Acute intoxication is the most likely culprit for an increased risk of violence or agitation, but personality, psychosis, and cognitive problems can all play a role. A skilled clinician can glean a great deal of information in a short period of time.
Interviewing and assessing patients in the emergency setting is challenging for many reasons. Privacy can be limited, the environment can be loud and chaotic, and there is pressure to make a decision quickly. Patients may be brought to the emergency department (ED) against their will; they may be agitated, distraught, or difficult to engage. A great deal of information needs to be obtained in a short span of time, and many EDs are not set up to manage psychiatric patients. A skilled clinician needs to be able not only to quickly gain rapport and elicit enough information to make a decision, but also to contain the patient.
The psychiatrist may serve as a liaison to the medical staff in the ED. Hospital systems, local mental health laws governing civil commitment, and available community resources all play a role in determining disposition and treatment; thus, clinicians should be familiar with the options available in their system. The decreased availability of inpatient beds and the greater use of EDs by psychiatric patients have led to increased dwell times and crowding in medical EDs.1,2 Given all of these issues, it is important for the ED psychiatrist to have a systematic and rational approach to assessment, but also to maintain a flexible and practical perspective, to adapt to a wide variety of patients and staff.
Safety of the patient and provider
Patients who are brought to the ED for psychiatric evaluation-or even patients who present voluntarily-are usually in crisis. Acute intoxication is the most likely culprit for an increased risk of violence or agitation, but personality, psychosis, and cognitive problems can all play a role. Many EDs do not have separate facilities for psychiatric patients, or patients may have an initial medical complaint that quickly escalates into a behavioral crisis. Medical EDs are full of dangers to the patient and to the staff simply because normal medical equipment is readily available.
CASE VIGNETTE 1
The on-call psychiatrist for a busy city hospital is asked to evaluate an intoxicated man in the ED. When she arrives, the patient is throwing packets of gauze and intravenous equipment at the staff and is screaming incoherently. “You don’t believe I’m going to kill myself?” he yells, as he picks up oxygen tubing and begins to wrap it around his neck.
While in many EDs staff routinely search and remove contraband or weapons from psychiatric patients, in many they do not. Furthermore, it is not routine to search all “medical” patients in most settings. Thus, clinicians should approach each situation with a measure of caution; become familiar with policies at their institution as to where patients are seen and how they are searched; and ask for assistance from other staff, security, or police when indicated. Although all clinicians who work in EDs can benefit from training in crisis management and verbal de-escalation techniques, in many EDs the psychiatrist is the clinician who has the most experience or training in dealing with acute agitation.3 Being able to quickly gain rapport and to function as a team leader in crisis situations are key skills in the effective evaluation of patients in crisis. In the case presented here, the initial task is to obtain the help of other staff and hospital security to engage the patient and move him to a safer setting.
Why is the patient here?
Often it is very clear why a patient has presented to the hospital, because the patient is able to tell his or her own story or is escorted by someone who can provide coherent information. In many cases, whether owing to the patient’s disorganization, intoxication, or desire to withhold or distort information, the reason for presentation is less clear. In some cases, a decision needs to be made very quickly whether there is any justification to hold the patient against his will long enough to do a more thorough evaluation.
CASE VIGNETTE 2
The psychiatrist on call to a community hospital ED is paged to see a patient who was brought in by the police for “psychotic behavior.” By the time the consultant arrives to see the patient, the officers who brought him are gone, and a nursing staff “sitter” is watching the patient. Members of the nursing triage team only know that the patient was brought in from the street, that his vital signs are stable, and that he refused to answer any of their questions. “They said he was acting bizarre,” one nurse states. The patient himself shrugs when asked why he was brought in and then begins to mumble to himself quietly. He appears disheveled and somewhat hostile; he is still in his street clothes and is eyeing the door.
Laws vary by state regarding grounds to hold a patient involuntarily even for an evaluation, but in this case the police took the patient to a hospital, which seems to indicate that there was enough concern about his behavior to justify a thorough evaluation. By gaining rapport with the patient, the clinician may be able to elicit information that helps clarify the situation. Frequently, collateral information is required from police officers or emergency medical services workers, family members, and other ED personnel (Table 1).
Even if patients are disorganized or reluctant, their perspective on why they are in the ED is important in determining their level of insight into the situation. Expressing concern, offering to help alleviate distress, and empathizing with their frustration or displeasure at being in the hospital can help engage the patient in the process of assessment. For patients who are reluctant or oppositional, gaining rapport or a common purpose can be helpful: “I can see you want to leave, but I need to be able to talk to you first. Can we work on this together?”
