David Alfandre

Reconsidering against medical advice discharges: embracing patient-centeredness to promote high quality care and a renewed research agenda.

ABSTRACTHospital discharges against medical advice (AMA) are common, costly, stigmatizing to patients, and are associated with excess morbidity and mortality. Achieving better quality care for patients discharged AMA has been limited both by the sparse research illuminating how best to care for this challenging patient population, as well as a lack of standards regarding this clinical practice. This paper will review elements of the AMA literature and highlight the gaps, including the predictors of AMA discharge, challenges to high quality informed consent in AMA discharges, problematic aspects of AMA discharge forms, and the stigma associated with patients discharged AMA. These gaps in the evidence base collectively limit the ability to adequately and completely address AMA discharges and improve health care quality. This paper will recommend future directions to answer remaining questions for the field, and offer guidance for providing ethically sound and high quality care for the affected population. Applying the widely accepted principles of patient-centered care and shared decision making to AMA discharges offers the opportunity to improve quality of care and promote ethical health care practice.Hospital discharges against medical advice (AMA) are common, costly, stigmatizing to patients, and are associated with excess morbidity and mortality. Achieving better quality care for patients discharged AMA has been limited both by the sparse research illuminating how best to care for this challenging patient population, as well as a lack of standards regarding this clinical practice. This paper will review elements of the AMA literature and highlight the gaps, including the predictors of AMA discharge, challenges to high quality informed consent in AMA discharges, problematic aspects of AMA discharge forms, and the stigma associated with patients discharged AMA. These gaps in the evidence base collectively limit the ability to adequately and completely address AMA discharges and improve health care quality. This paper will recommend future directions to answer remaining questions for the field, and offer guidance for providing ethically sound and high quality care for the affected population. Applying the widely accepted principles of patient-centered care and shared decision making to AMA discharges offers the opportunity to improve quality of care and promote ethical health care practice.

What Is Wrong With Discharges Against Medical Advice (and How to Fix Them)

It is estimated that as many as 2% of all US hospital discharges (approximately 500 000 per year) are designated as against medical advice1; that is, a patient chooses to leave the hospital before the treating physician recommends discharge. The risks to these patients are significant. Compared with patients discharged conventionally, readmission rates for patients discharged against medical advice are 20% to 40% higher, and their adjusted relative risk of 30-day mortality may be 10% higher.2 Furthermore, physicians and other health care staff report feeling distressed and powerless when patients choose suboptimal care, and disagreement over a discharge against medical advice can cause patient-physician and intrateam conflict.3 Although these harms have been well described, the stigmatizing effect on patients of discharges against medical advice has rarely been examined. Compared with how the profession handles clinical disagreements in other settings (eg, outpatient), an “against medical advice” designation is an outdated concept unsupportive of patients. In this Viewpoint, starting from a core value of patient centeredness, we aim to highlight the problematic aspects of discharges against medical advice and suggest a new approach. Designating a discharge as against medical advice is a clinical practice not subject to professional standards. There is no clear medicolegal rationale for this designation and no professional consensus on what constitutes a discharge as against medical advice. If a competent patient or his or her authorized surrogate declines further inpatient care, physicians should fulfill their legal and ethical obligations to obtain informed consent for the patient’s decision and document that decision and the patient’s reasons for it in the patient’s record. But the physician’s subsequent choice to designate the hospital discharge as against medical advice and pursue the formalized process associated with it (eg, specialized discharge forms) has no evidence-based utility for patient care, is not legally required, and has been shown to be associated with a reduced willingness for the patient to return for future care.4 Furthermore, there is no consensus about what clinical criteria warrant a “discharged against medical advice” designation. This lack of clarity leads to greater variability in its clinical use, lacks transparency, and impedes standardization of a common medical practice. Although a more specific definition of discharge against medical advice could improve research and clinical processes, the term is an anachronism that has outlived its usefulness in an era of patient-centered care. Recent studies have highlighted problematic informed consent practices for discharges against medical advice by identifying that a majority of house officers and attending physicians mistakenly believe and inform patients that if they sign out against medical advice, their insurance may not pay for the hospitalization. In a cross-sectional survey of physicians conducted by Schaefer et al,5 85% of residents and 67% of attending physicians reported that they informed patients about denial of insurance payment so that patients would reconsider remaining in the hospital. These studies suggest that the use of misleading information in discharges against medical advice threatens to undermine a patient’s voluntary choice and insinuates that coercion is an acceptable and oft-repeated practice. The use of specialized discharge forms that document a patient’s risks and liability is common hospital practice in discharges against medical advice. Despite apparent widespread use of these documents, there is no evidence that they advance patient care. Although health professionals generally support the use of discharge against medical advice forms because they believe it is required to protect themselves and their institutions from legal liability, these presumptions are not valid.6 Indeed, the contrary may be true. Malpractice claims are associated with poor physician communication and patient perceptions of feeling deserted or devalued.7 If discharges against medical advice occur when there are breakdowns in communication, it is possible that such discharges may contribute to increased liability. At a minimum, there is limited understanding of whether the desire to protect clinicians and institutions from legal liability by using a specialized discharge form interferes with the care of the patient. Because clinical care decisions for hospitalized patients are sensitive to patient preference, shared decision making (SDM) has a role in achieving more patientcentered care in decisions related to discharge against medical advice. Although SDM is well accepted in overtly value-laden clinical decisions such as prostate-specific antigen testing and mammography screening, the principles of SDM apply to a broad range of health care decisions, discharges against medical advice included. Contrary to the principles of SDM, a discharge against medical advice sends the undesirable message that physicians discount patients’ values in clinical decision making. Accepting an informed patient’s values and preferences, even when they do not appear to coincide with commonly accepted notions of good decisions about health, is always part of patient-centered care. The active engagement of the medical community will be necessary to reform the practice of discharges against medical advice. Physicians can begin with individual patients, but they also can support research in this area and in establishing standards for such discharges. VIEWPOINT

