The number of requests for psychiatrists to evaluate mental capacity in general hospitals in the United States and the United Kingdom has increased over the past 20 years.1–4 This trend, reported primarily in psychiatric journals, has been attributed to a confluence of factors: increased emphasis on patient autonomy, greater availability of life-extending technology, well-publicized applicable legal decisions,5,6 legislation in the United States and the United Kingdom,7–10 and the adverse impact on patient care of changes in the healthcare delivery system.1,4,11–14 Evaluations for mental capacity are most commonly used to determine whether patients have the legal right to reject their physician’s recommendations for diagnostic procedures, treatment, postdischarge care, and continued hospitalization, or to justify the assignment of a medical-care proxy.
In 1973, Albert and Kornfeld addressed the problems created by “the recalcitrant patient.”15 The usual wording of such consultation requests at that time was, “Please help with a patient who is threatening to sign out of the hospital against medical advice.” That study and later reports found that such threats were usually a manifestation of anger, fear, psychosis, or dementia.13,14 With appropriate psychiatric management (psychological and/or pharmacological), 78% of the patients in that study remained in the hospital, and the others returned or were followed in an outpatient clinic. Subsequent reports contained similar findings.4,12–14
The current study examined a prospective series of requests for psychiatric consultation to evaluate mental capacity in order to identify the circumstances that initiated them, and to determine the interventions required to respond adequately. We hypothesized that a mental-capacity evaluation, without other interventions, would be inadequate for most of these situations.
This study was conducted at the New York Presbyterian Hospital (Columbia), where all patients are managed by a house staff/attending team. Psychiatric consultants (“attendings” and PGY 4 residents) on the Psychiatric Consultation–Liaison Service of the New York Presbyterian Hospital (Columbia) completed a multiple-choice questionnaire at the completion of each consultation and associated follow-up visits. (Questionnaire available upon request).
Three periods were arbitrarily chosen: March 2003–June 2003, January 2004–April 2004, and September 2006–October 2006. The primary questions asked were 1) the reason for the consultation, as stated by the referring physician; 2) the psychiatric consultant’s determination of the reason; 3) the referring physician’s determination of the patient’s mental capacity at the time of the consultation; 4) the nature of the psychiatrist’s intervention; and 5) the clinical outcome (where known), and a brief case vignette.
Two of the authors (DSK, PRM), before reviewing these surveys, independently reviewed the hospital records of all the patients to determine clinical outcome and categorize each consultation, by consensus, into one of three categories:
Capacity Issue Only: The consultation required only the psychiatrist’s opinion of the patient’s decision-making capacity. Clinical example: An 83-year-old woman was admitted to the hospital after hitting her head in a fall. A CAT scan was negative; however she was confused and disoriented. She asked to return home immediately. The psychiatric consultant determined that she lacked the capacity to make that decision, and she remained in the hospital for further evaluation.
Capacity-Plus: The consultation required additional psychiatric interventions, for example, pharmacological treatment and/or psychological intervention with patients, their families, and medical staff. Clinical examples: First, a 72-year-old man with carcinoma of the colon refused preoperative testing. The consultant found him to be anxious, paranoid, and psychotic, and treated him with olanzapine. The paranoid symptoms subsided, and he permitted the needed testing. Second, a 52-year-old Spanish-speaking man with end-stage renal disease was admitted after a myocardial infarction. He refused hemodialysis and cardiac catheterization. With the assistance of an interpreter, the psychiatrist determined that he had capacity but was very frightened. With support from his family and the use of lorazepam, the patient’s anxiety diminished. He accepted the treatment recommendations.
Not About Capacity: The consultation did not require that the psychiatrist determine the patient’s decision-making capacity. Clinical example: A mental-capacity evaluation was requested for a 35-year-old woman with AIDS who refused to allow a phlebotomy for blood cultures. The patient told the consultant she would be more cooperative if addressed more politely. She also complained of severe diarrhea and painful perirectal sores. The psychiatrist communicated this information to the staff. After they addressed these needs, the patient cooperated with required procedures.
