28 - 23 Emergency Psychiatric Medicine

01 - 23.1 Suicide

23.1 Suicide

Emergency Psychiatric Medicine 23.1 Suicide Suicide is derived from the Latin word for “self-murder.” It is a fatal act that represents the person’s wish to die. There is a range, however, between thinking about suicide and acting it out. Some plan for days, weeks, or even years before acting, while others take their lives seemingly on impulse without premeditation. Lost in the definition are intentional misclassifications of the cause of death, accidents of undetermined cause, and so-called chronic suicide (e.g., deaths through alcohol and substance abuse and consciously poor adherence to medical regimens for addiction, obesity, and hypertension). For other terms in the literature on suicide, see Table 23.1-1. Table 23.1-1 Terms Comprising Suicidal Ideation and Behavior In psychiatry, suicide is the primary emergency, with homicide and failure to diagnose an underlying potentially fatal illness representing other, less common psychiatric emergencies. Suicide is to the psychiatrist as cancer is to the internist—the psychiatrist may provide optimal care, yet the patient may die by suicide nonetheless. Thus, suicide is impossible to predict, but numerous clues can be seen. There are also some generally accepted standards of care that facilitate risk reduction, as well as lessen the likelihood of successful litigation, should a patient death occur and a lawsuit be filed. Suicide also needs to be considered in terms of the devastating legacy that it leaves for those who have survived a loved one’s suicide, the impact it has on the treating physician, and the ramification for the clinicians who cared for the decedents. Perhaps the most important

concept regarding suicide is that it is almost always the result of mental illness, usually depression, and is amenable to psychological and pharmacological treatment. EPIDEMIOLOGY There are over 35,000 deaths per year (approximately 100 per day) in the United States attributed to suicide. This is in contrast to approximately 20,000 deaths annually from homicide. It is estimated that there is a 25 to 1 ratio between suicide attempts and completed suicides. Although significant shifts were seen in the suicide death rates for certain subpopulations during the past century (e.g., increase adolescent and decreased elderly rates), the rate remains fairly constant, averaging about 12 per 100,000 through the 20th century and into the first decade of the 21st century. Suicide is currently ranked the tenth overall cause of death in the United States, after heart disease, cancer, chronic lower respiratory diseases, cerebrovascular diseases, accidents, Alzheimer’s disease, diabetes, influenza and pneumonia, and kidney disease. Suicide rates in the United States are the midpoint of the rates for industrialized countries. Internationally, suicide rates range from highs of more than 25 per 100,000 persons in Lithuania, South Korea, Sri Lanka, Russia, Belarus, and Guyana to fewer than 10 per 100,000 in Portugal, the Netherlands, Australia, Spain, South Africa, Italy, Egypt, and others. A state-by-state analysis of suicides in the past decade revealed that New Jersey had the nation’s lowest suicide rate for both sexes and Montana had the nation’s highest rate. Montana and Wyoming had the highest rates for men, and Alaska and Idaho had the highest rates for women. The prime suicide site in the world is the Golden Gate Bridge in San Francisco, with 1,600 suicides committed there since the bridge opened in 1937. Risk Factors Gender Differences. Men commit suicide more than four times as often as women, regardless of age or race, in the United States—despite the fact that women attempt suicide or have suicidal thoughts three times as often as men. Although this disparity remains unclear, it may be related to the methods used. Men are more likely than women to commit suicide using firearms, hanging, or jumping from high places. Women, on the other hand, more commonly take an overdose of psychoactive substances or poison. The use of firearms among women, however, is increasing. In states with gun control laws, the use of firearms has decreased as a method of suicide. Globally, the most common method of suicide is hanging. Age. For all groups, suicide is rare before puberty. Suicide rates increase with age and underscore the significance of the midlife crisis. Among men, suicides peak after age 45; among women, the greatest number of completed suicides occurs after age 55. Rates of 29 per 100,000 population occur in men age 65 or older. Older persons attempt

suicide less often than younger persons, but are more often successful. Although they represent only 13 percent of the total population, older persons account for 16 percent of suicides. The suicide rate, however, is rising among young persons. Suicide is the third leading cause of death in those aged 15 to 24 years, after accidents and homicides. Attempted suicides in this age group number between 1 million and 2 million annually. Most suicides now are among those aged 35 to 64. Race. Suicide rates among white men and women are approximately two to three times as high as for African American men and women across the life cycle. Among young persons who live in inner cities and certain Native American and Alaskan Native groups, suicide rates have greatly exceeded the national rate. Suicide rates among immigrants are higher than those in the native-born population. Religion. Historically, Protestants and Jews in the United States have had higher suicide rates than Catholics. Muslims have much lower rates. The degree of orthodoxy and integration may be a more accurate measure of risk in this category than simple institutional religious affiliation. Marital Status. Marriage lessens the risk of suicide significantly, especially if there are children in the home. Single, never-married persons register an overall rate nearly double that of married persons. Divorce increases suicide risk, with divorced men three times more likely to kill themselves as divorced women. Widows and widowers also have high rates. Suicide occurs more frequently than usual in persons who are socially isolated and have a family history of suicide (attempted or real). Persons who commit so-called anniversary suicides take their lives on the day a member of their family did. Homosexual men and women appear to have higher rates of suicide than heterosexuals. Occupation. The higher the person’s social status, the greater the risk of suicide, but a drop in social status also increases the risk. Work, in general, protects against suicide. Among occupational rankings, professionals, particularly physicians, have traditionally been considered to be at greatest risk. Other high-risk occupations include law enforcement, dentists, artists, mechanics, lawyers, and insurance agents. Suicide is higher among the unemployed than among employed persons. The suicide rates increase during economic recessions and depressions and decrease during times of high employment and during wars. PHYSICIAN SUICIDES. The weight of current evidence supports the conclusion that both male and female physicians in the United States have elevated rates of suicide. It is estimated that approximately 400 physicians commit suicide each year in the United States. United Kingdom and Scandinavian data show that the suicide rate for male physicians is two to three times that found in the general male population of the same age. Female physicians have a higher risk of suicide than other women. In the United States, the annual suicide rate for female physicians is about 41 per 100,000, compared

with 12 per 100,000 among all white women over 25 years of age. Studies show that physicians who commit suicide have a mental disorder, most often depressive disorder, substance dependence, or both. Both male and female physicians commit suicide significantly more often by substance overdoses and less often by firearms than persons in the general population; drug availability and knowledge about toxicity are important factors in physician suicides. Among physicians, psychiatrists are considered to be at greatest risk, followed by ophthalmologists and anesthesiologists, but all specialties are vulnerable. Climate. No significant seasonal correlation with suicide has been found. Suicides increase slightly in spring and fall but, contrary to popular belief, not during December and holiday periods. Physical Health. The relation of physical health and illness to suicide is significant. Previous medical care appears to be a positively correlated risk indicator of suicide: About one third of all persons who commit suicide have had medical attention within 6 months of death, and a physical illness is estimated to be an important contributing factor in about half of all suicides. Factors associated with illness that contribute to both suicides and suicide attempts are loss of mobility, especially when physical activity is important to occupation or recreation; disfigurement, particularly among women; and chronic, intractable pain. Patients on hemodialysis are at high risk. In addition to the direct effects of illness, the secondary effects —for example, disruption of relationships and loss of occupational status—are prognostic factors. Certain drugs can produce depression, which may lead to suicide in some cases. Among these drugs are reserpine (Serpasil), corticosteroids, antihypertensives, and some anticancer agents. Alcohol-related illnesses, such as cirrhosis, are associated with higher suicide rates. Mental Illness. Almost 95 percent of all persons who commit or attempt suicide have a diagnosed mental disorder. Depressive disorders account for 80 percent of this figure, schizophrenia accounts for 10 percent, and dementia or delirium for 5 percent. Among all persons with mental disorders, 25 percent are also alcohol dependent and have dual diagnoses. Persons with delusional depression are at highest risk of suicide. A history of impulsive behavior or violent acts increases the risk of suicide as does previous psychiatric hospitalization for any reason. Among adults who commit suicide, significant differences between young and old exist for both psychiatric diagnoses and antecedent stressors. Diagnoses of substance abuse and antisocial personality disorder occurred most often among suicides in persons less than 30 years of age and diagnoses of mood disorders and cognitive disorders most often among suicides in those age 30 and above. Stressors associated with suicide in those under 30 were separation, rejection, unemployment, and legal troubles; illness stressors most often occurred among suicide victims over age 30. Psychiatric Patients. Psychiatric patients’ risk for suicide is 3 to 12 times that of nonpatients. The degree of risk varies, depending on age, sex, diagnosis, and inpatient

or outpatient status. Male and female psychiatric patients who have at some time been inpatients have five and ten times higher suicide risks, respectively, than their counterparts in the general population. For male and female outpatients who have never been admitted to a hospital for psychiatric treatment, the suicide risks are three and four times greater, respectively, than those of their counterparts in the general population. The higher suicide risk for psychiatric patients who have been inpatients reflects that patients with severe mental disorders tend to be hospitalized—for example, patients with depressive disorder who require electroconvulsive therapy (ECT). The psychiatric diagnosis with greatest risk of suicide in both sexes is a mood disorder. Those in the general population who commit suicide tend to be middle aged or older, but studies increasingly report that psychiatric patients who commit suicide tend to be relatively young. In one study, the mean age of male suicides was 29.5 years and that of women 38.4 years. The relative youthfulness in these suicide cases was partly attributed to two earlyonset, chronic mental disorders—schizophrenia and recurrent major depressive disorder—which account for just over half of these suicides, and so reflects an age and diagnostic pattern found in most studies of psychiatric patient suicides. A small, but significant, percentage of psychiatric patients who commit suicide do so while they are inpatients. Most of these do not kill themselves in the psychiatric ward itself, but on the hospital grounds, while on a pass or weekend leave, or when absent without leave. For both sexes, the suicide risk is highest in the first week of the psychiatric admission; after 3 to 5 weeks, inpatients have the same risk as the general population. Times of staff rotation, particularly of the psychiatric residents, are periods associated with inpatient suicides. Epidemics of inpatient suicides tend to be associated with periods of ideological change on the ward, staff disorganization, and staff demoralization. The period after discharge from the hospital is also a time of increased suicide risk. A follow-up study of 5,000 patients discharged from an Iowa psychiatric hospital showed that in the first 3 months after discharge, the rate of suicide for female patients was 275 times that of all Iowa women; the rate of suicide for male patients was 70 times that of all Iowa men. Studies show that one third or more of depressed patients who commit suicide do so within 6 months of leaving a hospital; presumably they have relapsed. The main risk groups are patients with depressive disorders, schizophrenia, and substance abuse and patients who make repeated visits to the emergency room. Patients, especially those with panic disorder, who frequent emergency services, also have an increased suicide risk. Thus, mental health professionals working in emergency services must be well trained in assessing suicidal risk and making appropriate dispositions. They must also be aware of the need to contact patients at risk who fail to keep follow-up appointments. DEPRESSIVE DISORDERS. Mood disorders are the ones most closely linked to suicide. Approximately 60 to 70 percent of suicide victims suffered a significant depression at the time of their deaths. The lifetime risk of death by suicide among individuals with bipolar disorder is approximately 15 to 20 percent, and suicide is more likely during depressed states rather than manic states. More patients with depressive disorders commit suicide early in the illness rather than later; more depressed men than women commit suicide; and the chance of depressed persons’ killing themselves increases if they are single, separated, divorced, widowed, or recently bereaved. Patients with depressive disorder in the community who commit suicide tend to be middle aged or older. Social isolation enhances suicidal tendencies among depressed patients. This finding is in accord with the data from epidemiological studies showing that persons who commit suicide may be poorly integrated into society. Suicide among

