Moving a Noncompliant Patient to Long-Term Care Against Her Wishes

Discussants: Alicia Arbaje, MD • Lindsay Branch, MD • Samuel Durso, MD • Edmund Duthie, MD • Natasha Harrison, MD • Ariba Kahn, MD • Jonny Macias, MD • Kyle Moylan, MD • Lilliana Oakes, MD • Nina Patel, MD This new section in Annals of Long-Term Care, which will appear periodically in the Journal, focuses on real-world, difficult cases that clinicians have encountered in their practice. Each of the articles in this section will include a case report, followed by a roundtable discussion about the case along with teaching points and clinical pearls from clinician faculty at collaborating medical schools. The case presentation in this article was written by the geriatrics fellows and faculty from Mayo Clinic, and the discussion that follows was compiled from a transcript of a monthly telephone conference hosted by Aurora Health Care. The “Wisconsin Star Method” was used to evaluate and address the most difficult aspects of the case. This method is described in this article.
__________________________________________________________________________________________________________________________________________

Case Presentation

A 75-year-old white woman with a history of hypertension, osteoporosis, and diabetes mellitus was brought to the emergency department (ED) by her son and a neighbor because of visual changes. The patient reported having problems with her visual fields, especially on the left side, and noted flashes of light and flickering. She did not have any headaches, focal weakness, dysarthria, dysphagia, or cardiac symptoms, such as chest pain or dyspnea. She had a history of recurrent hypoglycemia, for which she had been treated in the ED before. Otherwise, she received her primary care in the outpatient clinic.

The patient’s capillary blood glucose values were recently above 300 mg/dL, with some fluctuation, as taken by a home glucometer. Her neighbor was concerned that she had not been fully compliant with her medications, in particular her insulin. Her son described his previously unsuccessful efforts to get a court-appointed guardianship. The neighbor and son described her mental status as stable.

The patient had a cerebrovascular accident in 1991, with no residual focal weakness. She also had a history of diabetic retinopathy, and had previously declined medications to treat her hyperlipidemia. She underwent a psychiatric evaluation in 2004, which showed some unusual and illogical thinking. A psychiatric consultant saw her again in 2007, when she declined surgical intervention to treat a humeral fracture. At that time, the consultant wrote that “she probably has some degree of atypical psychosis” and that “she lacks decisional capacity.” The patient had no history of psychiatric hospitalization.

On physical examination, the patient was obese, well dressed, and well groomed. Her blood pressure was mildly elevated, but her other vital signs were normal. Her body mass index was 35. An eye examination was unremarkable, except for diabetic retinopathy, and the ophthalmologist who saw her in the ED noted no new visual changes since her prior visit. The rest of her examination, including cardiovascular, respiratory, and neurological assessments, was unchanged since her last ED visit. On psychiatric examination, she was cooperative and was in no acute distress. Her mood was euthymic and her affect was appropriate. She became irritable when asked about her medication compliance and insulin use. The patient’s speech was quiet, and she responded with brief “yes” and “no” answers at times. Her Mini-Mental State Examination score was 28 out of 30. Her thought flow was tangential, and her thought content was appropriate except for some paranoia. She denied having any delusions or perceptual disturbances. Her judgment and insight were limited.

Laboratory findings were normal, including a complete blood count, electrolytes, renal function, and urinalysis. Her electrocardiogram and cranial computed tomography (CT) scan were unchanged from her previous record.

Regarding her medications, she was not taking the losartan that had been prescribed, but was compliant with metoprolol 50 mg daily. She claimed to be taking calcium with vitamin D twice daily and 50 units of 70-30 insulin twice daily. She reported being “fine” with administering her own insulin, although her blood sugars were poorly controlled.

As far as her social issues were concerned, she lived alone. She was divorced and had three adult children: two sons, both of whom lived 6 hours away, and a daughter who lived out of state. She had no history of smoking, drug abuse, or alcohol use. Her son had previously admitted her to an assisted living facility for more help, but she did not like it and had returned to her home. She had meals regularly delivered to her home through the Meals-on-Wheels program.

