OBJECTIVES:

To validate the use of a brief suicide risk screening tool, the Ask Suicide-Screening Questions (ASQ) instrument, in pediatric inpatient medical and surgical settings.

METHODS:

Pediatric patients (10–21 years) hospitalized on inpatient medical and surgical units were recruited through convenience sampling for participation in a cross-sectional instrument validation study. The Suicidal Ideation Questionnaire was used as a standard criterion to validate the ASQ. Patient opinions about screening and parent consent to enroll in a suicide risk screening study were assessed to determine the feasibility of administering the ASQ in this venue.

RESULTS:

A total of 600 pediatric medical inpatients were screened. Compared with the gold standard, the ASQ had strong psychometric properties, with a sensitivity of 96.67% (95% confidence interval [CI]: 82.78 to 99.92), a specificity of 91.05% (95% CI: 88.40 to 93.27), a negative predictive value of 99.81% (95% CI: 98.93 to 99.99), and an area under curve of 0.94 (95% CI: 0.90 to 0.97). Only 3 participants (0.5%) had acute positive screen results on the ASQ, endorsing current suicidal ideation, whereas 77 participants (12.8%) screened nonacute positive, and 48 participants (8.0%) reported a past suicide attempt.

CONCLUSIONS:

The brief 4-item ASQ is a valid tool to detect elevated suicide risk in pediatric medical and surgical inpatients. Our findings also reveal that screening is feasible in terms of detection of suicidal thoughts and behaviors and is acceptable to parents and patients.

In the United States, a quarter of all deaths among youth aged 10 to 24 years are from suicide, resulting in the loss of 6807 young lives in 2018.1  Approximately 80% to 90% of youth who died by suicide had visited a health care professional in a medical setting within 12 months of their death.2,3  Underdetection of suicide risk in medical settings occurs because many patients presenting with somatic complaints do not disclose suicidal thoughts or plans unless asked directly,2  and most pediatric medical settings do not universally screen for suicide risk.4  Therefore, the ability to detect suicide risk in medical settings is an urgent priority.

Patients with physical illness are at increased risk for suicide.57  Suicide is the fourth most frequently reported sentinel event to The Joint Commission, with a significant percentage of these deaths occurring in nonbehavioral health care units, such as medical and surgical units, the ICU, and the emergency department (ED).8  In 2016, The Joint Commission issued Sentinel Event Alert 56, recommending that all medical patients be screened for suicide risk because medical settings provide an optimal venue for identifying youth at risk.9 

There is a lack of brief screening tools validated for use in inpatient medical and surgical units.10  The Ask Suicide-Screening Questions (ASQ) is a publicly available suicide risk screening tool that consists of 4 yes or no questions, takes ∼20 seconds to administer, and has been shown to have strong psychometric properties (eg, sensitivity, specificity, negative predictive value [NPV]) in pediatric ED patients.11  The ASQ was designed originally for pediatric ED medical and surgical patients but may have utility in other venues, such as the pediatric inpatient medical or surgical unit and the outpatient primary care clinic setting.

Ensuring the feasibility and acceptability of a screening program is critical to maintaining effective clinic flow and not overburdening a system of care.12  Although previous research among adolescents presenting to the ED13  and the outpatient primary care setting14  has demonstrated patient support for suicide risk screening, it is unknown how pediatric medical and surgical inpatients would respond to actually being screened.

Our aim for this study is to validate the ASQ among pediatric medical and surgical inpatients. The Suicidal Ideation Questionnaire (SIQ) and the Suicidal Ideation Questionnaire Junior (SIQ-JR) are used as the gold standard to validate the ASQ in this sample. A secondary aim is to describe the feasibility of screening pediatric inpatients for suicide risk in terms of (1) acceptability to patients and their parents, (2) a high enough prevalence to warrant screening, and (3) a low enough positive screen result rate to not overburden a busy inpatient workflow.

Between August 2013 and January 2019, we enrolled a convenience sample of pediatric medical and surgical inpatients at 3 large teaching hospitals in the United States. Patients were eligible to participate if they were between the ages of 10 and 21 years, English-speaking, and admitted to a designated general medical or surgical inpatient unit. Exclusion criteria were severe cognitive impairment, non–English-speaking or absence of an English-speaking parent to provide informed consent (for patients <18 years), and imminent discharge (not allowing for proper time to assess patients who screen positive for suicide risk). This study was approved by the institutional review boards at the National Institute of Mental Health, Nationwide Children’s Hospital, Boston Children’s Hospital, and Children’s National Hospital.

