BACKGROUND:

Increasingly, youth with mental health disorders and suicidality are presenting to emergency departments (EDs) and requiring hospitalization. For youth with suicidality, studies reveal seasonal variations in frequency of presentations but do not identify associated diagnoses or whether other primary mental health complaints also reveal seasonal variations.

METHODS:

Data were collected between January 2015 and December 2019 by a child and adolescent psychiatry consultation-liaison service in a pediatric ED and hospital. Descriptive analysis and multiple linear regression were performed to assess volume over time, seasonal trends, and associated diagnoses.

RESULTS:

A total of 2367 patients were included, with an average age of 13.9 years and female predominance (62.3%). During the study period, annual ED consultations increased 87.5% and hospital consultations increased 27.5%. Consultations revealed seasonal trends, with highest volumes during January, April, May, October, and November (P < .001; adjusted R2 = 0.59). The most frequent diagnostic categories were depressive disorders and trauma- or stressor-related disorders. Thirty-six percent of patients presented after a suicide attempt, with the highest rates in spring (P = .03; adjusted R2 = 0.19). Boarding rates revealed significant seasonality, with higher instances in February, March, April, May, and October (P = .009; adjusted R2 = 0.32).

CONCLUSIONS:

Mental health presentations to a pediatric ED and hospital reveal seasonal trends, with the highest volumes in fall and spring and the most common diagnoses being depressive and trauma-related disorders. Suicide attempts are highest in late spring. Knowledge of these trends should inform hospitals, mental health services, and school systems regarding staffing, safety, surveillance, and prevention.

It is widely recognized in the United States that children and adolescents are experiencing escalating rates of mental health disorders and suicidality,13  with concurrent increases in presentations to emergency departments (EDs) for these issues.47  This has resulted in increased admissions to pediatric hospitals for youth with mental health diagnoses.8  In many regions, outpatient and inpatient mental health services have not expanded to meet the increased need for psychiatric services.9  As a result, EDs and pediatric hospitals have experienced higher rates of youth “boarding,” or remaining in a medical hospital beyond medical necessity while awaiting psychiatric treatment.8,10  High rates of boarding present many challenges to EDs and pediatric hospitals, including safety concerns, significant financial cost, limited therapeutic benefit and delays to appropriate care.46,8,1013 

In adults, studies have identified seasonal trends related to suicidality,1417,18  but studies examining seasonal trends in other mental health diagnoses, such as depression or bipolar disorder, have been mixed.7,16,19,20  Maes et al17  found a peak in spring only for violent suicide deaths, whereas Ajdacic-Gross et al21  found cyclic patterns in some methods of violent and nonviolent deaths but not others. Others have described peaks in spring and fall for overdoses.14  Various hypotheses to account for seasonal variation have been proposed.18 

In youth and adolescents, clinicians have anecdotally observed increased numbers of youth with mental health complaints presenting to EDs and medical and psychiatric hospitals during spring and fall months. In 2 recent articles, the authors report increasing presentations for suicidal ideation and attempts between 2008 and 2015 and seasonal variations in the volumes of these presentations.8,22  Plemmons et al8  identified that during this period, the annual percentage of all pediatric ED and inpatient pediatric hospital encounters for suicidal ideation and suicide attempts almost tripled, increasing from 0.66% in 2008 to 1.82% in 2015. In this study, the authors reported the lowest percentage of cases in summer months and the highest percentage of cases during spring and fall months.8  Carbone et al22  noted similar seasonal variations in suicidality among children and adolescents but not among adults. In both studies, the authors noted that increased suicidal behaviors corresponded with the school year, trends not seen in adult populations. The authors of these studies described broad seasonal trends using national databases and did not identify underlying diagnoses or impact on hospital outcomes, such as boarding rates. Pediatric EDs have noted seasonal trends in presentations, including increased numbers of cases of respiratory illness in the winter and trauma cases in the summer.23  To our knowledge, no studies have identified which diagnoses are associated with seasonal trends in youth suicide-related presentations or examined seasonal trends in youth presenting with mental health diagnoses, such as bipolar disorder, psychotic disorders, anxiety disorders, or depressive disorders.

Identifying seasonal trends and the diagnoses associated with these trends are important for hospitals to predict rooming and staffing needs as well to advocate for mental health initiatives for youth. These include improving prevention at the school, primary care, and community level and strengthening and increasing the capacity of outpatient and inpatient mental health services.

In our academic teaching hospital (located in the Pacific Northwest) with 10 pediatric ED beds and 150 medical beds, our child and adolescent psychiatry consultation-liaison (CAP CL) service has tracked all psychiatric consultations on youth aged <18 years from January 2015 to the current time. Our objectives for this study were to examine changes in the volume of psychiatric consultations over a 5-year period and to examine whether there were seasonal variations related to the volume of consultations, boarding rates, and mental health diagnoses.

