Today's Ultrashort Stays Raise Questions About Effectiveness

Twenty-five years ago when Steven Sharfstein, M.D., came to the Sheppard and Enoch Pratt Hospital in Maryland, the average length of stay there was 80 days.

In that time, the patient received a diagnosis and an individualized treatment plan including medication and psychotherapy addressing acute symptoms as well as intrapsychic and psychosocial factors, with the active engagement of family members and the formulation of an aftercare plan. It was a protocol that was not atypical for many other freestanding psychiatric hospitals; at general hospitals, the length of stay was often 20 to 30 days.

But today, the same patient entering almost any hospital in the United States for psychiatric care will likely be out the door in five or six days, in what Sharfstein calls the “ultrashort stay.” Such treatment as can occur in that time focuses on crisis stabilization, relief of the most acute symptoms, and de-escalation of dangerousness.

Between Sharfstein's arrival at Sheppard Pratt in 1986 and today, a perfect storm of factors—managed care, the expansion of insurance coverage for outpatient treatment, and a belief in the efficacy of the least-restrictive therapeutic environment—has reduced psychiatric hospitalization to something that looks less like treatment than a kind of holding action or police function whose purpose is ensuring patient and public safety.

“When I give a talk today, I tell people that hospital treatment is an oxymoron,” Sharfstein said in an interview with Psychiatric News. “We no longer really do treatment. What we do is stabilize, evaluate, and keep the patient as safe as we can.”

In an “Open Forum” essay that appears in the February Psychiatric Services, Ira Glick, M.D., Sharfstein, and Harold Schwartz, M.D., argue that the ultrashort five- to six-day hospital stay may actually subvert the goals of recovery and may contribute to the criminalization of mentally ill individuals by releasing patients to the community with no real recovery-oriented, long-term treatment plan.

The authors offered a model for reform of psychiatric hospitalization that revives the therapeutic function of the hospital and leaves enough time for accurate assessment, real engagement with the patient and family, and formulation of an individualized treatment plan aimed at long-term recovery (see See also: How to Make Hospitalization Useful).

“We don't want a return to long-term hospitalization,” Sharfstein said, “but five or six days is too short. The purpose of the article is to raise the concern that in thinking about health reform, the hospital piece has been left out.

“Hospitalization is an opportunity, not a disaster,” he said. “It's an opportunity to bring high-tech resources to bear on the patient's illness and to come up with a better outpatient plan, one that will help the patient adhere to treatment and be better connected to psychotherapy and psychosocial interventions.”

Where Should Recovery Occur?

Yet the belief that recovery-oriented treatment should happen outside the hospital walls is persistent, and in an editorial accompanying the article, Psychiatric Services Editor Howard Goldman, M.D., Ph.D., argues that the appropriate role of inpatient care in the range of services for mental illness has yet to be resolved.

“Not all patients who need 24-hour supervision or confinement . . . need [hospital-level care],” Goldman wrote. “For some patients, freestanding psychiatric hospitals, affiliated with academic centers, are a more appropriate, lower-cost alternative to the general hospital. For many others, 24-hour alternatives may be more appropriate than the acute care hospital.”

In an interview with Psychiatric News, Marvin Herz, M.D., a longtime advocate for outpatient psychosocial interventions, said that while rigid adherence to a five- to six-day protocol serves no one well, he does not favor an expanded role for inpatient care as outlined by Glick and colleagues.

“I see the hospital as part of a broader system of care that ideally provides a continuum,” Herz told Psychiatric News. “In my opinion the definitive treatment in terms of helping the patient function should be in an ambulatory setting, not in the hospital. [Glick and colleagues] proposed an expanded role for the goals and methods of acute inpatient treatment that will inevitably increase length of stay and costs compared to the current inpatient model of crisis stabilization followed by appropriate ambulatory care.”

Anticipating those arguments, Glick and colleagues argued that lower-cost, low-tech models of care for patients who need 24-hour supervision do not now exist outside the hospital. In the meantime, they wrote, many patients have cognitive problems plus psychotic symptoms that prevent them from being “full partners on the treatment team” and from functioning in an outpatient setting.

Schwartz, psychiatrist in chief at the Institute of Living in Hartford, Conn., emphasized that the intensity of resources that can be brought to bear in an inpatient setting—and the crucial “holding environment” of the hospital—are especially important for influencing the trajectory of illness and recovery following a first-episode psychosis.

“We know that rapid, intense, and early intervention with the most resources possible is critical in the long-term outcome of first-episode psychosis,” he told Psychiatric News. “The evidence is very strong that an inadequately treated first episode predisposes to a second and to a downward trajectory. When we discharge people after five days because they are ‘safe,’ even if they remain psychotic and inadequately prepared for adjustment to life in the community, are we sending these patients on a downward course?”

How Much Hospitalization?

What everyone agrees is that research on the role of hospitalization and the appropriate number of days in a hospital for a given diagnosis is sorely lacking. (Goldman, in his editorial, noted that inpatient research “has disappeared.”)

What research exists tends to favor short-term over longer-term hospitalization. One study published in 1979 in Archives of General Psychiatry by Herz and colleagues looked at 175 newly admitted patients to the Community Service of the New York State Psychiatric Institute. Patients were randomly assigned to standard inpatient care, brief hospitalization followed by the availability of transitional day care, or brief hospitalization.

All patients were offered follow-up outpatient treatment. Initial length of stay was 11 days for both brief-hospitalization groups and 60 days for the standard-care group.

The long-term results indicated little differential effect between treatments, but when differences occurred, they generally favored the brief-care groups, according to the report. Similar results were found in a 1980 British study by Hirsch and colleagues published in the British Journal of Psychiatry.

Glick was principal investigator on studies in the 1970s and 1980s looking at long- and short-term hospitalization for patients with schizophrenia and those with other disorders. Generally, those studies showed that some subgroups of patients benefited from longer hospitalization, but that overall no differences in outcome were detectable when taking into account length of stay and diagnosis.

However, it is noteworthy that in all these studies what was considered “short term” at the time would be a long-term stay today.

“Today, patients admitted to inpatient care are either new cases or chronic patients who get readmitted because they are not complying with treatment,” Glick told Psychiatric News. “What happens with these ultrashort stays is that even the diagnosis is deferred—physicians are reluctant to render a diagnosis so the patient is classified as NOS (not otherwise specified). Then they get a blast of drugs—an antidepressant, antipsychotic, and antianxiety medication—told ‘good luck,’ and get sent out the door.

“What we are arguing for is spending the extra time to make a diagnosis, contact the previous doctor to get a careful history of what has been done or not done in the past, and prescribe an individualized treatment,” Glick said. “As in any other area of medicine, you have to do something active and therapeutic. In the case of psychiatric patients, the treatment team needs to include family, significant others, or a case worker—or it won't work.”

“Inpatient Psychiatric Care in the 21st Century: The Need for Reform” is posted at <http://psychservices.psychiatryonline.org/>. “Brief Hospitalization: Two-Year Follow-Up” by Herz and colleagues is posted at <http://archpsyc.ama-assn.org/cgi/reprint/36/6/701>.