depressed patients is likely at the onset or the end of a depressive episode. As with other psychiatric patients, the months after discharge from a hospital are a time of high risk. Regarding outpatient treatment, most depressed suicidal patients had a history of therapy; however, less than half were receiving psychiatric treatment at the time of suicide. Of those who were in treatment, studies have shown that treatment was less than adequate. For example, most patients who received antidepressants were prescribed subtherapeutic doses of the medication. SCHIZOPHRENIA. The suicide risk is high among patients with schizophrenia: Up to 10 percent die by committing suicide. In the United States, an estimated 4,000 patients with schizophrenia commit suicide each year. The onset of schizophrenia is typically in adolescence or early adulthood, and most of these patients who commit suicide do so during the first few years of their illness; therefore, those patients with schizophrenia who commit suicide are young. Thus, the risk factors for suicide among patients with schizophrenia are young age, male gender, single marital status, a previous suicide attempt, a vulnerability to depressive symptoms, and a recent discharge from a hospital. Having three or four hospitalizations during their 20s probably undermines the social, occupational, and sexual adjustment of possibly suicidal patients with schizophrenia. Consequently, potential suicide victims are likely to be male, unmarried, unemployed, socially isolated, and living alone—perhaps in a single room. After discharge from their last hospitalization, they may experience a new adversity or return to ongoing difficulties. As a result, they become dejected, experience feelings of helplessness and hopelessness, reach a depressed state, and have, and eventually act on, suicidal ideas. Only a small percentage committed suicide because of hallucinated instructions or a need to escape persecutory delusions. Up to 50 percent of suicides among patients with schizophrenia occur during the first few weeks and months after discharge from a hospital; only a minority commit suicide while inpatients. ALCOHOL DEPENDENCE. Up to 15 percent of all alcohol-dependent persons commit suicide. The suicide rate for those who are alcoholic is estimated to be about 270 per 100,000 annually; in the United States, between 7,000 and 13,000 alcohol-dependent persons commit suicide each year. About 80 percent of all alcohol-dependent suicide victims are male, a percentage that largely reflects the sex ratio for alcohol dependence. Alcohol-dependent suicide victims tend to be white, middle aged, unmarried, friendless, socially isolated, and currently drinking. Up to 40 percent have made a previous suicide attempt. Up to 40 percent of all suicides by persons who are alcohol dependent occur within a year of the patient’s last hospitalization; older alcohol-dependent patients are at particular risk during the postdischarge period. Studies show that many alcohol-dependent patients who eventually commit suicide are rated depressed during hospitalization and up to two thirds are assessed as having mood disorder symptoms during the period in which they commit suicide. As many as 50 percent of all alcohol-dependent suicide victims have experienced the loss of a close, affectionate relationship during the previous year. Such interpersonal losses and other types of undesirable life events are probably brought about by the alcohol dependence and contribute to the development of the mood disorder symptoms, which are often present in the weeks and months before the suicide. The largest group of male alcohol-dependent patients is composed of those with an associated antisocial personality disorder. Studies show that such patients are particularly likely to attempt suicide; to abuse other substances; to exhibit

impulsive, aggressive, and criminal behaviors; and to be found among alcohol-dependent suicide victims. OTHER SUBSTANCE DEPENDENCE. Studies in various countries have found an increased suicide risk among those who abuse substances. The suicide rate for persons who are heroin dependent is about 20 times the rate for the general population. Adolescent girls who use intravenous substances also have a high suicide rate. The availability of a lethal amount of substances, intravenous use, associated antisocial personality disorder, a chaotic lifestyle, and impulsivity are some of the factors that predispose substancedependent persons to suicidal behavior, particularly when they are dysphoric, depressed, or intoxicated. PERSONALITY DISORDERS. A high proportion of those who commit suicide have various associated personality difficulties or disorders. Having a personality disorder may be a determinant of suicidal behavior in several ways: by predisposing to major mental disorders such as depressive disorders or alcohol dependence; by leading to difficulties in relationships and social adjustment; by precipitating undesirable life events; by impairing the ability to cope with a mental or physical disorder; and by drawing persons into conflicts with those around them, including family members, physicians, and hospital staff members. An estimated 5 percent of patients with antisocial personality disorder commit suicide. Suicide is three times more common among prisoners than among the general population. More than one third of prisoner suicides have had past psychiatric treatment, and half have made a previous suicide threat or attempt, often in the previous 6 months. ANXIETY DISORDER. Uncompleted suicide attempts are made by almost 20 percent of patients with a panic disorder and social phobia. If depression is an associated feature, however, the risk of completed suicide rises. Previous Suicidal Behavior. A past suicide attempt is perhaps the best indicator that a patient is at increased risk of suicide. Studies show that about 40 percent of depressed patients who commit suicide have made a previous attempt. The risk of a second suicide attempt is highest within 3 months of the first attempt. The relation between a mood disorder, completed suicide, and attempts at suicide is shown in Figure 23.1-1.

FIGURE 23.1-1 Venn diagram summarizing suicide data and its relation to mood disorder and suicide attempts. (Courtesy of Alec Roy, M.D.) Depression is associated with both completed suicide and serious attempts at suicide. The clinical feature most often associated with the seriousness of the intent to die is a diagnosis of a depressive disorder. This is shown by studies that relate the clinical characteristics of suicidal patients with various measures of the medical seriousness of the attempt or of the intent to die. Also, intent-to-die scores correlate significantly with both suicide risk scores and the number and severity of depressive symptoms. Patients having high suicide intent are more often male, older, single or separated, and living alone than those with low intent. In other words, depressed patients who seriously attempt suicide more closely resemble suicide victims than they do suicide attempters. ETIOLOGY Sociological Factors Durkheim’s Theory. The first major contribution to the study of the social and cultural influences on suicide was made at the end of the 19th century by the French sociologist Emile Durkheim. In an attempt to explain statistical patterns, Durkheim divided suicides into three social categories: egoistic, altruistic, and anomic. Egoistic suicide applies to those who are not strongly integrated into any social group. The lack of family integration explains why unmarried persons are more vulnerable to suicide than married ones and why couples with children are the best protected group. Rural communities have more social integration than urban areas and, thus, fewer suicides. Protestantism is a less cohesive religion than Roman Catholicism, and so Protestants have a higher suicide rate than Catholics. Altruistic suicide applies to those susceptible to suicide stemming from their excessive integration into a group, with suicide being the outgrowth of the integration—for example, a Japanese soldier who sacrifices his life in battle. Anomic suicide applies to persons whose integration into society is disturbed so that they cannot follow customary norms of behavior. Anomie explains why a drastic change in economic situation makes persons more vulnerable than they were before their change in fortune. In Durkheim’s

theory, anomie also refers to social instability and a general breakdown of society’s standards and values. Psychological Factors Freud’s Theory. Sigmund Freud offered the first important psychological insight into suicide. He described only one patient who made a suicide attempt, but he saw many depressed patients. In his paper “Mourning and Melancholia,” Freud stated his belief that suicide represents aggression turned inward against an introjected, ambivalently cathected love object. Freud doubted that there would be a suicide without an earlier repressed desire to kill someone else. Menninger’s Theory. Building on Freud’s ideas, Karl Menninger, in Man against Himself, conceived of suicide as inverted homicide because of a patient’s anger toward another person. This retroflexed murder is either turned inward or used as an excuse for punishment. He also described a self-directed death instinct (Freud’s concept of Thanatos) plus three components of hostility in suicide: the wish to kill, the wish to be killed, and the wish to die. Recent Theories. Contemporary suicidologists are not persuaded that a specific psychodynamic or personality structure is associated with suicide. They believe that much can be learned about the psychodynamics of suicidal patients from their fantasies about what would happen and what the consequences would be if they commit suicide. Such fantasies often include wishes for revenge, power, control, or punishment; atonement, sacrifice, or restitution; escape or sleep; rescue, rebirth, reunion with the dead; or a new life. The suicidal patients most likely to act out suicidal fantasies may have lost a love object or received a narcissistic injury, may experience overwhelming affects like rage and guilt, or may identify with a suicide victim. Group dynamics underlie mass suicides such as those at Masada, at Jonestown, and by the Heaven’s Gate cult. Depressed persons may attempt suicide just as they appear to be recovering from their depression. A suicide attempt can cause a long-standing depression to disappear, especially if it fulfills a patient’s need for punishment. Of equal relevance, many suicidal patients use a preoccupation with suicide as a way of fighting off intolerable depression and a sense of hopelessness. A study by Aaron Beck showed that hopelessness was one of the most accurate indicators of long-term suicidal risk. Biological Factors. Diminished central serotonin plays a role in suicidal behavior. A group at the Karolinska Institute in Sweden first noted that low concentrations of the serotonin metabolite 5-hydroxyindoleacetic acid (5-HIAA) in the lumbar cerebrospinal fluid (CSF) were associated with suicidal behavior. This finding has been replicated many times and in different diagnostic groups. Postmortem neurochemical studies have reported modest decreases in serotonin itself or 5-HIAA in either the brainstem or the

frontal cortex of suicide victims. Postmortem receptor studies have reported significant changes in presynaptic and postsynaptic serotonin binding sites in suicide victims. Together, these CSF, neurochemical, and receptor studies support the hypothesis that reduced central serotonin is associated with suicide. Recent studies also report some changes in the noradrenergic system of suicide victims. Low concentrations of 5-HIAA in CSF also predict future suicidal behavior. For example, the Karolinska group examined completed suicide in a sample of 92 depressed patients who had attempted suicide. They found that 8 of the 11 patients who committed suicide within 1 year belonged to the subgroup with below-median concentrations of 5-HIAA in CSF. The suicide risk in that subgroup was 17 percent, compared with 7 percent among those with above-median concentrations of 5-HIAA in CSF (Fig. 23.1-2). Also, the cumulative number of patient-months survived during the first year after attempted suicide was significantly lower in the subgroup with low 5-HIAA concentrations. The Karolinska group concluded that low 5-HIAA concentrations in CSF predict short-range suicide risk in the high-risk group of depressed patients who have attempted suicide. Low 5-HIAA concentrations in CSF have also been demonstrated in adolescents who kill themselves. FIGURE 23.1-2 Cumulative suicide risk during first year after attempted suicide in patients with low versus high cerebrospinal fluid (CSF) concentrations of 5-hydroxyindoleacetic acid (5HIAA). Filled circles indicate CSF 5-HIAA concentrations below the sample median and filled squares indicate concentrations above the sample median (87 nM). (From Nördstrom P, Samuelsson M, Asberg M, Träskman-Bendz L, Aberg-Wistedt A, Nordin C, Bertilsson L. CSF concentrations 5-HIAA predicts suicide risk after attempted suicide. Suicide Life Threat Behav. 1994;24:1, with permission.) Genetic Factors. Suicidal behavior, as with other psychiatric disorders, tends to run in families. In psychiatric patients, a family history of suicide increases the risk of attempted suicide and that of completed suicide in most diagnostic groups. In medicine, the strongest evidence for involvement of genetic factors comes from twin and adoption studies and from molecular genetics. Such studies in suicide are reviewed below. Twin Studies. A landmark study in 1991 investigated 176 twin pairs in which one twin had committed suicide. In nine of these twin pairs, both twins had committed

