National Trends in Suicidal Thoughts and Suicide Attempts Among High School Students in the United States

The authors sought to estimate national trends in suicidal thoughts and attempts among high school students with and without health-risk behaviors.

Data were analyzed from the 2007 to 2021 Youth Risk Behavior Survey, a biennial cross-sectional survey of U.S. high school students (unweighted sample, N=119,654). Suicidal thoughts were defined as seriously considering attempting suicide or making a plan to attempt suicide in the past year. Suicide attempts included ≥1 attempt in the past year. Twenty-four health-risk behaviors across five categories (e.g., violence and sexual activities) were considered. Multivariate logistic regression identified health-risk behaviors associated with suicidal thoughts and attempts. National trends in past-year suicidal thoughts and attempts among students with and without health-risk behaviors were also estimated.

From 2007 to 2021, among 119,654 students (mean age, 16.0 years [SD=1.2]; 59,645 females [49.1%]; 53,120 non-Hispanic White [55.3%]), significant increases occurred in past-year suicidal thoughts (biennial average percent change [BAPC]=2.9%, 95% CI=2.3, 3.4), plans (BAPC=3.0%, 95% CI=2.4, 3.7), and attempts (BAPC=2.2%, 95% CI=1.3, 3.1). The rate of increase in suicidal thoughts was disproportionately high among females, non-Hispanic White and Black students, and students with depressive symptoms, while attempts rose disproportionately among non-Hispanic White and Black students and high school seniors. While suicidal thoughts and attempts were more common among students with multiple health-risk categories, trends did not significantly differ between students with ≤1 and those with 4-5 risk categories, and students with ≤1 risk category still reported elevated rates of suicidal symptoms. In subgroup analyses, there was a disproportionately large increase in suicidal thoughts and attempts among students who used cigarettes or cannabis.

Between 2007 and 2021, the trends were similar among students with and without established behavioral risks. These findings point to a need to identify adolescents with suicidal symptoms who may not be identified with targeted screening based on behavioral risks.