The ASQ11  is a 4-item self-report suicide risk screening tool that consists of 4 yes or no items used to assess recent suicidal ideation, burdensomeness, and lifetime suicidal behavior and a prompted fifth item used to assess imminent risk of suicidality (“Are you having thoughts of suicide right now?”) (see Fig 1). A patient who responds “no” to all of the first 4 questions is considered to have screened negative for suicide risk and is not asked the acuity item. Patients who respond “yes” to any of the first 4 questions, or refuse to answer, and respond no to the fifth item are considered to have screened nonacute positive and require further suicide risk assessment. Patients who respond yes to any of the first 4 questions, or refuse to answer, and also endorse the acuity item are considered to have screened acute positive and require immediate safety precautions. Internal consistency reliability, as assessed by Cronbach’s α, is .93. The ASQ was validated with a sensitivity of 96.9%, a specificity of 87.6%, and a NPV of 99.7% in pediatric medical and surgical patients in the ED.11  Recently, the ASQ has been found to have predictive validity because positive ASQ screen results were shown to predict suicide attempts in youth at 3-month follow-ups.15 

FIGURE 1

The ASQ. NIMH, National Institute of Mental Health.

FIGURE 1

The ASQ. NIMH, National Institute of Mental Health.

Close modal

The SIQ and SIQ-JR are gold standard measures of severity of suicidal ideation in adolescents.16  In this study, the 30-item SIQ was administered to participants 15 years or older, and the 15-item SIQ-JR was administered to participants 14 years or younger. In both versions of the SIQ, individuals were asked to rate the frequency with which a thought occurs on a 7-point continuum scale ranging from never (0) to almost every day (6). A cutoff of ≥41 on the SIQ and ≥31 on the SIQ-JR indicates the presence of clinically significant suicidal ideation warranting further assessment. The SIQ is often used as a criterion standard in pediatric studies of suicide risk because it has been shown to be a moderately or highly sensitive marker of broad suicidality and subsequent suicide attempts.11,1618  The SIQ has a high reliability (SIQ: r = 0.97; SIQ-JR: r = 0.94), validity, and predictive ability.16,19 

In addition to demographic information, patients responded to study evaluation questions used to assess their opinions about suicide risk screening (eg, “Do you think nurses should ask kids about suicidal thoughts while they are in the hospital? Why or why not?”).

On designated data collection days, the patient census was collected from the charge nurse on the preselected units, and all eligible patients were approached for study enrollment. Informed assent was obtained from participants 17 years or younger, and informed consent was obtained from their parent or guardian or directly from participants who were 18 years or older. Other individuals in the room were then asked to leave during administration of the study measures, which were verbally administered by trained data collectors in a counterbalanced manner, wherein the order of the measures was rotated to reduce the risk of fatigue effects. Each participant was given a $10 gift card as compensation for participating. Study participation ended after completion of the questionnaires. Parents of participants aged ≤17 were notified if their child screened positive for suicide risk. Participants aged ≥18 did not have their responses shared without their consent. All patients who screened positive on the ASQ and/or the SIQ or SIQ-JR (referred to hereinafter only as the SIQ) received a follow-up suicide risk assessment by a clinician. Patient safety was managed as clinically indicated and per standard of care for each institution.

On the basis of an estimated suicidal risk prevalence rate of between 4% and 17% in a pediatric nonpsychiatric population,11,20  a sample size of 600 would provide sufficient power to test the primary study aims. It was decided in advance that study recruitment would stop as soon as this threshold of 600 participants was reached.