The most difficult aspect of this case was how to effectively manage the patient’s chronic medical problems and ensure her well-being at home. As stated previously, her diabetes was poorly controlled, with multiple episodes of hypoglycemia and hyperglycemia, and she had not been compliant with her antihypertensive medications. While her blood pressure taken in the ED was relatively normal, several readings in the clinic were above 180 mm Hg systolic, and had reached as high as 190 mm Hg systolic.

The patient’s physicians attempted to convince her to comply with her medication regimen to avoid complications of uncontrolled hypertension and diabetes mellitus. As previously indicated, the son who had accompanied her to the ED had sought guardianship for her, but this had been denied, and his plans to move her into an assisted living facility were also unsuccessful. Both of her sons had alerted the county Department on Aging to assess the patient as a vulnerable senior; however, she had not allowed the social worker into her home for an assessment.

Should this noncompliant older patient be moved to an assisted living facility against her wishes?

Continued on next page

The case discussion that follows is based around the “Wisconsin Star Method” (see below for more on this method), which provides guidance on how to address the 5 realms of this patient’s complex, interacting problems, including medication compliance, medical issues, behavioral health issues, personality traits and personal values, and social issues. By using this method, we can approach the most difficult aspects of complex cases, such as that of the case patient, in a more systematic way.

Case Discussion

Duthie: Does the patient have any sort of bond or link with a primary care physician?

Hussain: Yes, she does have a primary care physician, and, interestingly enough, she has been following up regularly with her primary care physician, who also wanted the county Department on Aging to evaluate her.

Harrison: I notice she has a history of being noncompliant. Has she always been noncompliant with her medications, or was this a change for her?

Hussain: She has been noncompliant for the past 3 years, according to her primary care physician.

Patel: Did she take her own insulin? Do we know how good her vision was, with her history of retinopathy?

Hussain: Yes, that was the basic concern. She reported doing “fine” and that she has been administering her own insulin with ease. When we checked her records, her hemoglobin A1c was always elevated, and she had all the complications associated with uncontrolled diabetes mellitus. We doubted her compliance and proper administration of insulin.

Moylan: What did the patient think about her management of her medications? Would she have been surprised that we’re having a difficult case conference about her case? Would she have thought that she was noncompliant? What were her beliefs about her medical illness?

Hussain: It was very hard to maintain a good, healthy conversation with her. She seemed to get off the point, and she did not want to talk to me much about her medications. She had her own reasoning for not taking them. For example, she was given alendronate for her osteoporosis. For that, she said she had problems with her swallowing. For losartan, she did not cite any particular reason for not taking the medication. She just said, “I don’t want to take it…I don’t like it.” She had an excuse for whatever we recommended, or whenever we tried to convince her.

Macias: My question is regarding the decisional capacity of this patient. Would a second opinion help to guide us in guardianship?

Hussain: Actually, no second opinion was sought at that time. Maybe it would have been helpful in deciding her goals of care.

Kahn: How did you feel when you were in the room with her? I ask this question because it helps to give me a sense for personality disorders.

Hussain: Her behavior, especially in the ED, was very argumentative. She wasn’t depressed and was able to maintain eye contact; however, she didn’t want to listen to any reasoning, and she kept bringing up her own arguments. When we asked about her vision, she said that she had laryngitis and that was why she had visual problems. Then suddenly she said that no one cared about her health.

Kahn: Did you feel frustrated when you were trying to work with her?

Hussain: Yes, of course.

Arbaje: Did the patient herself call 911 on her trips to the ED? Secondly, had anyone asked her specifically what her expectations were when she was in the ED, like for receiving care? Finally, it looks like she had paranoia, but she was able to rationalize her thoughts. Did she have paranoid thoughts related to healthcare?

Hussain: The patient discussed her situation with her neighbor, who arranged for her transfer to the ED. I don’t think her expectations were discussed with her in the ED. Regarding her paranoia, she did not trust the people around her, although she did call her neighbor. In the past, she showed mistrust of her neighbor, her sons, and physicians.

