Providers’ Perceptions of Relationships and Professional Roles when Caring for Patients who Leave the Hospital Against Medical Advice
- 180 Downloads
Patients who leave hospitals against medical advice (AMA) may be at risk for adverse health outcomes. Their decision to leave may not be clearly understood by providers. This study explored providers’ experiences with and attitudes toward patients who leave the hospital AMA.
To explore providers’ experiences with and attitudes toward patients who leave the hospital AMA.
We conducted interviews with university-based internal medicine residents and practicing internal medicine clinicians caring for patients at a community hospital from July 2006 to August 2007. We approached 34 providers within 3 days of discharging a patient AMA. The semi-structured instrument elicited perceptions of care, emotions, and challenges faced when caring for patients who leave AMA. Using an editing analysis style, investigators independently coded transcripts, agreeing on the coding template and its application.
All 34 providers (100%) participated. Providers averaged 32.6 years of age, 22 (61%) were men, 20 (59%) were housestaff from three residency programs, 13 (38%) were faculty, hospitalist physicians, or chief residents serving as ward attendings, and one (3%) was a physician assistant.
Four themes emerged: 1) providers’ beliefs that patients lack insight into their medical conditions; 2) suboptimal communication, mistrust, and conflict; 3) providers’ attempts to empathize with patients’ concerns; and 4) providers’ professional roles and obligations toward patients who leave AMA.
Our study revealed that patients who leave AMA influence providers’ perceptions of their patients’ insight, and their own patient–provider communication, empathy for patients, and professional roles and obligations. Future research should investigate educational interventions to optimize patient-centered communication and support providers in their decisional conflicts when these challenging patient–provider discussions occur.
KEY WORDShospitalization patient discharge patient–provider communication difficult patient professionalism
Patients go to hospitals to receive care for illnesses they cannot manage at home. Appropriate inpatient care requires that providers understand and treat each patient’s presenting complaints and address comorbidities, health behaviors, and psychosocial risk factors that may contribute to poor outcomes. In their report Crossing the Quality Chasm, the Institute of Medicine described optimal inpatient care as patient-centered: “providing care that is respectful of and responsive to individual patient preferences, needs, and values, and ensuring that patient values guide all clinical decisions.”1
One in every 70 hospital discharges occurs against medical advice (AMA).2 Patients who leave AMA may be at risk for adverse health outcomes leading to more frequent hospital readmissions and increased morbidity and mortality.3, 4, 5, 6 Prior retrospective chart review studies have found that the likelihood of AMA discharges is associated with hospital factors such as size and location2 and patient factors such as mental illness,4,7 substance abuse,4,5,7, 8, 9 gender,2,5,7, 8, 9 income,2,5 lack of medical insurance,8,9 whether the individual is a Medicaid recipient,2,9 race,2,8, 9, 10 and young age2,5,7,9,11. One chart review found that providers document patients’ reasons for leaving AMA as personal issues, financial problems, and legal matters.12
A patient’s choice to leave AMA may represent a breakdown in communication, which in turn relates to a variety of patient, provider, and relationship factors.13, 14, 15 In addition, this decision may create conflict for providers in balancing their ethical obligations: respecting patient autonomy and self-determination, and protecting the patient from harm.16 While some information is known about patient factors associated with leaving AMA, no studies have assessed providers’ perspectives on their interactions with these patients. Since providers may have insight into the events leading to premature departure, our research focused on providers’: 1) actions when they found out a patient wished to leave prematurely; 2) feelings regarding the experience; and 3) lessons learned.
Study Design and Sample
We conducted one-on-one interviews to investigate providers’ experiences with and perspectives regarding patients who leave the hospital AMA. We chose a qualitative methodology, which allows greater in-depth exploration and emergence of themes unanticipated by the investigators than is possible using closed-ended questionnaires.17 We also attempted to telephone patients within 7 days of an AMA discharge to learn their reasons for leaving and perspectives on care. Although multiple attempts were made, we reached only four patients, which we deemed an insufficient number to analyze.