Descriptive statistics and univariate and multivariate analyses are reported. We calculated the proportion of true cases of suicide risk that screened positive (sensitivity), the percentage of true noncases that screened negative (specificity), the percentage of screened positives that were true cases (positive predictive value [PPV]), and the percentage of true noncases that screened negative (negative predictive value [NPV]). The likelihood ratio for positive test results (LR+) was calculated to indicate how much more likely the positive test results were to occur in participants with suicide risk compared with those without suicide risk. The likelihood ratio for negative test results (LR−) was calculated to represent how much less likely the negative test results were to occur in participants with suicide risk compared with participants without suicide risk. A Cohen’s κ was calculated because it is a commonly used measure of overall concordance between a screening scale and a gold standard test. The area under the curve (AUC) was also calculated as a measure of the extent to which the ASQ predicted the SIQ. If the ASQ is completely unrelated to the SIQ, the expected value of the AUC is 0.50 (ie, a 50-50 chance in distinguishing suicide risk from a non–suicide risk case); if the ASQ is a perfect predictor of the SIQ, the expected value of the AUC is 1.0. Sensitivity, specificity, NPV, and AUC are of particular interest for this study because a good suicide risk screening tool should be sensitive in case finding for further clinical assessment, should have good specificity so as to not overburden the system with false-positives, and should have a good concordance between the screen test and the gold standard. Multivariate binary logistic regression modeling was used to estimate associations between elevated suicide risk based on the ASQ tool and the demographic characteristics of sex, age, and race and/or ethnicity. Less than 0.2% of missing responses on ASQ and SIQ items were treated as “no” or “none.” Data analyses were conducted by using SAS software version 9.4 (SAS Institute, Inc, Cary, NC).21 

Across all 3 study sites, a total of 795 eligible pediatric medical and surgical inpatients were approached to participate in a study about suicide risk screening. After hearing about the study, 602 patients consented to participate in the study, although 2 participants provided invalid or incomplete data (ie, the participant had already participated in the study), resulting in a total of 600 participants who completed the study measures and an enrollment rate of 75.5% (600 of 795). The sample was 59.2% girls (355 of 600) and 55.2% white (331 of 600) and had a mean age of 15.4 years (SD = 2.8). Most patients (58%) were admitted for general medical conditions, such as fever and stomach pain, followed by admittance for general surgery (25%), injury (7%), general medical conditions and psychiatric reasons (3%), and psychiatric reasons (1%). For 6% of participants, the reason for admittance was unknown. See Table 1 for study sample characteristics.

TABLE 1

Sample Characteristics

CharacteristicsN = 600
n%
Sex   
 Female 355 59.17 
 Male 242 40.30 
 Unknown 0.50 
Race and/or ethnicity   
 White 331 55.17 
 African American 140 23.33 
 Hispanic 59 9.83 
 Asian American or Pacific Islander 10 1.67 
 Mixed 45 7.50 
 Other 11 1.83 
 Unknown 0.66 
Reason for admittance   
 General medicine 346 57.67 
 General surgery 152 25.33 
 Injury 42 7.00 
 General medicine and psychiatric reasons 16 2.67 
 Psychiatric reasons 1.17 
 Unknown 37 6.16 
Study site   
 Boston Children’s Hospital 200 33.33 
 Children's National Medical Center 200 33.33 
 Nationwide Children’s Hospital 200 33.33 
No English spoken at home 24 4.00 
Parents married 315 52.50 
Lives with both parents 312 52.00 
With private insurance 282 47.00 
Considers religion important 212 35.33 
Karnofsky/Lansky score (functional performance)   
 Low 1.17 
 Moderate 63 10.50 
 High 519 86.50 
 Unknown 11 1.83 
Endorsement of ASQ items   
 ASQ 1: “In the past few weeks, have you wished you were dead?” 43 7.17 
 ASQ 2: “In the past few weeks, have you felt that you or your family would be better off if you were dead?” 28 4.67 
 ASQ 3: “In the past week, have you been having thoughts about killing yourself?” 17 2.83 
 ASQ 4: “Have you ever tried to kill yourself?” 48 8.00 
 ASQ 5: “Are you having thoughts of killing yourself right now?” (ASQ acute positive) 0.50 
Endorsed any of ASQ items 1–4 (ASQ positive) 80 13.33 
SIQ or SIQ-JR result positive 30 5.00 
CharacteristicsN = 600
n%
Sex   
 Female 355 59.17 
 Male 242 40.30 
 Unknown 0.50 
Race and/or ethnicity   
 White 331 55.17 
 African American 140 23.33 
 Hispanic 59 9.83 
 Asian American or Pacific Islander 10 1.67 
 Mixed 45 7.50 
 Other 11 1.83 
 Unknown 0.66 
Reason for admittance   
 General medicine 346 57.67 
 General surgery 152 25.33 
 Injury 42 7.00 
 General medicine and psychiatric reasons 16 2.67 
 Psychiatric reasons 1.17 
 Unknown 37 6.16 
Study site   
 Boston Children’s Hospital 200 33.33 
 Children's National Medical Center 200 33.33 
 Nationwide Children’s Hospital 200 33.33 
No English spoken at home 24 4.00 
Parents married 315 52.50 
Lives with both parents 312 52.00 
With private insurance 282 47.00 
Considers religion important 212 35.33 
Karnofsky/Lansky score (functional performance)   
 Low 1.17 
 Moderate 63 10.50 
 High 519 86.50 
 Unknown 11 1.83 
Endorsement of ASQ items   
 ASQ 1: “In the past few weeks, have you wished you were dead?” 43 7.17 
 ASQ 2: “In the past few weeks, have you felt that you or your family would be better off if you were dead?” 28 4.67 
 ASQ 3: “In the past week, have you been having thoughts about killing yourself?” 17 2.83 
 ASQ 4: “Have you ever tried to kill yourself?” 48 8.00 
 ASQ 5: “Are you having thoughts of killing yourself right now?” (ASQ acute positive) 0.50 
Endorsed any of ASQ items 1–4 (ASQ positive) 80 13.33 
SIQ or SIQ-JR result positive 30 5.00 