The two senior authors classified a consultation as successful if the presenting problem, for example, a patient threatening to leave against medical advice remained in the hospital.
All consultations requests were submitted by house officers, who, in their PGY 1 year, received a brief discussion on the evaluation of mental capacity from one of the authors (PRM). The Consultation–Liaison (C–L) Psychiatry consultants received lectures/readings on the subject and group and individual supervision.
We summarized the data as frequencies and percentages. The chi-square test was used to compare categorical variables. The analysis of agreement between referring physicians and consultants in the determination of mental capacity used the kappa κ statistic. All analyses were performed with Stata 8.0 software.
This study was approved by the Institutional Review Board of the New York State Psychiatric Institute/Department of Psychiatry, Columbia University College of Physicians and Surgeons.
The Adult Medical/Surgical Units of the New York Presbyterian Hospital (Columbia) requested psychiatric consultation for 970 patients during the three study periods. Ninety-seven (10%) were requests to evaluate mental capacity (the “study group”). The percentage of such requests was the same in the three study periods. The patients in the study group were similar in age (mean: 59 years; range: 22–91) to those in the general consultation group (mean: 58 years; range: 17–86). The ratio of men to women in the study group (68% versus 32%), was significantly higher than in the general consultation group (56% versus 44%; p <0.001; Table 1).
TABLE 1.Sample Characteristics
| Add to My POL
|General-Consultation Group||Study Group||Patients Threatening SAMA or Refusing Treatment|
|Mean age, years (range)||58 (17–86)||59 (22–91)||57 (21–91)|
|SAMA: signing out against medical advice.|
The issues that prompted the consultation request are outlined in Table 2. Sixty-two (64%) of the 97 requests for capacity evaluation pertained to patients threatening to sign out of the hospital against advice (SAMA) and/or refusing medication or procedures.
TABLE 2.Reason for Capacity-Consultation Request (N = 96)
| Add to My POL
|Wants to sign out AMA (only)||29||30.2%|
|Refusing procedure (only)||25||26.0%|
|Refusing medication (only)||2||2.1%|
|Refusing procedure and medication||6||6.0%|
|Refusing nursing home||5||5.2%|
|Consent for DNR||4||4.0%|
|AMA: against medical advice; DNR: Do Not Resuscitate.|
In the SAMA sample, 16 (30%) were submitted as emergencies. The patients in the SAMA group were similar in age (mean: 57 years; range: 21–91) to those in the general consultation group. However, in the SAMA sample, the proportion of men to women was higher than in the general study sample, (73% versus 27%, as compared with 68% versus 32%; p <0.001; Table 1). Twenty percent of the complete sample was non–English-speaking. However, they represented only 13% of the SAMA sample and 32% of the non-SAMA sample: (13% versus 32%; p ≤0.02)
Two senior psychiatrists, (DSK, PRM), categorized 32% of all requests as requiring Capacity Evaluation Only; 31% as Capacity-Plus; and 37% as Not About Capacity. Among the subgroup of patients (N = 62) who were threatening SAMA and/or rejecting treatment/procedures, 23% were categorized as requiring Capacity Only; 40% as Capacity-Plus; and 37% as Not About Capacity (Table 3). The psychiatric consultants attributed this behavior to anger, fear, or both, in 44% of these patients, including 17.5% who were also psychotic or had delirium or dementia (Table 4). The consultants were successful in having 88% of the patients in the SAMA group remain in the hospital and/or accept treatment recommendations.
TABLE 3.Psychiatric Consultants’ Interventions
| Add to My POL
|Study Group||SAMA Group|
|Not About Capacity||36||37%||23||37%|
|SAMA: signing out against medical advice.|
TABLE 4.Patients Threatening to Sign Out AMA or Refusing Procedure/Treatment (N = 62)
| Add to My POL
|Anger and fear with psychosis and delirium||11||17.5%|
|Anger and fear without psychosis and delirium||16||26.5%|
|Psychosis, delirium, dementia alone||23||36.5%|
|Psychiatric consultants’ formulation of reason for capacity-evaluation request.|
AMA: against medical advice.