Characteristics of Patients Who Chose to Leave Against Medical Advice from a Connecticut Hospital Compared with Regional and National Characteristics*
Study patient demographics
Mean age, years
Male, n (%)
Race, n (%)
High school or greater education, %
Marital status (married), %
Employment rate, %
Median per capita income, $
From July 2006 to August 2007, we examined daily hospital discharge logs for patients who left AMA. We contacted each AMA patient’s medical providers within three days of discharge in order to obtain the most accurate account of experiences and emotions. One provider had two AMA patients but only the interview about the first patient was included in our analysis. Sampling was discontinued when thematic saturation was achieved.17
We obtained written consent from each participant. The Waterbury Hospital Institutional Review Board and the Yale University Human Investigation Committee approved the study protocol.
One investigator (DMW) conducted audiotaped, semi-structured interviews with consenting respondents in a private setting inside the hospital. Each interview began with the participant describing the medical circumstances surrounding the patient’s admission to the hospital. The interview instrument was in part derived from the Picker Patient Experience Questionnaire,19 a survey instrument with high construct validity and internal consistency designed to examine patients’ hospital experiences. We included the 10 items deemed most pertinent to communication with patients who choose to leave AMA (Appendix). The interviewer used reflective probes to encourage respondents to clarify and expand on their statements. We obtained patient demographic characteristics by chart review.
We analyzed transcripts using an “editing analysis style,” wherein the analysts seek meaningful segments of data to create codes, deriving the coding template from the data itself.17 The two study investigators independently analyzed six randomly selected transcripts to generate a list of codes summarizing the subjects’ statements. Each code could occur more than once in a transcript, and each quotation could be associated with more than one code. Investigators compared codes and negotiated discrepancies to create a consensus coding template, discussing and agreeing on the application of final coding categories. We independently applied this template to all remaining transcripts. Coder agreement was high (κ = 0.97). Themes presented were chosen by consensus based on the most frequent codes. We reviewed all available quotations for associated themes and chose representative quotations with the aim of presenting viewpoints from all provider types. To help establish validity of our themes, a random sample of 50% of provider respondents reviewed our data and agreed with our summary.
We managed and analyzed data using Atlas.ti 5.0 software (Atlas.ti GmbH, Berlin, Germany, 2005), a qualitative analysis software developed from a type of editing analysis style called the grounded theory approach.
Patients who left AMA were more likely to be older, male, unemployed, and uninsured compared to regional and national demographics (Table 1). Many patients smoked (75%), and some were involved with illicit drugs (38%).
Characteristics of 34 Physicians Interviewed Regarding Patients Who Choose to Leave Against Medical Advice from a Connecticut Hospital
Age, mean (SD)
Men, n (%)
Provider description, n (%)
Preliminary intern, primary care program
Intern, primary care program
Resident, primary care program
Resident, medicine/pediatrics program
Resident, traditional internal medicine program
Chief resident serving as ward attending
Providers’ Quantitative Characterizations of AMA Interactions
Summary of Actions Taken or Assessment of Patients by 34 Providers whose Patients Left the Hospital Against Medical Advice
# of providers
Patient asked questions about their condition or treatment
Patient had anxieties/fears about condition or treatment
Action taken after hearing patient wanted to leave
Went to patient bedside to discuss desire to leave
Contacted another provider (colleague or supervising physician)
Found out after the fact
Contacted risk management
Patient was asked to sign a form stating they were leaving AMA
Follow-up plans were made to ensure patient’s safety
Illness of patient at time of leaving
Treated patient with dignity and respect
Would change something that happened with the patient if could
Would change something in care of other patients given this experience
Results of Qualitative Analysis
Code Count from an Interview Study of 34 Providers Caring for Patients who Left Against Medical Advice from a Connecticut Hospital
# of codes
# of respondents
Patients’ lack of insight about their medical conditions
Providing medical information
Limited or no insight about medical conditions
Suboptimal communication, mistrust, and conflict
Withholding information: I just want to go home
Suspicion about true movies/substance abuse
Providers’ attempts to empathize with patients’ concerns
Upset about care
Communicating empathy and respect to patients
Spending time with patient
Showing respect and caring
Eliciting patient concerns
Negotiating to meet other concerns
Providers’ professional roles and obligations towards patients who leave AMA
Did all I could do
Theme 1: Providers’ beliefs that patients lack insight into their medical conditions
Regardless of patients’ stated motivations, providers felt that patients’ decisions to leave AMA demonstrated lack of comprehension about the dangers of their current medical conditions. This left providers attempting to provide their patients with medical information in hopes of convincing them to stay.
