Ethical Challenges in Geriatric Psychiatry

Dr. J. is a 76-year-old Caucasian, retired economics professor from the small college town where you practice. He is accompanied by his wife, B., of 50 years and their daughter, H., for a follow-up appointment regarding depression. He has been in your clinic for four years for a history of depression, which he has battled since his young adulthood. Dr. J. reports that his mood has been stable on 10 mg escitalopram daily and 7.5 mg mirtazepine at bedtime for augmentation. He continues activities with his book club and playing tennis with other retired faculty. He is bright and social on examination. He was the sole breadwinner at home and is the oldest of three siblings. His father died in a nursing home as a result of the progressive loss of function. At this visit, he reports progressive and insidious difficulty with memory, particularly word finding and recalling short-term events. He has also noticed a decreased sense of smell and taste of food. There is no obvious tremor or rigidity on examination, and his gait is fluid without bradykinesia or shuffling. The Montreal Cognitive Assessment (MoCA), administered today, reveals a score of 22 out of 30 (with points missed on visuospatial/executive abilities and delayed recall; this is a decline from his previous score of 26 from two years ago).

He has remained independent in most activities, although his daughter notes that he has overdrawn his checking account three times this year, which he has never done before. Two weeks ago, he was seen in the emergency room for a near-syncope episode and required intravenous fluids for orthostasis. His wife noticed that he was short two days on his antihypertensive medication after this. According to his wife, he continues to drive, although she reports that recently she has grown more concerned about being in the car when he is driving, because he has been driving more slowly and sometimes misses turns. There have been no motor vehicle accidents, according to the patient and the wife, and he continues to shop, bathe, cook, and manage his activities of daily living (ADLs) independently. His daughter reports that he helps pick up her children from school before she comes home from work, although he has arrived late on several occasions this year, whereas previously he was very punctual. He reports that he is careful and that he has had no concerns driving.

“America’s love affair with the automobile”—our enduring passion for cars, driving, the open road, and the freedom they symbolize—can present psychiatrists caring for older adults with challenging ethical dilemmas. Older adults are often reluctant to “retire” from driving. The emotional impact of this loss can be compounded by the difficulty in finding transportation when one doesn’t drive, especially because the infrastructure of many U.S. cities and towns depends heavily on the automobile for logistical activities.

The psychiatrist’s role in assessing the capacity to drive among individuals with neurocognitive disorders can create ethical tensions. Understandably, the idea of removing a prominent source of a patient’s sense of independence (and, often, a beloved activity) can be distressing to psychiatrists. The psychiatrist often feels caught in the intersection of principles of autonomy, beneficence, privacy, veracity, and fidelity. To complicate matters, many physicians are uncertain of their reporting responsibilities or feel that they lack the appropriate guidance and resources to handle this responsibility adequately (3).

There is a consistent agreement among physicians that individuals with moderate to severe neurocognitive disorders should not drive. However, there remains a great deal of debate about when individuals with mild cognitive impairment (MCI; or minor neurocognitive disorders) should retire from driving. At the margins—in cases in which the individual may still retain good social skills, family members do not report specific concerns, there is an absence of glaring evidence of impaired driving, but the individual shows clear signs of multidomain cognitive impairment—the psychiatrist needs some signposts for navigating this ethical dilemma.

Overall, individuals with dementia who drive are at increased risk of having a motor vehicle accident as the driver—with one study indicating that this risk may be highest in the several years preceding the dementia diagnosis (4). Moreover, other research has shown that individuals with neurocognitive disorders are at increased risk of driving slowly, not wearing seat belts, and having an elevated accident or collision risk (5). Several studies have demonstrated that, as neurocognitive disorders progress, the crash risk is significantly increased (6). Given the nebulous and, at times, fluctuating course of varying capacities of persons with neurodegenerative disease, it is no wonder that, in one study, most physicians reported not feeling competent to reliably determine driving safety (7).

There remains no definitive cognitive test or set of tests that can guide clinicians to determine a patient’s driving safety. The British Psychological Society considers on-road assessment the gold standard in assessing driving capacity (5). However, there are no controlled trials or prospective data to confirm that on-road or neuropsychological testing can reduce car accidents or support safe drivers to remain driving (8).

