Suicide is now the third leading cause of death during adolescence and the prevalence of completed suicide among 15- to 24-year-olds has increased two-fold over the last two decades (1). A recent policy statement from the American Academy of Pediatrics (AAP) emphasized the critical role for pediatricians in both prevention and intervention for suicidal youth (2). This role is likely to become more important as these physicians become the gatekeepers for referral to specialists, including psychiatrists (3).
To date, there has been no national training initiative for pediatricians or other primary care physicians involved in the assessment and care of suicidal children and adolescents. The implications for psychiatrists, who have traditionally been called upon to provide training of this nature, have not been discussed in light of what is already known about pediatric practices in this arena.
Though limited, objective data regarding current pediatric practices is of concern. For instance, Hergenroeder et al. reviewed 45 charts of children and adolescents seen in a fully staffed pediatric emergency room for evaluation of suicidal ideation and/or behavior (4). Of these, 24 were referred directly to the child psychiatry service, with only minimal involvement of the pediatric resident, or were managed only for medical conditions. In 10 of the remaining 21 cases, the residents' notes revealed no indication of past suicide attempts, current suicidal ideation, or any in-depth assessment of the patient's current level of functioning. The authors concluded that "pediatric programs need to emphasize training in suicide interventions and establish a role for the pediatric resident in the assessment process" (4, p. 791).
The perspective of graduating pediatric residents regarding their own experiences in training have not been addressed in the literature in this country. Nevertheless, a study from Great Britain suggests that resident-level general physicians feel that their training in assessing patients who have overdosed is inadequate (5). When a cohort of 20 junior medical staff (akin to house staff) were interviewed about their attitudes regarding patients treated in the emergency room for overdose, there was no relationship between their attitudes and the standard of their assessments. Only 25% of those interviewed felt their medical school experience prepared them for assessing suicidal patients as house officers, and a majority spontaneously noted the need for more training in this arena in anticipation of a change of policy under which general physicians would decide which patients were to be seen by a psychiatrist. Chart reviews of patients seen in the emergency room for overdose in Britain reveal similar deficiencies to those just cited (6).
With the continued rise in rates of suicide and suicide attempts, the shortage of child and adolescent psychiatrists (7), and the increased emphasis on cost containment necessitating fewer referrals for assessment by specialists, the skills of pediatricians in assessing suicidal children and adolescents are likely to be further tested. It appears that the practice of having pediatricians and other primary care physicians screen most suicidal patients and decide which ones need psychiatric evaluation is likely to soon be the norm in this country.
Nineteen of 22 third-year pediatric residents at 2 university training programs in the northeastern United States were interviewed. Permission to approach these residents about the project was obtained from the pediatric training directors of both programs. Ten of 12 residents in Program A and 9 of 10 in Program B participated in the interviews; the 3 residents who did not participate in the study could not be reached before moving out of the area at the close of their residencies and attempts to interview them by telephone were unsuccessful. The study was described at the start of the interview, and each participant signed a consent form. All interviews were conducted by the first author and lasted about 40 minutes. Each interview was completed in the 2 months prior to, or just after, residency graduation.
The programs were chosen partly because they both participate in the Triple Board Training Program, in which 2 or more residents are enrolled each year to complete 2 years of pediatric training, followed by 3 years of training in both adult and child psychiatry. One of the goals of the Triple Board Program is to integrate pediatric and psychiatric education with the implicit hope of having this integration generalize to the entire pediatric training process. In applying to become a Triple Board training site, these programs had to demonstrate adequate cooperation between their pediatric, adult psychiatry, and child psychiatry divisions. These programs were, therefore, deemed likely to be at the "cutting edge" of integrative training. Since all Triple Board residents in these programs start psychiatric training after 2 years, none was included in this survey. All of these third-year non-Triple-Board residents were thus completing otherwise "typical" pediatric training, though during their first 2 years they trained alongside residents with a special interest in psychiatric illness in programs selected for their cohesive approach to pediatric psychiatry.
In this descriptive study, we interviewed graduating third-year pediatric residents and focused on several pertinent questions: How much experience in assessing acutely suicidal patients do pediatric residents get during their training? How competent do residents feel in their ability to assess and manage these patients? Do residents feel they receive adequate training to comprehensively evaluate these patients? What other factors, beyond training, affect residents' attitudes toward these patients?