Lack of Insight
He did not seem to make the connection that dealing with his active inpatient issues was an important part of his recovery from his illness.
I don’t think that the patient ever really wanted to be in the hospital. It is almost like he thought that the hospital was like fast food. You are hungry and you go and get a meal really quickly, and then you get out. That is pretty much it.
I do not think that he had any real understanding of how sick he was. I think that, in his mind, what he was saying was true. He felt better. He was eating. His strength was improved, and there was no reason for him to stay there.
Providing Medical Information
Providing medical information to patients was a common strategy for trying to convince patients to stay in the hospital. Providers gave patients detailed medical information regarding the dangers of leaving, including health risks of deferring treatment, need for further diagnostic testing, and potential for serious illness or death.
I talked about the risks of him going home. Not being monitored by medical personnel. We talked about the risk of him being home with two young children and no adults. The risk of him going into a seizure and not having anyone to help him, especially when his medications are not therapeutic.
Theme 2: Suboptimal Communication, Mistrust, and Conflict
Overall, providers felt that the quality of provider–patient communication was suboptimal, either due to resistance on the patient’s part or miscommunication by providers.
Withholding Information: “I Just Want to Go Home.”
All she said was that she wanted to go home and sleep in her own bed.
He mentioned that he was just tired of being in the hospital. He did not give any particular irritating or inciting problems for those feelings. He said just very nonspecifically a phrase like, “I have things to do.” There was no particular thing that seemed to be bothering him.
Suspicion about True Motives
My thought is that he may not have been interested in doing the rehabilitation, as well as his need to do more substance abuse, which he cannot do in a hospital. When the social worker talked to him, he had no interest in quitting his cocaine or heroin....
I think that is a question that we as housestaff ask ourselves every time: ‘Why did this patient want to leave?’ Was he withdrawing from alcohol? He knows that if he goes outside that he will either drink again or can potentially get into serious problems with withdrawal.
[The patient] actively disagreed when you suggested that anything else might be wrong.
People were telling him that he had colitis, but he said, ‘Well, no, I do not have any diarrhea. They are diagnosing me wrong’.... [I replied,] ‘We are not saying that you have diarrhea, it is just that is what it is termed and so we need to investigate that.’
I felt just frustrated and taken advantage of. From the beginning when I heard her story it just seemed like this was a woman who was drug-seeking and then, after I met her and examined her, she seemed so fine. I really felt like she was trying to manipulate the system, and I resented it.
Theme 3: Providers’ Attempts to Empathize with Patients’ Concerns
Some providers expressed empathy for their patients’ feelings and decisions.
Providers acknowledged that some patients may have competing concerns that they prioritized over their health.
He stayed with his sister with two young children.... He felt that it was his responsibility to go back home and take care of those kids so that his sister could go back to work.
She said that she has some personal issues to take care of, and there was an issue with one of her children and court, and then her grandchildren that she takes care of, and she really needed to get home.
Other providers felt that patients chose to leave because of emotional distress with their medical conditions.
He definitely had some fears that he discussed even on the last day. He had multiple friends who had died from HIV-related complications.
Upset About Care
Some providers expressed understanding of why patients might have felt anger or frustration with the care they received in the hospital.
I think that he got frustrated with sitting in a hospital bed and not much was changing, and waiting for something to happen, and not knowing what decisions, advice, or suggestions were being debated among his caregivers.
Communicating Empathy and Respect to Patients
The capacity to understand patients’ competing concerns and emotional reactions to their health allowed some providers to try different approaches to discussing patients’ decisions to leave AMA.
Foremost, I spent a great deal of time with the person. I spent an hour with the patient. I took the time to inquire why the patient wanted to leave the hospital.
If you put in enough effort and communicate in an open, caring fashion, they tend to listen and respect your opinion.
Most of the time, it requires sitting down with someone and identifying their concerns.
