Patients’ Opinions About Suicide Screening in a Pediatric Emergency Department

Abstract

Objective

Understanding how children react to suicide screening in an emergency department (ED) can inform implementation strategies. This qualitative study describes pediatric patients’ opinions regarding suicide screening in that setting.

Methods

As part of a multisite instrument validation study, patients 10 to 21 years presenting with both psychiatric and nonpsychiatric complaints to an urban, tertiary care pediatric ED were recruited for suicide screening. Interviews with subjects included the question, “do you think ER nurses should ask kids about suicide/thoughts about hurting them-selves…why/why not?” Responses were transcribed verbatim and up-loaded into NVivo8.0 qualitative software for coding and content analysis.

Results

Of the 156 patients who participated in the study, 106 (68%) presented to the ED with nonpsychiatric complaints and 50 (32%) presented with psychiatric complaints. The patients’ mean (SD) age was 14.6 (2.8) years (range, 10–21 years), and 56% of the sample was female. All patients answered the question of interest, and 149 (96%) of 156 patients supported the idea that nurses should ask youth about suicide in the ED. The 5 most frequently endorsed themes were as follows: (1) identification of youth at risk (31/156, 20%), (2) a desire to feel known and understood by clinicians (31/156, 20%), (3) connection of youth with help and resources (28/156, 18%), (4) prevention of suicidal behavior (25/156, 16%), and (5) lack of other individuals to speak with about these issues (19/156, 12%).

Conclusions

Pediatric patients in the ED support suicide screening after being asked a number of suicide-related questions. Further work should evaluate the impact of suicide screening on referral practices and link screening efforts with evidence-based interventions.

Youth suicide is a global public health problem and the third leading cause of death among 10- to 24-year-olds in the United States.¹ According to the most recent Youth Risk Behavior Survey, 13.8% of high school students seriously thought about suicide in the past year, and 6.3% made an actual suicide attempt.²

Routine screening of pediatric patients in medical settings has been suggested as a way to identify youth with undetected mental health needs.³ As a result, universal suicide screening has been proposed in both primary care settings and emergency departments (EDs).⁴ For millions of children and adolescents, the ED is their only contact with health care providers.⁵ As such, the ED may be uniquely situated to rapidly detect children and adolescents with unmet mental health needs presenting to the ED for both psychiatric and nonpsychiatric chief complaints.⁴

The onus of assessing suicide risk in a pediatric ED, regardless of chief complaint, falls mainly on nonpsychiatric clinicians because most pediatric EDs are not staffed with mental health professionals.⁶ Unfortunately, brief instruments assessing suicidality in pediatric nonmental health patients are lacking. Although a large, multisite study aiming to develop such a tool to detect suicidality in nonpsychiatric patients is underway, it is essential to understand how children and adolescents will react to being questioned about suicide during an ED visit before implementing universal screening efforts. Whereas limited data suggest that pediatric patients are supportive of mental health screening⁷ and there is no evidence that asking about suicide increases thoughts of suicidal ideation⁸ or negative mood states,⁹ there is little qualitative research assessing adolescent patients’ opinions about suicide screening in the ED. Data on how pediatric patients respond and react to suicide screening in the ED are necessary to create the most effective, appropriate, and realistic strategies for intervention. Specifically, these data have the potential to inform screening practices and to provide assurance to nonpsychiatric clinicians that suicide assessments will be acceptable to patients. Therefore, the aim of this qualitative analysis was to describe the opinions of psychiatric and nonpsychiatric patients, aged 10 to 21 years, regarding universal suicide screening in the ED.

METHODS

Participant Population

As part of an ongoing, larger, multisite instrument validation study, a convenience sample of ED patients with both medical and psychiatric presenting complaints were recruited to participate in a suicide screening study. Participants included patients, aged 10 to 21 years, seeking care in an urban, tertiary care pediatric ED, with an annual census exceeding 80,000 visits. This age limit was chosen based on the ages of most patients seen at this particular pediatric ED. The study was conducted between September 2008 and April 2009. During the study period, trained study staff members were stationed in the ED during the week between the hours of 1:00 and 9:30 pm. To obtain target numbers of psychiatric and nonpsychiatric patients, every psychiatric patient and every second nonpsychiatric patient who entered the ED were approached for recruitment into the study.

The institutional review boards at the Children’s National Medical Center and the National Institute of Mental Health approved this study. For participants 17 years and younger, written informed consent was obtained from the parent or guardian, and written informed assent was obtained from the patient. Parents of nonpsychiatric patients were consented separately from their children, such that they could decide whether they wanted their child to be asked about suicide. All participants 18 years and older provided written informed consent. Interviewers were trained to administer the questionnaires in a standardized and consistent manner.

