Abstract
BACKGROUND: During the COVID-19 (SARS-CoV-2) pandemic, a major challenge for policymakers was finding the balance between containing outbreaks and supporting mental health. Although lockdowns slowed COVID-19 transmission, it appears that they also increased the prevalence of depressive symptoms. It is not yet clear whether different, hypothetical lockdown scenarios would have differently impacted the prevalence of depressive symptoms. METHODS: To that end, we developed an open-source microsimulation model, COMMA (COvid Mental health Model with Agents). With COMMA, we explored how lock down scenarios, individual characteristics, and behaviors during lockdowns related to the prevalence of depressive symptoms during the COVID-19 period. The characteristics of the population were synthesized from Lifelines, a prospective cohort study set in the northern Netherlands. RESULTS: COMMA simulations estimated that the actual lockdown scenario between June 2020 and June 2021 would result in 10.92% (95% CI: 10.68%-11.18%) of the population experiencing depressive symptoms, up from 3.49% (95% CI: 3.49–3.49) at baseline. Had there been a full or partial lockdown for the entirety of this period, 11.44% (95% CI: 11.18%-11.69%), or 10.63% (95% CI: 10.38%-10.87%), of the population, respectively, would have experienced depressive symptoms. CONCLUSIONS: Lockdown severity was an important predictor of the prevalence of depressive symptoms, with higher rates in full lockdowns than partial lockdowns. However, across all simulations, only a minority of the population ever developed depressive symptoms. These findings suggest that lockdowns, particularly full lockdowns, are not without consequences for mental health. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12889-026-27001-3.