In total, 5% (30 of 600) of participants screened positive for suicide risk on the SIQ, and 13.3% (80 of 600) of participants screened positive on the ASQ. The ASQ correctly identified all but 1 individual who screened positive on the SIQ. Of the 80 ASQ positive screen results, 96% (77 of 80) were deemed nonacute positive, and 3 patients (0.05% of the total sample) endorsed current suicide ideation, indicating an acute positive screen result. A total of 48 participants (8.0% of the total sample and 60% of ASQ positive screen results) reported a previous suicide attempt. Additionally, 35% (28 of 80) of those who screened positive on the ASQ met criteria for a positive screen result solely because of endorsement of a previous lifetime suicide attempt.

The clinical validity of the ASQ, by using the SIQ as a gold standard, is shown in Table 2. The ASQ had a sensitivity of 96.67% (95% confidence interval [CI]: 82.78% to 99.92%), a specificity of 91.05% (95% CI: 88.40% to 93.27%), a PPV of 36.35% (95% CI: 25.79% to 47.76%), and an NPV of 99.81% (95% CI: 98.93% to 99.99%). The LR+ was 11.18 (95% CI: 8.24 to 14.11), and the LR− was 0.04 (95% CI: −0.03 to 0.11). The κ was 0.49 (95% CI: 0.38 to 0.60), and the AUC was 0.94 (95% CI: 0.90 to 0.97).

TABLE 2

Clinical Validity Statistics

ASQ Positive Screen Result, n (%)SIQ or SIQ-JR Positive Result: Gold Standard, n (%)Sensitivity, % (95% CI)Specificity, % (95% CI)PPV, % (95% CI)NPV, % (95% CI)Log-Likelihood Ratioκ (95% CI)AUC, c (95% CI)
LR+ (95% CI)LR− (95% CI)
80 (13.33) 30 (5.00) 96.67 (82.78 to 99.92) 91.05 (88.40 to 93.27) 36.25 (25.79 to 47.76) 99.81 (98.93 to 99.99) 11.18 (8.24 to 14.11) 0.04 (−0.03 to 0.11) 0.49 (0.38 to 0.60) 0.94 (0.90 to 0.97) 
ASQ Positive Screen Result, n (%)SIQ or SIQ-JR Positive Result: Gold Standard, n (%)Sensitivity, % (95% CI)Specificity, % (95% CI)PPV, % (95% CI)NPV, % (95% CI)Log-Likelihood Ratioκ (95% CI)AUC, c (95% CI)
LR+ (95% CI)LR− (95% CI)
80 (13.33) 30 (5.00) 96.67 (82.78 to 99.92) 91.05 (88.40 to 93.27) 36.25 (25.79 to 47.76) 99.81 (98.93 to 99.99) 11.18 (8.24 to 14.11) 0.04 (−0.03 to 0.11) 0.49 (0.38 to 0.60) 0.94 (0.90 to 0.97) 

c, c-statistic.