The data in this study are retrospective, and it is clear that asking residents to recall numbers of patients seen and details attendant to these clinical encounters is reliant on potentially faulty memory. The first part of the interview encouraged the residents to recall the breadth of their past experiences in this arena and, as the interview progressed, residents were encouraged to create a narrative about their views of this type of patient and their clinical experience with them. Since these data were felt to be most indicative of these residents' states of mind as they were finishing training, no attempt to objectively validate these impressions with other means (e.g., chart reviews or paper and pencil measures) was made. The participants were asked not to talk to one another about the survey in the hope that the subjects already interviewed would not influence the answers of those not yet seen. Despite these caveats, the relatively uniform responses to the questions both within and between programs were encouraging.
A summary of responses to each of the questions asked will be presented.
How many suicidal patients (of any age) did you care for in medical school?
The average response in Program A was 5 and in Program B, 7. In each, the range was 1 to 20, although 60% reported caring for 3 to 10 patients.
How many suicidal patients did you care for in residency?
The average in Program A was 9, in Program B, 20, and the range 3 to 25 in Program A and 6 to 36 in Program B. Although we did not provide a strict definition of "suicidal," it is notable that the residents spoke only of patients treated for medical sequelae of suicide attempts, with the rare exception of seeing patients with suicidal ideation being screened for medical problems.
Did your medical school experience concerning suicide help you in residency?
In total, 12 of 19 residents (60% in each program) responded affirmatively. Most remarked that the experiences gave them some idea of the questions to ask suicidal patients and helped make them, as one resident put it, "more comfortable talking about feelings directly." The other 40% reported no benefit from medical school experience.
We asked the residents what percentage of suicidal patients they had cared for were seen in the emergency room (ER) vs. on the wards or in the clinic, intensive care unit (ICU), etc. In both programs, over 70% of suicidal patients were seen in the ER setting. The residents reported that child psychiatric coverage was often limited in these settings; a few residents in each program recalled having to make decisions regarding disposition of suicidal teens without benefit of supervision or psychiatric evaluation. The residents uniformly felt burdened by this responsibility.
On a scale of 1 to 10 (1 being least, 10 being most), how competent did you feel in evaluating these patients?
In Program A, the mean response was 6.0; in Program B, the mean was 6.6. Typically, the residents saw themselves as adequately trained to handle the medical care of these patients (e.g., the procedures for treating a child who has taken an overdose); however, 14 of the 19 total residents spontaneously noted discomfort with evaluation of a patient's psychiatric status and with decisions regarding disposition. These residents defined their roles as medical triage personnel and universally expressed a desire to defer further evaluation to psychiatric colleagues.
The residents recalled minimal involvement in the psychiatric evaluation process, either because they had other responsibilities in the ER or ward setting or because they chose not to. One resident summed up her feelings by saying, "I can tell you how to handle an OD (overdose), but I am not sure how serious the person was who took it."
There was no relationship between the total number of suicidal patients cared for in medical school and residency and each resident's ratings of his or her own competence (Program A: r= -0.39, NS; Program B: r = -0.19, NS).
Do you feel there is a need for more formal training regarding evaluation and treatment of suicidal ptients?
Eight of 10 in Program A and 9 of 9 in Program B responded affirmatively. When asked for suggestions on how to conduct this training, the residents focused on a combination of didactic presentations (at noon conferences or Grand Rounds) and case conferences. Most also believed that this training should begin in the first postgraduate year and continue throughout the training experience. This assertion was emphasized by several residents who alluded to a tendency by some of their colleagues to ask first-year residents to care for these patients, particularly in the ER, because of the time and effort it can take to treat an angry and/or resistant teen who has overdosed.
Several residents in Program B mentioned being helped by a questionnaire used in the ER over a 1-year period to study factors related to teen suicide attempts. They felt this questionnaire helped them to be more thorough in workup of these patients and more insightful into the etiologic factors involved in individual cases. One resident suggested having a suicide meeting following any interactions with a suicidal patient, in which the resident who dealt most closely with that patient presents the case, while pediatric and child psychiatry faculty facilitate discussion on management issues.
Have you had personal experience with suicidal individuals outside of the medical setting?
