The review identified 94 maternal and pregnancy outcomes, and 47 unique neonatal outcomes heterogeneously named and described in a large number of verbatims across 440 included studies. Mode of delivery, stillbirth, preterm birth, hypertensive disorders of pregnancy, and maternal death were among the most frequently reported maternal and pregnancy outcomes. These were gestational age at birth, congenital malformations of the nervous system, birth weight, neonatal admission to intensive care units, and neonatal death among neonatal outcomes .
The main strength of our study is the rigorous methods applied for conducting and reporting systematic reviews. Additionally, the external validity of our findings was expanded by the inclusion of alltypes of study designs, not only clinical trials, and use of no language restrictions. However, the review also has some limitations. The number of identified studies on effectiveness and efficacy was low, which hinders the understanding of the outcomes used in these studies, and the applicability of outcomes identified in this review to future interventional/product development studies. Furthermore, most of the studies addressed respiratory infections, mainly COVID-19 disease, which inevitably had an impact on frequency of reporting of identified outcomes, especially neonatal outcomes, which show some level of variation according to the mode of disease transmission. For COVID-19 studies, a sampling strategy was used to counteract the effect of their overrepresentation in the final list of outcomes. Although we did not formally check for saturation, a rapid exploration of COVID-19 studies not included in the review due to sampling strategy used, showed a high redundancy of outcomes. Admittedly, we could have missed some relevant studies published before 2015; but at the same time, we managed to cover several significant outbreaks that occurred in recent years, such as influenza. It should be noted that we identified a very high number of publications on COVID-19 disease which compensates for a relatively small number of included studies on other respiratory infectious disease outbreaks (e.g., Influenza 2009). There is also a possibility that we missed outcomes reported in smaller studies, that is those with less than 50 participants – and could potentially explain the very low number of studies that included covering hemorrhagic fever, zoonosis, or other recent outbreaks (e.g, Monkeypox). Finally, we failed to include studies reporting only on indirect effects of epidemics and pandemics (e.g., coverage or access to routine care such as number of prenatal visits), clinical presentation of disease – like fever, cough, headache – or risk factors. Other published COS, covering for example the COVID-19 disease and long-COVID condition, should be considered complementary to our efforts [27].
The top ranking maternal and pregnancy outcomes (mode of delivery, stillbirth, preterm birth/delivery, hypertensive disorders of pregnancy and maternal death), and neonatal outcomes (gestational age at birth, congenital malformations of nervous system, birthweight and neonatal admission to intensive care unit [related outbreak disease or other reason]) identified in this review includes outcomes commonly reported in maternal and perinatal health research, and does not reflect specific outcomes related to infectious diseases. This suggests that outcomes of interest may not depend on specific infectious diseases or mode of transmission. In fact, only a few top-rated outcomes identified in this review are disease specific or related to mode of transmission (maternal or neonatal confirmed infection and maternal pneumonia), and could indicate severity of infectious diseases (e.g., mortality, admission to ICU) or specific organ dysfunction (e.g., mechanical ventilation, need for oxygen support or respiratory failure). Although, relevance of certain outcomes may vary depending on the nature of the disease (e.g., zika infection and malformations). Surprisingly, mother-to-child transmission of infectious diseases was reported in relatively few (n = 13) studies included in the review. This could be related to challenges in defining vertical transmission early during the emergence of new infectious diseases [28].
Some of the unique pregnancy outcomes identified represent similar concepts and could be further combined. For example, spontaneous abortion/miscarriage, stillbirth, live birth, and neonatal death, represent a continuum of perinatal vital status [29]. However, depending on how studies are designed, it may not be possible to report outcomes across pregnancy trimester and after birth. This was evident during the COVID-19 pandemic where most studies covered late pregnancy and neonatal vital status, and few early pregnancy outcomes were covered [8, 30]. Other examples of overlap include low or abnormal APGAR score, reporting of actual APGAR score values, and birth asphyxia; or size for gestational age, which requires for its computation data on gestational age and birthweight. Other outcomes reported as maternal or neonatal outcomes represent similar outcomes, the only difference being whether these were reported in the obstetric or neonatal population in the studies included in this review. For example, preterm birth/delivery, gestational age at delivery or gestational age at birth all reflect gestational age at the end of a pregnancy and could be eventually considered as a unique outcome.
It is worth highlighting that among the most commonly reported outcomes were a few related to maternal mental health, such as depression, anxiety, and stress. All of these were reported in COVID-19 studies reflecting probably recent interest in maternal mental health and well-being, beyond survival and severe morbidity [31, 32]. For neonatal outcomes, a few outcomes related to newborn care (feeding and skin-to-skin) were reported in a small number of COVID-19 studies. The same applies for other outcomes of maternal cognitive or emotional functioning.
Several challenges remain unaddressed, such as the lack of comparability among studies and the difficulty in synthesizing data for meta-analyses. It is important to note that variations in reported outcomes in maternal and perinatal health are not exclusive to studies conducted in the context of epidemic threats; they have been observed across various maternal and perinatal conditions [33]. This is the case for outcomes such as stillbirths for which different gestational age and birthweight cut-off points are used, or variability in reporting vital status at birth, confounded by the need to appropriately assess gestational age at birth and birthweight. In consequence, misclassifications of perinatal status (stillbirths, live births and miscarriages) or prematurity may often occur [34]. Some other challenges relate to limited health workforce capabilities, availability of diagnostic technologies, and reporting systems, particularly in low-resource settings [35, 36].