Procedure

After initial triage assessment processes and while in examination rooms waiting to be seen by a clinician, participants were asked a series of questions about suicide risk, as well as an additional survey containing items on sociodemographic information, history of medical and psychiatric illness, and prior health care utilization. After this survey, patients were asked if they had previously been asked about suicide before this study visit. In addition, the open-ended question of interest, “Do you think ER nurses should ask kids about suicide/thoughts about hurting themselves…why/why not?” was also asked. Interviews were conducted without the parent or guardian in the room, but participants were told that if the data collectors had any concerns about their safety, their parents would be notified and their answers would be shared with the ED clinical staff. Individuals who reported suicidal ideation or depressive symptoms were offered follow-up appointments in an outpatient clinic and received follow-up telephone calls to see if they had any additional concerns.

Qualitative Analysis

Responses to the question, “Do you think ER nurses should ask kids about suicide/thoughts about hurting themselves…why/why not?” were transcribed verbatim by each data collector at the time of interview and were subsequently uploaded into NVivo8.0. NVivo8.0 is a qualitative software management package used for data management and to assist with thematic coding. ¹⁰ A group composed of a clinical psychologist, social worker, advanced graduate student, and research assistant used grounded theory and open coding, a technique of naming and categorizing phenomena, to analyze the data. Two of the investigators used the open coding procedure to code the comments and identify themes that emerged from the responses. In meeting with the larger study group, a constant comparison method was used to refine the codes into larger themes. The study team met to collapse and to refine the themes, and 5 final themes were established. Interrater agreement was 80%. All discrepancies were resolved by discussion and consensus.

Quantitative Analysis

Descriptive statistics are reported as proportions and means with SDs and ranges. Logistic regression was used to adjust for age, sex, and insurance status when comparing responses between patients with nonpsychiatric and psychiatric complaints.

RESULTS

Overall, 156 patients in the pediatric ED agreed to participate, with a response rate of 60%. Demographic data are presented in Table 1. Of the 156 participants, 106 (68%) entered the ED with nonpsychiatric presenting complaints, with 20% of this nonpsychiatric sample reporting gastrointestinal diseases such as abdominal pain or vomiting, 10% reporting hematologic diseases such as sickle cell disease, and 8% reporting trauma such as abrasions and fractures. Classification of presenting complaints was conducted in accordance with the new diagnostic grouping system for pediatric ED visits.¹¹ Of the 156 patients, 50 (32%) entered the ED with psychiatric presenting complaints.

TABLE 1.

Demographics of Study Participants (n = 156)

Sex, n (%)
Female	88 (56.4)
Male	68 (43.6)
Race or ethnicity, n (%)
African American	104 (66.7)
White	23 (14.7)
Hispanic/Latino	8 (5.1)
Mixed/other	21 (13.5)
Age, mean (SD), y	14.6 (2.76)
10–11 y, n (%)	20 (12.8)
12–17 y, n (%)	108 (69.2)
18–21 y, n (%)	28 (17.9)
Insurance status, n (%)
Public	81 (51.9)
Private	69 (44.2)
None	6 (3.8)

All participants answered the question of interest and 151/156 (97%) gave a more detailed answer than “I don’t know.” Responses to the two other forced-choice study evaluation questions are reported in Table 2. Of note, only 31% of non-psychiatric respondents reported that they had been asked about suicide in the past, compared to 44% for psychiatric patients. After adjustment for age, sex, and insurance status, the odds of a psychiatric patient having been asked about suicide previously were significantly greater than for a nonpsychiatric patient (adjusted odds ratio = 2.8; 95% confidence interval, 1.3–6.1; P = 0.01). Most participants in both groups agreed that nurses in the ED should ask patients about suicide, with no significant difference between nonpsychiatric and psychiatric patients.

TABLE 2.

Overall Reactions to Suicide Screening in a Pediatric ED

		Total (N = 156)	Nonpsychiatric Patients (n = 106)	Psychiatric Patients (n = 50)	Odds Ratio*	95% Confidence Interval
Have you ever been asked about suicide before?	Yes	55 (35)	33 (31)	22 (44)	2.81	1.29–6.11
Do you think it is a good idea for nurses to ask kids about suicide?	Yes	149 (96)	102 (96)	47 (94)	1.02	0.17–6.22

Values are n (%).

^*Adjusted for age, sex, and insurance status.

Reactions to Suicide Screening in the ED

Themes from the qualitative responses are presented in Table 3. Descriptions of these themes as well as quotes from participants follow.

TABLE 3.

Identified Themes Supporting Suicide Screening in the Pediatric Emergency Department

	Total (N = 156)	Nonpsychiatric Patients (n = 106)	Psychiatric Patients (n = 50)
Identification of at-risk youth	31 (20)	24 (23)	7 (14)
A desire for clinicians to know and understand their situation	31 (20)	17 (16)	14 (28)
Connection of youth with mental health help and resources	28 (18)	13 (12)	15 (30)
Prevention of suicidal behavior	25 (16)	17 (16)	8 (16)
Lack of other individuals to speak with about these issues	19 (12)	15 (14)	4 (8)

Values are n (%).