suicide. Seven of these nine pairs concordant for suicide were found among the 62 monozygotic pairs, whereas two pairs concordant for suicide were found among the 114 dizygotic twin pairs. This twin group difference for concordance for suicide (11.3 vs. 1.8 percent) is statistically significant (P <.01). Another study collected a group of 35 twin pairs in which one twin had committed suicide and the living co-twin was interviewed. Ten of the 26 living monozygotic cotwins had themselves attempted suicide, compared with 0 of the 9 living dizygotic cotwins (P <.04). Although monozygotic and dizygotic twins may have some differing developmental experiences, these results show that monozygotic twin pairs have significantly higher concordance for both suicide and attempted suicide, which suggests that genetic factors may play a role in suicidal behavior. Danish-American Adoption Studies. The strongest evidence suggesting the presence of genetic factors in suicide comes from adoption studies carried out in Denmark. A screening of the registers of causes of death revealed that 57 of 5,483 adoptees in Copenhagen eventually committed suicide. They were matched with adopted controls. Searches of the causes of death revealed that 12 of the 269 biological relatives of these 57 adopted suicide victims had themselves committed suicide, compared with only 2 of the 269 biological relatives of the 57 adopted controls. This is a highly significant difference for suicide between the two groups of relatives. None of the adopting relatives of either the suicide or control group had committed suicide. In a further study of 71 adoptees with mood disorder, adoptee suicide victims with a situational crisis or impulsive suicide attempt or both (particularly) had more biological relatives who had committed suicide than controls had. This led to the suggestion that a genetic factor lowering the threshold for suicidal behavior may lead to an inability to control impulsive behavior. Psychiatric disorders or environmental stress may serve “as potentiating mechanisms which foster or trigger the impulsive behavior, directing it toward a suicidal outcome.” Molecular Genetic Studies. Tryptophan hydroxylase (TPH) is an enzyme involved in the biosynthesis of serotonin. A polymorphism in the human TPH gene has been identified, with two alleles—U and L. Because low concentrations of 5-HIAA in CSF are associated with suicidal behavior, it was hypothesized that such individuals may have alterations in genes controlling serotonin synthesis and metabolism. It was found that impulsive alcoholics, who had low CSF 5-HIAA concentrations, had more LL and UL genotypes. Furthermore, a history of suicide attempts was significantly associated with TPH genotype in all the violent alcoholics; 34 of the 36 violent subjects who attempted suicide had either the UL or LL genotype. Thus, it was concluded that the presence of the L allele was associated with an increased risk of suicide attempts. Also, a history of multiple suicide attempts was found most often in subjects with the LL genotype and to a lesser extent among those with the UL genotype (Fig. 23.1-3). This led to the suggestion that the L allele was associated with repetitive suicidal behavior. The presence of one TPH*L allele may indicate a reduced capacity to hydroxylate

tryptophan to 5-hydroxytryptophan in the synthesis of serotonin, producing low central serotonin turnover and, thus, a low concentration of 5-HIAA in CSF. FIGURE 23.1-3 Relation between tryptophan hydroxylase (TPH) genotype and lifetime history of multiple suicide attempts. For each genotype, the fraction of subjects having each genotype (UU, squares; UL, circles; LL, triangles) is plotted against the number of suicide attempts they have made in their lives. (From Nielsen D, Goldman D, Virkkunen M, Tokola R, Rawlings R, Linnoila M. Suicidality and 5-hydroxyindoleacetic acid concentration associated with a tryptophan hydroxylase polymorphism. Arch Gen Psychiatry. 1994;51:34, with permission.) Parasuicidal Behavior. Parasuicide is a term introduced to describe patients who injure themselves by self-mutilation (e.g., cutting the skin), but who usually do not wish to die. Studies show that about 4 percent of all patients in psychiatric hospitals have cut themselves; the female-to-male ratio is almost 3 to 1. The incidence of self-injury in psychiatric patients is estimated to be more than 50 times that in the general population. Psychiatrists note that so-called cutters have cut themselves over several years. Self-injury is found in about 30 percent of all abusers of oral substances and 10 percent of all intravenous users admitted to substance-treatment units. These patients are usually in their 20s and may be single or married. Most cut delicately, not coarsely, usually in private with a razor blade, knife, broken glass, or mirror. The wrists, arms, thighs, and legs are most commonly cut; the face, breasts, and abdomen are cut infrequently. Most persons who cut themselves claim to experience no pain and give reasons for this behavior such as anger at themselves or others, relief of tension, and the wish to die. Most are classified as having personality disorders and are significantly more introverted, neurotic, and hostile than controls. Alcohol abuse and other substance abuse are common, and most cutters have attempted suicide. Selfmutilation has been viewed as localized self-destruction, with mishandling of aggressive impulses caused by a person’s unconscious wish to punish himself or herself or an

introjected object. PREDICTION Clinicians must assess an individual patient’s risk for suicide on the basis of a clinical examination. The predictive items associated with suicide risk are listed in Table 23.1-2. Suicide is grouped into high-risk–related and low-risk–related factors (Table 23.1-3). High-risk characteristics include more than 45 years of age, male gender, alcohol dependence (the suicide rate is 50 times higher in alcohol-dependent persons than in those who are not alcohol dependent), violent behavior, previous suicidal behavior, and previous psychiatric hospitalization. Table 23.1-2 Variables Enhancing Risk of Suicide among Vulnerable Groups

Table 23.1-3 Evaluation of Suicide Risk It is important that questions about suicidal feelings and behaviors be asked, often directly. Asking depressed patients whether or not they have had thoughts of wanting to kill themselves does not plant the seed of suicide. To the contrary, it may be the first opportunity a patient has had to talk about suicidal ideation that may have been present for some time. The American Psychiatric Association (APA) developed practice guidelines for treating patients with suicidal behaviors, and Table 23.1-4 lists a host of questions that can help the clinician assess suicide risk. Table 23.1-4 Questions about Suicidal Feelings and Behaviors

Treatment Most suicides among psychiatric patients are preventable, because evidence indicates that inadequate assessment or treatment is often associated with suicide. Some patients experience suffering so great and intense, or so chronic and unresponsive to treatment,

that their eventual suicides may be perceived as inevitable. Such patients are relatively uncommon, however (see discussion of inevitable suicide below). Other patients have severe personality disorders, are highly impulsive, and commit suicide spontaneously, often when dysphoric, intoxicated, or both. The evaluation for suicide potential involves a complete psychiatric history; a thorough examination of the patient’s mental state; and an inquiry about depressive symptoms, suicidal thoughts, intents, plans, and attempts. A lack of future plans, giving away personal property, making a will, and having recently experienced a loss all imply increased risk of suicide. The decision to hospitalize a patient depends on diagnosis, depression severity and suicidal ideation, the patient’s and the family’s coping abilities, the patient’s living situation, availability of social support, and the absence or presence of risk factors for suicide. Inpatient versus Outpatient Treatment Whether to hospitalize patients with suicidal ideation is the most important clinical decision to be made. Not all such patients require hospitalization; some can be treated on an outpatient basis. But the absence of a strong social support system, a history of impulsive behavior, and a suicidal plan of action are indications for hospitalization. To decide whether outpatient treatment is feasible, clinicians should use a straightforward clinical approach: Ask patients who are considered suicidal to agree to call when they become uncertain about their ability to control their suicidal impulses. Patients who can make such an agreement with a doctor with whom they have a relationship reaffirm the belief that they have sufficient strength to control such impulses and to seek help. In return for a patient’s commitment, clinicians should be available to the patient 24 hours a day. If a patient who is considered seriously suicidal cannot make the commitment, immediate emergency hospitalization is indicated; both the patient and the patient’s family should be so advised. If, however, the patient is to be treated on an outpatient basis, the therapist should note the patient’s home and work telephone numbers for emergency reference; occasionally, a patient hangs up unexpectedly during a late night call or gives only a name to the answering service. If the patient refuses hospitalization, the family must take the responsibility to be with the patient 24 hours a day. According to Edwin S. Shneidman, a clinician has several practical preventive measures for dealing with a suicidal person: reducing the psychological pain by modifying the patient’s stressful environment, enlisting the aid of the spouse, the employer, or a friend; building realistic support by recognizing that the patient may have a legitimate complaint; and offering alternatives to suicide. Many psychiatrists believe that any patient who has attempted suicide, despite its lethality, should be hospitalized. Although most of these patients voluntarily enter a hospital, the danger to self is one of the few clear-cut indications currently acceptable in all states for involuntary hospitalization. In a hospital, patients can receive antidepressant or antipsychotic medications as indicated; individual therapy, group

therapy, and family therapy are available, and patients receive the hospital’s social support and sense of security. Other therapeutic measures depend on patients’ underlying diagnoses. For example, if alcohol dependence is an associated problem, treatment must be directed toward alleviating that condition. Although patients classified as acutely suicidal may have favorable prognoses, chronically suicidal patients are difficult to treat, and they exhaust the caretakers. Constant observation by special nurses, seclusion, and restraints cannot prevent suicide when a patient is resolute. ECT may be necessary for some severely depressed patients, who may require several treatment courses. Useful measures for the treatment of depressed suicidal inpatients include searching patients and their belongings on arrival in the ward for objects that could be used for suicide and repeating the search at times of exacerbation of the suicidal ideation. Ideally, suicidally depressed inpatients should be treated on a locked ward where the windows are shatterproof, and the patient’s room should be located near the nursing station to maximize observation by the nursing staff. The treatment team must assess how much to restrict the patient and whether to make regular checks or use continuous direct observation. Vigorous treatment with antidepressant or antipsychotic medication should be initiated, depending on the underlying disorder. Some medications (e.g., risperidone [Risperdal]) have both antipsychotic and antidepressant effects and are useful when the patient has signs and symptoms of both psychosis and depression. Supportive psychotherapy by a psychiatrist shows concern and may alleviate some of a patient’s intense suffering. Some patients may be able to accept the idea that they are suffering from a recognized illness and that they will probably make a complete recovery. Patients should be dissuaded from making major life decisions while they are suicidally depressed, because such decisions are often morbidly determined and may be irrevocable. The consequences of such bad decisions can cause further anguish and misery when the patient has recovered. Patients recovering from a suicidal depression are at particular risk. As the depression lifts, patients become energized and, thus, are able to put their suicidal plans into action (paradoxical suicide). A further complication is the activating effect of serotonergic drugs, such as fluoxetine (Prozac), which are effective antidepressants, especially with suicidally depressed patients. Such agents may improve psychomotor withdrawal, thus permitting the patient to act on preexisting suicidal impulses because they have more energy. Sometimes, depressed patients, with or without treatment, suddenly appear to be at peace with themselves because they have reached a secret decision to commit suicide. Clinicians should be especially suspicious of such a dramatic clinical change, which may portend a suicide attempt. Although rare, some patients lie to the psychiatrist about their suicidal intent, thus subverting the most careful clinical assessment. A patient may commit suicide even when in the hospital. According to one survey, about 1 percent of all suicides were committed by patients who were being treated in general medical-surgical or psychiatric hospitals, but the annual suicide rate in

psychiatric hospitals is only 0.003 percent. Table 23.1-5 lists guidelines for selecting a treatment setting for suicidal patients. Table 23.1-5 Guidelines for Selecting a Treatment Setting for Patients at Risk for Suicide or Suicidal Behaviors Legal and Ethical Factors Liability issues stemming from suicides in psychiatric hospitals frequently involve questions about a patient’s rate of deterioration, the presence during hospitalization of clinical signs indicating risk, and psychiatrists’ and staff members’ awareness of, and response to, these clinical signs. In about half of the cases in which suicides occur while patients are on a psychiatric unit, a lawsuit results. Courts expect suicides to occur; do not require zero suicide rates, but do require periodic patient evaluation for suicidal risk, formulation of a treatment plan with a high level of security, and having staff members follow the treatment plan. Currently, suicide and attempted suicide are variously viewed as a felony and a misdemeanor, respectively; in some states, the acts are not considered crimes but

unlawful under common law and statutes. Aiding and abetting a suicide adds another dimension to the legal morass; some court decisions have held that although neither suicide nor attempted suicide is punishable, anyone who assists in the act may be punished. National Strategy for Suicide Prevention In 2001, Surgeon General David Satcher organized the National Strategy for Suicide Prevention, under the auspices of the National Institutes of Health (NIH). The National Strategy for Suicide Prevention has set specific goals and objectives to reduce suicide (Table 23.1-6). Table 23.1-6 Goals to Reduce Suicide The National Strategy for Suicide Prevention creates a framework for suicide prevention for the nation. It is designed to encourage and empower groups and individuals to work together. The stronger and broader the support and collaboration on suicide prevention, the greater the chance of success for this public health initiative. Suicide and suicidal behaviors can be reduced as the general public gains more understanding about (1) the extent to which suicide is a problem, (2) the ways in which it can be prevented, and (3) the roles individuals and groups can play in prevention efforts. SUICIDES INVOLVING OTHER DEATHS Victim-Precipitated Homicide The phenomenon of using others, usually police, to kill oneself is well known to law enforcement personnel. Described by Marvin Wolfgang, the classic situation is exemplified by a person holding up a gas station or all-night store and brandishing a