“Against Medical Advice” Discharges Among HIV-Infected Patients: Health and Health Services Outcomes

&NA; Nurses are often first to identify and manage a patient leaving against medical advice (AMA), and so they are critical contributors to the development of strategies to address this problem. We studied AMA discharge in order to help develop useful interventions. We performed a cross‐sectional analysis of 55,938 discharges from a single urban hospital for the years 2002‐2003 and 2012‐2013. AMA discharge rates were higher for HIV‐infected patients than for patients with alcohol‐related disorders or sickle cell anemia in both time periods, even after adjustment for age, race, sex, insurance status, and household income. For HIV infection, 25% of AMA discharges occurred in patients with multiple AMA discharges and 30‐day readmission rates were higher after an AMA discharge: odds ratio 1.57 (95% confidence interval 1.01 to 2.43, p = .046). Team‐based and nursing interventions that incorporate the treatment‐based values and preferences of HIV‐infected patients with prior AMA discharges may improve linkage to care and reduce readmissions.

Caring for ‘Very Important Patients’—Ethical Dilemmas and Suggestions for Practical Management

The care of Very Important Patients (VIPs) is different from other patients because they may receive greater access, attention, and resources from health care staff. Although the term VIP is used regularly in the medical literature and is implicitly understood, in practice it constitutes a wide and heterogeneous group of patients that have a strong effect on health care providers. We define a VIP as a very influential patient whose individual attributes and characteristics (eg, social status, occupation, position), coupled with their behavior, have the potential to significantly influence a clinicians judgment or behavior. Physicians, celebrities, the politically powerful, and philanthropists, may all become VIPs in the appropriate context. The quality of care may be inferior because health care professionals may deviate from standard practices when caring for them. Understanding the common features among what may otherwise be very different groups of patients can help health care providers manage ethical concerns when they arise. We use a series of vignettes to demonstrate how VIPs behavior and status can influence a clinicians judgment or actions. Appreciating the ethical principles in these varied circumstances provides health care professionals with the tools to manage ethical conflicts that arise in the care of VIPs. We conclude each vignette with guidance for how health care providers and administrators can manage the ethical concern.

Against Medical Advice Discharges

The “Things We Do for No Reason” (TWDFNR) series reviews practices which have become common parts of hospital care but which may provide little value to our patients. Practices reviewed in the TWDFNR series do not represent “black and white” conclusions or clinical practice standards, but are meant as a starting place for research and active discussions among hospitalists and patients. We invite you to be part of that discussion .