Moylan: Had anyone (her neighbor, medical professionals, or anybody else) seen the inside of her house? I wonder how well she was taking care of herself. What is the goal of obtaining guardianship at this point? Were we planning to take this woman out of her home and put her into a nursing home? Were we planning to give her insulin whether she wanted it or not? What would we do with the guardianship once we got it?

Hussain: We were unable to access her house. Her living conditions were also a major concern of ours. Obtaining guardianship was important to the son and the people around her because they were concerned about her medical problems. Their objective was to optimize her living situation. They felt that if she was unsafe and not taking her medications, then she needed to be sent to another living setting.

Malone: Was she depressed? Did she have a Geriatric Depression Scale administered?

Hussain: She did not have a Geriatric Depression Scale in the ED. She was well dressed and calm, not restless. She did not look depressed, and she did not verbalize being depressed. Her major mental problem was her inability to maintain a conversation. She usually got off track very easily.

Harrison: When she came to the ED, did she have good eye contact when she was talking about the reasons for her visit, and was there any previous personality disorder—history from the family, the son, or other children?

Hussain: She did have good eye contact while she was in the ED, and even looking back at her clinic visits, she seemed to have good eye contact at that time. According to her son, her responses relating to her medications, and her quiet speech and low conversation levels, have been like this at least for the last few years. There is no documented personality disorder in our records. Duthie: Were there any issues with financial mismanagement or exploitation?

Hussain: No, not as far as we know. Patel: Was there any formal neuropsychiatric assessment, including tests like the Clock Drawing Test?

Hussain: She did not have any neuropsychometric analysis. She only had psychological testing in 2004.

Durso: I want to know if someone is going to address the medical issues besides the psychosocial issues. I’m struck by the difference between her chief complaint in the ED and what we’re talking about. She came in describing visual changes and actually something rather specific—visual field flickering. In addition to the obvious problems she has with her behavior, I wonder what we know about this, namely, how long she had the symptom, and whether this was evaluated. Was her particular question addressed, and how was it addressed?

Hussain: An ophthalmologist in the ED saw her. Her visual acuity was unchanged and measured 20/40 for her right eye and 20/200 for her left eye. Ophthalmological testing was similar to her previous testing. An ophthalmologist had seen her on multiple occasions in the past. She had a CT scan performed, which was unremarkable.

Durso: So there was no feeling that she had a retinal tear or perhaps any problem with intraocular pressure or a new hemorrhage, or anything that would account for this rather specific symptom she described?

Hussain: No, I even called her recently to ask her how she was feeling. She told me that her vision was “fine.” She was bothered by her laryngitis, a complaint that has been there for almost 2 years.

Malone: Dr. Moylan, if this older person presented to your hospital ED, how would you approach this situation? How would you approach the issue of helping her with her numerous ED visits, treating her chronic medical conditions, and determining whether she could manage at home? How would you help the patient?

Moylan: This certainly is a tough case. I would ask first, what’s going on with her vision here, and is there anything acutely that we need to do to help? Obviously, the more overriding issue is whether this woman is safe to live alone. What is it that bothers us so much about her being independent at this point? Is it the frequent ED trips? Are we unhappy that her hemoglobin A1c isn’t lower? An 8.8% hemoglobin A1c value drawn recently isn’t terrible. So, do we guarantee that she would be better off in a nursing home with a hemoglobin A1c level of 7.5%? How would we handle this here? We would certainly be looking closely at where the patient lived in our community, and what services were possible to get in the home. It sounds as though this patient is very resistant and has not allowed social workers to make home visits. We do have some limited home practices where physicians will actually go into the home. Maybe that would be an option for her if that was available, in an effort to keep her out of the ED. But I’m not sure there is any definitive solution to keeping her out of the ED, and, personally, I probably would not pursue guardianship in this situation.

Harrison: We think it’s very important to define her goals of care. It appears as if she has the capacity to make her own decisions, whether or not we agree with them. With regard to her numerous ED visits, it may help her to follow up with her primary care physician more often. I don’t know whether that will work or not. There may be other psychiatric issues that we’re not addressing.

