facilitate a culture of review, learning, and improvement instead of making MPDSR just another task to get through.

Q: What advice would you give on sampling which perinatal deaths to review in settings where incidence is high, assuming the aim is to count all?

Roos: The WHO guide (which I previously mentioned) proposes different approaches. One approach is to analyse maternal and perinatal deaths if they occur together. Another possible approach is to analyse all deaths which, occur in the first week of each month, or having a thematic approach to analysing all deaths caused by a specific condition.

Kerber: If there is no existing system of death notification at community level, and existing stigma around reporting stillbirths that occur at home, even a sampling system is a resource-intensive undertaking. Getting a count on all deaths that occur in the facility and creating a culture around the importance of reporting all births and deaths is likely a good place to start.

In places where the burden of death is high, I love the idea of reviewing near-misses and positive outcomes as a morale-boosting opportunity, but also because there are important lessons to highlight in those cases around teamwork, preparedness and innovative thinking.

I also think it’s helpful to pick a theme – e.g. just review deaths in a certain birth-weight category, due to a particular cause, or those that occurred on the weekend – and vary this at each meeting, so that death review is not such a tedious process.

Q: What are your reflections on the feasibility and scope of community-level audits?

Kerber: I think it is essential to involve the community in understanding and implementing the recommendations from facility-based death reviews. This has been successful by having a community liaison sit on the mortality audit committee, under the same confidentiality rules as the other participants. I think it is more challenging to conduct the whole process at the community level (e.g. counting the deaths, investigating the cause, coming up with solutions) and I haven’t seen this done at wide scale in routine systems.

Segun: We know that many deaths take place in communities that are not captured or classified appropriately. A community-level audit will strengthen the continuum of care across the community and facility level. It will also build relationships between communities and health facilities. Improvements in CRVS systems can have a positive ripple effect.

Biswas: In my opinion, it is essential to give greater emphasis to commExpert Opinion_Perinatal death surveillance and response_Quote box 5unity-based audits. The findings can help governments better understand the medical and social factors associated with maternal and perinatal deaths.

An active surveillance system can generate a quick response from local health managers to implement solutions. In Bangladesh, deliveries commonly happen at home or on the way to the facility. [Moreover] many community cases are treated by unskilled individuals. Community-based audits will provide a clearer picture about why, how, when and where deaths occur. These findings can also help communities understand the importance of seeking adequate health care from facilities.

Rhoda: In South Africa, we have not started the open conversation of the burden of the perinatal deaths with our communities. In our experience with HIV, we need to think carefully about how to share such high figures with communities to avoid scare-mongering that may create a backlash. I do think that some sort of positive media campaign must be part of the department plan to address this problem.

Q: In your experience, can you describe some of the cultural and social constraints that challenge effective implementation of PDSR, and how they can be addressed or overcome?

Biswas: In Bangladesh, different social and belief systems must be taken into account, particularly in hard to reach areas (i.e. hill areas which are very different to, say, the coastal belt). In these areas, communities are influenced by myths, traditional practices and cultural beliefs that restrict them from seeking appropriate healthcare. Social autopsy is a key example of a method that can improve the knowledge and perception of the community around the factors contributing to a death. It can sensitise communities by dispelling myths, building knowledge and changing care seeking behaviour.  It can also engage community leaders in the discussion to ensure their buy in to lead change.

Kerber: We need advocacy to shift our norms around counting every birth and death, and strengthening vital registration systems in general.

Expert Opinion_Perinatal death surveillance and response_Image 2Q: How can blame be avoided or addressed when conducting perinatal (or maternal) audits?

Segun: In Nigeria, the approach to conducting MPDSR is on the basis of ‘no name, no blame’. Emphasis is placed on the ability to learn from each event to ultimately improve quality of care. This will encourage healthcare providers to continue to conduct reviews, ensuring their sustainability. The other aspect is to shield MPDSR findings entirely from legal processes to prevent findings from being subpoenaed. However if there is evidence of gross negligence, then there are other ways healthcare providers can be sanctioned i.e. through disciplinary committees of specific cadre councils (e.g. the Medical and Dental Council, the Nursing and Midwifery Council, and so on).

Kerber: I think blame is one of the biggest barriers. There must be legal protection in place for healthcare professionals, such that the discussions in mortality audit meetings are separate from any disciplinary or legal action. Anonymity may be difficult to guarantee in smaller facilities or on smaller teams, but confidentiality outside of the meetings is a requirement. Having a code of conduct that participants sign at each meeting may help reinforce the goal of the review meeting being a safe space.