Are there any acute medical issues?
There can be a great deal of overlap between patients who present for ostensibly psychiatric issues and those who actually have an acute “altered mental status” or delirium. A fluctuating level of consciousness, disorientation, visual hallucinations, and autonomic instability all raise concern for acute delirium or intoxication.4 Other medical issues of immediate concern are listed in Table 2. In general, sensory phenomena or other psychotic symptoms in a patient with no known psychiatric history who is older or younger than the normal age of onset for psychotic illness should raise suspicion of an occult medical cause. Delirium in elderly patients is underdiagnosed in emergency settings and is a risk factor for death in the following 6 months.5
CASE VIGNETTE 3
A call comes in to the psychiatric resident on call in a busy ED: “The patient is medically cleared; he’s only here for psych; we need you to come right away.” During the interview, the patient fluctuates between agitation and drowsiness. He is pale and diaphoretic and states that he is depressed. A chart review shows multiple visits for alcohol intoxication. Concerned about delirium, the resident conducts a brief cognitive screening and finds the patient thinks he is on the Korean front in 1951. He confides to the psychiatrist that he sees tiny soldiers crawling across the ceiling with grenades.
While ED providers may be more familiar with anxious patients or those with personality disorders who over-report their somatic symptoms, many patients with serious mental illness may under-report their medical complaints. Whether owing to frank paranoia, difficulty with self-care, apathy, or disorganization, patients with chronic psychotic illness are less likely to seek medical care, and they are more likely to have comorbid diabetes, hypertension, and metabolic problems than patients without chronic mental illness.6 Patients with substance use disorders may have significant medical comorbidities but may not seek treatment or report the problem because of denial. Because psychiatrists generally talk with patients for a longer time and in greater depth than other providers, they may uncover information that others are not able to obtain.
Taking a complete history
While it is frequently necessary to target the emergency assessment to the most pressing issue, it is also important to be thorough. Making assumptions about why patients are in the ED based on their appearance, apparent socioeconomic status, or stated chief complaint can lead to blind spots. Patients who are eager to leave, who are ashamed of aspects of their history, or who are concerned about legal repercussions of their behavior may obscure information or present a report that is not consciously untruthful but unconsciously skewed to their own viewpoint.
CASE VIGNETTE 4
A 40-year-old woman presents to the ED at midnight escorted by her friend; she reports worsening depression, poor self-care, and insomnia. The patient is in the middle of a divorce, which includes a contentious custody battle, and is employed as an executive. She was alone at home while her children were with her ex-husband for the weekend and says she called her friend because she needed company. She apologizes for wasting everyone’s time and says the friend “over-reacted” by bringing her here. She is in treatment with a psychiatrist but does not want to have her doctor notified about her ED visit, nor does she want the consultant to ask her friend why she was concerned enough to bring her to an ED. She pleads with the consultant to just let her go home so that her ex-husband does not find out that she was in the ED. After the interview, the psychiatrist realizes that he has not even asked her whether she has any substance abuse problems or past suicide attempts, nor has he attempted to verify who is taking care of her children.
The interview should at least attempt to cover all aspects of psychiatric history, while focusing on high-risk areas. The following areas are particularly important to cover in a thorough evaluation.
History of present illness. Focus on acute decompensation, most recent stressors, and changes in treatment.
Psychiatric history. Ask about episodes of suicidality or violence, previous inpatient and outpatient treatment, a longitudinal history of symptoms and illness, and any history of forensic or legal involvement.
Substance use. Staying abreast of local trends in substance abuse can be helpful, but in general patients should be asked directly about substance use or they may not offer this information. Toxicology studies are rarely able to detect all available drugs of abuse.
Social, developmental, and trauma history. A quick review of basic questions can establish a social context that can be very helpful in understanding the patient’s risk factors and supports: Do you have a place to live? How do you support yourself? How far did you get in school? Who raised you? Did you experience abuse or neglect in your childhood? Have you experienced violence or sexual abuse as an adult? While detailed descriptions of abuse history are probably inappropriate for an emergency setting, a trauma- informed understanding of the patient can be helpful in determining diagnosis and framing a therapeutic response. Clinicians should use their judgment about when it is appropriate to inquire about developmental history, but asking about basic social context (housing status and means of support) should be routine and framed in a nonjudgmental manner.