The most commonly identified response was that suicide screening would identify at-risk youth who would otherwise not report their thoughts and feelings. These responses often emphasized that a nurse would not know about a patient’s suicidal thoughts or behavior unless the nurse specifically asked about suicide.

“Because if you don’t ask them they may never tell you.”

—An 11-year-old male nonpsychiatric patient

“Just because they (kids) don’t show signs, doesn’t mean they aren’t thinking about it.”

—A 15-year-old female psychiatric patient

Another identified theme was the importance of a nurse understanding what was going on in a patient’s life. These responses included the patients’ desire to feel known, heard, and understood.

“So nurse can know what is going on with child mentally.”

—A 17-year-old female nonpsychiatric patient

“Because they (kids) should express their feelings to someone and let them know what is going on in their life.”

—A 13-year-old female nonpsychiatric patient

The connection of suicide screening to mental health intervention was another theme frequently mentioned. These responses suggested that the information gathered from suicide screening could help nurses link a patient with effective interventions.

“So you (nurses) can help stop them (patients) before it is too late.”

—A 15-year-old male nonpsychiatric patient

“Can save somebody’s life by asking and getting them the help they need.”

—A 16-year-old female psychiatric patient

“A lot of kids don’t know who to go to for help and so they can’t get help and resources they need.”

—A 15-year-old female psychiatric patient

Prevention of suicidal behavior and death by suicide was also identified as important.

“Because nurse could do something to save the kids life.”

—A 13-year-old male nonpsychiatric patient

“Cause if you don’t ask they will probably go along and do it.”

—A 12-year-old male nonpsychiatric patient

“Because it is their (nurses) job to make sure kids don’t get hurt.”

—A 12-year-old male psychiatric patient

Participants also identified that some children do not have people in their lives with whom they feel comfortable talking to about these issues. Specifically, participants spoke of the fact that children are not being asked about suicide in other environments or contexts.

“Because sometimes when no one asks them (kid/teen), they (kid/teen) feel no one cares- when someone asks they know someone cares about them.”

—A 12-year-old male nonpsychiatric patient

“Because it is very important, you never know what someone is going through/thinking, they might be thinking about hurting themselves but are scared to tell their parents and just need someone to ask them.”

—An 18-year-old female nonpsychiatric patient

In addition, participants mentioned that screening in a medical setting such as an ED could be a particularly appropriate place to be asked these questions because the nurses asking the questions are a “safe,” “caring,” noncritical person with whom to share this information.

“Because they (kids) may feel comfortable talking to someone who is not going to criticize them.”

—A 17-year-old female nonpsychiatric patient

“Because a lot of kids especially teenagers get sad and don’t have anyone to talk about it with, so if a kid/teen is already in ER and with people who are trained, it’s a good time to talk.”

—A 12-year-old female nonpsychiatric patient

Other Findings

A minority of the participants (4%) stated that they disagreed with suicide screening in children. Reasons given were fear of iatrogenic risk of asking children about suicide or simply believing that asking is “not right.” Other participants suggested conditional screening based on presenting complaint or age (>14 years).

Four nonpsychiatric patients also connected medical illness with emotional distress, pointing out that factors, such as suicidal ideation, could actually be related to the pain and effects of chronic medical illness.

“Because stuff that kids with serious illnesses go through is no joke and can make you lose your mind.”

—A 19-year-old female nonpsychiatric patient

“For their overall well-being, emotional problems could affect their health.”

—A 19-year-old female nonpsychiatric patient

Five nonpsychiatric patients connected suicide screening to personal experiences in their own lives. One shared that she had a friend who died by suicide and mentioned that screening efforts might have “protected” her friend.

Another psychiatric patient also pointed out that just the act of asking about suicide might help a patient, even if they do not receive help in the moment.

“Some will tell the truth and get help and for those that don’t tell the truth maybe it will help them realize they should tell someone.”

—A 14-year-old female psychiatric patient

DISCUSSION

Overall, most study participants agreed with suicide screening efforts and stated that children and adolescents should be asked about suicide in the pediatric ED. Moreover, only 35% of all participants reported that they had ever been asked about suicide in the past. When asked to elaborate further on their opinions, the participants stated beliefs that asking directly about suicidal thoughts could help identify youth at risk, address their desire to be understood, allow youth contemplating suicide to get help, aid in the prevention of suicidal behavior, and provide an unbiased listener with whom to speak. The few patients who disagreed with screening cited concerns about iatrogenic risks of suicide assessment; however, there is no evidence that asking about suicide will lead to an increase in suicidal thoughts or behavior in children or adolescents.⁸ Interestingly, participants were asked for their thoughts after experiencing a battery of suicide-related questions in the ED. Other research has explored adolescent beliefs in focus group discussions or other hypothetical scenarios, but this is the first study to evaluate screening in the ED after participants experienced standardized suicide screening.

