Reasons and trends in youth's suicide rates during the COVID-19 pandemic

Abstract

Background: The COVID-19 pandemic posed many mental health challenges to youth through unprecedented infection control measures such as nationwide school closures. Despite this, few studies have investigated trends in suicide among youth during the pandemic, let alone their reasons.

Methods: Population-level data on crude monthly suicide rates (2016-2021) and reasons of suicide (2018-2020) among youth aged 10-19 years were obtained from the Japanese Ministry of Health, Labour and Welfare and the National Police Agency, respectively. Using an event study design (with a Poisson regression model to calculate changes-in-changes (CiC) estimates) and interrupted time series analysis, we investigated changes in monthly suicide rates during the first 12 months of the pandemic (May 2020 to April 2021) compared to pre-pandemic levels (May 2016 to March 2020). Additionally, we investigated the changes in reasons of suicide (family-related, mental illness, social concerns, and academic concerns).

Findings: In the event study analysis, suicide rates among youth increased during the pandemic relative to pre-pandemic levels, especially between August-November 2020 (e.g., ratio of the suicide rate in November 2020 relative to previous years, 1.86; 95% confidence interval (CI), 1.30 to 2.66). Though suicide levels returned closer to pre-pandemic levels by December 2020, they remained slightly elevated into 2021. In the interrupted time series analysis, suicide rates increased from May to August 2020 (0.099 cases per 100,000 youth per month; 95% CI, 0.022 to 0.176), followed by a decrease from September to December 2020 (-0.086 cases per 100,000 youth per month; 95% CI, -0.164 to -0.009). We observed elevated suicide rates for all major reasons from summer to autumn 2020, especially suicides attributed to family-related problems and social concerns. Furthermore, rates of suicides attributed to mental illness remained higher than pre-pandemic levels into December 2020.

Interpretation: Suicide rates among youth remained slightly elevated compared to pre-pandemic levels into 2021. The reasons of the increase in suicide rates were multifactorial, including mental health issues and disruptions in social relationships. During a pandemic, interventions that provide mental support as well as opportunities for regular social interactions to youth may be beneficial.

Funding: Norwegian Agency for International Cooperation and Quality Enhancement in Higher Education.

Keywords: COVID-19; COVID-19 pandemic; Child and adolescent mental health; Child and adolescent psychiatry; Pediatrics; SARS-CoV-2; Suicide.

Figure 1

Trends in suicides per 100,000 population among youth. Shown in green are the monthly suicide rates from April 2020 to March 2021, in blue is the suicide rate in April 2021, and in red are means of monthly suicide rates from May 2016 to April 2020 and their 95% confidence intervals, assuming a normal distribution. CI, confidence interval.

Figure 2

Changes-in-changes (CiC) estimates of suicides per 100,000 population among 10- to 19-year-olds between May 2020 and March 2021 relative to pre-pandemic months. Shown are CiC estimates under a Poisson regression model using an event study design, 95% confidence intervals, and P-values. CI, confidence interval.

Figure 3

Results of the interrupted time series analysis. Shown are unadjusted results (solid line) and results adjusted for seasonality (dotted line). In both the adjusted and unadjusted results, suicide rates increased from May to August 2020, followed by a decrease from September to December 2020. From January 2021, the trend in suicide rates was similar to pre-pandemic. Both trends were analyzed using linear regression models. The detailed results of the interrupted time series analyses are available in the supplemental material.

Figure 4

Changes-in-changes (CiC) estimates of suicides per 100,000 population among 10- to 19-year-olds during the pandemic (May to December 2020) relative to pre-pandemic months, by reasons of suicide. Shown are CiC estimates under a Poisson regression model using an event study design, 95% confidence intervals, and P-values. Each pandemic month is compared to the corresponding pre-pandemic months in 2016 to 2019. For instance, the estimates for May 2020 represent the CiC estimates relative to May 2016, May 2017, May 2018, and May 2019. CI, confidence interval.