We further examined associations between endorsement of suicide risk, as measured by the ASQ, and the demographic variables of sex, age, and race (See Table 3). Girls were significantly more likely than boys to screen positive for suicide risk (odds ratio [OR] = 1.79; 95% CI: 1.01 to 3.18). With age stratified into 2 categories (≤12 years and ≥13 years), youth aged 13 years and older were 5 times more likely than youth aged 12 years and younger (OR = 5.22; 95% CI: 1.85 to 14.69) to screen positive for suicide risk. Suicide risk did not vary significantly across racial and ethnic groups (χ2 = 7.1; degree of freedom = 3; P = .068).

TABLE 3

The Association of Suicide Risk (Assessed by the ASQ Screen Tool) and Demographic Characteristics

% ASQ Result PositiveOR95% CIχ2 (df) [P]
Age group, y    4.4 (1) [.036] 
 10–12 3.5 Reference Reference — 
 13–21 15.7 5.22 1.85 to 14.69 — 
Sex    9.8 (2) [.002] 
 Female 16.3 1.79 1.01 to 3.18 — 
 Male 9.1 Reference Reference — 
Race and/or ethnicity    7.1 (3) [.068] 
 White 10.3 Reference Reference — 
 Black 17.1 1.79 1.01 to 3.18 — 
 Hispanic 18.6 2.24 1.05 to 4.81 — 
 Other 16.7 1.89 0.89 to 4.02 — 
% ASQ Result PositiveOR95% CIχ2 (df) [P]
Age group, y    4.4 (1) [.036] 
 10–12 3.5 Reference Reference — 
 13–21 15.7 5.22 1.85 to 14.69 — 
Sex    9.8 (2) [.002] 
 Female 16.3 1.79 1.01 to 3.18 — 
 Male 9.1 Reference Reference — 
Race and/or ethnicity    7.1 (3) [.068] 
 White 10.3 Reference Reference — 
 Black 17.1 1.79 1.01 to 3.18 — 
 Hispanic 18.6 2.24 1.05 to 4.81 — 
 Other 16.7 1.89 0.89 to 4.02 — 

df, degree of freedom; —, not applicable.

Feasibility of screening was assessed in the domains of acceptability, prevalence, practicality, and patient opinion. Of the 795 eligible patients who were told about the study’s focus on suicide risk screening, 75.5% of patient and parent dyads agreed to participate in a suicide risk screening study. Common reasons for declining to participate included acute medical symptoms (such as pain or fatigue), parental belief that the patient was not suicidal and would not be helpful to the study, and parental concerns about iatrogenic risk of asking about suicide.

Prevalence of suicidal thoughts and behaviors via the positive screen result rate was assessed to determine if suicidal thoughts or behaviors are common enough in medical and surgical inpatient populations to warrant screening. A total of 81 participants (13.5%) screened positive for suicide risk, as measured by the ASQ and the SIQ. For the total sample, 0.5% were at imminent risk because of endorsement of current suicidal ideation.

The practicality of screening for suicide risk was also considered. All patients who screened positive for suicide risk received an appropriate suicide risk assessment in accordance with the inpatient unit’s standard of care, which equated to roughly 1 extra hospital psychiatric consultation per week during the data collection period.

Finally, patient opinions of the screening were assessed by using an in-depth qualitative analysis approach as initially reported by Ross et al.22  In summary, most patients (83.0%) agreed that nurses should regularly screen pediatric patients for suicide risk in the hospital. Patients in favor of suicide risk screening identified the importance of screening for suicide prevention and the belief that routine screening can identify high-risk patients and allow them to receive needed mental health care. Patients opposed to suicide risk screening cited concerns for privacy, irrelevance of suicide risk screening to medical patients, and concerns about iatrogenic harm of asking about suicide.

This is the first study to reveal the validity and feasibility of administration of the ASQ as a brief 4-item suicide risk screening tool among pediatric medical and surgical inpatients. The ASQ was able to accurately detect suicide risk, as measured against the gold standard criterion SIQ in hospitalized medical and surgical patients, with excellent sensitivity, specificity, NPV, and AUC. The psychometric properties in this study are comparable with those of the original ASQ development study in the ED.13  In particular, the ASQ was found to have stronger specificity in the inpatient pediatric medical and surgical sample versus the pediatric ED sample (91.05% vs 87.6%).13 

The high sensitivity of the ASQ indicates that 96.67% of youth at risk for suicidal behavior, as assessed by the gold standard, were identified by having a positive response to at least 1 ASQ item. Regarding the specificity of the ASQ, 91.05% of those who were not at current risk for suicide were correctly identified as such. For a medical or surgical setting with limited mental health resources, this high specificity rate helps to ensure the practicality of using this measure without overburdening an inpatient medical or surgical setting. The NPV of 99.81% is significant in making sure that youth at risk for suicide are identified because this rate indicates the probability that a negative screen result is truly not at elevated risk for suicide.