The CAP CL service provides psychiatric consultations to medical teams in the pediatric ED, PICU, and pediatric unit at a tertiary care children’s hospital in the Pacific Northwest. The hospital does not contain an inpatient psychiatric unit, although it is one of the few hospitals in the region offering pediatric psychiatry consultation.

There are 2 acute psychiatric units for children and adolescents in the state, both within the metropolitan area of the hospital. Admissions to these units occur only through EDs or medical hospitals. There are several subacute and residential-level psychiatric programs for youth, which are lower levels of inpatient care accessed through an ED, hospital, or outpatient referral. During the study period, the number of inpatient psychiatric beds remained roughly the same.

Youth presenting to the ED with primary mental health issues without medical comorbidities generally remain in the ED until disposition is determined. Youth with comorbid medical and psychiatric issues, including eating disorders or suicide attempts requiring intensive medical management, are admitted to the hospital from the ED or may be transferred from an outside hospital. Additionally, youth admitted for primary medical issues (ie, cancer treatment) may have comorbid psychiatric concerns. Most consultations are for youth aged 5 to 17 but are occasionally for younger children.

The CAP CL team consists of an attending child and adolescent psychiatrist, a child and adolescent psychiatry fellow, and a social worker. The primary team requests consultation through the electronic medical record (EMR). When initially consulted, the attending psychiatrist enters data from the EMR into a secure electronic spreadsheet; these data include the patient’s medical record number, the patient’s sex, the patient’s age, the county of residence, the insurance type, and the referral source, that is, the unit that entered the initial referral (ED, PICU, or medical ward). Data on race and/or ethnicity and patient language were not included in the database because they were not available from the EMR. Clinical data, added after the initial psychiatric evaluation, include whether the patient presented with a suicide attempt; the primary diagnosis, based on Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) categories; the initial recommendation for disposition (discharge from the hospital versus referral to inpatient psychiatry); and the means used for suicide attempts (such as overdose, firearm, cutting, or hanging). Although suicidal ideation was not tracked in the original data set, it has been recorded since 2019. These data were not included in the analysis; however, they are noted in the Discussion section.

The CAP CL team usually remains involved throughout the patient’s hospitalization. On patient discharge, the spreadsheet is updated with the length of stay (based on admission and discharge dates) and whether the length of stay exceeded medical necessity, a proxy for boarding that is based on whether the patient remained in the hospital awaiting psychiatric placement after being medically cleared by the primary team. Finally, the discharge plan (outpatient versus inpatient psychiatry) documented by the child psychiatry consultation-liaison team is recorded.

A research assistant reviews the spreadsheet monthly for accuracy, with missing fields or discrepancies cross-checked against the EMR; the data are used regularly for quality improvement efforts. The analysis was conducted on data collected from January 2015 to December 2019. The university institutional review board designated the database not human research, waiving oversight.

Multiple regression was used to assess for temporal and seasonal trends. For each analysis, a time period index and individual months were used as the independent variables. The time period index was a numerical designation for each month, from January 2015 (1) to December 2019 (60). The dependent variable was the number of cases. Seasonality was evaluated by using the month of August as the intercept because it most often had the lowest number of cases. There were no controls; an independent regression was performed for each group, with a significance level of α = .05. For the total volume regression, a predictive model was generated. This predictive model reflects the anticipated number of cases that would be seen by using the seasonal and temporal coefficients produced by the regression without outside influence or variation. The variance in data points in the observed volume, compared to the predictive model, reveals the strength and fit of our data to a perfect seasonal and temporal model. The model can be used to predict future consultation volumes. Lastly, to determine if there were differences in suicidality (suicide attempt versus no suicide attempt) between DSM-5 diagnostic groups, a Pearson χ2 test was conducted by using a significance level of α = .05. Regression analyses were completed in Microsoft Excel 2016 by using the Analysis ToolPak. The Pearson χ2 test was conducted in SPSS version 24 (IBM SPSS Statistics, IBM Corporation).

A total of 2367 cases were included over this 60-month period (Table 1). The average age was 13.93 years (SD 2.88 years), and 62.3% of patients identified as female. Approximately half of patients had private insurance (52.4%), and just less than half had public insurance (45.3%). Slightly more than one-third presented for a suicide attempt (36.1%), the majority of which were overdoses (89.7%). The most common diagnosis was depressive disorder, with almost equal ED and hospital consultations.

