As declared by the American Academy of Pediatrics, pediatricians are entering an era of "the new morbidity," characterized, in part, by disrupted personal, family, and social patterns. Adolescent suicide is part of this domain. Pediatricians have traditionally prided themselves on the early recognition of threats to health, yet there exists a major gulf between the pediatrician's practice and adolescent suicide.
Suicide among adolescents has steadily increased since 1950 and is presently their third leading cause of death, following accidents and homicide.1 Figures from the Division of Vital Statistics of the National Center for Health Statistics indicate that teenage suicide rates have tripled since 1950 for males and doubled for females. Rates of completed suicide and attempted suicides are considered minimal estimates because suicidal incidents are often labeled as accidents by families wishing to avoid guilt and stigma.2 Adolescent suicide attempts occur approximately three times as often among females as males, while completed suicides occur three times as often among males. The rate for whites is twice the rate for blacks. For every completed suicide it is estimated that there are 50 to 200 attempts.3 This ratio of completed and attempted suicide is in contrast to the approximately 1:6 ratio seen in adults. One interpretation is that teenage suicide gestures may reflect a plea for help or an attempt to communicate an intolerable situation more than a true death wish.
Although the increased trend of adolescent suicide over the last several decades has been well documented, the difficulty in explaining this trend has led to several theories. Murphy and Solomon, in two separate studies using cohort analysis, showed that between the 1940s and 1970s, the suicide rate for each 5-year cohort increased not only within the cohort as it aged but across each subsequent cohort.4,5 Based on these results they predict an increasing suicide rate with time. Hollinger and Offer suggest an alternative explanation that suicide rates and other violent crimes among teenagers cot relate with the proportion of teenagers in the general population.6 The "baby boom" generation became adolescents during the 1960s and 1970s, causing the proportion of teenagers in the general population to increase. They postulated that increased numbers of adolescents led to increased competition for employment, causing less success, decreased self-esteem, and an increased failure rate. Decreased self-esteem is characteristic of depressed and suicidal adolescents.
The rise in adolescent suicide may also be associated with the decline of family stability, the transition of children to more aduit roles at younger ages, increased social violence, the lessened influence of religion, and the threat of nuclear war. These trends could lead to decreased self-esteem, increased stress, fewer coping skills, diminished support systems, and a loss of hope.
PEDIATRICIAN CONTACT WITH SUICIDAL ADOLESCENTS
Despite the heavy toll of morbidity and mortality resulting from adolescent suicidal behavior, pediatricians often lack involvement with suicidal adolescents. Reasons for this may include the pediatrician's lack of clinical contact with the suicidal adolescent, the unwillingness on the part of adolescents and their families to seek pediatrie help during a crisis and the lack of training among pediatricians in suicide prevention and intervention.
Hodgman and Roberts reported evidence of the low incidence of pediatricians' contact with suicidal adolescents in metropolitan Syracuse and Rochester.7 Between 1971 and 1980, 101 completed adolescent suicides occurred, but only a total of eight were known to the 48 surveyed pediatricians. Based on the estimate that suicide gestures occur 50 to 200 times as often as completed suicides and referring only to the 8 suicides known to the pediatricians, they shouicf have been aware of 400 to 1,600 gestures. They repotted 71. Only 10 of the 48 pediatricians routinely included suicidal symptoms in their history- taking. Pediatricians might potentially have a more active role in preventing or intervening with adolescent suicide if they conducted more comprehensive assessments of suicidal risk. In addition, parents and their adolescents need to be willing to rely on their pediatrician during the suicidal crisis.
Marks surveyed a population of predominantly white suburban teenagers relying on private practitioners for care, of whom 49% saw a pediatrician regularly.8 The majority of the students indicated that they would not choose to go to a private physician for care related to sexuality, substance abuse, or emotional upset, even though 25% were sexually active, 20% used illegal drugs, and 9% reported emotional upset for which they had never seen a health care provider. If pediatricians were perceived as more accessible and helpful to adolescents for their psychosocial needs, perhaps there would be a better utilized support system for adolescents in emotional crisis.
Another reason that pediatricians lack impact on adolescent suicide may be that they are not specifically trained. Talking with suicidal teens creates anxiety. They are angry. There is the fear that they will suddenly act out their suicidal ideation unpredictably. There is the feeling that the pediatrician will not be able to intervene effectively. Anxiety is generated from the feeling of responsibility for their safety. Pediatricians are often not trained to handle the suicidal adolescent, which may also contribute to their not being more involved.