Screening for suicide risk on a pediatric inpatient medical or surgical unit by using the ASQ was also feasible. Parents were generally accepting of screening, as demonstrated by a high enrollment rate. Some reasons for parents not enrolling their children in the study, such as patient being in too much pain or too acutely ill to be screened, can inform improved screening procedures. Parents should be educated about the common misconception around iatrogenic risk because numerous studies have revealed that asking young people questions about suicide does not lead to the onset or an increase in suicidal thoughts or behaviors.2326 

The positive screen result rate on the ASQ in this sample was 13.3%, but only 0.5% (3 of 600) of participants screened acutely positive and required full safety precautions. This detected rate was high enough to warrant screening but low enough to not overburden busy medical and surgical unit workflows or overwhelm the available mental health resources because positive screen results amounted to 1 or 2 extra psychiatric consultations per week. Notably, although youth >12 years were more likely to screen positive for suicide risk compared with youth <12 years, 3.5% of the preteenagers in this study screened positive for suicide risk, supporting the practice of universal screening beginning at least at age 10 years, perhaps even starting at 8 years.15,27  Girls in this study were also more likely to screen positive for suicide risk compared with boys, which is supported by previous research revealing that although there is a higher suicide completion rate among young boys, young girls are more likely to report suicidal ideation or past suicide attempts.28  Additionally, 35% of the total positive screen results consisted of a single yes to the question about previous suicide attempts, signifying that a sizeable portion of positive screen results may be related to past behavior and may not be a current safety issue, thereby not requiring immediate intervention during the hospital stay (eg, 1:1 observer, removal of clothing, etc). Finally, a majority of patients in our sample (83%) expressed support for screening and highlighted the importance of using the medical or surgical setting for suicide prevention, providing patients with an opportunity to speak openly about suicidal thoughts and connecting with mental health resources, which is in line with previous research revealing screening to be a critical step in suicide prevention.29 

These results should be interpreted with consideration of the following study limitations. First, all 3 study sites were in urban academic hospitals, so findings may not be generalizable. Second, a convenience sample was used, and no data were collected on participants who declined, which has the potential to add selection bias but suggests that positive screen result rates will likely be lower with universal implementation. Third, the timing of when measures were administered varied within each patient’s length of stay and was not recorded. Fourth, use of multiple measures of suicidality by repeatedly asking participants about suicide ideation may have an effect, although the order of administering the ASQ and SIQ or SIQ-JR was counterbalanced to mediate any possible fatigue effects. Lastly, the SIQ or SIQ-JR is primarily intended to identify risk for suicide ideation and may not be predictive of suicide behavior.

The ASQ is intended as a primary screening tool to identify patients who require further evaluation. Rational management of positive screen results can be conducted in a way that does not impede medical care, overwhelm the patient, or exhaust mental health resources. As recommended in the Youth Suicide Risk Clinical Pathway developed by Brahmbhatt et al,30  nonacute positive screen results should be followed-up with a brief suicide safety assessment31  to determine if any safety precautions or a more in-depth psychiatric evaluation is needed. Conversely, acute positive screen results, which are extremely rare among medical and surgical patients, do not require a brief suicide safety assessment because they are automatically considered an emergency and will require full safety precautions and full mental health and safety evaluation. In future research, the most appropriate minimum age for universal suicide risk screening should be established.

The ASQ has strong sensitivity, specificity, and NPV among pediatric medical and surgical inpatients, making it a valid brief suicide risk screener for youth in this setting. Screening with the ASQ was found to be acceptable by both parents and patients, is warranted on the basis of prevalence rates, and is not overly burdensome to manage on the pediatric medical or surgical unit. By using this brief screener as the first part of a 3-tiered clinical pathway, inpatient medical and surgical settings can be better equipped to detect and appropriately manage suicide risk among pediatric inpatients.