gun, which he threatens to use on the police when they arrive. They then shoot him, thinking that it is in self-defense. The psychology of such victims is not clear, except that they apparently believe this is the only way they can die. A 25-year-old white divorced father of twin 3-year-old boys had been threatening to his wife, and, consequently, she had an order of restraint placed on him. Nonetheless, one evening, he went to her home, carrying a realistic-looking toy pistol in his pocket “to give her a scare.” She refused to admit him, and, when he began to create a scene, she called the police. When three police officers arrived, he refused to leave, pointed the toy pistol at them and taunted them to shoot him. They drew their revolvers, ordered him to drop his “weapon” (which he did), and restrained him. They took him to a local emergency department, where the nurse’s admission note read: “divorced and angry man threatened others with a toy pistol.” The on-call psychiatrist saw him briefly; the patient denied suicidal or homicidal intent; and the psychiatrist concluded that it was safe to discharge him (as a “situational problem–marital issues”). The following day, he killed himself by using carbon monoxide. Although this was not a case of “completed” victim-precipitated homicide, hospital staff failed to perceive that this represented “attempted” victim-precipitated homicide and was an act of high risk. Noting that he “threatened others with a toy pistol” trivialized the gravity of pointing what appears to be a genuine gun at armed police and telling them to shoot. In effect, he had given up control over this life-threatening situation to the police, and only their self-restraint protected him from being killed that evening. Murder-Suicides Murder-suicides receive a disproportionate amount of attention because they are dramatic and tragic. Unless it is a pact between two truly consenting adults, such events testify to the enormous amount of aggression inherent in many suicides—in addition to the depression. Furthermore, what appears to be a pact is often, in fact, more of a coercion (or flat-out murder) than a true pact among equals. Pacts tend to be made more often by females or elderly couples. Terrorist Suicides. Terrorist-bomber suicides represent a special category of murder-suicides, one in which there is no question of willingness of the victim’s part and in which the victims are unknown to the perpetrators except in some generic, group sense (e.g., Jews, Westerners). Some suicide experts do not classify these as “true” suicides because they differ in so many domains from typical suicides (Table 23.1-7). Table 23.1-7 Differences between Terrorist-Bomber Suicides and Typical Suicides

Although many terrorists are recruited from poorer and less-educated classes, it is surprising that a very large proportion of terrorist bombers are instead from middle-class, well-educated, and possibly less-fundamentalist populations. Because suicide means to take one’s life, it is hard to exclude these terrorist deaths from such a classification. INEVITABLE SUICIDE Not all suicides are preventable; some may be inevitable. In fact, over one third of all completed suicides occur in persons who are receiving treatment for a psychiatric disorder, most commonly depression, bipolar disorder, or schizophrenia. It is not unreasonable to assume that some of those patients received the best care available, but that their suicides could not be prevented. Some clinicians believe that viewing certain suicides as being inevitable may lead to therapeutic nihilism, others feel it may cause both clinicians and patients to lose hope. But inevitable suicide can only be determined a posteriori, after all known facts of a particular suicide have been analyzed and synthesized. And if it cannot be predicted there is no reason for therapeutic nihilism or for treatment efforts to be influenced negatively; indeed viewing some suicides as possibly inevitable may encourage clinicians to increase their therapeutic zeal to prevent or postpone the inevitable from happening. Certain criteria must be met for a particular suicide to be considered inevitable. Most important is a strong genetic history of suicide in one or more family members as well as heavy genetic loading for mental illness. Although a strong genetic diathesis for suicide is associated with completed suicide, it is not, in and of itself, sufficient. Other risk factors must also be present, numerous, and at the extreme end of profound pathology. Among the many risk factors (as described above) are a history of physical, emotional, or sexual abuse, especially during childhood; divorce; unemployment; male gender; recent discharge from a psychiatric hospital; prior suicide attempts; alcoholism or other substance abuse; a history of panic attacks; and the presence of a medical illness. Persistent suicidal thoughts, especially coupled with a plan, are particularly dangerous. As mentioned above, inevitability presumes that these risk factors are numerous, severe, and present in severe degrees. Finally, to consider suicide inevitable the patient must have received the highest standard of treatment and that treatment must have failed. Inevitability assumes, among many other factors, that everything that could have been done was done—and done correctly—yet the patient died.

The case of Ernest Hemingway may be an example of inevitable suicide. Including Ernest, five people committed suicide in the Hemingway family. His father, brother, sister, and granddaughter all killed themselves. In addition, one of his sons suffered from major depression and underwent several courses of ECT during his lifetime. Toward the end of his life, Hemingway had several hospitalizations for depression accompanied by suicide attempts. His last hospitalization was in 1961 at the Mayo Clinic, where he had been admitted in a severely depressed state after yet another suicide attempt. He was delusional (thinking that people were following him with deadly intent), had cognitive difficulties that prevented him from writing creatively, was physically ill with cardiovascular disease, and had been drinking heavily. He was hospitalized for 7 weeks, during which time he was treated with antidepressants, ECT, and psychotherapy. On June 26, 1961, he was discharged from the hospital. As Hemingway was leaving the hospital, in a last conversation, he was purported to have said, “You and I both know what I am going to do to myself one day.” On July 2, 1961, at 7:30 in the morning, 6 days after discharge, Hemingway put a shotgun to his head and pulled the trigger. Hemingway had all of the biopsychosocial determinants of an inevitable suicide. There was heavy genetic loading for suicide, severe psychiatric disorder characterized by persecutory delusions, substance abuse, and other risk factors such as profound suicidal ideation and prior suicide attempts. In addition, Hemingway was the victim of severe childhood trauma, which increased his vulnerability to suicide. As yet, there are insufficient data to predict the inevitability of a particular suicide. The paradigm of inevitability, however, may serve as a stimulus to increase root cause analysis into this phenomenon. The history of medicine is replete with disorders that inexorably led to death but which are now curable and suicide may one day join those ranks. SURVIVING SUICIDE To be a suicide survivor refers to those who have lost a loved one to suicide, not to someone who has attempted suicide but lived. The toll on suicide survivors appears greater than that by other deaths, mainly because the opportunities for guilt are so great. Survivors feel that the loved one intentionally and willfully took his or her life and that if only the survivor had done something differently, the decedent would still be here. Because the decedent cannot tell them otherwise, survivors are at the mercy of their often merciless consciences. What is generally more accurate is that the decedents were not entirely willful but were themselves victims of their own genetic or lifetime experience predispositions to depression and suicide. For children, in particular, the loss of a parent to suicide feels like a shameful abandonment for which the child may blame himself or herself. For parents of children who have killed themselves, their grief is compounded not only by

02 - 23.2 Psychiatric Emergencies in Adults

23.2 Psychiatric Emergencies in Adults

having lost a part of themselves, but also by having failed in what they perceive as their responsibility for the total feelings of their child. To provide mutual support, survivors of suicide groups have appeared throughout the United States, generally led by nonprofessional survivors themselves. Therapists who have lost patients to suicide comprise another survivor group—one too often ignored and unsupported, despite their own considerable suffering and sense of guilt and compounded by the specter of litigation potentially being brought to bear. REFERENCES Allen MH, Abar BW, McCormick M, Barnes DH, Haukoos J, Garmel GM, Boudreaux ED. Screening for suicidal ideation and attempts among emergency department medical patients: Instrument and results from the Psychiatric Emergency Research Collaboration. Suicide Life Threat Behav. 2013;43:313. Betz ME, Miller M, Barber C, Miller I, Sullivan AF, Camargo CA, Boudreaux ED. Lethal means restriction for suicide prevention: beliefs and behaviors of emergency department providers. Depress Anxiety. 2013;30:1013. Betz ME, Sullivan AF, Manton AP, Espinola JA, Miller I, Camargo CA, Boudreaux ED. Knowledge, attitudes, and practices of emergency department providers in the care of suicidal patients. Depress Anxiety. 2013;30:1005. Cha CB, Najmi S, Park JM, Finn CT, Nock MK. Attentional bias toward suicide-related stimuli predicts suicidal behavior. J Abnorm Psychol. 2010;119:616. Figueroa S, Dalack GW. Exploring the impact of suicide on clinicians: A multidisciplinary retreat model. J Psychiatr Pract. 2013;19:72. Kohli MA, Salyakina D, Pfennig A, Lucae S, Horstmann S, Menke A, Kloiber S, Hennings J, Bradley BB, Ressler KJ, Uhr M, Müller-Myhsok B, Holsboer F, Binder EB. Association of genetic variants in the neurotrophic receptor–encoding gene NTRK2 and a lifetime history of suicide attempts in depressed patients. Arch Gen Psychiatry. 2010;67(4):348. Nordentoft M, Mortensen PB, Pedersen CB. Absolute risk of suicide after first hospital contact in mental disorder. Arch Gen Psychiatry. 2011;68(10):1058. Patorno E, Bohn RL, Wahl PM, Avorn J, Patrick AR, Liu J, Schneeweiss S. Anticonvulsant medications and the risk of suicide, attempted suicide, or violent death. JAMA. 2010;303:1401. Sadock BJ. Inevitable suicide: A new paradigm in psychiatry. J Psychiatr Pract. 2012;18:221. Shirey KG. Suicide and HIV. In: Loue S, ed. Mental Health Practitioner’s Guide to HIV/AIDS. New York: Springer Science+Business Media; 2013:405. Simon RI, Hales RE, eds. The American Psychiatric Publishing Textbook of Suicide Assessment and Management. 2nd ed. Washington, DC: American Psychiatric Publishing; 2012. Sudak HS. Suicide. In: Sadock BJ, Sadock VA, Ruiz P, eds. Kaplan & Sadock’s Comprehensive Textbook of Psychiatry. 9th ed. Philadelphia: Lippincott Williams & Wilkins; 2009:2717. Tidemalm D, Runeson B, Waern M, Frisell T, Carlström E, Lichtenstein P, Långström N. Familial clustering of suicide risk: A total population study of 11.4 million individuals. Psychol Med. 2011;41:2527. Vaz JS, Kac G, Nardi AE, Hibbeln JR. Omega-6 fatty acids and greater likelihood of suicide risk and major depression in early pregnancy. J Affect Disord. 2014;152–154:76–82. Vieta E. Suicide risk. In: Managing Bipolar Disorder in Clinical Practice. 3rd ed. New York: Springer Healthcare; 2013:63. 23.2 Psychiatric Emergencies in Adults