Reconsidering Against Medical Advice Discharges

T he Author Replies—I appreciate Dr. Meyer’s correction regarding the exclusion for Against Medical Advice (AMA) discharges in the Centers for Medicare and Medicaid Services (CMS) readmission calculation. I also agree that the current CMS readmission calculations may have an impact on hospital discharge practices. It is both reasonable and prudent to assume that hospitals that choose to reduce their AMA discharges may have a resultant increase in their readmissions as calculated by CMS. Currently however, we don’t have sufficient data to reliably predict that this policy will affect hospital practice in this way. This would be an area deserving of further study. As hospitals institute systems changes to make discharge practices more patient-centered (e.g., reducing AMA discharges), they should consider all the potential impacts on both the hospital and the patient. The CMS exclusion for AMA discharges in the readmission calculation raises a number of other questions. Most importantly, if there are no clear professional standards for how an AMA discharge is determined, there is likely to be significant variability in the clinical use of the term. How will this affect the quality and generalizability of CMS readmission data? Also, AMA discharges are associated with higher rates of hospital readmission up to 6 months later, suggesting that they may have persistent effects. What potential data related to this effect may be lost when patients discharged AMA are not included in the CMS calculations? Continued research in this area will help to address these and other questions.

The parallels between parenting and patient care.

Soon after my medicine residency, while on faculty at a large academic medical center, I crossed paths with a complicated patient. Middle aged, homeless, and struggling with substance abuse, he was as well known to the staff for his disruptive behavior as he was for being discharged AMA. His behavior continued after he was admitted to my inpatient service for alcohol withdrawal. Late Friday afternoon, his nurse called me, distressed. “This patient needs to be transferred off this unit. He’s just too disruptive.” Shortly thereafter when I arrived on the floor, he was at the nurses’ station, stumbling and screaming to be discharged. He was largely incoherent, unable to reason, and it became clear relatively quickly that he lacked decisional capacity to choose to leave the hospital. During our discussion, though he continued to yell, I remained calm and even-handed about how we were going to care for him. I told him why he was here, sick in the hospital, and I listened to his frustrations about the staff, his room, and the food. The entire encounter, which ended amicably, lasted about 5 minutes. As I left him and returned to the nurses’ station, his nurse came up to me and said, “You handled that so well. How come you didn’t get angry?” And I said “I’ve seen worse. I’m a parent.”

Bedside Management of Discharges Against Medical Advice

Physicians and other healthcare professionals have long struggled with discharges against medical advice (AMA), or when patients leave the hospital prior to a specified clinical end point over the recommendation of the physician. There is ample literature describing the risk factors associated with AMA discharge and the common reasons that patients choose to leave AMA. In this chapter, we review the best practice concepts of bedside management of an AMA discharge. We discuss components of a quality decisional capacity assessment and informed consent discussion as a prerequisite for a safe transition of care for AMA discharges. We also discuss cognitive and behavioral strategies that can form the basis for a practical approach to the professional, legal, and ethical obligations surrounding this type of discharge. Last, we discuss patient-centered strategies, shared decision-making models, and motivational interviewing and describe operational tools and processes to assist providers in negotiating a more productive health outcome for AMA discharges.

Ethical Considerations in Against Medical Advice Discharges: Value Conflicts Over Patient Autonomy and Best Interests

Patients who wish to leave the hospital against medical advice (AMA) present physicians with a common but challenging ethical dilemma. Physicians are obligated to respect patient’s autonomy when patients choose treatment consistent with their values, needs, and preferences. However, physicians also have a fiduciary obligation to promote the patient’s best interest by recommending medically appropriate care. This chapter will provide guidance in managing this ethical dilemma by reviewing the construct of medical advice giving, examining the central role of power in that process, describing the process of shared decision-making and how it promotes strong ethics practices in informed consent, and exploring the role of surrogates in the AMA process.