We thank all the patients, nurses, mental health teams, and medical teams who helped make the study run smoothly. We also thank the following people, who were instrumental to the success of the study: Paramjit Joshi, Stephen J. Teach, Sandra Rackley, Michael Lotke, Jichuan Wang, Adelaide Robb, Laura Nicholson, Nadirah Waites, Nicole Hedrick, Kerry Brodziak, Mariam Gregorian, Elizabeth Ballard, Deborah Snyder, Jeanne Radcliffe, Ian Stanley, Julian Lantry, William Booker, Michael Schoenbaum, Cristan Farmer, Sally Nelson, and Patricia Ibeziako.

Drs Horowitz, Wharff, Pao, and Bridge conceptualized and designed the study, coordinated and supervised data collection, and critically revised the manuscript for important intellectual content; Ms Mournet and Ms Lanzillo conducted analysis and interpretation of data, prepared the first draft of the manuscript, and critically reviewed the manuscript; Drs Ross, McBee-Strayer, White, and Solages coordinated and supervised data collection and critically revised the manuscript; Ms He and Dr Merikangas conducted analysis and interpretation of data and reviewed and revised the manuscript; Ms Bergdoll and Mr Powell contributed to the acquisition of data and critically revised the manuscript; and all authors approved the final manuscript and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Deidentified individual participant data will not be made available.

The views expressed in this article are those of the authors and do not reflect the official policy of the National Institute of Mental Health, National Institutes of Health, or the Department of Health and Human Services.

This trial has been registered at www.clinicaltrials.gov (identifier NCT02050867).

FUNDING: Supported in part by the Intramural Research Program of the National Institute of Mental Health (ZIAMH002922). This research was also supported in part by the American Foundation for Suicide Prevention. Funded by the National Institutes of Health (NIH).