A chart review was conducted by the authors at The Children's Orthopedic Hospital in Seattle in order to determine how involved pediatrie residents were in the medical and psychological evaluation of suicidal patients presenting to the Emergency Room. Charts of all patients presenting to the Emergency Room after a suicide attempt or with suicidal ideation between January 1984 and July 1985 were reviewed. Of the 45 charts, 21 indicated a referral directly to the Child Psychiatry Service and the pediatrie resident was minimally involved, or they were incapacitated by an overdose and the resident managed only their medical condition. In 10 of the remaining 24 cases, chart reviews of residents' notes revealed no indication of past suicide attempts, current suicidal ideation, OT that the resident explored in any depth the patient's current level of functioning. In this setting, it appeared that residents conducted suicide assessments in only about one fourth of the cases seen. The actual number of suicidal adolescents seen by each resident during residency and the amount of instruction or backup support by attending physicians in such cases is unknown. Although it is appropriate that child psychiatry house staff be involved in the emergency room evaluation, pediatrie programs need to emphasize training in suicide interventions and establish a role for the pediatrie resident in the assessment process. In numerous communities, pediatricians may be the only practitioners available to assess or intervene with the suicidal adolescent. Therefore, they should be prepared to do so with training and experience during their residency.
Over the years of practice, the pediatrician achieves a unique position to identify adolescents at risk for suicide. Psychosocial risk factors include: family dysfunction; depression; low self-esteem; feelings of anget, rejection, and isolation; and recent or long-standing life stress and interpersonal losses (especially parent loss before the age of 12 years). Recent studies have emphasized the relationship between adolescent suicide and family instability. In comparing suicidal adolescents to depressed and psychiatric controls, CohenSandler, Berman, and King documented that suicidal adolescents experienced more temporary or permanent separations from parents or grandparents, more rejection from peers, and generally more isolation and rejection in their lives since infancy.9 In a random sample of school children, Pfeffer found that suicidal children were distinguished by their depression, suicidal drive in their mothers, and feared or actual separation from their parents.10 Other studies have found that mothers of suicidal children were more likely to have a chronic illness, usually depression or alcoholism.11,12
These studies suggest that pediatricians may improve their detection of suicide risk in their patients by comprehensive history- taking. A history marked by separation or loss of parents and parent illness, especially alcoholism and depression, constitutes a red flag for suicidal risk in children. Although pediatricians do not routinely include parental depression as part of h is tory- taking, maternal depression may affect child rearing through decreased emotional involvement, decreased communication, decreased affection, and increased hostility displaced onto the child.13 Of course, an additional pathway between parental depression and adolescent suicide is the familial link of biologic depression.
Not all suicidal adolescents are depressed, however, and this erroneous expectation will overlook those suicidal adolescents who lack the dysphoria or depressed mood that makes a diagnosis of depression more obvious. Studies have reported that one third or more of adolescents who are suicidal are not depressed.8 Adolescents presenting with a number of problems, including conduct or adjustment disorders or conflicts with parents, may have suicidal ideation.14 Even normal appearing adolescents report suicidal ideation when asked by a skilled interviewer.9 This underscores the importance of a thorough history with emphasis on psychosocial functioning with any adolescent.
SUICIDE ASSESSMENTA PRACTICAL APPROACH
Any suspicion that a patient might be suicidal should be investigated through a standard interview (Checklist 1). Universally, the literature stresses that eliciting suicidal ideation does not promote further suicidal ideation or action. Rather, it is the first step in intervention. Questioning about ideation should be direct and not implied. Suicidal adolescents are frequently relieved that someone asked the question and is willing to take the time to listen. The pediatrician has the obligation to inform parents about their adolescent's suicidal ideation, but this can be postponed until rapport is established and time is taken to listen to how suicide became an option to the patient. Given that parents are usually involved in the adolescent's crisis or, at the very least, affected by his or her problems, the pediatrician can recommend involving the parents as a part of planning intervention and obtaining the necessary support. While interviewing, parental or judgmental attitudes should be avoided, as well as trying to talk the patient out of a suicide attempt. Emphasis should be placed first on understanding the patient's perspective and second on planning an alternative strategy for addressing the patient's predicament.
