As the SARS-CoV-2 virus spread around the world in spring 2020, one of the most drastic measures governments took was to shut schools, upending students’ lives and, for many, their mental health. As the consequences of those decisions continue to reverberate, scientists have tried to disentangle the mental health fallout caused by school closures from the general tumult of the pandemic.

Now, researchers in Germany have tackled the question by combining data on how long different groups of students were out of school with responses to an ongoing mental health survey. The results, published today in Science Advances, show that for German children between the ages of 11 and 17, additional weeks of home schooling in the early months of the pandemic correlated with worsening mental health measures, especially for younger children, boys, and children with limited living space.

“Schools were the first [institutions] to be closed and the last to reopen, and this affects the most vulnerable in terms of stability and mental health,” says Christina Felfe, an economist at the University of Konstanz who helped lead the study.

Previous findings on the mental health effects of school closures have been mixed. Some studies have shown dramatically increased depression and anxiety among young people during the pandemic. Others, in contrast, found a decrease in the demand for mental health care and in youth suicides while schools were closed. Some data also suggest bullying—including online—may have decreased during school closures.

Germany provided a natural experiment to help researchers distinguish the effects of school closures from the mental health impacts of the broader pandemic, as its 16 federal states have the power to decide their own school policies. All the states closed schools between 16 and 18 March 2020, but each followed its own schedule for sending students back. Within states, policies also varied by grade level: Students nearing graduation, whether from elementary or upper level schools, returned to in-person instruction as soon as late April, whereas other students remained at home through early June.

Felfe and her colleagues combined these data with answers to a survey of 1040 children ages 11 to 17 across Germany conducted by a nationwide consortium of researchers between 26 May and 10 June 2020. It aimed to gauge “health-related quality of life,” posing questions such as “Have you felt fit and well during the previous week?” and asking youth to rate on a scale how often they felt sadness, nervousness, and a range of other emotions.

By comparing the responses of children in the same grade levels who went back to school at different time points, the researchers found that for the average student, additional weeks at home led to increasing emotional health problems and depressive symptoms but did not affect anxiety. Overall, the team calculated, school closures accounted for a decline in average well-being of 11 percentage points.

But there were different effects in different groups: Younger children were most adversely affected, whereas children older than 15 showed little or no effect. Boys were more strongly affected than girls by additional weeks at home. The reason for that is unclear, says Ulrike Ravens-Sieberer, an expert in child public health at the University Medical Center Hamburg-Eppendorf who helped develop the mental health survey. She says one reason may be that girls tend to develop more effective coping strategies at younger ages. Unsurprisingly, children cooped up in relatively smaller living spaces also showed stronger negative effects—a finding that suggests the mental health burden of home schooling disproportionately hit poorer children.

The researchers also looked for effects of school closures in data from a nationwide mental health crisis hotline. They found that for any given location, when schools were closed, calls about conflicts or problems with teachers and peers fell off, whereas calls about family concerns increased. For students who returned to school after mid-May, the high number of family-related calls persisted until early August.

Combining all of these data sources is “a step in an interesting direction to identify the effects on overall mental health” that might not show up in statistics covering doctor visits or suicides, says Valentin Klotzbücher, an economist at the University of Freiburg who has studied helpline use during the pandemic among both youth and adults. However, he cautions, the number of students surveyed is very small, especially when subdivided by grade level and state of residence. Jonas Vlachos, an economist at Stockholm University who has studied the effects of school closures on both infections and mental health, agrees. Although the survey responses are valuable, he adds, they offer a single snapshot from early in the pandemic. “I’d like to see more evidence for the long-term implications” of school closures, he says.

That evidence is starting to show up, Ravens-Sieberer says. She and her colleagues have continued to collect survey responses and now have data from five time points, through fall 2022. The results suggest mental health scores hit a low in late 2020 and have since rebounded, though not quite to prepandemic levels. That may also be due to other events, such as the escalating climate crisis and the war in Ukraine, she notes. She is now working with colleagues in nine countries across five continents to compare the impacts of school closures on social inequalities affecting child health.

Ravens-Sieberer says the new data shouldn’t be used to rehash debates about the wisdom of school closures in response to COVID-19. “Hindsight is 20/20,” she says. In early 2020, “everyone made the best decisions they could at the time.” The next public health crisis, she says, might be different, and so it will be difficult to draw direct lessons from what happened during this pandemic. Instead, she says, the impact of the 2020 school closures on younger and poorer children highlights the need for schools and other institutions to help build children’s resilience and psychological health, “so that we are better prepared, no matter what crisis comes next.”