One of the most commonly identified themes supported by these data was the notion that directly and specifically asking patients about suicide was important. Participants believed that if they were not specifically asked, youth would not offer up this information unsolicited. This theme relates to current research detailing that many children with depression and suicidal ideation are not identified¹² and that screening efforts can detect mental health conditions in children with nonpsychiatric complaints. ^13,14 Focus group discussions with community adolescents and parents also suggest that parents do not believe that they are able to identify a suicidal adolescent,¹⁵ which led the authors to conclude that systematic screening may be indicated. Overall, these responses suggest that youth who experience suicidal ideation may not volunteer this information unless specifically asked.

Beyond the identification of at-risk youth, participants also suggested that screening was a way for patients to feel known, heard, and understood. These comments suggested that these participants believed that identifying suicide risk is part of obtaining a comprehensive picture of the patient. Moreover, a few of the participants connected the screening to personal experiences of friends displaying suicidal behavior, demonstrating that adolescent suicide is not an uncommon concern in their peer groups. In contrast, in parent-adolescent focus group discussions around the country, participants identified suicide as a nationwide concern but did not believe that adolescent suicide was a problem in their own communities.¹⁵

Another identified theme was that adolescents are not being asked about suicide elsewhere. Fewer than half of the patients reported that they had been asked about suicide before this ED visit, suggesting that while screening programs are being implemented into schools and primary care settings,⁴ there are still many children and adolescents who are not routinely assessed. Furthermore, as for disclosing these thoughts to adults, some participants felt that they would be criticized and/or did not seem comfortable telling their parents; rather, they viewed the ED nurse as someone impartial with whom they could discuss their suicidal thoughts and feelings. These comments suggested that the ED would be a particularly optimal setting for screening efforts.

The participants also connected screening efforts with a way to provide suicide prevention and intervention strategies with teens. They believed that through screening efforts, individuals who were at-risk could receive “help” that would prevent later suicidal behavior. This connection of screening to validated interventions has been stressed as an essential component of effective suicide prevention through screening.^4,16 Mental health screening in adolescent primary care settings has been shown to be associated with increased mental health services.¹⁷ However, identification of previously undetected mental health concerns in the ED in adolescents has not been shown to lead to an increase in psychiatric follow-up beyond the ED,¹⁸ suggesting that further work is needed in designing effective interventions for positive screens in the ED.

Another special consideration identified by participants included the appropriate age range for suicide screening. An analysis of suicide screening in children younger than 12 years suggests that younger children in the ED may endorse suicidal thoughts and behaviors when asked, but the validity of these responses warrant further investigation.¹⁹ Participants also raised concerns about the effects of chronic medical illness on emotional health in pediatric psychosomatic medicine literature^20–22 and whether individuals should be screened according to presenting complaint, a topic currently under investigation by researchers studying suicide screening^23,24 and depression screening.¹⁴ Further analyses within the larger multisite study will investigate the impact of chronic illness on suicide risk.

This study had several limitations including the fact that data were collected using a convenience sample of patients recruited from a single urban pediatric ED. Responses, therefore, may not generalize to other patient populations or settings. It should be noted that this particular ED provided full evaluations for psychiatric patients up to age 18 years only and arranged evaluations elsewhere for older patients.

In addition, participants and their parents agreed to participate in a suicide assessment study, which may have biased the patient population toward individuals who were more likely to agree that suicide screening was important. It is not known whether the 40% of patients and guardians who declined to participate would have agreed with suicide screening; however, most of the declines occurred because the parents were concerned about pressing medical concerns and chose not to participate in research. Further analysis of the decline rates is presented elsewhere. ²⁴ Parents were not in the room during this evaluation, but participants were told that safety concerns would be shared with parents, which also may have impacted responses. Lastly, as part of the study inclusion criteria, patients who would not be able to communicate in English as well as individuals without a legal guardian in the ED (eg, foster care populations) were excluded from data collection further limiting generalizability. However, this is the first known study of individuals in the pediatric ED who were asked their opinions about suicide screening after being asked a set of suicide-related questions. It is possible that, through their own experience of being asked about suicide, participants had a new perspective on whether screening should be implemented universally.

CONCLUSIONS

Pediatric patients in a busy urban pediatric ED agreed that ED nurses should screen patients for suicide risk and gave illuminating answers in their own words describing the importance of detection, prevention, and intervention through suicide screening. These data lend further evidence to the importance of identifying pediatric patients previously unknown to be at suicide risk. Future research focused on the impact of screening in the ED and its utility in connecting patients at risk with follow-up services is warranted.