If the patient endorses suicidal ideation, the next line of inquiry involves a question about a suicide plan. If the patient has chosen a highly dangerous method, such as a gun, and has access to the method as well as a clear suicidal plan envisioned, a prompt referral to a mental health care provider is indicated. The interview should also include questions about previous attempts, minor drug and alcohol overdoses, and single occupant motor vehicle accidents. Previous attempts significantly increase the severity of suicidal risk. The patient 's intent for suicide should be assessed to examine whether the ideation is an infrequent thought versus a detailed plan that the patient intends to carry out or feels is inevitable. If the patient states an intent for suicide and an unwillingness to accept intervention, involuntary treatment must be enforced. In cases of ideation without intent, every effort should be made to validate the patient 's experience of stress and consequent need for mental health referral. Any suicidal ideation needs to be taken seriously and requires further evaluation by a mental health care provider.
When the adolescent presents to the pediatrician after a suicide attempt, physical and mental status are examined, continued suicidal risk should be assessed, and a decision about appropriate disposition must be made Checklist 2). If the patient is medically unstable or psychotic, then the decision to hospitalize is obvious. In other cases, the decision to admit or follow up as an outpatient should be undertaken in consultation with a psychiatrist, if possible. Although seemingly benign suicidal behaviors such as an overdose of a few aspirin or minor wrist cutting are often minimized as "gestures," all such behaviors represent maladaptive problem-solving and are indications of psychosocial difficulty. If practitioners or parents ignore or treat this behavior as mere misbehavior, the patient may escalate the "cry to be heard" with more lethal attempts.
Repeat potential should be assessed with a review of the circumstances involved in the attempt and an examination of current ideation and intent. Weismann and Wortman developed a risk-rescue rating scale based on the premise that all suicide attempts reflect ambivalence, a desire to die as well as be rescued.15 Although inadequate as a single measure of repeat potential, it provides a framework for examining the current suicide attempt in terms of both how much risk the patient has taken related to the lethality of the method chosen and how much desire to be rescued can be inferred from the factors that might have enhanced the likelihood of rescue. As an example, an overdose of a few aspirin performed at home with family members present represents low risk and high rescue potential. The greater the evidence in favor of a serious desire to die and not to be rescued, the greater the indication for hospitalization- Even if the patient is without current ideation, discharge with an outpatient referral should only be made if the patient and parents are genuinely motivated and committed to the referral plan. If the parents are angry, indifferent, or lacking in support, the patient should be hospitalized until the family situation can be better assessed. A patient who is assessed as a low risk for repeat potential but returned to the home environment without intervention can easily become suicidal again.
In an emergency room study, it was documented that both hospitalized and discharged adolescent patients complied with a mental health referral at a rate of 39%. 16 Patients who were most likely to be compliant were those without any suicidal history. This underscores the importance of mobilizing existing resources to get the patient and family into effective counseling after the first evidence of suicidal tendencies. Compliance with follow-up referral will also be improved if there is strong liaison between the referring physician and mental health care provider.
PUBLIC HEALTH POLICY AND ADOLESCENT SUICIDE
Eisenberg stated that completed suicide rates can be influenced by: 1) limiting access to controllable means of self-destruction; 2) limiting the publicity and sensationalism given to suicide; and 3) improving the skill in diagnosis, referral, and treatment of depressed adolescents who are at particular risk of suicide.2
In the early 1960s, suicide by domestic gas exposure was the leading cause of completed suicide in the United Kingdom. At this same time, natural gas replaced coke gas as a domestic fuel, which diminished carbon monoxide concentrations and toxicity of fuels used in homes. This change coincided with a decrease in the overall suicide rates by one third. In the United States between 1971 and 1976, there was a decrease by one half of the numbers of suicides due to barbiturate overdose. Over the same period, the number of barbiturate prescriptions written by physicians had decreased from 40 to 20 million pet year.
In the United States the leading cause of completed suicide is firearms. Over the past decade, increased suicide rates have been due in large part to increased rates of suicide by firearms. The rate of completed suicide by firearms increased from 50% in 1970 to 57% in 1980. In contrast, suicide by firearms occurs in 6% of suicides in the United Kingdom. In the United States, handguns account for 83% of suicide by firearms according to one estimate. The recommendations of the National Committee on Criminal Justice Standards and Goals in 1973 included the statement that "The manufacture, sale, and private possession of handguns should be prohibited for all persons other than law enforcement and military personnel." High rates of gun ownership are directly correlated with high rates of gun violence. A handgun in the house is six times more likely to kill a family member than a burglar. The states with stricter gun control laws have lower overall rates of suicide. In July 1985, the Senate passed legislation allowing easier transport of handguns across state lines. Stricter hangun controls may decrease the available means to an easy method of selfdestruction and might influence the completed suicide rate.
