School Year and Suicidal Behaviors Among Youth: Insight From a National Dataset

Prim Care Companion CNS Disord 2022;24(4):21m03112

To cite: Baweja R, Patel RS, Tankersley WE, et al. School year and suicidal behaviors among youth: insight from a national dataset. Prim Care Companion CNS Disord. 2022;24(4):21m03112.
To share: https://doi.org/10.4088/PCC.21m03112

© 2022 Physicians Postgraduate Press, Inc.

^aDepartment of Psychiatry and Behavioral Health, Penn State College of Medicine, Hershey, Pennsylvania
^bDepartment of Psychiatry, Griffin Memorial Hospital, Norman, Oklahoma
^cDepartment of Psychiatry and Behavioral Science, Oklahoma State University, Tulsa, Oklahoma
*Corresponding author: Raman Baweja, MD, MS, Department of Psychiatry and Behavioral Health, Penn State College of Medicine, 500 University Drive, Hershey, PA 17033-0850 (rbaweja@pennstatehealth.psu.edu).

Suicide is the second leading cause of death among youth aged 10–18 years in the United States.¹ Multiple social and environmental factors increase the risk for suicidal behaviors among youth, including social isolation, substance abuse, bullying, difficulties in school, and school absenteeism and dropout.^2–4 Past reports found an increase in suicide attempts during the school year.^4–6 However, most of the previous studies about the temporal patterns of suicidal behaviors report about completed suicide⁴ or analyze a single academic site⁶ or selected number of children’s hospitals.⁵ No study has examined whether serious suicidal behaviors, which require inpatient hospitalization, are associated with school attendance in a national dataset. Furthermore, little is known about which youth are at higher risk for the apparent increase in suicidal behaviors during the school year. In this study, we aim to compare suicidal behaviors during the time of year when students are typically in school versus on break from school. We also examine demographic characteristics and comorbidities with suicidal behaviors that occur during these different time periods. Identification of youth who are at high risk for suicidal behaviors would allow for better allocation of scarce resources.

METHODS

We conducted a retrospective cross-sectional analysis of the nationwide inpatient sample (NIS).⁷ The health care cost and utilization project (HCUP) NIS acquires administrative data on inpatient discharges from about 4,400 nonfederal hospitals and 44 states in the United States, excluding rehabilitation and long-term care hospitals.⁸

We included 74,385 inpatients (aged 10–18 years) from the NIS (January–December 2014) with primary ICD-9 codes of suicidal ideation (V62.84) or suicide and self-inflicted injury including poisoning (E950.XX-959.XX). Back-to-school dates in the United States vary considerably across states and regions; around 86% of schools start between mid-August and early September.⁹ For this study, we further subgrouped based on school year (September to May) and school break (June to August). Given the variability in school start date and seasonal variability, we also looked specifically at suicidal behaviors in different seasons and when all youth (regardless of state or region) were in school (October [fall], January [winter], and May [spring]) and when all students were on summer break (July). Primary ICD-9 diagnoses were identified for mood disorders, anxiety disorders, disruptive behavior disorders, alcohol, and other substance-related disorders.

Individual patient identifiers are used to protect patient health information.⁷ The use of administrative databases under the HCUP and publicly available deidentified NIS database does not require institutional review board approval.⁸ We used χ² and t test to compare demographics, clinical diagnoses, and comorbid substance use disorder during school months versus school break. Next, we used the binomial logistic regression model to identify the predictors of suicidal behaviors during the school year (compared to school break as reference category). All the analyses were done using SPSS, and statistically significant P value was set at < .01.

RESULTS

Suicidal behaviors were higher during the school year (average of 6,761/month, 223/day) compared to school break (average of 4,512/month, 147/day) (Table 1). This pattern persisted when data were restricted to periods when all students were in school (October, January, and May, average of 208/day) versus when all students were on break (July, average of 136/day). Prevalence of suicidal behaviors was highest in October for both hospitalization (n = 7,515) and primary diagnosis of mood disorder (n = 6,800) (Figure 1).

Among the school year cohort, the rate of suicidal behaviors was higher in youth who were younger, females, Hispanics, those from high-income families, and those with mood disorders (91.6% vs 90%). Hospital care cost during inpatient stay was higher during the school year. During school break, the rate of suicidal behaviors was higher for White and African American youth, males, and those with disruptive behavior disorders (34.6% vs 31.5%) and comorbid alcohol (7.9% vs 5.7%) and other substance use disorders (21.7% vs 18.4%). There was no statistically significant difference between cohorts in terms of length of stay and anxiety disorder.