Mainly, I started with an opening question to see what was on his mind.... What is the reasoning around his wanting to leave?
I try to sit down and talk to them and try to understand what their frustrations are. I try to get to the core of why they want to leave AMA. I offer them support and try to get social work involved. If necessary, I try to do some give and take. Sometimes the patients may want something as simple as a pass to just leave the floor to see the light of day. Sometimes you can get someone to assist them to go down. Just simple things like that.
Theme 4: Providers’ Professional Roles and Obligations toward Patients Who Leave AMA
Providers reflected how their experiences with patients leaving AMA made them reconsider their professional roles and obligations toward challenging patient relationships.
Self-Doubt and Uncertainty about Missed Opportunities
Because he had something going on but we just could not figure out what it was....
I don’t know what else, what other resources we have to talk to the patient about. It is too complicated, patient care is, and the time we spend.... I am not an expert in social services, but maybe they can spend some time.
I just wish that we were able to know that he was leaving. I wish that somehow we could have identified that he was leaving the hospital and tried to intervene.
I did not feel very good about the way that he left. I think that it was mainly because we were not there and there was a lot going on that day.
The Dilemma of Respecting Patient Autonomy
Especially in these situations, it is hard to understand a patient’s rationales because, from a care providers’ perspective, health is the most important thing, because that is what we try to address. A lot of time, patients have other circumstances in life that are more important to them than health.
I think that it was a difficult situation, a troubling situation because I acknowledge and believe the importance of giving people determination over their own decision making.
Sometimes I feel that if you say to the patient, ‘You know you are a grown up and you are free to leave, but these are the reasons why I think that you should stay and this is what is going to happen to you if you leave’ then they often calm down and make the decision to stay.
Maintaining Responsibility versus Accepting Futility
I would make more of an effort to help patients understand their condition, the implications of their condition, and the prognosis, and then allow them to make a decision on whether or not to leave the hospital.
I would identify the reason that they are in the hospital from their perspectives and try to talk about it at length so the patients do not get disillusioned.
I feel like I covered all of the bases with my approach to him, and unfortunately, regardless of our efforts, we can not make a patient do what we want them to do all of the time.
Despite all of our efforts, there is no way to ensure 100% of the people are going to stay in the hospital. I think that the lesson to this is that, no matter what we do sometimes you can’t keep everyone in-house. It is unfortunate.
Conceptual Diagram of Patient–Provider Communication about Decisions to Leave AMA
Patients who leave the hospital prior to a desired endpoint present a challenge to many providers. These challenges often evoke strong feelings, including concern for a patient’s health and safety, frustration, and ineptitude.
Four themes emerged from our study. The first involved the providers’ assessment that patients lack insight into their medical condition. To address this issue, providers often spent time discussing the patient’s medical condition and treatment options. Despite attempts at patient education, providers’ and patients’ views toward health, disease, and illness may differ fundamentally. Arthur Kleinman distinguishes between the illness — the patient’s “lived experience” — and the disease — the provider’s biomedical explanation for what is occurring.20 A patient’s meaning of illness may be shaped by their symptoms, how those symptoms translate into knowledge about themselves and the culture of their wider psychosocial environment.20 Thus, the common reaction to leaving AMA — offering medical information to raise the patient’s awareness about their disease — may be insufficient by itself.
In our second theme, we found that suboptimal communication, mistrust, and conflict emerged as potential reasons patients left AMA. A study evaluating views of hospitalization by patients and providers found little agreement between the two groups with respect to justification of hospital stay and discharge planning.21 This discrepancy was felt to be due in part to poor patient-physician communication.21 Another study to determine what patients’ value most during hospitalization established that confidence and trust in providers, treatment with respect and dignity, and adequacy of involvement in care were the most important factors.22 These studies support the importance of effective communication and relationship building skills. Governing and accreditation bodies23, 24, 25 and the Institute of Medicine1 also recognize these skills as paramount; they have also been linked to important outcomes including improved diagnostic and clinical proficiency,26,27 reduced patient emotional distress,26 and increased patient and provider satisfaction.28, 29, 30
Many providers attempted to empathize with their patients in the third theme that emerged. In doing so, they made an effort to elicit concerns and communicate in an open, caring fashion. Some providers recognized a power struggle between themselves and their patients regarding individual needs. These providers identified the importance of negotiation and working together to find common ground as keys to persuading patients to stay in the hospital. Educators recommend that providers in challenging interactions take an approach characterized by collaboration through improved partnership, and appropriate use of power, negotiation, and empathy.31,32 Opening the lines of communication earlier and often during patient interactions may have prevented many of the misconceptions providers described in our study from occurring (Fig. 1).
