From Pre-Implementation to Institutionalization: Lessons From Sustaining a Perinatal Audit Program in South Africa.

INTRODUCTION: Maternal and perinatal death surveillance and response (MPDSR), or related forms of maternal and perinatal death audits, can strengthen health systems. We explore the history of initiating, scaling up, and institutionalizing a national perinatal audit program in South Africa. METHODS: Data collection involved 56 individual interviews, a systematic document review, administration of a semistructured questionnaire, and 10 nonparticipant observations of meetings related to the perinatal audit program. Fieldwork and data collection in the subdistricts occurred from September 2019 to March 2020. Data analysis included thematic content analysis and application of a tool to measure subdistrict-level implementation. This study expands on case study research applied to 5 Western Cape subdistricts with long histories of implementation. RESULTS: Although established in the early 1990s, the perinatal audit program was not integrated into national policy and guidelines until 2012 but was then excluded from policy in 2021. A network of national and subnational structures that benefited from a continuity of actors evolved and interacted to support uptake and implementation. Intentional efforts to demonstrate impact and enable local adaptation allowed for more ownership and buy-in. Implementation requires continuous efforts. Even in 5 subdistricts with long histories of practice, we found operational gaps, such as incomplete meeting minutes, signaling a need for strengthening. Nevertheless, the tool used to measure implementation may require revisions, particularly in settings with institutionalized practice. CONCLUSION: This article provides lessons on how to initiate, expand, and strengthen perinatal audit. Despite a long history of implementation, the perinatal audit program in South Africa cannot be assumed to be indefinitely sustainable or final in its current form. To monitor uptake and sustainability of MPDSR, including perinatal audit, we need research approaches that allow exploration of context, local adaptation, and underlying issues that support sustainability, such as relationships, leadership, and trust.

Implementation of the Every Newborn Action Plan: Progress and lessons learned.

Progress in reducing newborn mortality has lagged behind progress in reducing maternal and child deaths. The Every Newborn Action Plan (ENAP) was launched in 2014, with the aim of achieving equitable and high-quality coverage of care for all women and newborns through links with other global and national plans and measurement and accountability frameworks. This article aims to assess country progress and the mechanisms in place to support country implementation of the ENAP. A country tracking tool was developed and piloted in October-December 2014 to collect data on the ENAP-related national milestones and implementation barriers in 18 high-burden countries. Simultaneously, a mapping exercise involving 47 semi-structured interviews with partner organizations was carried out to frame the categories of technical support available in countries to support care at and around the time of birth by health system building blocks. Existing literature and reports were assessed to further supplement analysis of country progress. A total of 15 out of 18 high-burden countries have taken concrete actions to advance newborn health; four have developed specific action plans with an additional six in process and a further three strengthening newborn components within existing plans. Eight high-burden countries have a newborn mortality target, but only three have a stillbirth target. The ENAP implementation in countries is well-supported by UN agencies, particularly UNICEF and WHO, as well as multilateral and bilateral agencies, especially in health workforce training. New financial commitments from development partners and the private sector are substantial but tracking of national funding remains a challenge. For interventions with strong evidence, low levels of coverage persists and health information systems require investment and support to improve quality and quantity of data to guide and track progress. Some of the highest burden countries have established newborn health action plans and are scaling up evidence based interventions. Further progress will only be made with attention to context-specific implementation challenges, especially in areas that have been neglected to date such as quality improvement, sustained investment in training and monitoring health worker skills, support to budgeting and health financing, and strengthening of health information systems.

Early clinical signs in neonates with hypoxic ischemic encephalopathy predict an abnormal amplitude-integrated electroencephalogram at age 6 hours.

BACKGROUND: An early clinical score predicting an abnormal amplitude-integrated electroencephalogram (aEEG) or moderate-severe hypoxic ischemic encephalopathy (HIE) may allow rapid triage of infants for therapeutic hypothermia. We aimed to determine if early clinical examination could predict either an abnormal aEEG at age 6 hours or moderate-severe HIE presenting within 72 hours of birth. METHODS: Sixty infants ≥ 36 weeks gestational age were prospectively enrolled following suspected intrapartum hypoxia and signs of encephalopathy. Infants who were moribund, had congenital conditions that could contribute to the encephalopathy or had severe cardio-respiratory instability were excluded. Predictive values of the Thompson HIE score, modified Sarnat encephalopathy grade (MSEG) and specific individual signs at age 3-5 hours were calculated. RESULTS: All of the 60 infants recruited had at least one abnormal primitive reflex. Visible seizures and hypotonia at 3-5 hours were strongly associated with an abnormal 6-hour aEEG (specificity 88% and 92%, respectively), but both had a low sensitivity (47% and 33%, respectively). Overall, 52% of the infants without hypotonia at 3-5 hours had an abnormal 6-hour aEEG. Twelve of the 29 infants (41%) without decreased level of consciousness at 3-5 hours had an abnormal 6-hour aEEG (sensitivity 67%; specificity 71%). A Thompson score ≥ 7 and moderate-severe MSEG at 3-5 hours, both predicted an abnormal 6-hour aEEG (sensitivity 100 vs. 97% and specificity 67 vs. 71% respectively). Both assessments predicted moderate-severe encephalopathy within 72 hours after birth (sensitivity 90%, vs. 88%, specificity 92% vs. 100%). The 6-hour aEEG predicted moderate-severe encephalopathy within 72 hours (sensitivity 75%, specificity 100%) but with lower sensitivity (p = 0.0156) than the Thompson score (sensitivity 90%, specificity 92%). However, all infants with a normal 3- and 6-hour aEEG with moderate-severe encephalopathy within 72 hours who were not cooled had a normal 24-hour aEEG. CONCLUSIONS: The encephalopathy assessment described by the Thompson score at age 3-5 hours is a sensitive predictor of either an abnormal 6-hour aEEG or moderate-severe encephalopathy presenting within 72 hours after birth. An early Thompson score may be useful to assist with triage and selection of infants for therapeutic hypothermia.

Defining hypoxic ischemic encephalopathy in newborn infants: benchmarking in a South African population.

OBJECTIVES: There are few population-based studies of hypoxic ischemic encephalopathy (HIE) in sub-Saharan Africa, and the published criteria that are used to define and grade HIE are too variable for meaningful comparisons between studies and populations. Our objectives were (1) to investigate how the incidence of HIE in our region varies with different criteria for intrapartum hypoxia and (2) to determine how encephalopathy severity varies with different grading systems. METHOD: We reviewed the records of infants with a diagnosis of HIE born between September 2008 and March 2009 in public facilities in the Southern Cape Peninsula, South Africa.The incidence of HIE was calculated according to four definitions of intrapartum hypoxia and graded according to three methods. RESULTS: Depending on which defining criteria were applied,the incidence of HIE varied from 2.3 to 4.3 per 1000 live births, of mild HIE ranged from 0.4 to 1.3 per 1000 live births, and of moderate-severe HIE ranged from 1.5 to 3.7 per 1000 livebirths. Ninety-seven of the 110 (88%) infants reviewed had at least one intrapartum-related abnormality. Only 62 (56%) infants had a blood gas performed in the fi rst hour of life. CONCLUSION: The incidence and grade of HIE can vary more than 2-fold in the same population, depending on which defining criteria are used. Consensus definitions are needed for benchmarking.