Clinical and Research News
ED Physicians Often Misdiagnose Psychosis
Psychiatric News
Volume 41 Number 21 page 20-32

Physicians in emergency departments often default to an uncomplicated diagnosis of psychosis, even when the presence of substance abuse should make them more cautious.

Emergency department (ED) physicians frequently misdiagnose substance-induced psychotic disorder, an initial error often compounded by inappropriate follow-up treatment, according to a study in five New York hospitals.

A more detailed review of their cases found that 25 percent of patients in the study who were first diagnosed with a primary psychotic disorder actually had substance-induced psychotic disorder or no psychosis at all, wrote Bella Schanzer, M.D., M.P.H., now mental health director of the AIDS center at Montefiore Medical Center in New York, and four colleagues in the October Psychiatric Services. Schanzer was a research fellow in the Department of Psychiatry at Columbia University when she did this study.

Referral to inpatient hospitalization, use of antipsychotic medications, and referral to mental health or substance-abuse treatment varied according to the diagnosis made in the ED, despite later evidence that in some cases the initial diagnosis was incorrect.

The subjects received care in emergency areas that were solely focused on psychiatric illnesses, and all diagnoses were made by physicians, although not necessarily by psychiatrists, Schanzer told Psychiatric News.

Study results, the researchers said, “[highlight] the challenge of accurately diagnosing a first psychotic episode when it occurs in the context of substance use and underscore the potential for negative consequences if a diagnostic error is made.”


What might account for this pattern?

“Probably a combination of three factors: a complicated clinical presentation, lack of time in the ED, and gaps in physician training,” said Schanzer in an interview.

Patients coming to an emergency department in a psychotic state are hardly articulate reporters of their own medical history, including even whether they have taken drugs recently.

Also, the medical system doesn't give patients adequate time to come off drugs, said Schanzer. DSM-IV calls for a month of observation after the patient ceases substance use, and hospital admissions averaged only 16 days.

Finally, psychiatric training devotes too little time to learning about addiction, she said. “Psychiatrists too often jump to a diagnosis of primary psychosis, rather than substance-induced because that's how they were trained.”

In addition, patients were often not admitted to the same hospital where the ED was located. That limited the chance for feedback from the hospital's more deliberate evaluation to the psychiatrists in the ED.

“We can't even learn from the patients we admitted,” said Schanzer.

Schanzer and her colleagues studied patients from five Manhattan psychiatric emergency departments who were diagnosed with early-phase psychosis and had used drugs or alcohol in the previous 30 days.

They administered the Psychiatric Research Interview for Substance and Mental Disorders, the Positive and Negative Syndrome Scale, and urine toxicology screens at baseline, six months, and 12 months. Master's-level clinicians or a physician interviewed patients at all three assessments. Additional information came from ED records, inpatient hospital records, caregivers, and reports from outpatient follow-up referrals.

A second set of expert diagnosticians looked over that information and made a “best-estimate longitudinal diagnosis.” These diagnoses were divided into substance-induced psychotic disorder, primary psychotic disorder, or no psychotic disorder.

Of the 302 patients in the total sample, 223 (74 percent) were diagnosed in the ED with primary psychotic disorder, 53 (18 percent) with substance-induced psychotic disorder, and 26 (9 percent) with indeterminate symptoms.

The best-estimate diagnoses, however, found that only 195 patients (65 percent) had a primary psychotic condition, 101 (33 percent) had a substance-induced psychosis, and six (2 percent) had no psychotic disorder.

Agreement between the two sets of diagnoses was only fair, wrote the researchers. Fifty-six patients classified as having primary psychotic disorder in the ED (false positives) actually had substance-induced disorder (52) or no psychotic disorder (4), according to the best-estimate standards. On the other hand, of the 53 patients initially diagnosed with substance-induced psychotic disorder, 11 (21 percent) had primary psychotic disorder by the best-estimate procedure.

Schanzer and colleagues expressed surprise at these results. They thought that ED psychiatrists would more likely diagnose a substance-induced psychotic disorder, since all the patients were known to have used alcohol, marijuana, or cocaine during the previous month, and most had positive urine drug screens.

Schanzer suggested that better diagnostic tools and improved training might redress some of this imbalance.


“I don't know if we need a new diagnostic tool,” said Jon Berlin, M.D., medical director of crisis services at Milwaukee County Behavioral Health Division and an assistant clinical professor of psychiatry at the Medical College of Wisconsin. “The DSM-IV criteria are well-written and help in differentiating between substance-induced psychotic disorder and primary psychosis. We might spend more time with residents training them how to better distinguish the two by getting them to probe for symptoms that did or did not occur prior to substance use or are more pronounced than the drug would cause.”

Someone with either type of psychosis needs 24-hour care and antipsychotic medications, but the focus depends on the etiology, added Berlin in an interview. A urine drug screen should be mandatory and, if positive, can prompt the physician to add substance-abuse treatment. “Ideally you pick up the difference right away, but you can still shift gears.”

Diagnoses in the ED in Schanzer's study governed more than simply how patients were treated there. When the ED diagnosed a primary psychotic disorder, patients were more likely to be admitted for an inpatient stay or started on antipsychotic medications, and not to be referred to outpatient treatment for substance abuse. Those diagnosed with substance-induced psychosis were less likely to be hospitalized, to be prescribed antipsychotic medication, or to receive mental health treatment.

About 93 percent of patients thought to have primary psychotic disorder in the ED were hospitalized, and 80 percent were prescribed antipsychotic drugs at discharge, but only 66 percent of those with substance-induced disorder were sent for hospital admission, and 32 percent received antipsychotics at discharge.

Many such patients would benefit from hospitalization, to permit more accurate diagnoses, said Berlin. However, if they are no longer psychotic and can function, they usually are released without a definitive diagnosis for insurance reasons or because inpatient beds are limited. Services available to these patients are poorly integrated in most places.

“People with substance-induced psychosis need antipsychotic medications, but drug abuse [treatment] programs often don't have prescribers for antipsychotics,” he said. “Plus, some psychiatrists don't want to treat substance-abuse patients, and some substance-abuse centers don't want psychiatric patients.”

The cultural gap between physicians and drug abusers should also be addressed in training, he said. “Physicians have a hard time being sympathetic with users,” he said. “They're turned off by their need for instant gratification. It's the opposite of how doctors think.”

“Diagnosing Psychotic Disorders in the Emergency Department in the Context of Substance Use” is posted at<http://psychservices.psychiatryonline.org/cgi/content/full/57/10/1468>.

Interactive Graphics


Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).