Running a Code in My Ice Skates, and Other Tales

“Is there a doctor here? We need someone!” The impassioned call for medical help outside the hospital has for me always generated a troubling mix of obligation and dread. In the 10 years since I finished my postgraduate training, I’ve responded to patients in distress numerous times and in various public settings. I now work in an outpatient setting, and the medical problems I manage are important but usually not emergent. In responding to sick patients outside the hospital, I’m driven by a strong obligation to serve, which I know originates from my long socialization and professionalism as a student and trainee. But my dread is something deeper and more complicated. I don’t fear liability or the possibility of committing a medical error in a very public setting. It is something more varied and subtle that has become clearer to me over the years with the (many) new patients I was called to treat. Recently I’ve reflected on these situations and considered how my professional identity has been affected. During a much-needed night off from my medical internship, I was at a cocktail party when the host called me to help a middle-aged man who was having chest pain. In the adjacent room, a throng of people had formed around the patient. My history taking was brief, the examination even briefer. He was diaphoretic and anxious. One of the man’s friends was kneeling beside him, intently gripping both of his thumbs and reassuring him that she was helping to stop a heart attack. After I took his radial pulse to assess its rate and regularity and to estimate his systolic blood pressure, I asked the host to find some aspirin and call for an ambulance, which arrived 15 minutes later. Crouching next to him and looking into his frightened eyes, I realized there was a stark contrast between how I had been caring for patients in the hospital. In the hospital, I had been shielded from uncertainty and powerlessness by my stat medications, labs, radiological studies, even my white coat. When called to help outside the hospital, I was stripped of the usual tools, technology, resources, and an existing prior relationship to sustain the care. Instead, I had to rely primarily on those rudimentary skills of doctoring, of careful observation, basic physical examination skills, and simple communication. On top of this was what felt like a blurring of my professional boundaries—dressed “unprofessionally” and introducing myself simply as David. When I had attended that cocktail party, I had already cared for similarly sick CCU patients and thought I felt comfortable dealing with a patient’s death in a professional manner. But I began to realize how this changing boundary altered how I subsequently responded to, and reflected on, these types of “public” patients. It was this “out-of-hospital” experience that transported me back to my intense pain and sadness when my father died 4 weeks after I started medical school. Shortly after the patient left in the ambulance and as I waited in the lengthening line for the buffet dinner, I found myself washed over with grief and beginning to sob. As I had been trained, I provided competent, efficient medical care in the face of significant uncertainty. But this alien patient-care environment lowered my intellectual defense to confronting the sometimes halting fear and pain of caring for critically ill patients. On the commuter railway that takes me between my home and my work in the city, I have responded numerous times to the overhead call for assistance to a passenger in distress. Even though I no longer routinely manage emergencies like seizures, I reason that when it happens outside the hospital, better that I am there rather than someone with no or less medical training. But more often than not, the problem is not nearly as dire, and it is obvious how I could be helpful once I arrive and assess the patient. In many cases, I find an utterly medically stable patient. In those cases I have the opportunity to provide confident reassurance while we wait together for fully equipped emergency personnel. In assessing the patients, I feel their forehead; I take their pulse and observe if they are frightened, I hold their shoulder, and I reassure them that they will be okay. It’s easy to conclude that I’ve provided little actual medical care. But care is measured and perceived by patients and others in memorable ways. For me, people’s gratitude for my service has been enormously rewarding—from the patient, of course, but also from the healthy passengers who heartily thanked me for the care I’d provided. Are they grateful that someone will help if they ever fall ill in public? On one of these busy commuter trains, I found myself entangled in the dilemma of patients who decline recommended care conflicting with the wants of other passengers. After one of my medical evaluations, the patient declined to accept the ambulance transport off the train and, in doing so, delayed the train for other passengers. In working with this patient, I tried to provide her with the best medical advice I could given my limited information. And in this case, my primary recommendation was for her to accept ambulance transport for further evaluation at the hospital. But once I made my recommendation and she chose to remain on board for her own reasons, I saw as my obligation to support her in her choice and do my best to advocate for her in a way she could accept. This patient’s choice was now inconveniencing other passengers because ambulance personnel took 20 minutes to evaluate and treat her on board the stopped train, rather than transporting her to the hospital. The other passengers began to yell at the patient and pressure her to get off the train. Here was a case, and a public one at that, of what felt like the physician’s dual responsibility to the public and to an individual patient. A PIECE OF MY MIND