The coronavirus disease 2019 (COVID-19) pandemic is the most serious global public health crisis of the 21st century. As of July 29, 2021, more than 190 million people worldwide have been infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) (1). Coronavirus infection affects pregnant women more than the general population, and infected pregnant women are more likely to need intensive care treatment than non-pregnant women (2, 3). The most common reported adverse perinatal outcomes related to COVID-19 are preterm birth and cesarean delivery (3–5).

Another concern with COVID-19 infection during pregnancy is that of the potential effects on early childhood development (6), which might occur via two possible pathways. The first is the pathophysiological effects of SARS-CoV-2 infection. Although, there is insufficient evidence of vertical SARS-CoV-2 infection transmission (7), mothers infected during pregnancy have higher interleukin-6 (IL-6) levels than non-pregnant women (8). IL-6 is an indicator of maternal systemic inflammation that is potentially related to infant brain development (9) and inversely associated with offspring cognition at 12 months of age (10). The second pathway is mother–infant separation caused by hospitalization. In the early phase of the pandemic, due to insufficient understanding of the transmission dynamics of COVID-19, and to treat infected pregnant women promptly and protect their newborns, mothers were separated from their infants for long periods (11). Mother–infant separation in early childhood has adverse effects on child development (12). Unfortunately, there is no evidence of whether SARS-CoV-2 infection during pregnancy increases the risk of neurodevelopmental delay in early childhood.

Our team has reported the clinical characteristics of pregnant women infected with SARS-CoV-2 (in the third trimester, mainly in the late trimester) and assessed the early development of their children (4, 13) using the National Epidemic Reporting System (NERS) of the National Health Commission of China. In this study, we used the previous data as an infection cohort, and established a non-infection cohort in Wuhan, China, recruiting pregnant women and their infants who had never been infected with SARS-CoV-2, to compare the early childhood development of the two cohorts and explore the potential secondary effects of COVID-19 pandemic on early childhood development.

This is the first cohort study to evaluate the potential direct and indirect effects of SARS-CoV-2 infection during pregnancy (mainly in the third trimester) on early childhood development. We followed 57 mother–infant dyads in an infection cohort and 78 dyads in a non-infection cohort until 3 months after delivery. Our findings showed that SARS-CoV-2 infection during pregnancy did not increase the risks of social–emotional (RR = 0.87, 95% CI: 0.51–1.49) and overall (RR = 1.02, 95% CI: 0.60–1.73) developmental delay. However, the length of mother–infant separation was significantly negatively correlated with the gross motor score (β = −0.09, 95% CI: −0.16 to −0.02). Furthermore, using a mediation model, we found that SARS-CoV-2 infection had a significant indirect effect on the gross motor score by increasing the length of mother–infant separation.

Betacoronaviruses, including SARS-CoV, MERS-CoV, and SARS-CoV-2, have caused severe global pandemics. A study after the 2003 SARS epidemic showed that SARS-CoV infection during pregnancy did not increase the risk of developmental delay in children compared with children of mothers who were not infected with SARS-CoV during pregnancy (19). There was no evidence that MERS-CoV infection during pregnancy affected early offspring development. Current evidence suggests that SARS-CoV, MERS-CoV, and SARS-CoV-2 affect fewer children, cause fewer symptoms, and less severe disease, and are associated with much lower case-fatality rates, compared with adults (20, 21). Studies have proposed a potential mechanism of SARS-CoV-2 infection during pregnancy on early offspring development, including maternal–fetal interface destruction, maternal systemic immune response, and direct brain infection interfering with brain development (22); however, these hypotheses have not been verified. Our findings provide epidemiological evidence that SARS-CoV-2 infection during pregnancy does not increase the risk of early developmental delay in offspring, compared with non-infected infants. In addition, it should be noted that the mothers in the infection cohort in the present study were mainly infected in the third trimester (accounting for 93.0%). Our results cannot explain the impact on the neurodevelopment of offspring when the infection occurs in the early pregnancy, especially in the first trimester, which needs to be explained by more studies in the future.