TABLE 1

Sample demographics

Value
Total consultations, N (%) 2367 (100) 
Age, mean (SD), y 13.93 (2.88) 
 0–5 y old, n (%) 41 (1.7) 
 6–11 y old, n (%) 348 (14.7) 
 12–14 y old, n (%) 719 (30.4) 
 15–17 y old, n (%) 1259 (53.2) 
Length of stay, mean (SD), d 4.03 (7.13) 
Gender identity, n (%)  
 Female 1475 (62.3) 
 Male 817 (34.5) 
 Transgender: feminine 3 (0.1) 
 Transgender: masculine 72 (3.0) 
Insurance, n (%)  
 Public 1072 (45.3) 
 Private 1240 (52.4) 
 Uninsured 46 (1.9) 
 Unknown 9 (0.4) 
Primary diagnosis, n (%)  
 ADHD or disruptive, impulse-control disorder 166 (7.0) 
 Anxiety disorder 232 (9.8) 
 Autism spectrum disorder 89 (3.8) 
 Bipolar disorder 49 (2.1) 
 Depressive disorder 940 (39.7) 
 Feeding or eating disorder 104 (4.4) 
 Medication-induced movement disorder or other adverse effects of medication 22 (0.9) 
 Neurocognitive disorder 84 (3.5) 
 Neurodevelopmental disorder 25 (1.1) 
 Obsessive-compulsive or related disorder 22 (0.9) 
 Personality disorder 3 (0.1) 
 Schizophrenia spectrum or other psychotic disorder 53 (2.2) 
 Somatic symptom or other related disorder 45 (1.9) 
 Substance-related and addictive disorder 79 (3.3) 
 Trauma- and stressor-related disorder 454 (19.2) 
Presented for suicide attempt, n (%) 855 (36.1) 
 Overdose 767 (32.4) 
 Hanging, strangulation, or suffocation 20 (0.8) 
 Cutting 16 (0.7) 
 Jumping 9 (0.4) 
 Non-overdose ingestion 8 (0.3) 
 Firearm 3 (0.1) 
 Other 9 (0.4) 
 Unknown 23 (1.0) 
Referral unit, n (%)  
 ED consultation 1111 (46.9) 
 Non-ED consultation (PICU, medical admission) 1256 (53.1) 
Value
Total consultations, N (%) 2367 (100) 
Age, mean (SD), y 13.93 (2.88) 
 0–5 y old, n (%) 41 (1.7) 
 6–11 y old, n (%) 348 (14.7) 
 12–14 y old, n (%) 719 (30.4) 
 15–17 y old, n (%) 1259 (53.2) 
Length of stay, mean (SD), d 4.03 (7.13) 
Gender identity, n (%)  
 Female 1475 (62.3) 
 Male 817 (34.5) 
 Transgender: feminine 3 (0.1) 
 Transgender: masculine 72 (3.0) 
Insurance, n (%)  
 Public 1072 (45.3) 
 Private 1240 (52.4) 
 Uninsured 46 (1.9) 
 Unknown 9 (0.4) 
Primary diagnosis, n (%)  
 ADHD or disruptive, impulse-control disorder 166 (7.0) 
 Anxiety disorder 232 (9.8) 
 Autism spectrum disorder 89 (3.8) 
 Bipolar disorder 49 (2.1) 
 Depressive disorder 940 (39.7) 
 Feeding or eating disorder 104 (4.4) 
 Medication-induced movement disorder or other adverse effects of medication 22 (0.9) 
 Neurocognitive disorder 84 (3.5) 
 Neurodevelopmental disorder 25 (1.1) 
 Obsessive-compulsive or related disorder 22 (0.9) 
 Personality disorder 3 (0.1) 
 Schizophrenia spectrum or other psychotic disorder 53 (2.2) 
 Somatic symptom or other related disorder 45 (1.9) 
 Substance-related and addictive disorder 79 (3.3) 
 Trauma- and stressor-related disorder 454 (19.2) 
Presented for suicide attempt, n (%) 855 (36.1) 
 Overdose 767 (32.4) 
 Hanging, strangulation, or suffocation 20 (0.8) 
 Cutting 16 (0.7) 
 Jumping 9 (0.4) 
 Non-overdose ingestion 8 (0.3) 
 Firearm 3 (0.1) 
 Other 9 (0.4) 
 Unknown 23 (1.0) 
Referral unit, n (%)  
 ED consultation 1111 (46.9) 
 Non-ED consultation (PICU, medical admission) 1256 (53.1) 

ADHD, attention-deficit/hyperactivity disorder.

There was a positive trend over time in the volume of consultations, with an increase of 51.9% between 2015 and 2019 (Fig 1). This positive trend was most pronounced in ED consultations, which increased by 87.5% during that time, with consistent increases each year. During that time, the total annual pediatric ED volume increased by 8%. Consultations for patients admitted to the pediatric hospital increased by 27.5%, whereas the total hospital volume slightly decreased.

FIGURE 1

Yearly consultation volume. a The majority of patients have 1 consultation for each hospital admission. Infrequently, there are patients with complex medical problems for whom psychiatry is consulted more than once.

FIGURE 1

Yearly consultation volume. a The majority of patients have 1 consultation for each hospital admission. Infrequently, there are patients with complex medical problems for whom psychiatry is consulted more than once.