Several attempts have been made to aid physicians in identifying at-risk drivers. In 2010, the American Academy of Neurology published practice parameters for identifying at-risk drivers. The systematic review identified scores >.5 on the Clinical Dementia Rating scale, as well as the caregiver’s rating of a patient’s unsafe driving, as the highest level of evidence of identifying at-risk driving. In their report, a history of citations, crashes, reduced miles driven, or avoidance of driving; a Mini-Mental State Examination score of <24; and impulsive or aggressive personality traits were somewhat less useful in predicting driving risk; the patient’s self-report of driving ability was not useful (9). Recently, Rapoport et al. compiled a consensus recommendation for the assessment, identification, and reporting of individuals with MCI/mild dementia for providers (10). The authors recommended that caregiver concern about the patient’s driving, as well as abnormalities on the clock drawing test, a lower score on the MoCA, or a prior crash, should trigger a report to transportation authorities and/or a recommendation for specialized on-road testing. In addition, prolonged completion time for the Trail Making Test, Part B (with several errors), should also raise the consideration of reporting (10). Although there is mixed evidence regarding driving evaluations in terms of accident risk, Carr and O’Neill have suggested that there may be situations in which providers should consider referral for DMV or occupational therapist (often called driver rehabilitation specialists)–administered driving tests to stratify risk and inform driving retirement recommendations (11). The consensus remains that providers should weigh several aspects of the patient’s history and assessment to inform recommendations about driving cessation. Similarly, a recent Danish study argued for using the combination of the clinical interview, a neuropsychological battery, and a driving simulator to identify at-risk drivers more accurately (12). Carr and O’Neill have presented a reasonable approach to evaluating driving risk in aging adults with neurocognitive disorders, including questions that providers may ask to foster conversations about driving safety and techniques that family members could use to ensure that patients who are unsafe drivers no longer drive (13). Carr and O’Neill have also included several resources for providers and patients, including driving evaluation services and alternative transportation agencies.

The patient undergoes an occupational therapy evaluation. On the basis of the results of this evaluation, the family’s report, and the psychiatrist’s own assessment, the psychiatrist files a report with DMV. The patient returns in one month, having received a notice from DMV stating that he may no longer drive. The patient is angry at the psychiatrist, stating that driving was a primary source of independence and pleasure for him. He does not recall that the psychiatrist informed him that he would have to report him to DMV. He asks the psychiatrist to contact DMV to try to get them to reverse the decision. The wife, who is in the room for this appointment, remains quiet but appears distressed.

In a survey of 239 physicians, several factors were predictors of the decision to address driving with patients, the strongest being the physician’s perception that addressing driving should be part of their role, as well as number of years in practice. Knowledge regarding the legal requirement to report to DMV was somewhat associated; moreover, physicians who were aware of the American Medical Association’s Physician’s Guide to Assessing and Counseling Older Drivers were significantly more likely to address the issue. Of note, physician specialty was not generally associated with a greater likelihood of addressing driving with patients (14).

The National Highway Traffic Safety Administration provides state-by-state information on physician reporting, in addition to a brief cognitive test called the Assessment of Driving-Related Skills, which assesses vision, cognition, and motor function (as well as the Trail-Making Test, Part B) but which the agency admits “does not predict crash risk.” Of note, states vary with respect to requirements for reporting to motor vehicle regulation boards. For instance, in California, physicians are considered mandated reporters and can be held liable for damages.

Although the intention in reporting individuals with neurocognitive disorders to regulating bodies such as DMV is to ensure the safety of the patient and his or her family, as well as that of fellow drivers and the general public, filing such reports and subsequent driving cessation do not come without consequence. Patients are often unhappy about the physician’s decision to report, and caregivers face additional burdens of transport and arranging activities (15). Despite unhappiness on the part of the patient or family members/caregivers, the physician’s ethical (and, in many cases, legal) obligation is to consider the patient’s and the public’s safety first. Referring the patient and family members to useful resources (see Box 1), as well as initiating these conversations early in the disease course, may help soften the blow.

Mr. H. is a 78-year-old man with a major neurocognitive disorder resulting from Alzheimer’s disease (AD), which was diagnosed two years ago. He and his wife of 52 years live together and have three grown children, all of whom all live out of state. Mr. H.’s psychiatric symptoms have consisted of anxiety and occasional insomnia, and the couple has had no need for in-home assistance. His psychiatric medications have included only 10 mg donepezil per day, 20 mg memantine per day, and 50 mg trazodone at bedtime as needed. At today’s visit, Mrs. H. reports that Mr. H. has become increasingly difficult to redirect at home. He has started to accuse her of infidelity, and at times he becomes confused about her identity as his wife. Furthermore, he has begun to swear at her as often as several days per week over the past month. His insomnia has worsened so that Mrs. H. has taken to sleeping in another room. She reports hearing him pacing and talking at night, sometimes angrily. Mr. H. himself denies any problems with his mood or behavior. However, during the session, he appears irritable and gets up out of his chair repeatedly, needing to be redirected or distracted. At one point, when his wife tries to reassure him that “We’re here to get some help from the doctor,” Mr. H. states angrily, “I don’t need any help, and you sure aren’t helping! Let’s get out of here!” Mrs. H. manages to calm him down by changing the subject but looks very stressed herself. She becomes tearful and states, “I think I need something to knock him out, you know what I mean? I don’t know what else to do.”

