Background The COVID-19 pandemic may have caused a sudden change in the parental socioeconomic composition of live births. Consequently, any future differences in health or socioeconomic outcomes between the cohort conceived and born during the pandemic and earlier cohorts could be attributable to changes in parental socioeconomic composition. Our objective was to compare the observed socioeconomic composition of babies born between December 2020 and December 2021 with their counterfactual composition had the pandemic not occurred.

Methods We used population-wide register data covering over 76.5 million live births from January 2015 to December 2021 from Austria, Brazil, Colombia, Ecuador, England, Finland, Mexico, the Netherlands, Scotland, Spain, the United States, and Wales to estimate level changes in the number of weekly or monthly live births by pre-pandemic parental socioeconomic characteristics (household income; education; area-level deprivation) from December 2020 onwards using interrupted time series Poisson regressions. Based on the results, we estimated the counterfactual number of births for each socioeconomic group had the pandemic not happened. We inferred compositional change from the differences between observed proportions and counterfactual proportions of each group.

Results Compared with counterfactual proportions, the December 2020-December 2021 birth cohorts had a higher proportion of babies born to parents in the least deprived areas in Scotland (percentage point difference=1%[95%CI:0.4%;1.5%]) and England (0.7%[95%CI:0.6%;0.9%]), in the highest fifth of household income in Finland (0.9%[95%CI:0.2%;1.6%]) and Netherlands (0.6%[95%CI:0.2%;1.0%]), and with post-secondary or tertiary education in Spain (2.5%[95%CI:2.2%;2.9%]), the United States (1.0%[95%CI:0.9%;1.1%]), and Austria (0.8%[95%CI:0.2%;1.3%]). In contrast, in Ecuador (-3%[95%CI:-3.3%;-2.7%]) and Brazil (-0.7%[95%CI:-0.7%;-0.8%]), the proportion of babies born to parents in the least deprived areas was lower than their counterfactuals. In Colombia (-0.6%[95%CI:-0.8%;-0.4%]) and Mexico (-0.3%[95%CI:-0.4%;-0.1%]), the proportion of babies born to mothers with upper secondary education or higher was lower than their counterfactuals.

Conclusion In all included countries, the COVID-19 pandemic produced changes in the socioeconomic composition of births conceived during the pandemic. Our results may imply that in contexts where material barriers and sociocultural norms made it more difficult for women in socioeconomically disadvantaged positions to postpone pregnancy, the composition changed towards a more disadvantaged birth cohort. In contexts with higher agency in fertility decision making, the composition changed towards a less disadvantaged birth cohort. These compositional changes may cause between-cohort differences in health and socioeconomic outcomes that are affected by the socioeconomic position of parents even if in-utero or early life exposure to the pandemic had no direct effect on these outcomes.