Close modal

During the study period, there were statistically significant seasonal trends in overall consultations (F12 = 7.96; P < .001; adjusted R2 = 0.59; Table 2). A forecasting model highlights distinct peaks in volume during the months of January, April, May, October, and November compared to the baseline month of August (Fig 2).

TABLE 2

Multiple Regression Results for Total, ED, and Admitted Patient Volume

Predictor VariableTotal Volume, N = 2367ED Volume, n = 1111Admitted Volume, n = 1256
B (95% CI)PB (95% CI)PB (95% CI)P
Time index .38 (0.28 to 0.48) <.001 .26 (0.20 to 0.32) <.001 .12 (0.04 to 0.20) 0.004 
January 8.84 (0.61 to 17.07) .04 4.22 (−1.07 to 9.51) .12 4.62 (−1.89 to 11.13) .16 
February 6.46 (−1.76 to 14.69) .12 3.56 (−1.73 to 8.84) .18 2.90 (−3.60 to 9.41) .37 
March 8.09 (−0.13 to 16.30) .054 4.90 (−0.38 to 10.18) .07 3.19 (−3.31 to 9.69) .33 
April 18.31 (10.10 to 26.52) <.001 8.24 (2.94 to 13.52) .003 10.07 (3.57 to 16.57) .003 
May 16.33 (8.12 to 24.54) <.001 7.38 (2.10 to 12.66) .007 8.95 (2.46 to 15.45) .008 
June 4.55 (−3.65 to 12.76) .27 2.52 (−2.75 to 7.79) .34 2.03 (−4.46 to 8.53) .53 
July 1.98 (−6.23 to 10.18) .63 −0.34 (−5.61 to 4.93) .90 2.32 (−4.17 to 8.81) .48 
August 20.33 (13.73 to 26.94) <.001 6.89 (2.64 to 11.13) .002 13.44 (8.22 to 18.67) <.001 
September 3.82 (−4.38 to 12.03) .35 0.74 (−4.53 to 6.01) .78 3.08 (−3.41 to 9.57) .34 
October 11.05 (2.84 to 19.25) .009 9.88 (4.61 to 15.15) <.001 1.17 (−5.33 to 7.66) .72 
November 8.47 (0.26 to 16.68) .04 1.02 (−4.26 to 6.30) .70 7.45 (0.95 to 13.94) .03 
December 3.49 (−4.72 to 11.70) .40 2.36 (−2.92 to 7.64) .37 1.13 (−5.37 to 7.63) .73 
Predictor VariableTotal Volume, N = 2367ED Volume, n = 1111Admitted Volume, n = 1256
B (95% CI)PB (95% CI)PB (95% CI)P
Time index .38 (0.28 to 0.48) <.001 .26 (0.20 to 0.32) <.001 .12 (0.04 to 0.20) 0.004 
January 8.84 (0.61 to 17.07) .04 4.22 (−1.07 to 9.51) .12 4.62 (−1.89 to 11.13) .16 
February 6.46 (−1.76 to 14.69) .12 3.56 (−1.73 to 8.84) .18 2.90 (−3.60 to 9.41) .37 
March 8.09 (−0.13 to 16.30) .054 4.90 (−0.38 to 10.18) .07 3.19 (−3.31 to 9.69) .33 
April 18.31 (10.10 to 26.52) <.001 8.24 (2.94 to 13.52) .003 10.07 (3.57 to 16.57) .003 
May 16.33 (8.12 to 24.54) <.001 7.38 (2.10 to 12.66) .007 8.95 (2.46 to 15.45) .008 
June 4.55 (−3.65 to 12.76) .27 2.52 (−2.75 to 7.79) .34 2.03 (−4.46 to 8.53) .53 
July 1.98 (−6.23 to 10.18) .63 −0.34 (−5.61 to 4.93) .90 2.32 (−4.17 to 8.81) .48 
August 20.33 (13.73 to 26.94) <.001 6.89 (2.64 to 11.13) .002 13.44 (8.22 to 18.67) <.001 
September 3.82 (−4.38 to 12.03) .35 0.74 (−4.53 to 6.01) .78 3.08 (−3.41 to 9.57) .34 
October 11.05 (2.84 to 19.25) .009 9.88 (4.61 to 15.15) <.001 1.17 (−5.33 to 7.66) .72 
November 8.47 (0.26 to 16.68) .04 1.02 (−4.26 to 6.30) .70 7.45 (0.95 to 13.94) .03 
December 3.49 (−4.72 to 11.70) .40 2.36 (−2.92 to 7.64) .37 1.13 (−5.37 to 7.63) .73 

August was set as the intercept for each analysis. CI, confidence interval.

FIGURE 2

Total consultation volume seasonality forecasting model. a The majority of patients has 1 consultation for each hospital admission. Infrequently, there are patients with complex medical problems for whom psychiatry is consulted more than once.

FIGURE 2

Total consultation volume seasonality forecasting model. a The majority of patients has 1 consultation for each hospital admission. Infrequently, there are patients with complex medical problems for whom psychiatry is consulted more than once.

