Periodic Reporting for period 1 - LoCo-Effect (The Lockdown Cohort-Effect: Does the unequal impact of the COVID-19 pandemic on fertility lead to long-term consequences for health and public services provision?)
Période du rapport: 2024-07-01 au 2026-06-30
Résumé du contexte et des objectifs généraux du projet
The COVID-19 pandemic placed a heavy social and health burden on populations. Yet, because of potential social inequalities in the pandemic’s impact on fertility behaviour, babies conceived during the pandemic may be the most socially advantaged and healthiest birth cohort of the last decades. This counterintuitive but realistic consequence of the pandemic was hypothesised as Lockdown-cohort effect (LoCo-effect) in 2022 supported by preliminary results from Scotland. The LoCo-effect hypothesis is the main scientific motivation for this MSCA-funded project and hypothesises that the COVID-19 pandemic had an unequal effect on fertility across population groups resulting in a birth cohort with different parental characteristics compared with previous birth cohorts. That is, the unequal impact of the pandemic on families may have abruptly changed the socioeconomic and demographic composition of parents. For example, while disadvantaged groups might have postponed or abandoned plans for having a baby due to more precarious current and anticipated future circumstances, more advantaged groups might have taken the opportunities afforded by working from home and reduced travel to realise or bring their initial plans forward. Importantly, changes in parental socioeconomic composition that may produce the LoCo’s health advantage in birth weight and prematurity – and these outcomes themselves – will have long-run effects on health, developmental, and socioeconomic outcomes throughout this cohort’s life.
Thus, the LoCo-effect will likely have long-term consequences across the social and health sciences, and domains of public policy. First, if the pandemic-induced sudden compositional changes are confirmed, describing and disseminating understanding of this phenomenon will prevent a wide range of flawed inferences on COVID-19 consequences and misguided policy recommendations from future studies unaware of the LoCo’s unique composition. Second, being born in, on average, more socioeconomically advantaged circumstances, the LoCo may need less support from public services than previous birth cohorts. If, however, postponed pregnancies among disadvantaged groups are eventually realised in following birth cohorts, health and educational systems may need to prepare for an abrupt increase in service demand.
To better understand the consequences of the COVID-19 pandemic and future public health emergencies on parental composition and birth outcomes, this MSCA-funded project carried out large-scale comparative research with individual-level total population birth data currently covering 20 countries across sub-projects.
Thus, the LoCo-effect will likely have long-term consequences across the social and health sciences, and domains of public policy. First, if the pandemic-induced sudden compositional changes are confirmed, describing and disseminating understanding of this phenomenon will prevent a wide range of flawed inferences on COVID-19 consequences and misguided policy recommendations from future studies unaware of the LoCo’s unique composition. Second, being born in, on average, more socioeconomically advantaged circumstances, the LoCo may need less support from public services than previous birth cohorts. If, however, postponed pregnancies among disadvantaged groups are eventually realised in following birth cohorts, health and educational systems may need to prepare for an abrupt increase in service demand.
To better understand the consequences of the COVID-19 pandemic and future public health emergencies on parental composition and birth outcomes, this MSCA-funded project carried out large-scale comparative research with individual-level total population birth data currently covering 20 countries across sub-projects.
Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent
At the end of the project funding, two main achievements of this project stand out: i) creating a large-scale international collaboration using individual-level full population birth data, and ii) a landmark empirical and theoretical contribution to how macro-level shocks affect parental composition across national contexts published in a top generalist journal (Nature Communications).
First, using the opportunities afforded by this fellowship, I built a unique international multidisciplinary collaboration on comparative work using individual-level birth data analysis far beyond what was aimed for in the initial proposal. Instead of doing comparative analysis using register-based data from Finland, Scotland, Spain, Austria, the US and Brazil (5 countries) as planned, the first study on changes in the parental socioeconomic composition during the pandemic used full-population birth data from Austria, Brazil, Colombia, Denmark, Ecuador, England, Mexico, the Netherlands, Scotland, South Australia, Spain, Sweden, United States, and Wales (15 countries) covering about 78 million births between 2015-2021. For the second study on how social inequalities in birthweight distributions changed during the pandemic, the collaboration grew further and now continues to grow after the funding of this project ended. For this second study, we are currently using individual-level full population birth data from Austria, Brazil, Canada, Denmark, Ecuador, Finland, Hungary, Italy, Netherlands, New South Wales (Australia), Scotland, South Australia (Australia), Spain, Sweden, Taiwan, and the United States. The collaboration has gathered experience in its workflows to share statistical software code and exchange knowledge about country-specific pandemic experiences and data infrastructure. Moreover, as the coordinator of this collaboration, I have now gathered invaluable and outstanding experience in working with register-data infrastructures for scientific use in many countries across the globe. This and the scientific network created by this fellowship project puts me in a perfect position for further ground-breaking comparative social sciences research using individual-level register-based data.