Where data were available (75/97 patients), there was a 67% overall concordance rate between the referring physicians and the psychiatric consultants in their determination of patients’ mental capacity. However, when the referring physicians thought the patient had capacity, the concordance rate was 37.5%; when the referring physicians thought the patient lacked capacity, the concordance rate was 80% (κ: 0.19; Table 5). In the nonconcordant group in which the referring physicians thought patients had capacity, 47% were patients threatening to SAMA. In the nonconcordant group in which referring physicians thought patients lacked capacity, only 10% were threatening to SAMA. The sample size precluded statistical analysis.
TABLE 5.Concordance in Assessment of Capacity Between Referring Physician and Psychiatric Consultant (N = 75)
| Add to My POL
|Patient Lacks Capacity N||Patient Has Capacity N|
This study was conducted at a single major academic medical center, and some data were based on the clinical assessments of the C–L consultants and the senior authors.
However, the clinical findings are compatible with earlier reports.1–4,11,12 All three study periods had essentially the same data for percentage of consultation requests for capacity, male patient versus female patient ratio, and percentage of non–English-speaking patients.
Sixty-four percent of the sample was for patients threatening to sign out of the hospital against medical advice or refusing treatment and/or procedures, which is higher than in five other published reports (mean: 44%; range: 26%–83%).1–4,11 It is interesting to note that the report of the highest incidence (83%) came from another New York City hospital.
The data confirm our hypothesis that psychiatric consultation requests to evaluate mental capacity most often arise from problems that require other and/or additional interventions. The determination of mental capacity alone resolved only 32% of the problems. Additional interventions were required in 31%, and 37% did not require an evaluation.
In contrast with a 1973 publication from this institution,15 not one of the current consultation requests was framed as a request for psychiatric help with a difficult problem. All were presented simply as requests for a mental-capacity evaluation. It may be that some house staff are unaware of the potential for resolution of these problems; some may be seeking a quick solution; or perhaps this form of request is a reflection of changes in the culture of hospital practice.
Our data support the assumption that a physician’s emotional state can affect his or her judgment of a patient’s mental capacity at the time such a determination is made. Analysis of the concordance between the opinions of the referring physicians and the psychiatric consultants found a 67% overall concordance rate, which is similar to findings of earlier studies.1,3,12 However, the concordance rate was 80% in patients thought not to have decision-making capacity by the referring physicians, but only 37.5% when the referring physicians believed that patients had adequate decision-making capacity (κ: 0.19). Thus, referring physicians were more than twice as likely as the psychiatric consultants to record that patients had capacity. Our data suggest this was more apt to occur when patients were threatening to sign out against medical advice; however the sample size was too small for statistical analysis. Similar findings have been reported and attributed to the referring physician’s conscious or unconscious wish that a troublesome or “undesirable” patient have capacity and thus be permitted to leave against medical advice.1,13,14 This may also reflect the reluctance of physicians to use coercion, albeit legally, to enforce hospitalization. Unfortunately, hospital discharges against medical advice can lead to serious medical consequences.20,21
As in earlier reports, the threat to SAMA or the refusal of treatment/procedures accounted for a major portion (64%) of requests for mental-capacity evaluation.1–4 Within this group of patients, only 23% of the problems were resolved with a mental-capacity evaluation alone; 40% required additional interventions; and, in 37%, mental capacity evaluation was considered to be unnecessary (Table 3). The psychiatric consultants successfully resolved 88% of these problems by utilizing psychotherapeutic and/or psychopharmacological interventions, confirming earlier reports of similar success.1,4,12–15
In the “SAMA” subset of patients, as in earlier reports, there was a higher percentage of men than women (73% versus 27%; p <0.001).3,4,13 This suggests that men may be more likely to challenge medical authority. There were also fewer consultations for non–English-speaking patients (13% versus 32%; p <0.02). This suggests that these patients may be more compliant or that physicians may spend additional time with interpreters and families to resolve the underlying problems.