Close modal

The number of youth referred to the CAP CL team from the ED was statistically significantly higher in April, May, and October (F12 = 8.43; P < .001; adjusted R2 = 0.60). Referrals from the PICU and pediatric ward occurred more frequently in April, May, and November (F12 = 2.50; P = .013; adjusted R2 = 0.23).

There were no significant monthly variations in average length of stay that persisted with the removal of outliers of ≥30 days’ length of stay. However, from January 2015 to December 2018, data recorded included whether youth remained in the hospital past medical necessity (or boarded) (Table 3). In this limited data set, in addition to an overall increase in the number of youth boarding over time, there were significantly higher instances of boarding in February, March, April, May, and October (F12 = 2.80; P = .009; adjusted R2 = 0.32).

TABLE 3

Multiple Regression Results for Boarding, Suicide Attempt, and Depression Diagnosis Patient Volume

Predictor VariableBoarding, n = 596Suicide Attempt, n = 855Depression, n = 940
B (95% CI)PB (95% CI)PB (95% CI)P
Time index .13 (0.04 to 0.21) .003 .06 (0.00 to 0.13) .04 .07 (0.01 to 0.13) .03 
January 4.13 (−1.20 to 9.45) .13 4.45 (−0.71 to 9.62) .09 5.89 (0.78 to 11.00) .03 
February 6.25 (0.93 to 11.57) .02 6.39 (1.23 to 11.55) .016 6.82 (1.72 to 11.92) .010 
March 6.13 (0.81 to 11.44) .03 5.72 (0.57 to 10.88) .03 7.35 (2.25 to 12.45) .006 
April 5.75 (0.44 to 11.06) .04 8.06 (2.91 to 13.21) .003 11.48 (6.39 to 16.58) <.001 
May 7.88 (2.57 to 13.18) .005 9.39 (4.24 to 14.54) .001 11.41 (6.32 to 16.50) <.001 
June 4.00 (−1.30 to 9.30) .14 1.73 (−3.42 to 6.88) .50 2.34 (−2.75 to 7.43) .36 
July −1.88 (−7.18 to 3.43) .48 2.86 (−2.28 to 8.01) .27 2.87 (−2.22 to 7.96) .26 
August 5.50 (1.21 to 9.79) .014 7.73 (3.59 to 11.88) <.001 7.36 (3.26 to 11.45) .001 
September 1.88 (−3.43 to 7.18) .48 3.54 (−1.61 to 8.68) .17 4.73 (−0.36 to 9.82) .07 
October 6.25 (0.95 to 11.55) .02 4.47 (−0.68 to 9.62) .09 10.86 (5.77 to 15.95) <.001 
November 3.38 (−1.93 to 8.68) .21 4.81 (−0.34 to 9.96) .07 6.59 (1.50 to 11.68) .012 
December 2.50 (−2.81 to 7.81) .35 3.14 (−2.01 to 8.29) .23 3.72 (−1.38 to 8.81) .15 
Predictor VariableBoarding, n = 596Suicide Attempt, n = 855Depression, n = 940
B (95% CI)PB (95% CI)PB (95% CI)P
Time index .13 (0.04 to 0.21) .003 .06 (0.00 to 0.13) .04 .07 (0.01 to 0.13) .03 
January 4.13 (−1.20 to 9.45) .13 4.45 (−0.71 to 9.62) .09 5.89 (0.78 to 11.00) .03 
February 6.25 (0.93 to 11.57) .02 6.39 (1.23 to 11.55) .016 6.82 (1.72 to 11.92) .010 
March 6.13 (0.81 to 11.44) .03 5.72 (0.57 to 10.88) .03 7.35 (2.25 to 12.45) .006 
April 5.75 (0.44 to 11.06) .04 8.06 (2.91 to 13.21) .003 11.48 (6.39 to 16.58) <.001 
May 7.88 (2.57 to 13.18) .005 9.39 (4.24 to 14.54) .001 11.41 (6.32 to 16.50) <.001 
June 4.00 (−1.30 to 9.30) .14 1.73 (−3.42 to 6.88) .50 2.34 (−2.75 to 7.43) .36 
July −1.88 (−7.18 to 3.43) .48 2.86 (−2.28 to 8.01) .27 2.87 (−2.22 to 7.96) .26 
August 5.50 (1.21 to 9.79) .014 7.73 (3.59 to 11.88) <.001 7.36 (3.26 to 11.45) .001 
September 1.88 (−3.43 to 7.18) .48 3.54 (−1.61 to 8.68) .17 4.73 (−0.36 to 9.82) .07 
October 6.25 (0.95 to 11.55) .02 4.47 (−0.68 to 9.62) .09 10.86 (5.77 to 15.95) <.001 
November 3.38 (−1.93 to 8.68) .21 4.81 (−0.34 to 9.96) .07 6.59 (1.50 to 11.68) .012 
December 2.50 (−2.81 to 7.81) .35 3.14 (−2.01 to 8.29) .23 3.72 (−1.38 to 8.81) .15 

August was set as the intercept for each analysis. CI, confidence interval.