A psychiatric emergency is any disturbance in thoughts, feelings, or actions for which immediate therapeutic intervention is necessary. For a variety of reasons—such as the growing incidence of violence, the increased appreciation of the role of medical disease in altered mental status, and the epidemic of alcoholism and other substance use disorders—the number of emergency patients is on the rise. The widening scope of emergency psychiatry goes beyond general psychiatric practice to include such specialized problems as the abuse of substances, children, and spouses; violence in the form of suicide, homicide, and rape; and such social issues as homelessness, aging, competence, and acquired immune deficiency syndrome (AIDS). The emergency psychiatrist must be up to date on medicolegal issues and managed care. This section provides an overview of psychiatric emergencies in general and in adults in particular. Section 23.3 covers psychiatric emergencies in children. TREATMENT SETTINGS Most emergency psychiatric evaluations are done by nonpsychiatrists in a general medical emergency room setting, but specialized psychiatric services are increasingly favored. Regardless of the type of setting, an atmosphere of safety and security must prevail. An adequate number of staff members—including psychiatrists, nurses, aides, and social workers—must be present at all times. Additional personnel to help out in times of overcrowding should be available. Specific responsibilities, such as the use of restraints, should be clearly defined and practiced by the entire emergency team. Clear communication and lines of authority are essential. The organization of the staff into multidisciplinary teams is desirable. Children and young adolescents are best served in a pediatric setting (see Section 23.3). Unless there is a risk of behavioral problems or of their leaving the hospital against advice, they need not be sent to the adult psychiatric emergency service. Immediate access to the medical emergency room and to appropriate diagnostic services is necessary because one third of medical conditions present with psychiatric manifestations. The full spectrum of psychopharmacological options should be available to the psychiatrist. Violence in the emergency service cannot be condoned or tolerated. The code of conduct expected of staff members and patients must be posted and understood from the time of the patient’s arrival in the emergency room. Security is best managed as a clinical issue by the clinical staff, not by law enforcement personnel. Whenever possible, agitated and threatening patients should be sequestered from the nonagitated. Seclusion and restraint rooms should be located close to the nursing station for close observation. The entire staff must understand that patients in physical and emotional distress are fragile and that various expectations and fantasies, often unrealistic, influence their responses to treatment. For example, a man with impaired reality testing who is brought in by the police against his will may not understand that the clinician is interested in helping him. Other patients, influenced by previous unsatisfactory treatment experiences, may be hostile. A high percentage of patients believe that

psychiatrists can read minds or are only interested in admitting patients to lock them away. Such people see little point in openly discussing their problems. Many people have an inaccurate understanding of their rights as patients. All clinical interventions must take those expectations and attitudes into account to minimize the possibility of misunderstanding and consequent problems. EPIDEMIOLOGY Psychiatric emergency rooms are used equally by men and women and more by single than by married persons. About 20 percent of these patients are suicidal, and about 10 percent are violent. The most common diagnoses are mood disorders (including depressive disorders and manic episodes), schizophrenia, and alcohol dependence. About 40 percent of all patients seen in psychiatric emergency rooms require hospitalization. Most visits occur during the night hours, but usage difference is not based on the day of the week or the month of the year. Contrary to popular belief, studies have not found that use of psychiatric emergency rooms increases during the full moon or the Christmas season. EVALUATION The primary goal of an emergency psychiatric evaluation is the timely assessment of the patient in crisis. To that end, the physician must make an initial diagnosis, identify the precipitating factors and immediate needs, and begin treatment or refer the patient to the most appropriate treatment setting. In view of the unpredictable nature of emergency room work, with many patients presenting both physical and emotional complaints, and in view of the limited space and the competition for ancillary services, a pragmatic approach to the patient is required. Sometimes, moving the patient out of the emergency room into the most appropriate diagnostic or treatment setting is best for the patient. Medical emergencies are generally better managed elsewhere in the system. Keeping the number of emergency patients in one place to a minimum reduces the chance of agitation and violence. The standard psychiatric interview—consisting of a history, a mental status examination, and, when appropriate and depending on the rules of the emergency room, a full physical examination and ancillary tests—is the cornerstone of the emergency room evaluation. The emergency room psychiatrist, however, must be ready to introduce modifications as needed. For example, the emergency psychiatrist may have to structure the interview with a rambling manic patient, medicate or restrain an agitated patient, or forgo the usual rules of confidentiality to assess an adolescent’s risk of suicide. In general, any strategy introduced in the emergency room to accomplish the goal of assessing the patient is considered consistent with good clinical practice as long as the rationale for the strategy is documented in the medical record. What constitutes a psychiatric emergency is highly subjective. The emergency room has increasingly come to serve as an admitting area, a holding room, a detoxification center, and a private medical office. Such medical conditions as head

traumas, acute intoxications, withdrawal states, and AIDS encephalopathies may present with acute psychiatric manifestations. The emergency psychiatrists must rapidly assess and distinguish the truly emergency psychiatric patients from those who are less acutely ill and from nonpsychiatric emergencies. A triage system using psychiatrists, nurses, and psychiatric social workers is an efficient and effective way to identify emergency, urgent, and nonurgent patients, who can then be prioritized for care (Fig. 23.2-1). FIGURE 23.2-1 Evaluation and treatment of psychiatric emergencies. In one model, every patient who comes to the emergency room is assessed by a triage nurse on arrival to ascertain the patient’s chief complaint, clinical condition, and vital signs. The psychiatrist then briefly meets with the patient and other significant people involved in the case—family members, emergency medical service technicians, and police—to assign the patient to one of the three categories—emergency, urgent, and nonurgent—or to refer the patient to an appropriate treatment setting, such as the medical emergency room. Having a senior clinician perform that task ensures rapid identification of the most urgent and troublesome cases, an appropriate allocation of resources, and an answer to the most common question heard in the emergency room: “When am I going to see a doctor?” The psychiatrist then assigns clinical responsibility for each patient to the appropriate personnel. As the evaluation often stretches over more than one shift, a careful procedure to transfer responsibility and to pass along information from tour to tour must be built into the system by using visual, oral, and written communications. A request for old records should be made automatically for every patient who is assigned to the emergency room. Each emergency should be judged on its own merits, but information from previous records and from workers in the field and family members can be of crucial importance in assessing patients, especially patients who are psychotic, frightened, or otherwise unable or unwilling to cooperate in giving a good history. A multilingual staff and a hospital language bank that lists bilingual staff members and other translation services should be readily available to the psychiatrist. The use of the patient’s friends or family members as translators is not desirable because of the possibility of unconscious or deliberate denial or distortion of the clinical picture stemming from their involvement with the patient. An initial assessment of the patient’s total biopsychosocial needs is optimal, but the patient’s emergency status, other patients waiting to be seen, and the constraints of the emergency room setting often make such a full assessment a moot point. At a minimum, the emergency evaluation should address the following five questions before any

disposition is decided on: (1) Is it safe for the patient to be in the emergency room? (2) Is the problem organic, functional, or a combination? (3) Is the patient psychotic? (4) Is the patient suicidal or homicidal? (5) To what degree is the patient capable of self-care? Table 23.2-1 provides a general strategy for evaluating patients. Table 23.2-1 General Strategy in Evaluating the Patient Patient Safety Physicians should consider the question of the patient’s safety before evaluating every patient. The answer must address the issues of the emergency room’s physical layout, staffing patterns and communication, and patient population. Psychiatrists must then take stock of themselves: Are they in the proper frame of mind to conduct an evaluation? Do any issues in the case spark countertransference reactions? The selfassessment should go on throughout the evaluation. The physical and emotional safety of the patient takes priority over all other considerations. If verbal interventions fail or are contraindicated, the use of medication or restraints must be considered and, if necessary, ordered. Careful attention to the possible outbreak of agitation or disruptive behavior beyond acceptable limits is often the best insurance against untoward occurrences.

Medical or Psychiatric? The most important question for the emergency psychiatrist to address is whether the problem is medical, psychiatric, or both. Medical conditions—such as diabetes mellitus, thyroid disease, acute intoxications, withdrawal states, AIDS, and head traumas—can present with prominent mental status changes that mimic common psychiatric illnesses (Fig. 23.2-2). Such conditions may be life-threatening if not treated promptly. Generally, the treatment of a medical illness is more definitive and the prognosis is better than for a functional psychiatric disorder. The psychiatrist must consider all casual possibilities. FIGURE 23.2-2 Bellevue Hospital emergency ward: a drug addict brought in after having taken an overdose. (Courtesy of Leonard Freed for Magnum Photos, Inc.) Once patients are labeled psychiatric, their complaints may not be taken seriously by nonmental health professionals, however, and such patients’ conditions may deteriorate, especially if they have a major Axis I syndrome. Because of such factors as deinstitutionalization, homelessness, and chronic alcoholism, the mentally ill are at great risk of tuberculosis, vitamin deficiencies, and other easily overlooked, but easily treated, conditions. Symptoms such as paranoia, internal preoccupation, and acute psychosis can make a routine medical diagnosis exceedingly difficult. Each patient must be assessed for the possibility that an organic illness is combined with an underlying psychiatric illness. A young man who comes to the emergency room intoxicated or in alcohol withdrawal two or three times a month may one day come with a subdural hematoma as a result of a fall. Table 23.2-2 lists features that point to a medical cause

of a mental disorder. Table 23.2-2 Features that Point to a Medical Cause of a Mental Disorder SPECIFIC INTERVIEW SITUATIONS Psychosis Whether the patient is psychotic refers not so much to the diagnosis as to the severity of the patient’s symptoms and the degree of life disruption. The patient’s degree of withdrawal from objective reality, level of affectivity, intellectual functioning, and degree of regression are other important parameters. Impairment in any of those areas may lead to difficulties in conducting an evaluation. Agitated, assaultive behavior or failure to comply with treatment recommendations may also result. A paranoid, hypervigilant patient may misperceive a staff member’s offer of help as an attack and may lash out in self-defense. Command auditory hallucinations may cause a patient to deny symptoms and to throw prescriptions in the garbage immediately after leaving the emergency room. The psychiatrist should be alert to the complications that can arise with patients whose reality testing is impaired and the psychiatrist should modify the approach accordingly. All communication with patients must be straightforward. All clinical interventions should be briefly explained in language the patient can understand. Psychiatrists should not assume that the patient trusts or believes them or even wants their help. Clinicians must be prepared to structure or to terminate an interview to limit the potential for agitation and regression. Depression and Potentially Suicidal Patients The clinician should always ask about suicidal ideas as part of every mental status examination, especially if the patient is depressed. The patient may not realize that such symptoms as waking during the night and increased somatic complaints are related to depressive disorders. The patient should be asked directly, “Are you or have you ever been suicidal?” “Do you want to die?” “Do you feel so bad that you might hurt yourself?” Eight of ten persons who eventually kill themselves give warnings of their intent. If the patient admits to a plan of action, that is a particularly dangerous sign. If a patient who has

been threatening suicide becomes quiet and less agitated than before, that may be an ominous sign. The clinician should be especially concerned with the factors listed in Table 23.2-3. Table 23.2-3 History, Signs, and Symptoms of Suicidal Risk A suicide note, a family history of suicide, or previous suicidal behavior on the part of the patient increases the risk of suicide. Evidence of impulsivity or of pervasive pessimism about the future also places the patient at risk. If the physician decides that the patient is in imminent risk for suicidal behavior, the patient must be hospitalized or otherwise protected. A difficult situation arises when the risk does not seem to be immediate but the potential for suicide is present as long as the patient remains depressed. If the psychiatrist decides not to hospitalize the patient immediately, the doctor should insist that the patient promise to call whenever the suicidal pressure mounts. Violent Patients Patients may be violent for many reasons, and the interview with a violent patient must attempt to ascertain the underlying cause of the violent behavior, because cause determines intervention. The differential diagnosis of violent behavior includes psychoactive substance-induced organic mental disorder, antisocial personality disorder, catatonic schizophrenia, medical infections, cerebral neoplasms, decompensating obsessive-compulsive personality disorder, dissociative disorders, impulse control disorders, sexual disorders, alcohol idiosyncratic intoxication, delusional disorder, paranoid personality disorder, schizophrenia, temporal lobe epilepsy, bipolar disorder, and uncontrollable violence secondary to interpersonal stress. The psychiatric interview must include questions that attempt to sort out the differential for violent behavior and questions directed toward the prediction of violence. The best predictors of violent behavior are (1) excessive alcohol intake; (2) a history of violent acts, with arrests or criminal activity; and (3) a history of childhood abuse. Table 23.2-4 lists some of the most significant factors in assessing and predicting violence. Table 23.2-4 Assessing and Predicting Violent Behavior

Rape and Sexual Abuse Rape is the forceful coercion of an unwilling victim to engage in a sexual act, usually sexual intercourse, although anal intercourse and fellatio can also be acts of rape. As with other acts of violence, rape is a psychiatric emergency that requires immediate, appropriate intervention. Rape victims may suffer sequelae that persist for a lifetime. Rape is a lifethreatening experience in which the victim has almost always been threatened with physical harm, often with a weapon. In addition to rape, other forms of sexual abuse include genital manipulation with foreign objects, infliction of pain, and forced sexual activity. Most rapists are male, and most victims are female. Male rape does occur, however, often in institutions where men are detained (e.g., prisons). Women between the ages of 16 and 24 years are in the highest risk category, but female victims as young as 15 months and as old as 82 years have been raped. More than a third of all rapes are committed by rapists known to the victim, 7 percent by close relatives. A fifth of all rapes involve more than one rapist (gang rape). Typical reactions in both rape and sexual abuse victims include shame, humiliation, anxiety, confusion, and outrage. Many victims wonder whether they are partly responsible and somehow invited the assault. In fact, victim behavior is less important in precipitating a rape than it is in precipitating a homicide or a robbery. Rape and sexual abuse victims are often confused after the assault. Clinicians should be reassuring, supportive, and nonjudgmental. Inform the patient about the availability of medical and legal services and about rape crisis centers that provide multidisciplinary services. If possible, a female clinician should evaluate the patient, because the victim may find it easier to talk with a woman than with a man. The evaluation should take place in private. When rape or sexual abuse has not been acknowledged openly, it is usually because many victims hesitate to discuss the assault and thus avoid the topic. If the patient appears to be anxious when questioned about sexual history and avoids the discussion, it is important to validate the patient’s avoidance. Recognize that the rape victim has undergone an unanticipated, life-threatening stress. It is legally and therapeutically important to take a detailed and complete history of the attack. With the patient’s written consent, collect evidence, such as semen and pubic hair, that may be used to identify the rapist. Take photographs of the evidence, if possible. The medical record may be used as evidence in criminal proceedings; therefore, meticulous objective documentation of all aspects of the evaluation is essential.