Second, the first study of this project on changes in the socioeconomic composition of parents during the COVID-19 pandemic makes a major empirical and theoretical contribution to our understanding of how macro-level shocks affect parental composition of birth cohorts in differently across political contexts and pre-existing fertility regimes. We found that, compared with their counterfactual compositions, the December 2020-December 2021 birth cohort has a higher proportion of babies born to socioeconomically advantaged parents in Austria, England, Finland, the Netherlands, Scotland, Spain, Wales, and the United States while we observe the opposite change for Brazil, Colombia, Ecuador, and Mexico. These changes in cohort composition may cause between-cohort differences in life course outcomes that are influenced by parental socioeconomic circumstances even if early life exposure to the pandemic had no direct effect on this birth cohort. At the end of 2025 this study was published in a top generalist journal (Nature Communications).
Two more comparative studies are in progress after the project-funding has ended. In one study, which will be submitted before 2027, we estimate changes in mean and variance of birthweights across population groups across 16 countries using full population individual-level birth data. In this study we also suggest a new theoretical framework for analysing how social inequalities in birthweight change during macro-level shocks. This framework considers differential effects through socially stratified selection in-utero and selection at conception mechanisms and considers not only potential changes in the mean birthweights of population groups but also changes in the variance between and within population groups defined by demographic and socioeconomic characteristics: household income, parental education, marital status, maternal age, and parity. We found between-country differences in mean birthweight of up to 400 grams and similarly large within-country differences between population groups with different between-group variation across countries. Differences in grand mean birthweight between unexposed and exposed cohorts ranged from -15g to +40g with large within-country variation between population groups. These within-country differences in changes in mean birthweight were often patterned such that population groups with the lowest pre-pandemic mean birthweights had the largest decreases (or smallest increases) in birthweights or the highest increases (or smallest decreases) in birthweights within a country. These patterns in group-specific responses to the pandemic could emerge, disappear, or even reverse depending on whether female or male babies were analysed or whether pre-pandemic cohorts were compared to exposed babies conceived before or during the pandemic.
In the other study, we built on a finding of our previous results and estimate long-term trends in (maternal-age standardised) birthweight variation between socioeconomic groups putting our research on the COVID-19 pandemic into a longer time horizon. Our initial findings are that that more socioeconomically disadvantaged population groups had consistently higher variation in birthweight (with lower mean birthweights) but also that this inequality in variance has, after years of convergence, recently started to increase again in some countries.
First, using the opportunities afforded by this fellowship, I built a unique international multidisciplinary collaboration on comparative work using individual-level birth data analysis far beyond what was aimed for in the initial proposal. Instead of doing comparative analysis using register-based data from Finland, Scotland, Spain, Austria, the US and Brazil (5 countries) as planned, the first study on changes in the parental socioeconomic composition during the pandemic used full-population birth data from Austria, Brazil, Colombia, Denmark, Ecuador, England, Mexico, the Netherlands, Scotland, South Australia, Spain, Sweden, United States, and Wales (15 countries) covering about 78 million births between 2015-2021. For the second study on how social inequalities in birthweight distributions changed during the pandemic, the collaboration grew further and now continues to grow after the funding of this project ended. For this second study, we are currently using individual-level full population birth data from Austria, Brazil, Canada, Denmark, Ecuador, Finland, Hungary, Italy, Netherlands, New South Wales (Australia), Scotland, South Australia (Australia), Spain, Sweden, Taiwan, and the United States. The collaboration has gathered experience in its workflows to share statistical software code and exchange knowledge about country-specific pandemic experiences and data infrastructure. Moreover, as the coordinator of this collaboration, I have now gathered invaluable and outstanding experience in working with register-data infrastructures for scientific use in many countries across the globe. This and the scientific network created by this fellowship project puts me in a perfect position for further ground-breaking comparative social sciences research using individual-level register-based data.