Program A, 4 of 10 responded yes; in Program B, the figure was 9 of 9. When asked whether this experience affected their approach to these patients, 9 of the 13 who responded affirmatively again responded yes. All of these residents recalled being more sensitive to the issue of suicidality after their personal experience. One resident who had been close to a friend who attempted suicide in college stated, "Each new patient now strikes a deep chord." Another who had been an emergency medical technician in college stated that her experiences "heightened my awareness of the seriousness of suicide attempts." Yet another stated that his experience gave him "a closer realization of people' vulnerabilities."
How much contact did you have with the families of suicidal patients?
In Program A, 6 of 10 reported no contact or minimal contact with families. In Program B, it was 4 of 9. Most of these residents recalled passing this responsibility on to child psychiatric or social work supports. Those residents in each program who did participate in family evaluations did so 50% to 80% of the time. Although most residents recalled seeing teens who are accompanied by family at the time of the interaction, those who reported no contact or minimal contact with families stated that, essentially, they did so by choice. Most of the residents who reported frequent contact with families of suicidal patients found the experience educational but unsettling. One resident who routinely had contact with the families of suicidal patients was frustrated by her lack of confidence, which led to an inability to help the family understand. Another resident remembered getting involved with the family of one ICU patient who had taken a serious overdose. He stated flatly, "I couldn't deal with it," and he remembers shying away from being involved with other families of suicidal patients after that particular experience.
What do you remember as the most positive and the most negative experiences in dealing with these patients?
Three residents in each program noted nothing positive about treating these patients. However, all residents from each program recounted negative aspects of this experience. Of those who remembered positive experiences, these experiences were always along the lines of being able to reach a patient struggling with depression and suicidal ideation. The residents recounted feeling good about, as one resident put it, "being able to make a difference." However, this experience was counteracted by more frequent experiences of being unable to reach some of these teens and being frustrated by the system that seemed ill-equipped to change the things in these children's lives that seemed related to their suicidal feelings. Several of the residents across the two programs worried that many of the children and adolescents they had seen were, in their words, "lost causes."
This descriptive study was designed to assess how well pediatric residents feel they are trained in the assessment of suicidal children and adolescents. Residency training provides most residents with considerable clinical experience with suicidal patients, especially in the emergency room. Yet residents endorsed a subjective sense of incompetence with regard to evaluating suicidal patients, particularly in assessing the patient's state of mind. Ninety percent desired more training on the subject of suicide in children and adolescents.
This study was limited to the third-year residents at two teaching hospitals in New England involved in the Triple Board training initiative. It documents only resident perceptions of their training experience. The results are consistent with previously published surveys documenting inadequate assessments of patients screened in the ER, though the retrospective and descriptive nature of these data limit their generalizability. However, given the requirements for Triple Board programs to have effective cooperation and collaboration between child psychiatry and pediatrics, it is not likely that the findings from these institutions overestimate the problem. If these problems exist in pediatric programs that are involved in Triple Board training, then they are likely to exist elsewhere as well.
Perhaps the first question to address is whether pediatricians (and other primary care physicians) are fundamentally uncomfortable with this population of patients and are therefore unable to experience a subjective sense of competence in treating them. Enzer et al. have documented "fundamental attitudinal differences" between the ways that pediatricians and psychiatrists view childhood (8). It is possible that pediatricians' more optimistic outlook may predispose them to be uncomfortable with the problem of suicide and adversely influence their approach to these patients despite adequate training. However, most of the two-thirds of residents in this sample who acknowledged personal experiences with friends or family members who attempted suicide believed these experiences made them more interested in the problem and more empathic with the patients. Furthermore, even psychotherapists rank working with suicidal patients to be the "most stressful" of all clinical situations (9).
The possibility that clinical experience in medical school engenders confidence in residency was not supported in this study. This group of residents' medical school experiences clearly did not fully prepare them to deal with suicidal children and adolescents, even though 60% found these early experiences to be of some later use. The fact that there is no correlation between number of patients seen and subjective ratings of competence in this cohort suggests that experience alone is insufficient to engender residents' confidence in their own assessments. For new physicians, experience without supervision or debriefing may actually potentiate anxiety and discomfort. Residents in both programs noted that their hospitals had no specific policy for managing suicidal patients. It was up to the residents and their immediate supervisors to decide which patients needed further psychiatric evaluation.
