Clinical and Research News
Psychiatrists Often Reluctant To Encourage PADs
Psychiatric News
Volume 40 Number 14 page 17-29

Psychiatric advance directives can help patients express treatment preferences and guide physicians in moments of crisis, but both groups need more help understanding and using them, said a panel of researchers at APA's 2005 annual meeting in May.

A psychiatric advance directive (PAD) is an alternative to coercion, a document that allows competent persons to state their psychiatric treatment preferences for times when they are not competent, said Marian Butterfield, M.D., M.P.H., an associate professor of psychiatry at Duke University. The directive covers hospital admission, medication administration, use of other interventions such as electroconvulsive therapy, and may authorize a surrogate agent to act on the patient's behalf. Twenty states now have laws authorizing some form of PAD.

"A psychiatric advance directive is completed before a mental health crisis in which the patient loses the capacity to make decisions," said Butterfield. "The PAD is then consulted by treating physicians when the patient is in an impaired condition."

PADs are intended to help patients by giving them a voice in their treatment and support in times of crisis. Physicians, too, could benefit by better coordinating care with family and primary care doctors and targeting treatment more exactly, knowing what has or hasn't work in the past.

That's the theory. In practice, complications arise in creating and implementing PADs. Patients may need help in understanding their value and in filling them out. Some physicians fear that a PAD will restrict their treatment choices. Law-enforcement and hospital authorities may be unfamiliar with them and ignore them in the very times of crisis when they are intended to help.

PADs bear similarities and differences with general medical advance directives, said former APA President Paul Appelbaum, M.D., chair of the Department of Psychiatry at the University of Massachusetts and chair of APA's Council on Psychiatry and Law. Both types of documents meet with general approval by the public, but few individuals actually have them. Neither can cover every contingency, so that designating a health care proxy may be just as important. They are also often unavailable when they are most needed, and it may be hard to get officials or medical personnel to even read them, much less follow their directions, he noted.


Unlike general health directives, the question of patient competence hangs over a PAD, although there is a presumption that the patient was competent when signing it.

Physicians also fear that the content of a PAD will be "crazy" and that patients will use them to refuse treatment. "In fact, patients are doing a good job of filling them out, especially if facilitators are involved," said Appelbaum.

Picking the right time to fill out a PAD is another key issue. Following passage of the federal Patient Self-Determination Act of 1990, hospitals began asking patients to complete advanced directives at admission, the worst time for any patient but especially for psychiatric patients, said Appelbaum. Completing a directive at the first outpatient visit or upon hospital discharge would be better.


A team from Duke University School of Medicine studied 1,000 outpatients in mental health clinics in Chicago, Tampa, San Francisco, Durham, N.C., and Worcester, Mass., and found that few (4 percent to 13 percent) had PADs but that 65 percent to 77 percent said they wanted one, once the idea was explained to them, said study leader Jeffrey Swanson, Ph.D., an associate professor of psychiatry at Duke. The study will be published shortly, he said.

"There's a large, latent demand," said Swanson. "The PAD gives patients some sense of empowerment at very vulnerable moments in their lives."

Patients interested in PADs tended to be female, nonwhite, suicidal, with a low sense of mastery, and a history of feeling pressured to take medications and an arrest in the prior six months, according to the Duke study.

Patients in the study who had completed a directive were more likely to have high insight (odds ratio, 1.9), a representative payee managing their money for them (3.0), been transported by the police to treatment (2.3), and feel pressure from clinicians and others to keep their appointments at community mental health clinics (1.8).

Eric Elbogen, Ph.D., an assistant professor of psychiatry at Duke, has studied 121 advanced-instruction documents produced by patients randomized from public-sector settings. The patients were assigned a bachelor's-level facilitator who helped explain their options in completing a PAD.

Most of the patients made clear the options they preferred, Elbogen noted.

For instance, 98 percent alerted doctors to at least one crisis symptom." I want to hurt myself because the television is talking to me," said one. "I have racing thoughts and become aggressive, especially in the emergency room," wrote another. This kind of prior knowledge of violent history can help hospital staff and reduce risk when the patient appears again, said Elbogen.

Seventy-five percent of the patients said they wanted to be treated with respect and listened to if brought to the hospital. Eighty-eight percent mentioned at least one hospital to which they were willing to go, while 62 percent documented a refusal to go to a specific hospital, usually giving reasons for doing so.


PADs may help guide treatment as well. All subjects mentioned at least one factor likely to cause a relapse. Also, 77 percent rejected at least one medication, but 94 percent gave advanced consent for treatment with at least one drug. "No one refused all medications, but no one liked Haldol," said Elbogen.

Half of the patients studied instructed staff on how to avoid use of restraint and seclusion, and three-quarters listed side effects they experienced on particular medications. Sixty-two percent refused to have electroconvulsive therapy. All gave emergency contacts, usually family members.

The facilitators helped focus the documents by asking open-ended questions, Elbogen said.

"There were few medically inappropriate treatment requests in the directives, and they included much relevant information that would be valuable to clinicians," said Elbogen. The most unrealistic response was," I need a cigarette in intake because I can't calm down without a smoke."

The authority of the PAD, however, is not absolute, said Elbogen. It does become part of the patient's chart, and if one section of the PAD can't be carried out, the rest is not invalidated. Furthermore, involuntary commitment overrides the PAD.

Clinicians have little experience with PADs so far, and they are ambivalent about them, said Marvin Swartz, M.D., professor and head of social and community psychiatry at Duke. Swartz and Elbogen are co-investigators with Swanson on the Duke study.

"About two-thirds of psychiatrists say they would honor a PAD, although many think that patients would use it to inscribe treatment refusal," he said, drawing from his random sample of psychiatrists in North Carolina, where PADs have been authorized since 1997.

Responses from the psychiatrists surveyed varied depending on hypothetical circumstances, said Swartz. Asked if they would follow a PAD if a patient were concerned about drug side effects, 70 percent of the doctors said yes. But if the PAD as written sounded "psychotic," and the patient had a history of violence, only 40 percent said they were likely to follow it. Public-sector psychiatrists were less positive about PADs and more certain that they would have no impact on their practice.

In the end, the value of a PAD may lie as much in its role as a form of communication with future clinicians as in the fact that it is a legally binding document, said Appelbaum. A purely informational system does exist in the United Kingdom, he said, where "crisis cards" serve as nonbinding directives, telling clinicians in effect: "If I get in trouble, here's what helps me...."

Information about psychiatric advance directives is posted online at<http://ps.psychiatryonline.org/cgi/content/full/56/5/592?> and<http://ps.psychiatryonline.org/cgi/content/full/55/7/751>.

Psychiatr Serv200556592

Psychiatr Serv200455751

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).
Related Articles