1
Centers for Disease Control and Prevention
.
Welcome to WISQARS
.
2018
. Available at: https://www.cdc.gov/injury/wisqars/index.html. Accessed February 1, 2020
2
Pan
YJ
,
Lee
MB
,
Chiang
HC
,
Liao
SC
.
The recognition of diagnosable psychiatric disorders in suicide cases’ last medical contacts
.
Gen Hosp Psychiatry
.
2009
;
31
(
2
):
181
184
3
Rhodes
AE
,
Khan
S
,
Boyle
MH
, et al
.
Sex differences in suicides among children and youth: the potential impact of help-seeking behaviour
.
Can J Psychiatry
.
2013
;
58
(
5
):
274
282
4
Habis
A
,
Tall
L
,
Smith
J
,
Guenther
E
.
Pediatric emergency medicine physicians’ current practices and beliefs regarding mental health screening
.
Pediatr Emerg Care
.
2007
;
23
(
6
):
387
393
5
Kuo
CJ
,
Chen
VC
,
Lee
WC
, et al
.
Asthma and suicide mortality in young people: a 12-year follow-up study
.
Am J Psychiatry
.
2010
;
167
(
9
):
1092
1099
6
Rodway
C
,
Tham
SG
,
Ibrahim
S
, et al
.
Suicide in children and young people in England: a consecutive case series
.
Lancet Psychiatry
.
2016
;
3
(
8
):
751
759
7
Thibault
DP
,
Mendizabal
A
,
Abend
NS
,
Davis
KA
,
Crispo
J
,
Willis
AW
.
Hospital care for mental health and substance abuse in children with epilepsy
.
Epilepsy Behav
.
2016
;
57
(
pt A
):
161
166
8
The Joint Commission
.
Summary data of sentinel events reviewed by The Joint Commission
. Available at: https://www.jointcommission.org/assets/1/18/Summary_4Q_2017.pdf. Accessed February 1, 2020
10
Thom
R
,
Hogan
C
,
Hazen
E
.
Suicide risk screening in the hospital setting: a review of brief validated tools
.
Psychosomatics
.
2020
;
61
(
1
):
1
7
11
Horowitz
LM
,
Bridge
JA
,
Teach
SJ
, et al
.
Ask Suicide-Screening Questions (ASQ): a brief instrument for the pediatric emergency department
.
Arch Pediatr Adolesc Med
.
2012
;
166
(
12
):
1170
1176
12
Horowitz
LM
,
Boudreaux
ED
,
Schoenbaum
M
,
Pao
M
,
Bridge
JA
.
Universal suicide risk screening in the hospital setting: still a Pandora’s box?
Jt Comm J Qual Patient Saf
.
2018
;
44
(
1
):
1
3
13
Horowitz
L
,
Ballard
E
,
Teach
SJ
, et al
.
Feasibility of screening patients with nonpsychiatric complaints for suicide risk in a pediatric emergency department: a good time to talk?
Pediatr Emerg Care
.
2010
;
26
(
11
):
787
792
14
Wintersteen
MB
.
Standardized screening for suicidal adolescents in primary care
.
Pediatrics
.
2010
;
125
(
5
):
938
944
15
DeVylder
JE
,
Ryan
TC
,
Cwik
M
, et al
.
Assessment of selective and universal screening for suicide risk in a pediatric emergency department
.
JAMA Netw Open
.
2019
;
2
(
10
):
e1914070
16
Reynolds
WM
.
Suicidal Ideation Questionnaire (SIQ)
.
Odessa, FL
:
Psychological Assessment Resources
;
1987
17
Horowitz
LM
,
Wang
PS
,
Koocher
GP
, et al
.
Detecting suicide risk in a pediatric emergency department: development of a brief screening tool
.
Pediatrics
.
2001
;
107
(
5
):
1133
1137
18
King
CA
,
O’Mara
RM
,
Hayward
CN
,
Cunningham
RM
.
Adolescent suicide risk screening in the emergency department
.
Acad Emerg Med
.
2009
;
16
(
11
):
1234
1241
19
Keane
EM
,
Dick
RW
,
Bechtold
DW
,
Manson
SM
.
Predictive and concurrent validity of the Suicidal Ideation Questionnaire among American Indian adolescents
.
J Abnorm Child Psychol
.
1996
;
24
(
6
):
735
747
20
Kann
L
,
McManus
T
,
Harris
WA
, et al
.
Youth risk behavior surveillance - United States, 2017
.
MMWR Surveill Summ
.
2018
;
67
(
8
):
1
114
21
SAS Institute, Inc
.
SAS 9.4 Statements: Reference
.
Cary, NC
:
SAS Institute Inc
;
2013
22
Ross
AM
,
White
E
,
Powell
D
,
Nelson
S
,
Horowitz
L
,
Wharff
E
.
To ask or not to ask? Opinions of pediatric medical inpatients about suicide risk screening in the hospital
.
J Pediatr
.
2016
;
170
:
295
300
23
Gould
MS
,
Marrocco
FA
,
Kleinman
M
, et al
.
Evaluating iatrogenic risk of youth suicide screening programs: a randomized controlled trial
.
JAMA
.
2005
;
293
(
13
):
1635
1643
24
Crawford
MJ
,
Thana
L
,
Methuen
C
, et al
.
Impact of screening for risk of suicide: randomised controlled trial
.
Br J Psychiatry
.
2011
;
198
(
5
):
379
384
25
Mathias
CW
,
Michael Furr
R
,
Sheftall
AH
,
Hill-Kapturczak
N
,
Crum
P
,
Dougherty
DM
.
What’s the harm in asking about suicidal ideation?
Suicide Life Threat Behav
.
2012
;
42
(
3
):
341
351
26
DeCou
CR
,
Schumann
ME
.
On the iatrogenic risk of assessing suicidality: a meta‐analysis
.
Suicide Life Threat Behav
.
2018
;
48
(
5
):
531
543
27
Ballard
ED
,
Cwik
M
,
Van Eck
K
, et al
.
Identification of at-risk youth by suicide screening in a pediatric emergency department
.
Prev Sci
.
2017
;
18
(
2
):
174
182
28
Peña
JB
,
Caine
ED
.
Screening as an approach for adolescent suicide prevention
.
Suicide Life Threat Behav
.
2006
;
36
(
6
):
614
637
29
Plemmons
G
,
Hall
M
,
Doupnik
S
, et al
.
Hospitalization for suicide ideation or attempt: 2008-2015
.
Pediatrics
.
2018
;
141
(
6
):
e20172426
30
Brahmbhatt
K
,
Kurtz
BP
,
Afzal
KI
, et al
;
PaCC Workgroup
.
Suicide risk screening in pediatric hospitals: clinical pathways to address a global health crisis
.
Psychosomatics
.
2019
;
60
(
1
):
1
9
31
National Institute of Mental Health
.
Ask Suicide-Screening Questions (ASQ) toolkit
. Available at: https:www.nimh.nih.gov/ASQ. Accessed February 1, 2020

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.