Biswas: The issue of blame is a key challenge in a death review system at the community level – something that was discovered during Bangladesh’s piloting of maternal and perinatal death review (MPDR) in 2010. Healthcare and family planning professionals that were conducting MPDRs were well known in the communities where they worked. This meant that they had to be assured by higher-level officials that they would not be blamed for notifying deaths or assessing their causes. This reassurance approach enabled health workers to accurately report and review each death, as well as reducing underreporting, because they did not fear the repercussions. Health personnel, during MPDR trainings, were trained in strategies to avoid blame while conducting death reviews in communities. When reviewing a maternal or perinatal death, the health workers were trained to explain to the community the purpose for the review and collection of information. If blame were to arise during the review process, then the health workers would bring the focus back to the medical and social factors affecting each death.

Rhoda: Our best example of a well-run PNMM is in the rural areas, 200km outside of Cape Town. A senior clinician does outreach to the peripheral hospitals and part of his duty is to lead the PNMM meetings. Staff flock to his meeting as they know him and he addresses them as equals. He uses examples as a teaching opportunity to prevent recurrence of avoidable factors. He specifically focuses on the near misses and morbidity issues, and not only deaths. His meetings are regular –  notification is one year in advance so people can plan ahead to be present (e.g. third Wednesday of every month at 08:00) – and on some occasions he also covers a topic as part of the Continuing Medical Education programme which is an incentive. Staff know that they will be heard and that their opinions are valued.

Q: What, in your opinion, are the priority (i.e. key research and evidence) gaps in our knowledge for the successful implementation of MPDSR?

Rhoda: We have to first start with ensuring that issues with data quality are addressed.  The PPIP, for example, is a free download and while the limitation is [the need for] a computer, it can be a useful tool for data analysis and for managers to monitor their units.

8429654207_d59e9f8556_zIn terms of financial and human resources to oversee and run data systems, the institutionalisation of PPIP should be mandatory. We have numerous examples locally where PPIP champions (e.g. provincial data coordinators) are no longer available and as a result, both the quality and quantity of the data falls dramatically.  However, I think, institutionalisation must be the end goal for any health system to make maternal, perinatal and neonatal death surveillance and response sustainable in the reduction of mortalities and morbidities. Clinical insight (i.e. good clinical leadership and understanding of epidemiology) in the analysis of data is important to ensure that appropriate actions are made to prevent the recurrence of deaths.

Biswas: In Bangladesh, more research is needed on the response component of MPDSR implementation, particularly around how data collected is used to inform solutions and drive change. Lessons learnt on this topic could help inform the gaps in knowledge and improve the implementation of MPDSR. On a global level, more research on community-level audits is necessary as there is a greater focus on facility-based audits.

Segun: Using Nigeria as an example, MPDSR is at the beginning, so mapping what exists is a priority to better understand the scope of the system across the country, including challenges and lessons learnt.

Kerber: There are three questions I would like answered:

  1. Is there a cost-effective way of collecting data on perinatal deaths at the community level within routine systems in high-burden settings?
  2. How do we involve communities in the review of deaths at facility level while maintaining a safe space for both health professionals and affected families?
  3. How do we move from data collection and discussion of gaps to implementing recommendations?

REFERENCES

Pattinson R , Kerber K, Waiswa P, Day LT, Mussell F  Asiruddin S, Blencowe H, Lawn JE.  (2009) Perinatal mortality audit: Counting, accountability, and overcoming challenges in scaling up in low- and middle-income countries. International Journal of Gynaecology and Obstetrics 107 (2009) S113–S122

World Health Organization (WHO). (2016). Making Every Baby Count: Audit and review of stillbirths and neonatal deaths. Geneva: World Health Organization.

Acknowledgments: This piece was compiled by Sara Nam, Technical Specialist, and Jenna de St. Jorre, Technical Assistant, of Evidence for Action (E4A) based on interviews and feedback from our five expert contributors.

[1] World Health Organization (WHO). (2016). Making Every Baby Count: Audit and review of stillbirths and neonatal deaths. Geneva: World Health Organization.

[2] Kerber, K.J., Mathai, M., Lewis, G., Flenady, V., Erwich, J.J.H.M., Segun, T., Aliganyira, P., Abdelmegeid, A., Allanson, E., Roos, N., Rhoda, N., Lawn, J.E., Pattinson, R. (2015). Counting every stillbirth and neonatal death through mortality audit to improve quality of care for every pregnant woman and her baby. BMC Pregnancy and Childbirth, 15(Suppl 2): S9.