A total of 855 of the 2367 youth (36.1%) presented after a suicide attempt. In addition to an overall positive trend in volume over time, this population showed significant seasonal trends, with the highest number of suicide attempts seen in February, March, April and May (F12 = 2.15; P = .03; adjusted R2 = 0.19).

When examined by primary diagnosis, the most frequently presenting diagnostic category was depressive disorders, followed by trauma- or stressor-related disorders. There was a positive trend over time in numbers of youth with a depressive disorder. A higher number was seen in October through May, except in December, with the highest numbers on average seen in April (F12 = 4.28; P < .001; adjusted R2 = 0.40). Among those with a trauma- or stressor-related disorder, there was a positive trend in the volume over time, without seasonal trends (F12 = 4.30; P < .001; adjusted R2 = 0.40). There were no trends among the other primary diagnoses in Table 1. In another analysis, after we removed youth with a depressive diagnosis and/or youth who were presenting for a suicide attempt, we found no significant seasonal variation, although an overall increase over time was observed (F12 = 4.28; P < .001; adjusted R2 = 0.40).

The percentage of participants that presented for a suicide attempt differed by diagnosis ( = 468.73; P < .001). The effect size, Cramer’s V, was moderate (Cramer’s V = 0.45). Youth with a depressive disorder were more likely to present for a suicide attempt (57.8%) than to present without a suicide attempt (42.2%), whereas youth with other diagnoses were more likely to present for a non–suicide-related mental health complaint (Table 1).

In this study, we found increasing rates and seasonal trends over 5 years among youth with mental health concerns presenting to a pediatric ED and admitted to a pediatric hospital.

During a 5-year period, volumes of psychiatric consultations in our pediatric ED and hospital increased >50%. Most notably, ED consultations increased by 87.5%. During this time, the population in the metropolitan area where the hospital is located increased by an average 1.3% annually. The number of inpatient psychiatric beds did not substantially change, nor did ED or hospital volume change enough to account for these trends. Reports of increasing mental health presentations from hospitals across the state, as well as published reports from EDs nationwide, are consistent with our findings. This increase in cases overall is alarming. The finding that suicide attempts, depressive disorders and trauma-related disorders increased in volume over time suggests that they are likely the drivers of overall increased presentations. Further research is needed to understand why youth are presenting more frequently with depression, suicidality, and trauma-related disorders and to understand how to treat these concerns more effectively. In addition, improved early prevention and detection efforts are needed.

Similar to findings of seasonal trends in suicidality for youth,8,22  we found seasonal trends in suicide attempts in this study. Our initial data set did not include suicidal ideation; however, in 2019 data, it was found that of 568 youth presenting during this 12-month period, 146 (25.7%) reported suicidal ideation. When combined with the 34% of patients presenting after a suicide attempt, in total, 60% of cases involved suicidality, a concerning proportion. Future researchers should investigate these trends and strengthen interventions for patients with suicidal ideation to prevent subsequent suicide attempts.

In addition to suicide attempts, we examined volume and seasonality for all mental health diagnoses in this study. Depressive disorder diagnoses revealed seasonal trends, which overlapped with seasonal trends seen in suicide attempts; additionally, patients with depressive disorder diagnoses were more likely to present with a suicide attempt than without one, reinforcing the role of depression in youth suicidality. There were no seasonal trends observed for other mental health diagnoses. It is unclear why depression, and not other diagnoses, revealed seasonality; possible explanations include the impact of the school year on depressive disorders, the impact of seasonal sunlight changes on mood but not other disorders, and the relatively small number of youth presenting with disorders such as bipolar disorder, not allowing for detection of seasonality. More research is needed.

In this study, we also found seasonal trends in boarding, with rates highest in fall and spring. These trends largely correspond with months with the highest rates of suicide attempts; this likely reflects that during these months, youth present with higher acuity and require inpatient hospitalization, compared to months when lower-acuity youth can be discharged from the hospital with outpatient follow-up. Hospitals should anticipate greater numbers of youth presenting with mental health issues during fall and spring months, potentially leading to increased rates of boarding in areas with limited youth mental health services. Strategies to address this issue include providing intensive transitional care to support some high-risk youth in discharging safely from EDs and hospitals to the community.24  In high-volume months, hospitals and EDs may need to plan for increased resources to ensure safety and potentially initiate treatment of boarding youth.

