Background <p>Schools are important settings for respiratory virus transmission and a major focus of pandemic control measures in the absence of effective pharmaceutical interventions. During the COVID-19 pandemic, decisions regarding school closures and openings were highly debated due to uncertainties about their population-level impact.</p> Methods <p>In this retrospective mathematical modelling study, we used an age-stratified transmission model fitted to hospitalization and seroprevalence data in a Bayesian framework to assess the effect of school closures and openings on SARS-CoV-2 transmission in Portugal before mass vaccination. We compared the observed epidemic trajectory with two counterfactual scenarios: one in which schools remained open during the first lockdown of spring 2020 and another in which they closed for in-person education following the summer holidays in autumn 2020.</p> Results <p>We find that keeping schools open during the first lockdown would have led to hospitalizations rising far beyond observed levels, unless school contacts reached at most 50% of their pre-pandemic levels. However, closing schools after the summer holidays, when adult-driven community transmission was already high, would not have been sufficient to prevent the autumn wave of hospitalizations in 2020, though it could have mitigated its impact. Elasticity analyses quantify the age group contributions to transmission, showing that adults primarily drove the 2020 pandemic.</p> Conclusions <p>These findings underscore that the epidemiological impact of school closures is highly context-dependent, influenced by existing reductions in school contacts and the intensity of transmission outside them. Our results suggest that school closures remain a viable option in future pandemics, but are most effective when integrated with broader community interventions.</p>