Patel: We feel that she doesn’t have intact executive function, and that this needs to be assessed first. So we would actually want to admit her and further assess her for possible delirium or psychotic behavior, even though many features described appear to be chronic. We recommend that she be evaluated with neuropsychometric testing to allow us to create a plan for her.

Durso: It has already been stated almost precisely the way I think about this. She came in complaining about her eye, so I’d want to make sure that there wasn’t an urgent or emergent problem that needed treatment. I’m a little concerned about that, but it sounds like it was thoroughly evaluated. I also don’t see or hear evidence that she’s delirious or that she lacks the capacity to take care of herself. I agree she’s certainly a vulnerable person, having mental illness and living alone in the community. I think there were several suggestions of services that might appeal to her or might be acceptable to her, including home visits, or more frequent checks with her primary care physician. At this stage, I can’t justify something as intrusive as taking away her right to live on her own, or forcing her into a hospital.

Branch: I completely agree with Dr. Durso. I was thinking how similar this patient sounds to many of the patients in our elder house-call program. There are a lot of patients whose decisions we may not agree with, but I tend to err on the side of maximizing their freedom and ability to live on their own. So I would be hesitant to pursue guardianship to institutionalize her to make her laboratory results look better. I definitely would want to pursue her goals of care. We should do everything possible to support her at home, and let her enjoy her life.

Oakes: I’m still unsure of how well she’s really able to process more complex decision-making. As I approach these challenging cases, I have patients describe their understanding of their illness and their plan of care. When I hear patients’ voices, I get a better feel for their level of understanding of their illness.

Continued on next page

Teaching Comments

Drs. Howell and Roche were asked to review this case using the Wisconsin Star Method (see below for more on this method) and to prepare teaching comments to share with the group.

Howell: I agree that this is a very common scenario. When I use the Wisconsin Star Method, as in this case, I begin with a review of the medications. It appears to be fairly straightforward: make sure that she is not using over-the-counter remedies with diphenhydramine for insomnia or borrowing medications from other people. In terms of her medical problems, she has a lot of risk factors for stroke and, indeed, has had a stroke, which might account for some of her presentation. I agree with assessing her executive capacity. The general psychiatrist or geriatric psychiatrist who previously assessed her decisional capacity would have needed to take into account that she may have decisional capacity but lacks executive capacity. In other words, a patient can know what needs to be done, but then when he or she goes home, it does not happen, and that is what creates some of the risks. A good way to assess this patient, in addition to or instead of neuropsychological testing, would be to have an occupational therapy evaluation. This functional evaluation allows us to observe what she can and cannot do without prompts. This further provides evidence of the extent of her need for prompting, supervision, or assistance. The other issue to address is her paranoia. It sounds like she is not frankly delusional, but she is clearly mistrusting. There is some question about whether she has always been this way. Her son indicated that she has been this way for a few years. It’s very common to see evidence of a personality change, where people may develop some paranoia that wasn’t there before. Sometimes there is an amplification of premorbid suspiciousness that becomes more frank. Other personality changes you want to assess would be: labile affect, lack of motivation (“apathy”), aggression, or disinhibition. The clinicians would need to assess if she has always been like this, or if there has been a change or an intensification of her premorbid traits.

Hussain: According to the son, she has been like this for years. Relating to her personality, there has not been any recent change in her personality.

Howell: That is something you might go back and review. Sometimes family members might not notice the change until it’s called to their attention. Other times a family member who has had rather little contact with the patient over the years comes back and says, “Mom’s not the same as she used to be.” So that could be a red flag for a personality change. We often see a combination of personality changes and a decline in executive function in late life in the absence of memory changes. In terms of the goals, I’m totally in agreement with trying to get an inventory of the patient’s individual values and preferences. Then work out ways (short of guardianship) that we can achieve an optimum balance between safety and autonomy. The patient may say, “I’m aware of the fact that my diabetes is not under the best control. I can appreciate the risks, but this is the way I’d rather do it.” Then you have a better sense that working with her collaboratively would be reasonable. The clinician needs to make sure the patient has an appreciation of the risks and how congruent her goals are with what is actually happening. This case calls for some creative brainstorming on the part of the interdisciplinary team, to figure out what interventions, short of guardianship, would maintain this optimum balance between safety and patient autonomy. Then having worked that out, some months or years later, the situation may change again. In that case you have to be flexible enough to reassess and come up with some new plans.

