DOI: 10.1111/pcn.70117 ISSN: 1323-1316

Newly recorded psychiatric diagnoses among 21 million youth before, during, and after COVID ‐19 restrictions: A nationwide interrupted time‐series analysis

Naomi Matsumoto, Hisashi Ishida, Shunsaku Hayase, Takahiro Asada, Yoshie Shigeyasu, Ayumi Okada, Tomoya Hirota, Takashi Yorifuji

Aim

The COVID‐19 pandemic had substantial effects on youth mental health. Most previous studies focused on the initial pandemic years, leaving the post‐restriction period underexplored. We examined trajectories of newly recorded psychiatric diagnoses among Japanese youths across two policy shifts: March 2020 school closures and May 2023 reclassification of COVID‐19 as an influenza‐equivalent Category V disease.

Methods

We conducted an interrupted time‐series analysis using the National Database of Health Insurance Claims, covering over 99% of the Japanese population. The study included 21 million individuals aged under 19 years to analyze monthly trends in first recorded psychiatric diagnoses from January 2018 to March 2024. Segmented negative binomial regression was used to estimate shifts in levels and trends following both policy interventions.

Results

Following the 2020 closures, eating disorders showed sex‐specific patterns: males showed an immediate 28.8% decrease (IRR 0.712, 95% CI 0.619–0.819), whereas females showed a 28.4% increase (IRR 1.284, 95% CI 1.045–1.576). Newly recorded ADHD diagnoses showed an 8.1% immediate decrease (IRR 0.919, 95% CI 0.847–0.997). During the restriction period, anxiety/stress‐related and mood disorders showed apparent surge patterns following infection waves. After restrictions were lifted in 2023, eating disorders among males showed an upward level shift (IRR 1.214, 95% CI 1.076–1.370). Neurodevelopmental disorders showed preliminary downward trends in the post‐restriction period.

Conclusion

This nationwide study of 21 million youths suggested differential diagnostic patterns across pandemic phases. These contrasting trajectories highlight the need for subgroup‐level post‐pandemic diagnostic surveillance.

More from our Archive