TREATMENT OF EMERGENCIES Psychotherapy In an emergency psychiatric intervention, all attempts are made to help patients’ selfesteem. Empathy is critical to healing in a psychiatric emergency. The acquired knowledge of how biogenetic, situational, developmental, and existential forces converge at one point in history to create a psychiatric emergency is tantamount to the maturation of skill in emergency psychiatry. Adjustment disorder in all age groups may result in tantrum-like outbursts of rage. These outbursts are particularly common in marital quarrels, and police are often summoned by neighbors distressed by the sounds of a violent altercation. Such family quarrels should be approached with caution, because they may be complicated by alcohol use and the presence of dangerous weapons. The warring couple frequently turns their combined fury on an unwary outsider. Wounded self-esteem is a major issue, and clinicians must avoid patronizing or contemptuous attitudes and try to communicate an attitude of respect and an authentic peacemaking concern. In family violence, psychiatrists should note the special vulnerability of selected close relatives. A wife or husband may have a curious masochistic attachment to the spouse and can provoke violence by taunting and otherwise undermining a partner’s selfesteem. Such relationships often end in the murder of the provoking partner and sometimes in the suicide of the other partner—the dynamics behind most so-called suicide pacts. As with many suicidal patients, many violent patients require hospitalization and usually accept the offer of inpatient care with a sense of relief. More than one psychotherapist or type of psychotherapy is frequently used in emergency therapy. For example, a 28-year-old man, depressed and suicidal after a colostomy for intractable colitis, whose wife was threatening to leave him because of his irritability and their constant altercations, may be referred to a psychiatrist for supportive psychotherapy and antidepressant medication, to a marital therapist with his wife to improve their marital functioning, and to a colostomy support group to learn ways of coping with a colostomy. Emergency psychiatric clinicians are pragmatic; they use every necessary mode of therapeutic intervention available to resolve the crisis and facilitate value exploration and growth, with less concern than usual about diluting a therapeutic relationship. Emergency therapy emphasizes how various psychiatric modalities act synergistically to enhance recovery. No single approach is appropriate for all persons in similar situations. What does a doctor say to a patient and a family experiencing a psychiatric emergency, such as a suicide attempt or a schizophrenic break? For some, a genetic rationale helps; the information that an illness has a strong biological component relieves some persons. For others, however, this approach underlines a lack of control and increases depression and anxiety. All feel helpless because neither the family nor the patient can alter the behavior to minimize the likelihood of recurrence. Some persons may benefit from an explanation of family or individual dynamics. Others only want someone to listen to

them; in time, they reach their own understanding. In an emergency situation as in any other psychiatric situation, when a clinician does not know what to say, the best approach is to listen. Persons in crisis reveal how much they need support, denial, ventilation, and words to conceptualize the meaning of their crisis and to discover paths to resolution. Pharmacotherapy The major indications for the use of psychotropic medication in an emergency room include violent or assaultive behavior, massive anxiety or panic, and extrapyramidal reactions, such as dystonia and akathisia as adverse effects of psychiatric drugs. Laryngospasm is a rare form of dystonia, and psychiatrists should be prepared to maintain an open airway with intubation if necessary. Persons who are paranoid or in a state of catatonic excitement require tranquilization. Episodic outbursts of violence respond to haloperidol (Haldol), βadrenergic receptor antagonists (β-blockers), carbamazepine (Tegretol), and lithium (Eskalith). If a history suggests a seizure disorder, use clinical studies to confirm the diagnosis and an evaluation to ascertain the cause. If the findings are positive, anticonvulsant therapy is initiated or appropriate surgery is provided (e.g., in the case of a cerebral mass). Conservative measures may suffice for intoxication from drugs of abuse. Sometimes, drugs such as haloperidol (5 to 10 mg every half-hour to an hour) are needed until a patient is stabilized. Benzodiazepines may be used instead of, or in addition to, antipsychotics (to reduce the antipsychotic dosage). When a recreational drug has strong anticholinergic properties, benzodiazepines are more appropriate than antipsychotics. Persons with allergic or aberrant responses to antipsychotics and benzodiazepines are treated with amobarbital (Amytal; mg orally or intramuscularly [IM]), paraldehyde, or diphenhydramine (Benadryl; 50 to 100 mg orally or IM). Violent, struggling patients are subdued most effectively with an appropriate sedative or antipsychotic. Diazepam (Valium; 5 to 10 mg) or lorazepam (Ativan; 2 to 4 mg) may be given slowly intravenously (IV) over 2 minutes. Clinicians must give IV medication with great care to avoid respiratory arrest. Patients who require IM medication can be sedated with haloperidol (5 to 10 mg IM). If the furor is caused by alcohol or is part of a postseizure psychomotor disturbance, the sleep produced by a relatively small amount of an IV medication may go on for hours. On awakening, patients are often entirely alert and rational and typically have complete amnesia about the violent episode. If the disturbance is part of an ongoing psychotic process and returns as soon as the IV medication wears off, continuous medication may be given. It is sometimes better to use small IM or oral doses at half-hour to 1-hour intervals (e.g., haloperidol, 2 to 5 mg, or diazepam, 20 mg) until the patient is controlled than to use large dosages initially, which can result in an overmedicated patient. As the disturbed behavior is brought under control, successively smaller and less frequent doses should be used. During the preliminary treatment, a patient’s blood pressure and other vital signs should be

monitored. Restraints Restraints are used when patients are so dangerous to themselves or others that they pose a severe threat that cannot be controlled in any other way. Patients may be restrained temporarily to receive medication or for long periods if medication cannot be used. Usually, patients in restraints quiet down after a time. On a psychodynamic level, such patients may even welcome the control of their impulses provided by restraints. Table 23.2-5 provides a summary of the use of restraints. Table 23.2-5 Use of Restraints Disposition In some cases, the usual option of admitting or discharging the patient is not considered optimal. Suspected toxic psychoses, brief decompensations in a patient with a personality disorder, and adjustment reactions to traumatic events, for example, may be best managed in an extended-observation setting. Allowing the patient additional time in a secure environment can result in sufficient improvement or clarification of the issues to make traditional inpatient treatment unnecessary. It can also spare the patient the trauma and stigma of a psychiatric admission and can free up bed space for needier

patients. Crisis intervention for victims of rape and other traumas can also be done in an extended-observation setting. When the decision is to admit the patient to the hospital, it is preferable to do so on a voluntary basis. Allowing patients that option gives them a sense of control over their lives and of participation in the treatment decisions. Patients who clearly meet involuntary admission criteria on the basis of dangerousness to themselves or to others cannot leave the hospital without further review and can always be converted to involuntary status if warranted. Because the initial evaluation is often inconclusive, definitive treatment is best deferred until the patient can be further assessed in the inpatient unit or in the outpatient department. When the diagnosis is clear, however, and the patient’s response to previous treatment is known, nothing is gained by delay. For example, a patient with chronic schizophrenia that has decompensated after discontinuing the usual regimen of antipsychotic medication is best served by prompt resumption of treatment. Even if patients feel comfortable coming to the emergency room in times of need, the emergency psychiatrist should always direct or redirect them to the most appropriate treatment setting. Patients in the psychopharmacology clinic who have missed their regular appointments should be given only enough medication to sustain them until they can be seen in the clinic. Feedback to others treating them should be a matter of course. The emergency room is often the gateway to the department of psychiatry or the general hospital. First impressions carry a great deal of weight. The kind of attention and concern shown to patients on arrival in the emergency room strongly affects how they will respond to staff members and treatment recommendations and even their treatment compliance long after they have left the emergency room. Documentation In the interests of good care, respect for patients’ rights, cost control, and medicolegal concerns, documentation has become a central focus for the emergency physician. The medical record should convey a concise picture of the patient, highlighting all pertinent positive and negative findings. Gaps in information and their reason should be mentioned. The names and the telephone numbers of interested parties should be noted. A provisional diagnosis or differential diagnosis must be made. An initial treatment plan or recommendations should clearly follow from the findings of the patient’s history, mental status examination and other diagnostic tests, and the medical evaluation. The writing must be legible. The emergency physician has unusual latitude under the law to perform an adequate initial assessment; however, all interventions and decisions must be thought out, discussed, and documented in the patient’s record. Specific Psychiatric Emergencies Table 23.2-6 outlines common psychiatric emergencies in alphabetical order. Readers are referred to the index and to specific chapters of this textbook for a thorough

discussion of each disorder. Table 23.2-6 Common Psychiatric Emergencies

REFERENCES Agar L. Recognizing neuroleptic malignant syndrome in the emergency department: A case study. Perspect Psychiatr Care. 2010;46:143. Baron DA, Dubin WR, Ning A. Other psychiatric emergencies. In: Sadock BJ, Sadock VA, eds. Kaplan & Sadock’s Comprehensive Textbook of Psychiatry. 9th ed. Vol. 2. Philadelphia: Lippincott Williams & Wilkins; 2009:2732. Bienvenu OJ, Neufeld KJ, Needham DM. Treatment of four psychiatric emergencies in the intensive care unit. Crit Care Med. 2012;40:2662. Bruckner TA, Yonsu K, Chakravarthy B, Brown TT. Voluntary psychiatric emergencies in Los Angeles County after funding of California’s Mental Health Services Act. Psychiatr Serv. 2012;63(8):808. D’Onofrio G, Jauch E, Jagoda A, Allen MH, Anglin D, Barsan WG, Berger RP, Bobrow BJ, Boudreaux ED, Bushnell C, Chan YF, Currier G, Eggly S, Ichord R, Larkin GL, Laskowitz D, Neumar RW, Newman-Toker DE, Quinn J, Shear K, Todd KH, Zatzick D. NIH roundtable on opportunities to advance research on neurologic and psychiatric emergencies. Ann Emerg Med. 2010;56(5):551. Douglass AM, Luo J, Baraff LJ. Emergency medicine and psychiatry agreement on diagnosis and disposition of emergency department patients with behavioral emergencies. Acad Emerg Med. 2011;18:368. Georgieva I, Mulder CL, Wierdsma A. Patients’ preference and experiences of forced medication and seclusion. Psychiatr Q. 2012;83:1. Lin M-T, Burgess JF Jr, Carey K. The association between serious psychological distress and emergency department utilization among young adults in the USA. Soc Psychiatry Psychiatr Epidemiol. 2012;47:939. Polevoi SK, Shim JJ, McCulloch CE, Grimes B, Govindarajan P. Marked reduction in length of stay for patients with psychiatric emergencies after implementation of a comanagement model. Acad Emerg Med. 2013;20:338.