Publicity and sensationalism given to suicide is another public policy concern. In 1774, Goethe published a romantic novel entitled The Sorrows of Young Werther, about a gifted young man who, after an unsuccessful romance, shot himself. The book was widely believed to have set off a wave of imitative suicides across Western Europe. Waves of imitative suicides following publicity given the original suicide have been called the "Werther Effect." A series of well publicized teen serial suicides has occurred in this country in the past decade. Phillips has shown that suicide rates and motor vehicle fatality rates increase after publicized suicide stories.17 Over-reporting and sensationalizing suicide by the media can affect impressionable adolescents. Articles that stress the normalcy of the child who has committed suicide accentuate the mystery of adolescent suicide and make the adolescent reader feel vulnerable. Restraint in reporting the dramatic aspects of suicide seems indicated. Pediatricians consulting with the media can advise accordingly and instead emphasize the importance of public education about suicide.
How suicide is handled locally at schools may play a role in the contagious aspect of suicide. The practitioner comfortable with adolescents and the suicide issue can provide a valuable resource to schools. Following a suicide by a student, discussions in small groups of students and faculty who were close or directly involved with the student may be advised. It is inappropriate to involve the entire school in an assembly or give the school the day off to attend the funeral, as some schools have done. Anything that overdramatizes the event or overidealizes the student who has committed suicide is counterproductive. Yet, denial and avoidance can also be destructive and the needs of the students and faculty close to the student must be addressed.
Finally, the last item regarding public health policy in teen suicide involves the primary care practitioner and the adolescent at risk for suicide. The gulf between primary care practitioners and the suicidal adolescent needs narrowing. Greater diagnostic skill in detecting those adolescents at risk for suicide is needed. Residency training programs should make this a priority in training. In addition, local pediatrie societies might provide more continuing medical education for those in practice for dealing with this difficult but important cause of teenage mortality.
1. Offer D, Ostrow E, Howard Kl: The Adolexent. A Psychological Self-Profile New York. Basic Books, 1981, pp 123-124.
2. Eisenbern L: The epidemiology of suicide Ln adolescents. Pediarr Ann 1984; 13:47-53.
3. Mclnlire MS, Angle CR, Wilcoff RL, et al: Recurrent adolescent suicide behavior. Pediamcs 1977; 60:605-608.
4. Murphy GE, «feizel RD: Suicide risk by birth cohort in the United Stales. 1949 to 1974. Arch Gen Psychiatry 1980; 37:519-52}.
5. Solomon Ml, Hellon CP: Suicide and age m Alberta. Canada, 1951 to 1977: A cohort analysis. Arch Gen Psyaaany 1980; 37:511-513.
6. Holinger PC, Offer D: Prediction of adolescent suicide; A population model. Am J Psychiairi 1982; 139:302-306.
7. Hodgman CH, Roberts FN: Adolescent suicide and the pediatrician. 1 Pediatr 1962; 101:118-123.
8. Marks A, Malizio], Hoch j, étal: Assessment of health needs and willingness to utilue health care resources of adolescents in a suburban population. } Pedían 1983; 102:456-460.
9. Cohen-Sandler C, Bertnan A, King RA: Follow-up study of hospitalized suicidal children. J Am Acad CUJ PrycWn? 1982; 4:398-403.
10. Pfeffer C: Suicidal fantasies in normal children. J Nerv Ment Dis 1985; 173:7S-84.
11. Friedman R, Com R, Hurt S, et al: Family history of illness in the seriously suicidal adolescent: A life cycle approach. Am J Orthopsychiatry 1984; 54:390-397.
12. SaIk L. Lipsett L. Stumer W, et al: Relationship of maternal and perinatal conditions to eventual adolescent suicide. Lancet 1985; 3:624-627.
13. Sylvester C. Kastner L: Obtaining family psychiatric history in family practice and pediatrie practice. J Fom Pracr 1984; 18:135-141.
14. Carlion GA, Cantwell DP: Unmasking masked depression in children and adolescents. AmJ Prychiany 1980; 137:445-449.
15. Weisman AD, Worden JW: Risk-rescue raring in suicide assessment. Arch Gen Psychiatry 1972; 26:553-560.
16. Litt IF, Cuskev WR, Rudd S: Emeigency room evaluation of the adolescent who attempts suicide: Compliance with follow-up. J Adolsec Health COTÍ 1983, 4.106-108.
17. Phillips DP: The influence of suggestion on suicide: Substantive and theoretical implications of the Wferther Effect. Am iSat Ret' 1974; 39.-340-354.