[3] Kerber, K.J., Mathai, M., Lewis, G., Flenady, V., Erwich, J.J.H.M., Segun, T., Aliganyira, P., Abdelmegeid, A., Allanson, E., Roos, N., Rhoda, N., Lawn, J.E., Pattinson, R. (2015). Counting every stillbirth and neonatal death through mortality audit to improve quality of care for every pregnant woman and her baby. BMC Pregnancy and Childbirth, 15(Suppl 2): S9.

[4] World Health Organization (WHO). (2016). Making Every Baby Count: Audit and review of stillbirths and neonatal deaths. Geneva: World Health Organization.

[5] Child mortality estimates. New York (NY): United Nations Children’s Fund; 2015 (http://www.childmortality.org/  index.php?r=site/index, accessed 02 September 2016)

[6] Child mortality estimates. New York (NY): United Nations Children’s Fund; 2015 (http://www.childmortality.org/  index.php?r=site/index, accessed 02 September 2016)

[7] Heazell, A.E.P., Siassakos, D., Blencowe, H., Burden, C., Bhutta, Z.A., Cacciatore, J., Dang, N., Das, J., Flenady, V., Gold, K.J., Mensah, O.K., Millum, J., Nuzum, D., O’Donoghue, K., Redshaw, M., Rizvi, A., Roberts, T., Saraki, H.E., Storey, C., Wojcieszek, A.M. & Downe, S. (2016). Stillbirths: economic and psychosocial consequences. Early online publication

[8] Heazell, A.E.P., Siassakos, D., Blencowe, H., Burden, C., Bhutta, Z.A., Cacciatore, J., Dang, N., Das, J., Flenady, V., Gold, K.J., Mensah, O.K., Millum, J., Nuzum, D., O’Donoghue, K., Redshaw, M., Rizvi, A., Roberts, T., Saraki, H.E., Storey, C., Wojcieszek, A.M. & Downe, S. (2016). Stillbirths: economic and psychosocial consequences. Early online publication.

[9] Bhutta ZA, Das JK, Bahl R, Lawn JE, Salam RA, Paul VK et al. Can available interventions end preventable deaths in mothers, newborn babies, and stillbirths, and at what cost? Lancet. 2014;384(9940):347–70. doi:10.1016/S0140–6736(14)60792–3

[10] Bhutta ZA, Das JK, Bahl R, Lawn JE, Salam RA, Paul VK et al. Can available interventions end preventable deaths in mothers, newborn babies, and stillbirths, and at what cost? Lancet. 2014;384(9940):347–70. doi:10.1016/S0140–6736(14)60792–3

[11] Heazell, A.E.P., Siassakos, D., Blencowe, H., Burden, C., Bhutta, Z.A., Cacciatore, J., Dang, N., Das, J., Flenady, V., Gold, K.J., Mensah, O.K., Millum, J., Nuzum, D., O’Donoghue, K., Redshaw, M., Rizvi, A., Roberts, T., Saraki, H.E., Storey, C., Wojcieszek, A.M. & Downe, S. (2016). Stillbirths: economic and psychosocial consequences. Early online publication.

[12] Lawn JE, Blencowe H, Oza S, You D, Lee ACC, Waiswa P et al.; for The Lancet Every Newborn Study Group. Progress, priorities, and potential beyond survival. Lancet. 2014;384(9938):189–205. doi:10.1016/S0140–6736(14)60496–7.

[13] Kerber, K.J., Mathai, M., Lewis, G., Flenady, V., Erwich, J.J.H.M., Segun, T., Aliganyira, P., Abdelmegeid, A., Allanson, E., Roos, N., Rhoda, N., Lawn, J.E., Pattinson, R. (2015). Counting every stillbirth and neonatal death through mortality audit to improve quality of care for every pregnant woman and her baby. BMC Pregnancy and Childbirth, 15(Suppl 2): S9.

[14] Pattinson R , Kerber K, Waiswa P, Day LT, Mussell F  Asiruddin S, Blencowe H, Lawn JE.  (2009) Perinatal mortality audit: Counting, accountability, and overcoming challenges in scaling up in low- and middle-income countries. International Journal of Gynecology and Obstetrics 107 (2009) S113–S122

[15] United Nations. (2016). Sustainable Development: Knowledge Platform. Retrieved September 29, 2016, from https://sustainabledevelopment.un.org/sdg3

[16] World Health Organization (WHO) & UNICEF. (2014). Every Newborn: An Action Plan to End Preventable Deaths. Geneva: World Health Organization.

[17] Allanson & Pattinson (2015) Quality-of-care audits in South Africa. Bull World Health Organ 2015;93:424–428|