Studies from the US1,2 and Canada3,4 have reported increasing numbers of children and adolescents receiving treatment for eating disorders during the COVID-19 pandemic. Most patients are girls,1,2,4 and the predominant diagnosis is anorexia nervosa.1,2 There is insufficient information about the magnitude of this increase on a population level. In this cohort study, we analyzed trends in diagnoses of eating disorders among children and adolescents in Norway before vs during the pandemic.
We obtained primary care data from the Norwegian Registry for Primary Health Care and specialist care data from the Norwegian Patient Registry.5 Reporting to these registries is mandated by law and linked to the national reimbursement systems for health services. The entire Norwegian population is covered. This study followed the STROBE reporting guideline and was approved by the Norwegian Regional Committees for Medical and Health Research Ethics. Because the study used existing registry data, informed consent was not required.
Individuals aged 6 to 16 years living in Norway on January 1, 2020 (pandemic cohort), were observed from January 2019 to December 2021. The comparison group included this age group living in Norway on January 1, 2018 (prepandemic cohort), observed from January 2017 to December 2019. We assessed changes in the percentage of individuals with recorded eating disorders since onset of the pandemic in March 2020 in the pandemic cohort and after March 2018 in the prepandemic cohort. Changes were compared by difference-in-difference models.6 We calculated monthly percentages of individuals with eating disorders using event study models to estimate relative changes. Analyses were done by sex and age group (6-12 and 13-16 years) using Stata, version 16.0. The eMethods in the Supplement gives additional information.
The number of boys with eating disorder diagnoses was low; thus, girls were analyzed. The pandemic cohort included 348 187 girls (mean [SD] age, 11.03 [3.13] years), and the prepandemic cohort, 353 848 girls (mean [SD] age, 10.96 [3.15] years) (Table). For girls aged 6 to 12 years, we observed larger relative increases in the percentage with eating disorder diagnoses in the pandemic cohort: 66.90% (95% CI, 33.12%-100.67%) in primary care and 278.30% (95% CI, 160.44%-396.16%) in specialist care. For girls aged 13 to 16 years, the relative increase was 126.54% (106.48%-146.59%) in primary care and 95.96% (95% CI, 79.54%-112.38%) in specialist care. Increases were attributable to new cases.
The monthly percentage of girls with eating disorder diagnoses increased over time in the prepandemic cohort (Figure). The pattern was similar in the pandemic cohort but with a disruption after onset of the pandemic. The monthly percentage of girls aged 13 to 16 years with an eating disorder diagnosis ranged from 0.05% to 0.08% before the pandemic and from 0.15% to 0.20% after onset in primary care and from 0.27% in February 2019 to 0.74% in December 2021 in specialist care. The relative increase ranged from −11.89% (95% CI, −47.23% to 23.46%) in March 2020 to 194.63% (95% CI, 160.27%-228.99%) in February 2021 for primary care and from 11.06% (95% CI, −2.84% to 24.96%) in May 2020 to 150.70% (95% CI, 128.87%-172.53%) in November 2021 for specialist care and was statistically significant throughout 2021.
Among girls aged 6 to 12 years, the percentage with eating disorder diagnoses was lower but increased after March 2021. In specialist care, the percentage did not exceed 0.10%, but the relative increase remained greater than 200% after May 2021.
We found a substantial increase in the number of girls diagnosed with eating disorders in Norway starting after onset of the COVID-19 pandemic. The timing of the trend disruption suggests that the increase was associated with societal changes induced by the pandemic, including restrictions placed on youth’s lives, education, and activities. Limitations were that follow-up was incomplete for teenagers older than 16 years, we could not distinguish between eating disorder subtypes, and diagnostic data were not validated. Our findings are similar to those from North America,1-4 suggesting that the increase in eating disorders occurred internationally.