Objectives To assess the rates of SARS-CoV-2 positivity in babies born to mothers with SARS-CoV-2 infection, the timing of mother-to-child transmission and perinatal outcomes, and factors associated with SARS-CoV-2 status in offspring. Design Living systematic review and meta-analysis. Data sources Major databases between 1 December 2019 and 3 August 2021. Study selection Cohort studies of pregnant and recently pregnant women (including after abortion or miscarriage) who sought hospital care for any reason and had a diagnosis of SARS-CoV-2 infection, and also provided data on offspring SARS-CoV-2 status and risk factors for positivity. Case series and case reports were also included to assess the timing and likelihood of mother-to-child transmission in SARS-CoV-2 positive babies. Data extraction Two reviewers independently extracted data and assessed study quality. A random effects model was used to synthesise data for rates, with associations reported using odds ratios and 95% confidence intervals. Narrative syntheses were performed when meta-analysis was inappropriate. The World Health Organization classification was used to categorise the timing of mother-to-child transmission (in utero, intrapartum, early postnatal). Results 472 studies (206 cohort studies, 266 case series and case reports; 28 952 mothers, 18 237 babies) were included. Overall, 1.8% (95% confidence interval 1.2% to 2.5%; 140 studies) of the 14 271 babies born to mothers with SARS-CoV-2 infection tested positive for the virus with reverse transcriptase polymerase chain reaction (RT-PCR). Of the 592 SARS-CoV-2 positive babies with data on the timing of exposure and type and timing of tests, 14 had confirmed mother-to-child transmission: seven in utero (448 assessed), two intrapartum (18 assessed), and five during the early postnatal period (70 assessed). Of the 800 SARS-CoV-2 positive babies with outcome data, 20 were stillbirths, 23 were neonatal deaths, and eight were early pregnancy losses; 749 babies were alive at the end of follow-up. Severe maternal covid-19 (odds ratio 2.4, 95% confidence interval 1.3 to 4.4), maternal death (14.1, 4.1 to 48.0), maternal admission to an intensive care unit (3.5, 1.7 to 6.9), and maternal postnatal infection (5.0, 1.2 to 20.1) were associated with SARS-CoV-2 positivity in offspring. Positivity rates using RT-PCR varied between regions, ranging from 0.1% (95% confidence interval 0.0% to 0.3%) in studies from North America to 5.7% (3.2% to 8.7%) in studies from Latin America and the Caribbean. Conclusion SARS-CoV-2 positivity rates were found to be low in babies born to mothers with SARS-CoV-2 infection. Evidence suggests confirmed vertical transmission of SARS-CoV-2, although this is likely to be rare. Severity of maternal covid-19 appears to be associated with SARS-CoV-2 positivity in offspring. Systematic review registration PROSPERO CRD42020178076. Readers’ note This article is a living systematic review that will be updated to reflect emerging evidence. Updates may occur for up to two years from the date of original publication.

A novel learning process that does not require stimulus associations was explored in humans. The hypothesis was that two contextual stimuli taught in separate settings, with different stimuli, become equivalent if they accomplish identical functions with regard to the relations between the stimuli presented with them. The procedure consisted of : (a) first teaching an AB conditional discrimination (e.g., match A1 to B1 and A2 to B2) and then teaching a second-order XAB conditional discrimination in which X1 indicated performing the same selections as in AB and X2 indicated selecting the alternative comparison (e.g., match A1 to B2 and A2 to B1); (b) repeating the procedure with completely new stimuli, YHJ, in which the functions of the Y stimuli were identical to those of X; and (c) conducting a final probe under extinction to verify the equivalence between the X and the Y stimuli. Three experiments were conducted to explore the process and to rule out the influence of alternative variables. Out of these, 13 of the 14 participants matched the stimuli to the same contextual functions. Thus, the hypothesis was verified. These results demonstrate that humans are able to match stimuli according to their functions in relation to other stimuli. This process may be very much involved in language; for example, understanding that words or clauses that have been learned in separate contexts and with separate stimuli share the same meaning. Understanding this process may help to identify learning or developmental problems, such as those shown by persons with autism, and help to treat them.