We report the findings of the Malawian National Maternal Surveillance platform, and its characterisation of the clinical manifestation and outcomes of all pregnant and recently pregnant women hospitalised with confirmed COVID-19 in Malawi between 2nd June 2020 and 1st December 2021 across all district and central hospitals in Malawi.

Notably, 31% (76/246) women with confirmed COVID-19 in this cohort had a severe maternal outcome, of which 19% (47/246) died, and 12% (29/246) suffered a near-miss event. 19% (21/111) of women who delivered during their episode of admission had adverse perinatal outcomes, including 8% (9/111) of women who delivered having a stillbirth and 12% (12/101) of babies dying before discharge from hospital.

The findings in this cohort are stark evidence on the differing impacts and huge disparities of COVID-19 related maternal and neonatal outcomes across the world. Outcomes for COVID-19 positive women are far worse in this cohort than those expected in global data. A population-based cohort of 673 South African (Middle income Country) women admitted to hospital with COVID-19, reported that 32/39 (82%) of deaths were due to COVID-19 infection (5). The perinatal outcomes, 3% (5/148) stillbirths, 3% (5/143) neonatal deaths, from this South African study were also much lower than what we have reported. In addition, it is notable that even in high income settings, Black women had worse outcomes following infection (16).

The Malawian National Maternal Surveillance platform only reported women with COVID-19 who were hospitalised. This enabled the efficient and reliable reporting at scale of those women who were suffering from the most severe disease, but clearly only represents a small proportion of all those women infected. The stigma associated with COVID-19 and challenges in accessing and affording care means it is likely that only those with the most severe symptoms, or who also required hospital care for other pregnancy related reasons were included in this cohort (17). A phone survey of 4641 women and men from 27 districts across all three regions of Malawi, reported that in Malawi, 81% felt they would be treated poorly if they contracted COVID-19 (17). The very high rates of adverse outcomes must also be understood in the context of a maternity care system that even outside of the pandemic was under severe human and physical resource constraints, with maternal mortality rates at 349 per 100,000 births (18), well short of the required Sustainable development goal target of 70 deaths per 100,000 live births (19). In addition to adverse outcomes being directly caused by COVID-19, we also report women for whom COVID-19 was not the primary cause for admission, but nevertheless this diagnosis could have further complicated their care and contribute to the overall high rates of severe outcomes. For example, it could have delayed access to required surgical care or limited access to intensive care capacity as patients with confirmed COVID-19 were cared for in separate clinical areas for infection prevention and control purposes.

Though Malawi prioritised response to the pandemic, the wider impacts throughout the healthcare system were still profound with many preventative and community services disrupted, reduced access to care, as well as staff shortages due to sickness and isolation after being in contact with an individual with COVID-19 (20). In Malawi, there was no widespread community testing for asymptomatic pregnant or recently pregnant women, nor is it possible to disaggregate the national testing data to separately identify those women who were pregnant or recently pregnant. Therefore, the overall levels of asymptomatic or less severe disease in pregnant and recently pregnant women from the wider population during this time period is not known. It was also notable in our cohort that most women admitted to hospitals with COVID-19 did not have known comorbidities. This of course reflects the younger age of this population, and also widespread under ascertainment of chronic disease. Nevertheless, it underlines that health messages on vaccination and other COVID-19 infection prevention measures should target the whole population and that women without known co-existing diseases should not be complacent.

To the best of our knowledge, this is the first national facility-based cohort study of the impact of COVID-19 on pregnant women in SSA. It comprehensively describes the maternal characteristics and clinical outcomes in pregnant and recently pregnant women using data systematically and prospectively collected. It is also important evidence that national surveillance and real-time data provision for government and decision-makers from pregnant women is feasible in a low resource country and can be rapidly established. The success of the implementation of this approach at scale and at speed was due to the commitment of the Ministry of Health, its collaboration with technical partners and the integration with the existing well established Maternal Death Surveillance and Response network (21).

This was also efficient as the data was collected in every facility by the existing “safe motherhood coordinator” who already had similar responsibilities as part of their role for maternal death reporting and reviews. We sought to optimise the quality of data collection by providing in-person training alongside follow-up mentorship and real-time monitoring and feedback of data timeliness and quality. However, as the data relied on existing staff taking on these additional duties it is susceptible to underreporting. This may have been particularly problematic at times of increased service strain such as during the COVID waves and the extent of this effect cannot be readily quantified. This initiative leaves an important legacy in that maternal adverse outcomes beyond those caused by COVID-19 are now electronically recorded and the data rapidly accessible.

During the course of the data collection there was the introduction of lateral flow testing at many of the facilities, which was an important adjunct given the restricted access to molecular testing. These rapid tests were sometimes used at the bedside, but also sometimes used within the hospital laboratory. The type of COVID test undertaken may therefore not have been reliably reported by staff, and lateral flow test results were not universally confirmed with molecular testing, which might have led to misclassification of cases. This paper does not report vaccination status of the women because COVID-19 vaccines were only rolled out in March 2021 and recommended for pregnant women since August 2021 in Malawi. However, this information is now being recorded and will be reported in future. The MATSurvey platform required the daily review of patient whilst they remained in hospital, however, there was no provision for following up patient after discharge, so important outcomes such as neonatal outcomes were only available for those women who delivered during that hospital admission.

In conclusion, this paper addresses a critical gap in the global and regional COVID-19 data, and its impact on the important group of pregnant and recently pregnant woman. The study ascertains that maternal mortality and perinatal outcomes are alarmingly high in a cohort of pregnant and recently pregnant women admitted to Malawian healthcare facilities; a low income country in Africa. In addition, it demonstrates that national surveillance and real-time data provision from pregnant women is feasible in a low resource country and can be rapidly and successfully established.