In the further mediation model, we found that SARS-CoV-2 infection during pregnancy had significant indirect effects on the ASQ-3 gross motor domain score by increasing the length of mother–infant separation. After separating the indirect effects, the remaining direct effect was not significant, which was consistent with the results of the multivariable linear regression models. To prevent neonatal infection (11), infants in the infection cohort experienced longer separation from their mothers than the non-infection cohort, which means that the infants in the infected group did not receive enough support and response from their mothers to meet their developmental needs (12). Early separation interrupted skin-to-skin contact, breastfeeding, and mother–infant interactions, which had adverse effects on the physiological development (23), innate and specific immunity (24), and neurodevelopment (including language, cognition, motor, and social development) of the children (25). In particular, we observed that the breastfeeding rate of the non-infected cohort at 1 and 3 months was 100%, which was significantly higher than that of the infection cohort. Better breastfeeding would be more conducive to early childhood development. Although this study did not find its significant effect, it still deserves continuous attention. Thus, we confirmed the hypothesis of the mediation model that mother–infant separation may have a potential impact on early childhood development. Furthermore, our results also indicated that although the impact of SARS-CoV-2 infection during late pregnancy on early childhood development is limited, the secondary effects of the COVID-19 pandemic (such as mother–infant separation) may still affect early childhood development. It is also worth noting that a published study showed that when adequate protective measures are taken, the risk of infant infection caused by rooming-in practice is limited (26). Combined with our results, it is necessary to focus on the benefit of protected rooming-in, on the premise that the risk of infant infection can be effectively controlled.

The COVID-19 pandemic has caused both a health crisis and economic and social crises (27–29), which have profoundly affected sustainable development goals (30). The mother–infant separation discussed here is only the tip of the iceberg of this global crisis. The World Bank Group estimates that the global economy contracted 4.3% in 2020; the pandemic raised poverty rates to between 9.1 and 9.4%, back to levels last seen in 2017 (31). Poverty is the source of many risk factors for early childhood development, including malnutrition, caregiver stress and depressive symptoms, low caregiver responsiveness, low-quality nurturing care, and community environment, causing them to miss early education (32). The COVID-19 pandemic is accelerating the deterioration of these external factors, pushing the poorest and most vulnerable people and their children into worse situations. We call for global attention to the plight of these families and their children in the current pandemic.

Our study had some limitations. First, the study population was small, since NERS reported only 138 pregnant women confirmed with COVID-19 as of April 30, 2020, which might affect the power of our study. Second, we found that only the length of mother–infant separation was negatively associated with the ASQ-3 gross motor domain score. Therefore, we used the gross motor domain score in the mediation model instead of the classification result. Therefore, we found only that SARS-CoV-2 infection had a potential indirect negative effect on the development of the children, which did not mean that this indirect effect would increase the risk of developmental delay in the offspring. In addition, the recruitment and follow-up were conducted through the internet and by telephone, and participant compliance was not as satisfactory as with face-to-face follow-up. Since our follow-up lasted only 3 months, our results cannot represent longer-term effects. Finally, we used screening scales instead of diagnostic scales to access early childhood development, which might cause evaluation bias.

In conclusion, we found that SARS-CoV-2 infection during late pregnancy did not increase the risk of offspring developmental delay at 3 months. However, we found that the length of mother–infant separation was significantly negatively correlated with the ASQ-3 gross motor domain score. We found that SARS-CoV-2 infection indirectly affected the gross motor score by increasing the length of mother–infant separation, mediating 84.73% of the total effect. Future researchers should conduct longer follow-up studies of children whose mothers were infected with SARS-CoV-2 during pregnancy, to evaluate the long-term effects of COVID-19 on early childhood development. Future researchers also need to pay more attention to potential mediators and indirect effects between COVID-19 and early childhood development as the pandemic is now changing the external environment of child development.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Peking University Third Hospital Medical Science Research Ethics Committee (No. IRB00006761–M2020127). The patients/participants provided their written informed consent to participate in this study.

JQ, XW, YZ, and YWe conceived and designed the study. TW, LC, and YWa conducted the statistical analyses and drafted the initial manuscript. TW, LC, YWa, HS, JN, XY, ML, CT, HJ, and DZ conducted the follow-up surveys. HS and XW contributed to the interpretation of the data. All authors reviewed and revised the article. All authors read the final manuscript and approved submission.

This study was funded by the Chinese Academy of Engineering (2020-KYGG-01-06), the National Natural Science Foundation of China (72042013), the Peking University Health Science Center (BMU2020HKYZX001), and the National Key Research and Development Program of China (2016YFC1000307-2).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.