03 - 23.3 Psychiatric Emergencies in Children

23.3 Psychiatric Emergencies in Children

Rodnitzky RL. Movement disorder emergencies. In: Roos KL, ed. Neurology Emergencies. New York: Springer Science+Business Media; 2012:259. Sevransky JE, Bienvenu OJ, Neufeld KJ, Needham DM. Treatment of four psychiatric emergencies in the intensive care unit. Crit Care Med. 2012;40(9):2662. Simpson SA, Joesch JM, West II, Pasic J. Risk for physical restraint or seclusion in the psychiatric emergency service (PES). Gen Hosp Psychiatry. 2014;36(1):113–118. Weiss AP, Chang G, Rauch SL, Smallwood JA, Schechter M, Kosowsky J, Hazen E, Haimovici F, Gitlin DF, Finn CT, Orav EJ. Patient- and practice-related determinants of emergency department length of stay for patients with psychiatric illness. Ann Emerg Med. 2012;60:162. Zun LS, ed. Behavioral Emergencies for the Emergency Physician. New York: Cambridge University Press; 2013. 23.3 Psychiatric Emergencies in Children Few children or adolescents seek psychiatric intervention on their own, even during crisis; thus, most of their emergency evaluations are initiated by parents, relatives, teachers, therapists, physicians, and child protective service workers. Some referrals are for the evaluation of life-threatening situations for the child or for others, such as suicidal behavior, physical abuse, and violent or homicidal behavior. Other urgent but non–life-threatening referrals pertain to children and adolescents with exacerbations of clear-cut serious psychiatric disorders, such as mania, depression, florid psychosis, and school referral. Less diagnostically obvious situations occur when children and adolescents present with a history of a wide range of disruptive, aberrant behaviors, and are accompanied by an overwhelmed, anxious, and distraught adult who perceives the child’s actions as an emergency, despite the absence of life-threatening behavior of an obvious psychiatric disorder. In those cases, the spectrum of contributing factors is not immediately clear, and the emergency psychiatrist must assess the entire family or system involved with the child. Familial stressors and parental discord can contribute to the evolution of a crisis for a child. For example, immediate evaluations are sometimes legitimately indicated for a child caught in the crossfire of feuding parents or in a seemingly irreconcilable conflict between a set of parents and a school, therapist, or protective service worker regarding the needs of the child (Table 23.3-1). Table 23.3-1 Familial Risk Factors An emergency setting is often the site of an initial evaluation of a chronic problem behavior. For example, an identified problem—such as severe tantrums, violence, and destructive behavior in a child—may have been present for months or even years. Yet the initial contact with the mental health system in the emergency room or private

office may be the first opportunity for the child or adolescent to disclose underlying stressors, such as physical or sexual abuse. In view of the integral relation of severe family dysfunction to childhood behavioral disturbance, the emergency psychiatrist must assess familial discord and psychiatric disorder in family members during an urgent evaluation. One way to make the assessment is to interview the child and the individual family members, both alone and together, and to obtain a history from informants outside the family whenever possible. Noncustodial parents, therapists, and teachers may add valuable information regarding the child’s daily functioning. Many families, especially those with mental illness and severe dysfunction, may have little or no inclination to seek psychiatric help on a nonurgent basis; therefore, the emergency evaluation becomes the only way to engage them in an extensive psychiatric treatment program. LIFE-THREATENING EMERGENCIES Suicidal Behavior Assessment. Suicidal behavior is the most common reason for an emergency evaluation in adolescents. Despite the minimal risk for a complete suicide in a child less than 12 years of age, suicidal ideation or behavior in a child of any age must be carefully evaluated, with particular attention to the psychiatric status of the child and the ability of the family or the guardians to provide the appropriate supervision. The assessment must determine the circumstances of the suicidal ideation or behavior, its lethality, and the persistence of the suicidal intention. An evaluation of the family’s sensitivity, supportiveness, and competence must be done to assess their ability to monitor the child’s suicidal potential. Ultimately, during the course of an emergency evaluation, the psychiatrist must decide whether the child may return home to a safe environment and receive outpatient follow-up care or whether hospitalization is necessary. A psychiatric history, a mental status examination, and an assessment of family functioning help establish the general level of risk. Management. When self-injurious behavior has occurred, the adolescent likely requires hospitalization in a pediatric unit for treatment of the injury or for the observation of medical sequelae after a toxic ingestion. If the adolescent is medically clear, the psychiatrist must decide whether the adolescent needs psychiatric admission. If the patient persists in suicidal ideation and shows signs of psychosis, severe depression (including hopelessness), or marked ambivalence about suicide, psychiatric admission is indicated. An adolescent who is taking drugs or alcohol should not be released until an assessment can be done when the patient is in a nonintoxicated state. Patients with high-risk profiles—such as late-adolescent males, especially those with substance abuse and aggressive behavior disorders, and those who have severe depression or who have made prior suicide attempts, particularly with lethal weapons—warrant hospitalization. Young children who have made suicide attempts, even when the attempt had a low

lethality, need psychiatric admission if the family is so chaotic, dysfunctional, and incompetent that follow-up treatment is unlikely. Violent Behavior and Tantrums Assessment. The first task in an emergency evaluation of a violent child or adolescent is to make sure that both the child and the staff members are physically protected so that nobody gets hurt. If the child appears to be calming down in the emergency area, the clinician may indicate to the child that it would be helpful if the child recounted what happened and may ask whether the child feels in sufficient control to do so. If the child agrees and the clinician judges the child to be in good control, the clinician may approach the child with the appropriate backup close at hand. If not, the clinician may either give the child several minutes to calm down before reassessing the situation or, with an adolescent, suggest that a medication may help the adolescent relax. If the adolescent is clearly combative, physical restraint may be necessary before anything else is attempted. Some rageful children and adolescents brought to an emergency setting by overwhelmed families are able to regain control of themselves without the use of physical or pharmacological restraint. Children and adolescents are most likely to calm down if approached calmly in a nonthreatening manner and given a chance to tell their side of the story to a nonjudgmental adult. At this time, the psychiatrist should look for any underlying psychiatric disorder that may be mediating the aggression. The psychiatrist should speak to family members and others who have been witnessing the episode to understand the context in which it occurred and the extent to which the child has been out of control. Management. Prepubertal children, in the absence of major psychiatric illness, rarely require medication to keep them safe, because they are generally small enough to be physically restrained if they begin to hurt themselves or others. It is not immediately necessary to administer medication to a child or an adolescent who was in a rage but is in a calm state when examined. Adolescents and older children who are assaultive, extremely agitated, or overtly self-injurious and who may be difficult to subdue physically may require medication before a dialogue can take place. Children who have a history of repeated, self-limited, severe tantrums may not require admission to a hospital if they are able to calm down during the course of the evaluation. Yet the pattern, no doubt, will reoccur unless ongoing outpatient treatment for the child and the family is arranged. For adolescents who continue to pose a danger to themselves or others during the evaluation period, admission to a hospital is necessary. Fire Setting Assessment. A sense of emergency and panic often surrounds the parents of a child who has set a fire. Parents or teachers often request an emergency evaluation, even for a very young child who has accidentally lit a fire. Many children, during the course of normal development, become interested in fire, but in most cases, a school-aged child who has set a fire has done so accidentally while playing with matches and seeks help to put it out. When a child has a strong interest in playing with matches, the level of

supervision by family members must be clarified, so that no further accidental fires occur. The clinician must distinguish between a child who accidentally or even impulsively sets a single fire and a child who engages in repeated fire setting with premeditation and subsequently leaves the fire without making any attempt to extinguish it. In repeated fire setting, the risk is obviously greater than in a single occurrence, and the psychiatrist must determine whether underlying psychopathology exists in the child or in the family members. The psychiatrist should also evaluate family interactions, because any factors that interfere with effective supervision and communication—such as high levels of marital discord and harsh, punitive parenting styles—can impede appropriate intervention. Fire setting is one of a triad of symptoms—enuresis, cruelty to animals, and fire setting—that were believed, some years ago, to be typical of children with conduct disorders; however, no evidence indicates that the three symptoms are truly linked, although conduct disorder is the most frequent psychiatric disorder that occurs with pathological fire setting. Management. The critical component of management and treatment for fire setters is to prevent further incidents while treating any underlying psychopathology. In general, fire setting alone is not an indication for hospitalization, unless a continued direct threat exists that the patient will set another fire. The parents of children with a pattern of fire setting must be emphatically counseled that the child must not be left alone at home and should never be left to take care of younger siblings without direct adult supervision. Children who exhibit a pattern of concurrent aggressive behaviors and other forms of destructive behavior are likely to have a poor outcome. Outpatient treatment should be arranged for children who repeatedly set fires. Behavioral techniques that involve both the child and the family are helpful in decreasing the risk for further fire setting, as is positive reinforcement for alternate behaviors. Child Abuse: Physical and Sexual Assessment. Physical and sexual abuse occurs in girls and boys of all ages, in all ethnic groups, and at all socioeconomic levels. The abuses vary widely with respect to severity and duration, but any form of continued abuse constitutes an emergency situation for a child. No single psychiatric syndrome is a sine qua non of physical or sexual abuse, but fear, guilt, anxiety, depression, and ambivalence regarding disclosure commonly surround the child who has been abused. Young children who are being sexually abused may exhibit precocious sexual behavior with peers and present a detailed sexual knowledge that reflects exposure beyond their developmental level. Children who endure sexual or physical abuse often display sadistic and aggressive behaviors themselves. Children who are abused in any manner are likely to have been threatened with severe and frightening consequences by the perpetrator if they reveal the situation to anyone. Frequently, an abused child who is victimized by a family member is placed in the irreconcilable position of having either to endure continued abuse silently or to defy the abuser by disclosing the experiences and be responsible for destroying the family and risk being disbelieved or abandoned by the family. In cases of suspected abuse, the child and other family members must be interviewed individually to give each member

a chance to speak privately. If possible, the clinician should observe the child with each parent individually to get a sense of the spontaneity, warmth, fear, anxiety, or other prominent features of the relationships. One observation is generally not sufficient to make a final judgment about the family relationship, however; abused children almost always have mixed emotions toward abusive parents. Physical indicators of sexual abuse in children include sexually transmitted diseases (e.g., gonorrhea); pain, irritation, and itching of the genitalia and the urinary tract; and discomfort while sitting and walking. In many instances of suspected sexual abuse, however, physical evidence is not present. Thus, a careful history is essential. The physician should speak directly about the issues without leading the child in any direction, because already frightened children may be easily influenced to endorse what they think the examiner wants to hear. Furthermore, children who have been abused often retract all or part of what has been disclosed during the course of an interview. The use of anatomically correct dolls in the assessment of sexual abuse can help the child identify body parts and show what has happened, but no conclusive evidence supports sexual play with dolls as a means of validating abuse. Neglect: Failure to Thrive Assessment. In child neglect, a child’s physical, mental, or emotional condition has been impaired because of the inability of a parent or caretaker to provide adequate food, shelter, education, or supervision. Similar to abuse, any form of continued neglect is an emergency situation for the child. Parents who neglect their children range widely and may include parents who are very young and ignorant about the emotional and concrete needs of a child, parents with depression and significant passivity, substanceabusing parents, and parents with a variety of incapacitating mental illnesses. In its extreme form, neglect can contribute to failure to thrive—that is, an infant, usually under 1 year of age, becomes malnourished in the absence of an organic cause (Figs. 23.3-1 and 23.3-2). Failure to thrive typically occurs under circumstances in which adequate nourishment is available, but a disturbance within the relationship between the caretaker and the child results in a child who does not eat sufficiently to grow and develop. A negative pattern may exist between the mother and the child in which the child refuses feedings and the mother feels rejected and eventually withdraws. She may then avoid offering food as frequently as the infant needs it. Observation of the mother and the child together may reveal a nonspontaneous, tense interaction, with withdrawal on both sides, resulting in a seeming apathy in the mother. Both the mother and the child may seem depressed.

FIGURE 23.3-1 A 3-month-old baby suffering from failure to thrive secondary to caloric deprivation. Weight is only 1 ounce over birth weight. (Courtesy of Barbon Schmitt, M.D., Children’s Hospital, Denver, CO.) FIGURE 23.3-2 The same infant as in Figure 23.3-1, 3 weeks later, after hospitalization. (Courtesy of Barbon Schmitt, M.D., Children’s Hospital, Denver, CO.)