If experience alone is not enough to engender comfort in working with these patients, then perhaps adequate training is useful for both the practical purpose of adequate evaluations and the associated sense of competence that completing these evaluations might foster. This cohort uniformly endorsed the need for more training in this area. These findings are consistent with a recent survey of 236 program directors of residencies in multiple disciplines in which over 50% of the directors endorsed or strongly endorsed the need for training in emergency psychiatric intervention (10). Despite this acknowledged need, almost 75% of these programs did not provide reading assignments on emergency psychiatric intervention to their trainees and 70% of the programs included no lectures or seminars on this topic during training. Pediatric programs were below emergency medicine and family practice programs in both perceived need for and the delivery of this training. The reasons why such training is not being integrated into these programs have not been addressed in the literature. This shortcoming may be fundamentally an issue of resources and departments of pediatrics and adult and child psychiatry will be forced to decide whether this kind of training is warranted in the current funding climate.
The burden for providing this training is likely to fall to psychiatrists, a concept long supported by the National Institute of Mental Health (11) and one that has grown in importance as primary care physicians become "gatekeepers" for referral. Fortunately, recent reviews regarding training of primary care physicians by psychiatrists suggest that this training can be effective in changing physician behavior and, perhaps, patient outcome (12). The results of this survey suggest some possible ways to build a successful training program specifically focusing on suicidal children and adolescents. Many of these interventions appear to be relatively low cost, though others may require significant teaching time.
First, these residents report that neither program had an assessment protocol for evaluating suicidal patients. This lack of a minimum standard, as cited in the introduction, appears to be commonplace. Residents, particularly those with the least experience in this area, may not have the knowledge to perform adequate evaluations. It is likely not feasible to have every child with suicidal ideation seen by a psychiatrist; however, it may be feasible to have all first-year pediatric residents learn about important questions to ask in assessing the severity of a suicide attempt, allowing a more informed approach to referral.
An important incidental finding in this study points out the usefulness of this approach. A standardized set of questions developed for a research project at Program B was inadvertently an educational intervention in that the structure and comprehensiveness of the questions made the residents who used the tool feel more comfortable and less avoidant. This is consistent with the findings of McIntire et al. who used a structured interview with a pediatric perspective to train paraprofessionals to estimate lethality of suicide attempts, risk of recidivism, and current stress. These estimates correlated at 95% with independent ratings by psychiatrists (13). Similarly, Patterson found that third-year medical students given a lecture and taught an acronym encompassing various risk factors for suicide lethality (the SAD-PERSONS) were significantly better at judging suicidal risk from a videotaped interview than another group of medical students who were given the lecture alone (14). Teaching simple techniques takes relatively little time and appears to be of considerable benefit.
Second, several of the residents in this sample suggested that a curriculum regarding assessment of suicidal patients be instituted. This curriculum was commonly conceptualized as involving both pediatric and child psychiatric staff. Aside from providing didactic training using tools like suicide intent questionnaires, it was also suggested as important to allow residents to discuss these difficult cases, thereby diminishing the discomfort often experienced after caring for these patients. The use of a "suicide team" to discuss each case encountered by a resident may represent a way of accomplishing the dual goal of "debriefing" and learning from the clinical presentation. This forum would provide an opportunity for the psychiatrist to emphasize details like interview style, which may play a major role in the adequacy of the evaluation (15). Having pediatric residents watch or participate in an interview with a suicidal patient would further emphasize these points. This would also provide a forum to discuss the importance of routinely interviewing all adolescent patients about suicidal ideation (2). Presentations on the scope of the problem, associated risk factors, principles of psychiatric evaluation, and reviews of the research literature could be incorporated. Emphasis on the signs and symptoms of depression and substance abuse in teenagers could also be stressed.
Implementing comprehensive training initiatives focusing on adequate assessment of suicidal children and adolescents is likely to diminish physicians' subjective sense of incompetence in this arena. These training initiatives are particularly important as the skills of primary care providers in assessment are tested by the emerging trends in patient care. In time, increasing the number of well-trained providers will serve to improve efforts at prevention and intervention for a growing number of children and adolescents at risk for significant morbidity and mortality from attempted or completed suicide.