Roche: I would have looked at her appearance in the ED. What was her general condition when she was seen? I was interested to make sure that she could actually see. It’s helpful to know that she had 20/40 vision in her right eye and 20/200 vision in her left eye. It would be helpful to ask the ED nurse to check how she actually draws up her insulin. Is it done correctly? I would make sure the patient can actually read. I understand she’s had some education, but it’s important to make sure she has no problem with reading medication bottles, for example. It would be helpful to evaluate her electrocardiogram to assess if she has end-organ damage or left ventricular hypertrophy from long-standing hypertension. Obviously, it sounds like she has had quite a few distressing conversations in the ED, so this may not be the best place to assess her blood pressure control. I’d also like to pursue her hypoglycemic episodes. She seems pretty clear that she never wants it below 200 mg/dL, so it may actually have been a very frightening episode for her to be at home and to find that she was suddenly hospitalized for a hypoglycemic episode. I’d ask her to tell me a little bit more about how this hypoglycemic attack occurred.

I’m not sure I agree with the diagnosis of “psychosis.” We all have seen noncompliant patients. She definitely has been nonadherent, or noncompliant. The question sometimes for the geriatrics fellows is, “If she were 22 years old, would I want to pursue guardianship?” When people are older, we sometimes assume that because they really don’t agree with our plan of care, we need to consider guardianship. Many patients think differently than we do about their healthcare.

Regarding the social aspects of this case, I’d ask how close she is with her neighbor, and I would ask further about her driving. I’d also want to know about her relationship with her daughter. Maybe she’s closer to her daughter than her sons. I would like to understand her relationship with the son who pursued guardianship and tried to put her in an assisted living facility.

I would like to know more about her ex-husband and other close family members. How was her ex-husband’s health and how did he interact with healthcare professionals? This may provide a context to her perception of the medical system.

As for some overall teaching comments, I think it’s very important for this patient to have a close relationship with her primary care physician, because of the transitory nature of the ED. I agree with frequent scheduled visits with her primary care physician, as this would provide the opportunity to discuss options when she has an acute issue. The focus should be to define her goals and her values—try to get her to describe her short-term goals as you begin to manage her medical care.

The patient has full capacity currently, so I’d encourage her to address designating a durable power of attorney. It would be important for her to make that decision now rather than someone making it for her later. Then I would want to determine the reasons for her nonadherence. Is it because it’s difficult for her to give herself the insulin, or that she just hates injections? Is it access and/or finances? Is her reluctance the result of her prior experience of hypoglycemia? Is her glucometer working properly?

Finally, we know that a major issue is nonadherence, so I would focus on her medications. She is obese, with a body mass index of 35 and a weight of 183 lb. She may be a good candidate for metformin. Her insulin dosage does not seem to be working that well for her, as evidenced by her hypoglycemic episodes, so maybe that would be a consideration. Eliminating injections may simplify things for her. She also takes two blood pressure medicines that are not maximally dosed. I would try one blood pressure medicine first, at a higher dose if needed, and aim for once-daily dosing.

I agree with having home healthcare. She has agreed to home-delivered meals, so that relationship did work. I would explore further if she would accept home healthcare, especially to help her with her medicines, blood pressure control, and glucose monitoring. This might be very effective for her.

Disclaimer: The opinions expressed in this article are those of the authors and do not necessarily reflect the views of the Veterans Administration or the University of Wisconsin School of Medicine and Public Health.