Second, the first study of this project on changes in the socioeconomic composition of parents during the COVID-19 pandemic makes a major empirical and theoretical contribution to our understanding of how macro-level shocks affect parental composition of birth cohorts in differently across political contexts and pre-existing fertility regimes. We found that, compared with their counterfactual compositions, the December 2020-December 2021 birth cohort has a higher proportion of babies born to socioeconomically advantaged parents in Austria, England, Finland, the Netherlands, Scotland, Spain, Wales, and the United States while we observe the opposite change for Brazil, Colombia, Ecuador, and Mexico. These changes in cohort composition may cause between-cohort differences in life course outcomes that are influenced by parental socioeconomic circumstances even if early life exposure to the pandemic had no direct effect on this birth cohort. At the end of 2025 this study was published in a top generalist journal (Nature Communications).
Two more comparative studies are in progress after the project-funding has ended. In one study, which will be submitted before 2027, we estimate changes in mean and variance of birthweights across population groups across 16 countries using full population individual-level birth data. In this study we also suggest a new theoretical framework for analysing how social inequalities in birthweight change during macro-level shocks. This framework considers differential effects through socially stratified selection in-utero and selection at conception mechanisms and considers not only potential changes in the mean birthweights of population groups but also changes in the variance between and within population groups defined by demographic and socioeconomic characteristics: household income, parental education, marital status, maternal age, and parity. We found between-country differences in mean birthweight of up to 400 grams and similarly large within-country differences between population groups with different between-group variation across countries. Differences in grand mean birthweight between unexposed and exposed cohorts ranged from -15g to +40g with large within-country variation between population groups. These within-country differences in changes in mean birthweight were often patterned such that population groups with the lowest pre-pandemic mean birthweights had the largest decreases (or smallest increases) in birthweights or the highest increases (or smallest decreases) in birthweights within a country. These patterns in group-specific responses to the pandemic could emerge, disappear, or even reverse depending on whether female or male babies were analysed or whether pre-pandemic cohorts were compared to exposed babies conceived before or during the pandemic.
In the other study, we built on a finding of our previous results and estimate long-term trends in (maternal-age standardised) birthweight variation between socioeconomic groups putting our research on the COVID-19 pandemic into a longer time horizon. Our initial findings are that that more socioeconomically disadvantaged population groups had consistently higher variation in birthweight (with lower mean birthweights) but also that this inequality in variance has, after years of convergence, recently started to increase again in some countries.
Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent)
This project yielded results beyond the state of the art. We built our comparative research on full-population individual-level data analysis through international collaboration. This means we used data that is often restricted-access and needs to be applied and paid for at national statistical offices. We applied for data access specifically for this project for some countries and made use of existing access by collaborators with whom we share statistical software code to carry out our analysis. As of August 2026, this collaboration managed to analyse such data from 20 countries ranging from Brazil to Taiwan. The research carried out by this project is thus an example for future large-scale comparative research built on individual-level full population data. Thus, it is not only the scientific results of this project that will have an impact on social sciences but also the way this research was conducted that will hopefully influence a new branch of comparative social science.
Scientific results of the published first sub-study will inform future social and health science research that includes individuals born 2020-2021 (the Lockdown Cohort). This is because the changes in parental socioeconomic composition shown by this project are likely to be associated with a wide range of developmental, socioeconomic, and health outcomes throughout the life course of this birth cohort. For example, a future study might find that individuals born in 2020 and 2021 have a higher % of universities degrees than previous and subsequent birth cohorts. Our comparative study supports such studies in reaching conclusions about the long-term effects of the COVID-19 pandemic. Results of the second sub-study will provide an analytical framework for studying the effects of macro-level shocks on population-level birth outcomes combining demographic and epidemiological theory with cutting-edge statistical techniques. This framework tries to push future research to consider macro-level shocks’ effects on means and variances between and within population groups by using Bayesian multilevel modelling.
Scientific results of the published first sub-study will inform future social and health science research that includes individuals born 2020-2021 (the Lockdown Cohort). This is because the changes in parental socioeconomic composition shown by this project are likely to be associated with a wide range of developmental, socioeconomic, and health outcomes throughout the life course of this birth cohort. For example, a future study might find that individuals born in 2020 and 2021 have a higher % of universities degrees than previous and subsequent birth cohorts. Our comparative study supports such studies in reaching conclusions about the long-term effects of the COVID-19 pandemic. Results of the second sub-study will provide an analytical framework for studying the effects of macro-level shocks on population-level birth outcomes combining demographic and epidemiological theory with cutting-edge statistical techniques. This framework tries to push future research to consider macro-level shocks’ effects on means and variances between and within population groups by using Bayesian multilevel modelling.