A rare form of failure to thrive in children who are at least several years old and are not necessarily malnourished is the syndrome of psychosocial dwarfism. In that syndrome, marked growth retardation and delayed epiphyseal malnutrition accompany a disturbed relationship between the parent and the child, along with bizarre social and eating behaviors in the child. Those behaviors sometimes include eating from garbage cans, drinking toilet water, binging and vomiting, and diminished outward response to pain. Half of the children with the syndrome have decreased growth hormone. Once the children are removed from the troubled environment and placed in another setting, such as a psychiatric hospital with appropriate supervision and guidance regarding meals, the endocrine abnormalities normalize, and the children begin to grow at a more rapid rate. Management. In cases of child neglect, as with physical and sexual abuse, the most important decision to be made during the initial evaluation is whether the child is safe in the home environment. Whenever neglect is suspected, it must be reported to the local child protective service agency. In mild cases, the decision to refer the family for outpatient services, as opposed to hospitalizing the child, depends on the clinician’s conviction that the family is cooperative and willing to be educated and to enter into treatment and that the child is not in danger. Before a neglected child is released from an emergency setting, a follow-up appointment must be made. Education for the family must begin during the evaluation; the family must be told, in a nonthreatening manner, that failure to thrive can become life-threatening, that the entire family needs to monitor the child’s progress, and that they will receive some help in overcoming the many possible obstacles interfering with the child’s emotional and physical wellbeing. Anorexia Nervosa Anorexia nervosa occurs in females about ten times as often as in males. It is characterized by the refusal to maintain body weight, leading to a weight at least 15 percent below the expected weight, by a distorted body image, by a persistent fear of becoming fat, and by the absence of at least three menstrual cycles. The disorder usually begins after puberty, but it has occurred in children of 9 to 10 years of age, in whom expected weight gain does not occur, rather than a loss of 15 percent of body weight. The disorder reaches medical emergency proportions when the weight loss approaches 30 percent of body weight or when metabolic disturbances become severe. Hospitalization then becomes necessary to control the ongoing process of starvation, potential dehydration, and the medical complications of starvation, including electrolyte imbalances, cardiac arrhythmias, and hormonal changes. Acquired Immune Deficiency Syndrome Assessment. Acquired immune deficiency syndrome (AIDS), which is caused by the human immunodeficiency virus (HIV), occurs in neonate through perinatal transmission from an infected mother, in children and adolescents secondary to sexual abuse by an infected person, and in adolescents through intravenous drug abuse with an infected person or through intravenous drug abuse with infected needles and through sexual activities with infected partners. Child and adolescent hemophiliac patients may contract AIDS through tainted blood transfusions. Children and adolescents may present for emergency evaluations at the urging of a family member of a peer; in some

cases, they take the initiative themselves when they are faced with anxiety or panic about high-risk behavior. Early screening of high-risk persons may lead to the treatment of asymptomatic infected patients with such drugs as azidothymidine (AZT) and possibly other new medications that may slow the course of the disease. During the assessment of the risks for HIV infection, an educational process can be initiated with both the patient and the rest of the family so that an adolescent who is not infected, but exhibits high-risk behavior, can be counseled about that behavior and about safe-sex practices. In children, the brain is often a primary site for HIV infection; encephalitis, decreased brain development, and such neuropsychiatric symptoms as impairment in memory, concentration, and attention span may be present before the diagnosis is made. The virus can be present in the cerebrospinal fluid before it shows up in the bloodstream. Changes in cognitive function, frontal lobe disinhibition, social withdrawal, slowed information processing, and apathy constitute some common symptoms of the AIDS dementia complex. Organic mood disorders, organic personality disorder, and frank psychosis can also occur in patients infected with HIV. URGENT NON–LIFE-THREATENING SITUATIONS School Refusal Assessment. Refusal to go to school may occur in a young child who is first entering school or in an older child or adolescent who is making a transition into a new grade or school, or it may emerge in a vulnerable child without an obvious external stressor. In any case, school refusal requires immediate intervention, because the longer the dysfunctional pattern continues, the more difficult it is to interrupt. School refusal is generally associated with separation anxiety, in which the child’s distress is related to the consequences of being separated from the parent, so the child resists going to school. School refusal can also occur in children with school phobia, in which the fear and the distress are targeted on the school itself. In either case, a serious disruption of the child’s life occurs. Although mild separation anxiety is universal, particularly among very young children who are first facing school, treatment is required when a child actually cannot attend school. Severe psychopathology, including anxiety and depressive disorders, is often present when school refusal occurs for the first time in an adolescent. Children with separation anxiety disorder typically present extreme worries that catastrophic events will befall their mothers, attachment, or themselves as a result of the separation. Children with separation anxiety disorder may also exhibit many other fears and symptoms of depression, including such somatic complaints such as headaches, stomachaches, and nausea. Severe tantrums and desperate pleas may ensue when preoccupation that a parent will be harmed during the separation is frequently verbalized; in adolescents, the stated reasons for refusing to go to school are often physical complaints. As part of an urgent assessment, the psychiatrist must ascertain the duration of the patient’s absence from school and must assess the parents’ ability to participate in a treatment plan that will undoubtedly involve firm parental guidelines to ensure the child’s return to school. The parents of a child with separation anxiety disorder often exhibit excessive separation anxiety or other anxiety disorders themselves, thereby compounding the child’s problem. When the parents are unable to participate in a treatment program from home, hospitalization should be considered. Management. When school refusal caused by separation anxiety is identified during an emergency evaluation, the underlying disorder can be explained to the family, and an intervention can be started immediately. In severe cases, however, a multidimensional, long-term family-oriented treatment plan is necessary. Whenever

possible, a separation-anxious child should be brought back to school the next school day, despite the distress, and a contact person within the school (counselor, guidance counselor, or teacher) should be involved to help the child stay in school while praising the child for tolerating the school situation. When school refusal has been going on for months or years or when the family members are unable to cooperate, a treatment program to move the child back to school from the hospital should be considered. When the child’s anxiety is not diminished by behavioral methods alone, tricyclic antidepressants, such as imipramine (Tofranil), are helpful. Medication is generally prescribed not at the initial evaluation but after a behavioral intervention has been tried. Munchausen Syndrome by Proxy Assessment. Munchausen syndrome by proxy, essentially, is a form of child abuse in which a parent, usually the mother, or a caretaker repeatedly fabricates or actually inflicts injury or illness in a child for whom medical intervention is then sought, often in an emergency setting. Although it is a rare scenario, mothers who inflict injury often have some prior knowledge of medicine, leading to sophisticated symptoms; the mothers sometimes engage in inappropriate camaraderie with the medical staff regarding the treatment of the child. Careful observation may reveal that the mothers often do not exhibit appropriate signs of distress on hearing the details of the child’s medical symptoms. Prototypically, such mothers tend to present themselves as highly accomplished professionals in ways that seem inflated or blatantly untrue. The illnesses appearing in the child can involve any organ system, but certain symptoms are commonly presented: bleeding from one or may sites, including the gastrointestinal (GI) tract, the genitourinary system, and the respiratory system; seizures; and central nervous system (CNS) depression. At times, the illness is simulated, rather than actually inflicted. OTHER CHILDHOOD DISTURBANCES Posttraumatic Stress Disorder Children who have been subjected to a severe catastrophic or traumatic event may present for a prompt evaluation because they have extreme fears of the specific trauma occurring again or sudden discomfort with familiar places, people, or situations that previously did not evoke anxiety. Within weeks of a traumatic event, a child may re-create the event in play, in stories, and in dreams that directly replay the terrifying situation. A sense of reliving the experience may occur, including hallucinations and flashback (dissociative) experiences, and intrusive memories of the event come and go. Many traumatized children, over time, go on to reproduce parts of the event through their own victimization behaviors toward others, without being aware that those behaviors reflect their own traumatic experiences. Dissociative Disorders Dissociative states—including the extreme form, multiple personality disorder—are believed most likely to occur in children who have been subjected to severe and repetitive physical, sexual, or emotional abuse. Children with dissociative symptoms may be referred for evaluation because family members or teachers observe that the children sometimes seem

to be spaced out or distracted or act like different persons. Dissociative states are occasionally identified during the evaluation of violent and aggressive behavior, particularly in patients who truly do not remember chunks of their own behavior. When a child who dissociates is violent or self-destructive or endangers others, hospitalization is necessary. A variety of psychotherapy methods have been used in the complex treatment of children with dissociative disorders, including play techniques and, in some cases, hypnosis. REFERENCES Ballard ED, Stanley IH, Horowitz LM, Cannon EA, Pao M, Bridge JA. Asking youth questions about suicide risk in the pediatric emergency department: Results from a qualitative analysis of patient opinions. Clin Pediatr Emerg Med. 2013;14:20. Cashman M, Pasic J. Pediatric psychiatric disorders in the emergency department. In: Zun LS, ed. Behavioral Emergencies for the Emergency Physician. New York: Cambridge University Press; 2013:211. Ceballos-Osorio J, Hong-McAtee I. Failure to thrive in a neonate: A life-threatening diagnosis to consider. J Pediatr Heath Care. 2013;27:56. Dolan MA, Fein JA. Pediatric and adolescent mental health emergencies in the emergency medical services system. Pediatrics. 2011;127(5):e1356. Flaherty LT. Models of psychiatric consultation to schools. In: Weist MD, Lever NA, Bradshaw CP, Owens JS, eds. Handbook of School Mental Health: Issues in Clinical Child Psychology. 2nd ed. New York: Springer Science+Business Media; 2014:283. Frosch E, Kelly P. Issues in pediatric psychiatric emergency care. Emerg Psychiatry. 2013:193. Gilbert SB. Beyond acting out: managing pediatric psychiatric emergencies in the emergency department. Adv Emerg Nurs J. 2012;34:147. Ginnis KB, White EM, Ross AM, Wharff EA. Family-based crisis intervention in the emergency department: A new model of care. J Child Fam Stud. 2013;10.1007/s10826-013-9823-1. Grupp-Phelan J, Delgado SV. Management of the suicidal pediatric patient: An emergency medicine problem. Clin Pediatr Emerg Med. 2013;14:12. Hamm MP, Osmond M, Curran J, Scott S, Ali S, Hartling L, Gokiert R, Cappelli M, Hnatko G, Newton AS. A systematic review of crisis interventions used in the emergency department: recommendations for pediatric care and research. Pediatr Emerg Care. 2010;26:952. Jaffee SR. Family violence and parent psychopathology: Implications for children’s socioemotional development and resilience. In: Goldstein S, Brooks RB, eds. Handbook of Resilience in Children. 2nd ed. New York: Springer Science+Business Media; 2013:127. Kalb LG, Stuart EA, Freedman B, Zablotsky B, Vasa R. Psychiatric-related emergency department visits among children with an autism spectrum disorder. Pediatr Emerg Care. 2012;28:1269. Magallón-Neri EM, Canalda G, De la Fuente JE, Forns M, García R, González E, Castro-Fornieles J. The influence of personality disorders on the use of mental health services in adolescents with psychiatric disorders. Compr Psychiatry. 2012;53(5):509. Maunder RG, Halpern J, Schwartz B, Gurevich M. Symptoms and responses to critical incidents in paramedics who have experienced childhood abuse and neglect. Emerg Med J. 2012;29:222. Miller AB, Esposito-Smythers C, Weismoore JT, Renshaw KD. The relation between child maltreatment and adolescent suicidal behavior: A systematic review and critical examination of the literature. Clin Child Fam Psychol Rev.

2013;16:146. Ougrin D, Tranah T, Leigh E, Taylor L, Asarnow JR. Practitioner review: Self-harm in adolescents. J Child Psychol Psychiatry. 2012;53:337. Reading R. Weight faltering and failure to thrive in infancy and early childhood. Child Care Health Devel. 2013;39:151. Tenenbein M. Urine Drug Screens in Pediatric Psychiatric Patients. Pediatr Emerg Care. 2014;30(2):136–137.