Moderator, Discussant, and Attendee Affiliations: Mayo Clinic, Rochester, MN (Syed Wasif Hussain, MD, Shahzad Siddique, MD; Aimee Yu-Ballard, MD); Aurora Health Care, Milwaukee, WI (Ariba Kahn, MD, Jonny Macias, MD, S. Soryal, MD); University of Texas Southwestern Medical Center, Dallas (Natasha Harrison, MD, Vivyenne Roche, MD); Geriatric Research Education and Clinical Center, (GRECC), Madison VA Hospital, Madison, WI (Timothy Howell, MD); Johns Hopkins University School of Medicine, Baltimore, MD (Alicia Arbaje, MD, Lindsay Branch, MD, Samuel Durso, MD); Medical College of Wisconsin, Milwaukee (Edmund Duthie, MD); University of Missouri-Columbia, Columbia (Kyle Moylan, MD); University of Texas San Antonio, San Antonio (Lilliana Oakes, MD, Nina Patel, MD).

The Wisconsin Star Method

The Wisconsin Star Method is a simple, user-friendly tool for addressing complex clinical situations in geriatrics (Figure) providing clinicians with a quick and comprehensive approach to patient care with greater clinical integrity.¹ It consists of a “low-tech” graphic user interface—a small 5-pointed star on a clear surface—to map clinical data onto a single field with 5 domains: medication, medical, behavioral, personal, and social issues. The medication arm lists the patients’ medications. The medical and behavioral arms list known diagnoses, functional impairments, and any symptoms. The personal arm highlights patients’ individual traits, cultural values, and coping styles. The social arm outlines their interpersonal and environmental problems and assets. The primary clinical challenge is written in the center of the star.

The Wisconsin Star Method requires clinicians to modulate their professional egos in the interest of better outcomes—the data has to be written down. Because the effective carrying capacity of our human brains is at most 5 to 7 simultaneous interacting variables,² one cannot handle complex cases well by keeping all of the data “in your head.”

The resulting map becomes an extension of the user’s working memory,³ enhancing executive function for problem-solving. Writing down the data creates a small but significant distance between the user(s) and the problems, providing not only cognitive and affective perspective, but also facilitating attention to multiple interacting variables. One simply goes “around the star”—assessing and highlighting the elements listed in each arm that may be contributing to the challenge in the star’s middle—as well as identifying patterns of interactions between the salient factors (italicized in the Figure) and missing relevant data.

By integrating holistic and linear-causal perspectives into an ecological approach, the Wisconsin Star Method can enhance recognition of diagnostic patterns, facilitate identification of vicious cycles, and promote novel problem-solving by generating hypotheses, prioritizing and sequencing interventions, and integrating clinical “pearls”⁴ with evidence-based guidelines.⁵ Identification of patients’ personal traits, values, and coping methods, helps to appreciate the anxieties that may underlie their behaviors, thereby cultivating more collaborative relationships. It has the potential to enhance proficiency at providing comprehensive care, and also to reduce the cognitive and emotional burdens and subsequent errors associated with challenging cases.⁶ Regular use of the Wisconsin Star Method can help clinicians and teams address the complexities of the interacting physical, emotional, and social issues of their older patients, with greater sensitivity and specificity to the uniqueness of each one.⁷

More information on the Wisconsin Star Method and its evidence-based components is available at www.WGPI.org.

Definitions of Several Tests Mentioned in this Article Clock Drawing Test: Can be used as a part of a neurological test or to screen for dementia. Requires the individual to draw a clock, including the face, numbers, and arms (https://bit.ly/ClockDrawingTest). Geriatric Depression Scale (GDS): GDS is a 30-item self-report assessment used to identify depression in the elderly. Questions (eg, “Do you often feel helpless?”) are answered as “yes” or “no,” and 0 or 1 point is assigned to each answer (https://bit.ly/GeriatricDepressionScale). Mini-Mental State Examination (MMSE): Also known as the Folstein test, MMSE is a brief 30-point examination that is used to screen for cognitive impairment. It assesses orientation, recall ability, language, and basic motor skills (https://bit.ly/FolsteinTest.