Mental status including cognitive screening. By the end of the interview, the clinician should have enough information to document as complete a mental status as possible. Cognitive screening instruments such as the Montreal Cognitive Assessment or Mini-Mental State Examination may be helpful as well if the patient is able to participate.
What to ask collateral contacts. Health care professionals who have treated the patient should have information about the patient’s history and recent functioning, and they may have an opinion about disposition. Friends and family members who can provide specific information about history, recent behavior changes, and stressors can be helpful. It is unfair to ask family members, “Do you think this patient would kill himself?” because they are not professionals and should not be put in the position of making an incorrect prediction. It is more helpful to ask, for example, “Has the patient been talking about suicide? For how long? Any specific plans? Is the patient more withdrawn or depressed lately?” In other words, it is better to ask about symptoms rather than judgments.
While it is not always possible to make a definitive diagnosis in one interaction with a patient-unless substantial information about their longitudinal history is available-it is important to try to formulate at least a general sense of what the most prominent issues are, because this influences disposition and risk assessment. Reviewing all psychiatric symptom clusters is an important component of a complete interview, particularly if the clinician has never met the patient before. It is not safe to assume, for example, that the patient has no psychotic symptoms if one does not ask. Framing questions in a nonjudgmental manner can help elicit more information. It is probably less effective to ask the patient, “This may sound weird, but do you hear voices?” than to ask, “Some people tell me that they hear people talking to them when no one is there; has that ever happened to you?”
Disposition, documentation, and communication
Local mental health laws regarding involuntary commitment standards heavily influence disposition decisions; however, in general a primary aim of emergency assessment is to determine clinically whether patients can safely continue outside the hospital or are so acutely dangerous to themselves or others that they require inpatient care. Patients with more social support or outpatient resources may be able to avoid admission, but friends and family are not a substitute for a monitored inpatient setting if there is concern about immediate or serious danger. A general principle of interviewing and assessment in the ED is that the goal should be to acquire enough information to make a well-reasoned decision, while recognizing that the portrait may be somewhat incomplete. A question to ask oneself is simply: Do I have enough information to make a decision that I feel comfortable with? While documentation requirements vary by institution, the goal should be to communicate a decision and the reasoning behind it, as well as to delineate a plan for how to proceed. Thinking through the case as if presenting it to another colleague may help the clinician consider the risk assessment more clearly.
Communicating a disposition decision to a patient requires care, particularly when bad news is conveyed. Patients may become agitated if the decision is contrary to their expectations or plans. Having a sense of what the patient wants can prepare the clinician for delivering bad news. Bringing in additional staff, moving the patient to a secure area, or preparing the patient’s property for discharge can all be helpful, depending on the nature of the situation.
Psychiatric assessment in the ED is challenging for many reasons. Providers must adapt to demanding environments and work with patients who are not pleased to see them and staff who may not be familiar or comfortable with psychiatric pathology. It is also a dynamic and exciting setting that provides endless variety and opportunities for learning through experience.
Dr Maloy is Associate Director, Emergency Psychiatry Service, Bellevue Hospital, New York; she is also Clinical Assistant Professor of Psychiatry, New York University School of Medicine. She reports no conflicts of interest concerning the subject matter of this article.
1. Hazlett SB, McCarthy ML, Londner MS, Onyike CU. Epidemiology of adult psychiatric visits to US emergency departments. Acad Emerg Med. 2004;11:193-195.
2. Chang G, Weiss AP, Orav EJ, Rauch SL. Predictors of frequent emergency department use among patients with psychiatric illness. Gen Hosp Psychiatry. 2014;36:716-720.
3. Richmond JS, Berlin JS, Fishkind AB, et al. Verbal de-escalation of the agitated patient: consensus statement of the American Association for Emergency Psychiatry Project BETA De-Escalation Workgroup. West J Emerg Med. 2012;13:17-25.
4. Simon B, Hughes D, Smith S. Delirium. In: Glick RL, Berlin JS, Fishkind AB, et al, eds. Emergency Psychiatry: Practice and Principles. Philadelphia: Lippincott Williams & Wilkins; 2008:chap 20.
5. Han JH, Shintani A, Eden S, et al. Delirium in the emergency department: an independent predictor of death within 6 months. Ann Emerg Med. 2010; 56:244-252.
6. Jones DR, Macias C, Barreira PJ, et al. Prevalence, severity and co-occurrence of chronic physical health problems of persons with serious mental illness. Psychiatr Serv. 2004;55:1250-1257.