A prospective cohort study included pregnant and postpartum women (up to the 14th day of postpartum), asymptomatic or diagnosed with flu syndrome and/or suspected COVID-19, regardless of age, admitted to the Gynaecology and Obstetrics service of the Central Hospital of Maputo (HCM), Mozambique, from 20 October 2020 to 22 July 2021.

The HCM is a teaching and referral maternity hospital for the region and the country, with comprehensive obstetric care. The screening for SARS-CoV-19 infection in pregnant and postpartum women is similar to the general population, focused on symptomatic individuals or those with a history of contact with a positive case. We intentionally estimated a sample size of 300 participants (pairs of pregnant women and newborn) as the evidence on the effects of COVID-19 on pregnancy was paucity when we were implementing our study.

The study protocol included pregnant (regardless of the gestational age) and postpartum women (up to 14th day of puerperium) who attended the HCM obstetrical and gynaecological services and provided or signed the consent form. At hospital admission or soon after, the research team (nurses, resident doctors and consultant obstetrician) identified, invited and assessed for eligibility criteria all potentials participants (in the emergency room and/or patient wards) after giving complete study information, including procedures.

After reading and signing the informed consent form, the participants were asked to provide upper respiratory specimens for laboratory screening of SARS-CoV-2 infection. We excluded all women with invalid telephone numbers who did not accept providing upper respiratory specimens or withdrew their consent form during the study.

We collected nasopharyngeal and oropharyngeal specimens through swabs. For asymptomatic patients, we collected specimens in duplicate. After collecting the specimens, they were placed in a viral transport medium (VTM) containing antifungal and antibiotic supplements. We storage and shipped the specimens in cooler boxes on ice (at 2–8 °C) to the local laboratory for viral detection. All sample viral detection was done via GeneXpert platforms for COVID-19, and the results were available within 24 h (2 h for symptomatic participants and 24 h for asymptomatic participants).

The laboratory detention virus followed two approaches: the symptomatic participants and/or severe acute respiratory illness and high-risk contacts were individually tested. Conversely, samples from asymptomatic participants with no history of positive contact for COVID-19 were tested using a pool testing strategy. Pool testing is a technique in which specimens collected from different participants are organised into groups (‘pools’) and tested together [20]. At the time of study implementation, the data from the Mozambican obstetric population suggested that the prevalence of COVID-19 was around 6% [21]. Therefore, we estimated a pool of nine samples (P9S3) analysed in three stages [20, 22]. The pool tests positive was further divided into sub-pools of three specimens before retesting each specimen in the pool individually to determine which individual(s) are positive.

The specimens’ collection, processing, and testing were carried out by health professionals previously trained for this purpose and according to the standards recommended by the Ministry of Health Mozambique and the World Health Organization for collecting and handling clinical specimens for COVID-19 testing.

Subsequently, the included participants were allocated into two groups according to the test result. The COVID-19 positive group consisted of pregnant and postpartum women (up to the 14th day) with a positive test for SARS-CoV-2 infection. The second group (COVID-19 negative) consisted of pregnant and postpartum women with a negative test. During the follow-up, participants with a negative test could move to the COVID-19 positive group if they were positive for SARS-CoV-2 infection COVID-19 when retested.

During inclusion and follow-up (until the 6th week postpartum), data on sociodemographic and obstetric characteristics (including usual means of transportation, alcohol consumption, source of antenatal care, and underlying medical condition), clinical characteristics of SARS-CoV-2 infection, adverse maternal events, maternal and perinatal outcomes were collected by the research team. Data were collected through in person or /and telephone interview and medical record review. Moreover, all study data were collected and managed by the research team using REDCap (Research Electronic Data Capture) electronic data capture tools installed in smartphones (tablets) hosted at Eduardo Mondlane University, Maputo, Mozambique [23, 24].

The primary outcome was the severe maternal outcome (maternal death, SARS and UCI admission). Secondary outcomes were: pregnancy outcomes (abortion, foetal death), preterm birth, preeclampsia/ eclampsia, mode of delivery, Apgar, NICU admission, neonatal death, congenital anomaly and any composite of adverse pregnancy outcome (NICU admission, preterm birth, foetal death, neonatal death, miscarriage/abortion). In addition, we have considered potential confounders variables, other viral respiratory syndromes, history of adverse pregnancy outcomes, and all factors related to the three-delay model in obstetric care.

We describe and compare the sociodemographic, obstetric and clinical characteristics of pregnant and postpartum women included in the study according to exposure (COVID-19 positive and COVID-19 negative groups). Likewise, we estimated the prevalence of COVID-19 in the general population, in the symptomatic and asymptomatic groups, and compared the clinical and severity characteristics in the group of symptomatic women according to the exposure group and estimated the level of significance (we considered Two-sided p-value < 0.05 as statistically significant).

We additionally have considered the time of symptom onset before admission, the duration of symptoms, the most prevalent symptoms, the type of management at the time of admission, admission to the intensive care unit and the presence of the severe acute respiratory syndrome. Finally, we estimated the unadjusted relative risk with a 95% interval to evaluate the risk of adverse maternal and perinatal outcomes. For comparisons of categorical variables, we used the Chi-square test or Fisher’s exact test when indicated. Statistical analyses were performed using the IBM SPSS statistic program (version 27.0).

The study protocol was approved by the Mozambique National Review Board (Letter of approval number 61/CNBS/2020). Moreover, all participants were fully informed regarding the study procedure and provided written or oral consent before their inclusion in the study. In addition, all participants had adequate clinical management (for SARS-CoV-2 positive cases) and psychological support when needed.