The psychiatric consultants attributed 44% of the SAMA requests to a patient’s fear, anger, or both. This should not be surprising, since hospitalization, for most patients, provokes intense anxiety as they confront their vulnerability to serious illness and death. Unfortunately, in the hospital setting of today, patients may have to cope with these emotions without the support of an effective doctor–patient relationship.4,11–14 This may be a major contributing factor to the increase in these requests.1–4 Indeed, such consultation requests are less likely to occur on private services13,14 and in community hospitals,11 where patients’ personal physicians are more likely to be involved in their care. There are no reports on the prevalence of such requests in institutions where care is provided by hospitalists.
The authors believe that four changes in the healthcare delivery system have contributed to a weakening of the doctor–patient relationship: 1) the emphasis on shortening the length of hospital stay, which reduces the time physicians spend with patients; 2) the duty/hour limits required for house staff schedules, which diminish continuity of care;22,23 3) the increased number of subspecialties, which disperses clinical responsibility; and 4) the absence of patients’ personal physicians in their hospital care.26,27 We believe that these factors are likely to remain embedded in the fabric of medical practice into the foreseeable future. Physicians, by self-selection, have adapted to this team-based care, but apparently not all patients can do so.
There are only approximately 250 psychiatric consultation–liaison services among the more than 5,500 hospitals in the United States. Consequently, most hospitals are not prepared to respond to an urgent request for a psychiatric consultation or provide a consultation–liaison psychiatrist trained to deal with these situations. Therefore, C–L psychiatrists should use these consultations as an opportunity to teach referring physicians, who may not have such a service available to them in the future, what can be done to prevent or resolve these problems.
A basic lesson to be taught is that the determination of decision-making capacity does not always require a psychiatric evaluation. In most jurisdictions, any physician can do so. Therefore, the applicable legal criteria can be taught—and how they may vary with the urgency and severity of the clinical situation. House officers can be referred to the literature for additional information and assistance.16–19
A more difficult problem is the need to overcome the institutionalized obstacles to the establishment of effective patient–physician relationships. Simple measures can help. House officers can be told of the value of sitting down at the bedside.28,29 Without necessarily spending more time, this simple act can send a powerful message: My attention is now totally devoted to you, which, in itself, can reduce a patient’s anxiety. It also provides an opportunity to assuage specific fears, correct misconceptions, remedy complaints, and identify previously undetected mental-status problems.
Physicians may learn that a patient’s fears were based on a misunderstanding of what was said at the bedside or perhaps overheard, accompanied by a reluctance to ask a “stupid” question.1,30 Patients can be told at the outset that physicians know that they often use medical jargon that patients cannot be expected to understand and therefore no question is “stupid.”
As medical students, house officers were taught the basic principles of psychodynamics. These situations provide an opportunity to learn how such knowledge is applicable in everyday medical practice. On reflection, physicians can understand that serious illness and hospitalization cause regression and a dependency akin to childhood, that they and the nurses are unconsciously perceived as parents, and that these unconscious dynamics can produce inexplicable behaviors as patients struggle to deal with their sense of powerlessness. That understanding provides a physician with an opportunity to devise a means of restoring a patient’s sense of control; for example, agreeing to a request for some benign modification in their care. The medical literature is available to provide additional guidance on the management of common patient responses to illness.31–35
Most house officers regret their inability to establish more effective patient–physician relationships. However, they will be concerned that that some of these recommendations may require the expenditure of that most valuable commodity, time. The C–L psychiatrist can point out that when effectively applied, these techniques can avert more time-consuming problems.
Psychopharmacological treatment may be necessary to reduce anxiety or treat delirium, psychosis, or depression. In some cases, these drugs may restore a patient’s decision-making capacity. The consultation provides the C–L psychiatrist an opportunity to teach the effective use of psychotropic drugs in medical situations.
A mental-capacity evaluation alone was insufficient to resolve two-thirds of the problems that elicited such requests. Sixty-four percent involved patients who were refusing treatment/procedures and/or threatening to leave the hospital against medical advice. We believe that the current culture of hospital practice creates obstacles to the establishment of an effective patient–physician relationship, which contributes to the observed increase in capacity–consultation requests. C–L psychiatrists have the opportunity to provide their colleagues with the understanding and knowledge that can help them to function more effectively within these constraints.