As per the logistic regression model, Hispanics were at higher odds of hospitalization for suicidal behaviors during the school year compared to Whites. Sex and geographic region did not have significant association to predict the risk for suicidal behaviors by school period. Also, the odds of hospitalization for suicidal behaviors during the school year increases with median household income and is a significant predictor as shown in Supplementary Table 1. Psychiatric diagnoses and comorbid substance use were not significant predictors for suicidal behaviors during the school year when compared to school break.

DISCUSSION

In this retrospective study, suicidal behaviors requiring medical attention were higher (1.5 times) during the school year compared to school break. There were also differential patterns for suicidal behaviors during the school year versus when on break. For example, mood disorders were higher during the school year compared to school break, while disruptive behavior disorders and substance use showed the opposite pattern.

It is unclear why suicidal behaviors are higher during the school year, but several hypotheses can be suggested based on known risk factors. When school is in session, bullying, sleep deprivation, and academic stress, which are known risk factors for suicide behaviors, are more common.^2,4,10,11 In addition to mood disorders, the association between adverse experiences and suicidal behaviors among adolescents has been well documented.¹²

Temporal fluctuations in suicidal behaviors may be due to a variety of factors besides school schedules. Similar to previous reports of seasonal pattern of suicidal behaviors among youth, with highest in fall and spring and lowest during summer months,⁵ we found higher suicidal behaviors in October and April than in June and July in this study. Similarly, mood disorders were more frequent in October and April. It is possible that the apparently increased rates of mood disorders and suicidal behaviors during the early months of the school year (September to November) were related to stress due to returning to school. For example, students may be less likely to encounter in-person bullying when they are out of school, and the resumption of school may prompt greater stress that triggers suicidal behaviors or mood problems. Alternatively, increased mood disorders and suicidal behaviors at the end of the school year (March–May) could be associated with increased stress from increased high-stakes testing. These are speculative explanations that warrant further investigation.

For youth with disruptive behavior disorders including attention-deficit/hyperactivity disorder (ADHD), school may provide a structured environment that reduces high-risk behaviors while also increasing use of ADHD medication and behavioral health services at school. Given the high prevalence of drug holidays for central nervous system stimulants during the school break¹³ and unstructured days, youth with disruptive disorders may be more likely to have impulsive acts at the time of conflict or stress. There are reports of high risk of substance use disorders and legal troubles including juvenile arrests among high school dropouts, which independently increase the risk for suicidal behavior.^3,14,15

This study has several limitations given the nature of the database used and that it only included a sample of suicide attempts that required inpatient hospitalization. Information about specific academic stressors and bullying and details regarding other clinical variables including medication use is not available. Other limitations include variability in school year schedules across school districts in the United States, information about type of schooling (attending in person, in cyber, or home schooled), and ambiguity about accidental versus intentional overdose. Additionally, this was a cross-sectional retrospective study, which lacks data on causal relationship.

Suicide among youth is a public health concern, and the rate of suicide for youth has increased significantly over the last 2 decades.¹ The recent pandemic has had a significant impact on child mental health, exemplified by the increase in mental health–related emergency visits for suicide attempts.^16,17 There is a clear need to identify risk factors associated with worsening suicidal behaviors in youth. This is the first large sample study, to our knowledge, to use objective indices to examine the pattern of suicidal behaviors between school attendance and school break. There are at least 2 actionable implications of this study. First, given the finding that suicidal behaviors are higher with mood disorders during the school year, schools should implement universal depression and suicide screening.¹⁸ School-based mental health services should prioritize management of depression. Second, as youth with disruptive behavior disorders and substance use disorders are at higher risk for suicidal behaviors during school break, increased outreach and monitoring during extended breaks seem warranted for these high-risk youth during unstructured times. There is a need for a large prospective study to evaluate the various school stressors and impact on mental health and suicidal behaviors.

Submitted: August 23, 2021; accepted December 9, 2021.
Published online: July 21, 2022.
Relevant financial relationships: Dr Waxmonsky has received research funding from National Institutes of Mental Health, Supernus, and Pfizer in the past 3 years. Drs Baweja, Patel, Tankersley, and Waschbusch report no conflicts of interest related to the subject of this article.
Funding/support: None.
Supplementary material: See accompanying pages.

Clinical Points

• The rate of suicide among youth has increased significantly over the past 2 decades.
• Suicidal behaviors were higher in youth during the school year compared to school break.
• Mood disorders were higher during the school year, while disruptive behavior disorders and substance use were higher during school break.