In this study, we identified seasonal trends in mental health presentations different from seasonal trends in medical hospitalizations, in particular peaks for respiratory illnesses in winter months. In the region where the hospital is located, school begins in late August or early September and ends in early to mid-June. In this study, we found that trends in pediatric mental health presentations, specifically depression and suicidality, coincide with the school year, peaking toward the end of the school year in April and May, with numbers decreasing in June, July, August, and December, when school vacations occur. Research has linked both academic pressure and bullying to increased stress and suicidality among youth.2528  Notably, many schools have increased surveillance and prevention efforts for students with mental health concerns; decreased numbers of youth presenting with mental health complaints during summer months could reflect decreased supervision and identification of high-risk youth. Other explanations for these trends could include seasonal variations in sunlight impacting adolescent depression and suicidality. During the academic year, school personnel, primary care providers, and mental health providers should be aware that youth may be more likely to struggle with depression and suicidality in the fall and spring and should target increased prevention and support during those times.

Study limitations include the study’s retrospective nature. Our database is maintained for quality, and we expect it is accurate, but prospective data collection is important for future studies. Second, although evaluations were conducted by a board-certified child and adolescent psychiatrist using DSM-5 criteria, they were not performed primarily for research purposes in a controlled environment with standardized diagnostic tools and thus may have been subject to clinical bias or error in assigning primary diagnoses. Third, the CAP CL service may not be consulted on youth with lower-acuity presentations, such as mild depression; therefore, some cases may have been missed. Notably, our data captured only presentations for suicide attempts, not suicidal ideation. We attempted to estimate the impact of suicidal ideation using one year of data; however, these numbers are projections. Finally, although our study reflects trends in our region, they are not generalizable to a broader population. With pediatric mental health ED visits and admissions increasing nationally, we speculate that trends identified are not isolated to the Northwest. However, generalizability to nonpediatric EDs and hospitals will be limited. Our hospital is one of few hospitals statewide with a dedicated psychiatric team; increasing patient volume may reflect growing community knowledge of this service.

In this study, we found increases over time and seasonal variation in numbers of youth with mental health disorders seen in the ED and admitted to a pediatric hospital over 5 years. The months associated with increased volume correspond with the school year, which may suggest a school-related component to mental health presentations. Further research should be used to better characterize the nature of school-related mental health crises, including potential underlying stressors (such as bullying, academic stress, and social media use), and clarify whether decreased presentations during summer months are related to lower rates of mental health crises or to decreased detection. Policy efforts should be used to identify strategies for preventing and monitoring these concerns among youth.

We thank James Marshall, PhD, for his contributions.

Dr Marshall conceptualized and designed the study, collected the initial data, interpreted the data, and drafted the initial manuscript; Ms Ribbers conducted the initial data analyses, interpreted the data, and reviewed and revised the manuscript; Dr Sheridan reviewed and interpreted the data and revised the manuscript; Dr Johnson collected the initial data, reviewed and interpreted the data, and revised the manuscript; and all authors approved the final manuscript as submitted.

FUNDING: No external funding.