Patients with dementia commonly experience behavioral and psychological symptoms, including depressive symptoms, anxiety, irritability, sleep problems, psychotic symptoms, and agitation. The term behavioral and psychological symptoms of dementia is now used to encompass the full range of noncognitive symptoms of people with dementia. These symptoms are thought to occur nearly universally among people with dementia at some point during the illness (16).

Severe forms of behavioral disturbance have been described since Dr. Alois Alzheimer’s initial report of the disease’s characteristics. Since that time, agitation, delusions, and hallucinations have been noted to be common—perhaps even core—symptoms of the disease (17). The literature has repeatedly confirmed a substantial prevalence of delusions and hallucinations (17). Ropacki and Jeste provided a systematic review of psychotic symptoms in people with AD in diverse settings (18). They reported a median prevalence of psychotic symptoms as follows: any psychosis, 41.1% (range 12.2%–74.1%); hallucinations, 18% (range 4%–41%); delusions, 36% (range 9.3%−63%); and other psychotic symptoms (such as misidentification), 25.6% (range 3.6%−38.9%). Thus, psychotic symptoms are quite common for people with AD.

Aggressive behavior can also be further defined as either physical (including hitting, kicking, scratching, or biting) or nonphysical (pacing, wandering, or disrobing). Furthermore, patients also may exhibit inappropriate verbal behavior, such as screaming or swearing. Some recent studies have found that over half of dementia patients exhibit some of these signs at a given time (16, 19), and other reports describe a risk after two years of almost 100% (20).

The morbidity associated with agitation, aggression, and psychosis is significant. Not only do these symptoms cause indignity to the patient and jeopardize the safety and well-being of caregivers, they also appear to be predictors of more rapid cognitive decline, institutionalization, and hastened death (21). Caregiver burden related to these symptoms is such a major problem that caregiver health is often affected.

Mrs. H. enrolls the patient in a local adult day program, which he attends five days per week. For approximately six months, Mr. H. is more able to control his behavior at home. Mrs. H. attends a caregiver support group and learns skills for managing the patient’s outbursts, and she appreciates the supportive environment. However, Mr. H.’s functioning declines further; he is becoming increasingly confused and requires more help with bathing, feeding, grooming, and toileting than Mrs. H. can provide. She and her children make the difficult decision to place him in a residential care facility three miles from their home. Within the first week, the facility calls the psychiatrist, stating that the patient is hitting staff when they try to help him with routine care, is yelling out for hours at a time, and is up most of the night. The facility is asking for medication to help manage the patient’s behavior.

Although studies have found that some medications have modest effectiveness in the treatment of some of these behaviors (17), there is no Food and Drug Administration–approved medication regimen. First-line therapy should be nonpharmacologic to address the patient’s environment and caregiving supports, and clinicians should evaluate and address underlying medical etiologies of the behavior (22, 23). Recently published, the American Psychiatric Association Practice Guideline on the Use of Antipsychotics to Treat Agitation or Psychosis in Patients With Dementia recommends nonemergent (i.e., routine or scheduled) use of these medications only “when symptoms are severe, are dangerous, and/or cause significant distress to the patient” (24).

The guidelines also stress the importance of weighing potential risks and benefits (including the possibility of no or limited benefit) of antipsychotic medication and discussing these “with the patient (if clinically feasible) and with the patient’s surrogate decision maker (if relevant) with input from family or others involved with the patient” (24). The guidelines further emphasize the importance of discussing alternatives with the family or other caregivers to ensure full support with any decision involving pharmacologic approaches.

When the patient is experiencing severe distress, when the patient’s and others’ safety is a primary concern, and when alternative strategies have been attempted without any improvement, then pharmacologic approaches can be ethically justified on the basis of beneficence (attempting to improve the patient’s quality of life), although the tension with nonmaleficence remains. Given the potential risks of antipsychotic and other off-label uses of psychotropic medications in treating behavioral disturbances in dementia, psychiatrists should conduct a thorough discussion of risks, benefits, and alternatives with the surrogate decision maker (unless the patient can provide consent, which is almost never the case in such scenarios). Protecting others (family and/or caregiving staff and other providers) from an aggressive patient should also factor into the clinician’s ethical considerations. Finally, veracity and fidelity—ensuring that the family understands not only the goals of treatment but also the limits of what we know about treating these symptoms and remaining faithful to the patient’s interests while balancing these against other concerns—remain paramount.