1
Mojtabai
R
,
Olfson
M
,
Han
B
.
National trends in the prevalence and treatment of depression in adolescents and young adults
.
Pediatrics
.
2016
;
138
(
6
):
e20161878
2
Lindsey
MA
,
Sheftall
AH
,
Xiao
Y
,
Joe
S
.
Trends of suicidal behaviors among high school students in the United States: 1991-2017
.
Pediatrics
.
2019
;
144
(
5
):
e20191187
3
Bitsko
RH
,
Holbrook
JR
,
Ghandour
RM
, et al
.
Epidemiology and impact of health care provider-diagnosed anxiety and depression among US children
.
J Dev Behav Pediatr
.
2018
;
39
(
5
):
395
403
4
Cooper
JL
,
Masi
R
.
Child and Youth Emergency Mental Health Care: A National Problem
.
New York, NY
:
National Center for Children in Poverty, Columbia University Mailman School of Public Health
;
2007
.
5
Mapelli
E
,
Black
T
,
Doan
Q
.
Trends in pediatric emergency department utilization for mental health-related visits
.
J Pediatr
.
2015
;
167
(
4
):
905
910
6
Sheridan
DC
,
Spiro
DM
,
Fu
R
, et al
.
Mental health utilization in a pediatric emergency department
.
Pediatr Emerg Care
.
2015
;
31
(
8
):
555
559
7
Chiu
M
,
Gatov
E
,
Vigod
SN
, et al
.
Temporal trends in mental health service utilization across outpatient and acute care sectors: a population-based study from 2006 to 2014
.
Can J Psychiatry
.
2018
;
63
(
2
):
94
102
8
Plemmons
G
,
Hall
M
,
Doupnik
S
, et al
.
Hospitalization for suicide ideation or attempt: 2008-2015
.
Pediatrics
.
2018
;
141
(
6
):
e20172426
9
Substance Abuse and Mental Health Services Administration
.
Behavioral Health, United States, 2012
.
Rockville, MD
:
Substance Abuse and Mental Health Services Administration
;
2013
10
Kalb
LG
,
Stapp
EK
,
Ballard
ED
,
Holingue
C
,
Keefer
A
,
Riley
A
.
Trends in psychiatric emergency department visits among youth and young adults in the US
.
Pediatrics
.
2019
;
143
(
4
):
e20182192
11
Nolan
JM
,
Fee
C
,
Cooper
BA
,
Rankin
SH
,
Blegen
MA
.
Psychiatric boarding incidence, duration, and associated factors in United States emergency departments
.
J Emerg Nurs
.
2015
;
41
(
1
):
57
64
12
Claudius
I
,
Donofrio
JJ
,
Lam
CN
,
Santillanes
G
.
Impact of boarding pediatric psychiatric patients on a medical ward
.
Hosp Pediatr
.
2014
;
4
(
3
):
125
132
13
Gallagher
KAS
,
Bujoreanu
IS
,
Cheung
P
, et al
.
Psychiatric boarding in the pediatric inpatient medical setting: a retrospective analysis
.
Hosp Pediatr
.
2017
;
7
(
8
):
444
450
14
Beauchamp
GA
,
Ho
ML
,
Yin
S
.
Variation in suicide occurrence by day and during major American holidays
.
J Emerg Med
.
2014
;
46
(
6
):
776
781
15
Fernández-Niño
JA
,
Astudillo-García
CI
,
Bojorquez-Chapela
I
,
Morales-Carmona
E
,
Montoya-Rodriguez
AA
,
Palacio-Mejia
LS
.
The Mexican cycle of suicide: a national analysis of seasonality, 2000-2013
.
PLoS One
.
2016
;
11
(
1
):
e0146495
16
Valtonen
H
,
Suominen
K
,
Partonen
T
,
Ostamo
A
,
Lönnqvist
J
.
Time patterns of attempted suicide
.
J Affect Disord
.
2006
;
90
(
2–3
):
201
207
17
Maes
M
,
Cosyns
P
,
Meltzer
HY
,
De Meyer
F
,
Peeters
D
.
Seasonality in violent suicide but not in nonviolent suicide or homicide
.
Am J Psychiatry
.
1993
;
150
(
9
):
1380
1385
18
Woo
JM
,
Okusaga
O
,
Postolache
TT
.
Seasonality of suicidal behavior
.
Int J Environ Res Public Health
.
2012
;
9
(
2
):
531
547
19
Dove
J
,
Mistry
A
,
Werbeloff
N
,
Osborn
D
,
Turjanski
N
.
Weekday and seasonal patterns in psychiatric referrals in three major London A&E departments, 2012-2014
.
BJPsych Bull
.
2018
;
42
(
1
):
5
9
20
Heboyan
V
,
Stevens
S
,
McCall
WV
.
Effects of seasonality and daylight savings time on emergency department visits for mental health disorders
.
Am J Emerg Med
.
2019
;
37
(
8
):
1476
1481
21
Ajdacic-Gross
V
,
Wang
J
,
Bopp
M
,
Eich
D
,
Rössler
W
,
Gutzwiller
F
.
Are seasonalities in suicide dependent on suicide methods? A reappraisal
.
Soc Sci Med
.
2003
;
57
(
7
):
1173
1181
22
Carbone
JT
,
Holzer
KJ
,
Vaughn
MG
.
Child and adolescent suicidal ideation and suicide attempts: evidence from the Healthcare Cost and Utilization Project
.
J Pediatr
.
2019
;
206
:
225
231
23
McDermott
KW
,
Stokes
C
,
Freeman
WJ
.
Overview of pediatric emergency department visits, 2015: statistical brief #242
. In:
Healthcare Cost and Utilization Project (HCUP) Statistical Briefs [Internet]
.
Rockville, MD
:
Agency for Healthcare Research and Quality (US)
;
2006–2018
24
Marshall
R
,
Ribbers
A
,
Lin
AJA
,
Sheridan
DC
.
Addressing psychiatric boarding in the emergency department: a novel program to support youth and families after ED discharge. In: Proceedings from the American Academy of Child and Adolescent Psychiatry Annual National Meeting; October 14–19, 2019; Chicago, IL
25
Zaborskis
A
,
Ilionsky
G
,
Tesler
R
,
Heinz
A
.
The association between cyberbullying, school bullying, and suicidality among adolescents
.
Crisis
.
2019
;
40
(
2
):
100
114
26
Arun
P
,
Garg
R
,
Chavan
BS
.
Stress and suicidal ideation among adolescents having academic difficulty
.
Ind Psychiatry J
.
2017
;
26
(
1
):
64
70
27
Shain
B
;
Committee on Adolescence
.
Suicide and suicide attempts in adolescents
.
Pediatrics
.
2016
;
138
(
1
):
e20161420
28
Koyanagi
A
,
Oh
H
,
Carvalho
AF
, et al
.
Bullying victimization and suicide attempt among adolescents aged 12-15 years from 48 countries
.
J Am Acad Child Adolesc Psychiatry
.
2019
;
58
(
9
):
907
918.e4

Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.