Our fourth theme addressed professional roles and obligations toward patients who leave AMA. The code of medical ethics,33 which emphasizes respecting the rights of patients and their right to self-determination, appears to conflict with what providers learn regarding the legality of patients who leave AMA. Literature about AMA patients often focuses on the provider’s duty to determine patient capacity and qualifications for involuntary commitment.34,35 Scholarship in this domain discusses risk management strategies, noting that improving patient-physician communication can lead to decreased patient complaints and fewer malpractice claims.36,37 Articles also articulate the importance of careful and thorough documentation of the actions taken by the patient and the provider in the sequence of events that lead up to and include leaving the hospital AMA.34 These legal and procedural themes echo loudly in providers’ thoughts as they try to balance patient autonomy, potential harm, and beneficence. This leaves some providers conflicted about their roles and obligations.
What strategies could help providers struggling with the four themes described above? Interpersonal conflicts may be remedied if providers embrace different responsibilities in their care. Carrese discusses the exploration of treatment refusal in order to better understand the patient’s beliefs, expectations, fears, and personal needs.16 He recommends exploring “religious beliefs, cultural background, various psychosocial factors, previous interactions with the health care system, influential personal experiences, or the preferences of family members or friends.”16 Beginning in the first interview, this process may reveal that patients don’t actually lack insight, but are informed about their health in different ways. If, despite this effort, a patient still desires to leave before treatment is deemed completed, Swota recommends maximizing patient autonomy through mandatory post-hospital follow-up contact.38 These two methods use a patient-centered approach to care, and move beyond conflict, and allow both parties to effectively carry out their individual roles and obligations.
Assessing Your Own Capacity: Eight Questions Providers Should Ask Themselves When a Patient Wants to Leave Against Medical Advice
Theme encountered when a patient wishes to leave AMA
Questions for self-reflection
Belief that the patient lacks insight into their medical conditions
1. Do I have full insight into this patient’s beliefs about health and illness?
2. How can I use my insight to counsel this patient in a patient-centered way?
Suboptimal communication, mistrust, or conflict
3. What factors are making it difficult for us to communicate?
What is it about:
4. How can I re-establish my trust in this patient and this patient’s trust in me?
Empathizing with the patient’s concerns
5. What are this patient’s concerns about remaining in the hospital?
6. How can I acknowledge those concerns and communicate empathy and respect for this patient?
Professional roles and obligations towards the patient who wishes to leave AMA
7. How does this patient’s desire to leave make me feel about myself?
8. How does this patient’s desire to leave make me feel about my professional responsibilities toward this patient?
Our study has several limitations. First, as a qualitative study limited to clinicians at a single institution, our findings may not apply to health care providers in other hospitals or settings. At the same time, our study design was strengthened by our high response rate and our inclusion of providers from a variety of levels. Second, our study did not include the perspectives of the patients who left AMA, their significant others, or their other care providers in the hospital. Although soliciting patients’ perspectives was an additional aim of this study, the investigators were unable to recruit sufficient numbers of participants for theoretical saturation. Given the fact that many of these challenges are interpersonal in nature, it is especially important that future research gather data from all stakeholders in the decision to leave AMA.
In conclusion, our study revealed that patients who leave AMA raise questions for providers about their patients’ level of insight, quality of communication, need for empathy and professional roles and obligations. Future research should investigate educational interventions to optimize patient-centered communication and support providers in their decisional conflicts when these challenging patient–provider discussions occur.
None. No financial disclosures.
Conflicts of Interest
- 1.Institute of Medicine. Crossing the quality chasm: a new health system for the 21st century. Washington, D.C.: National Academies Press; 2001.Google Scholar
- 7.Seaborn Moyse H, Osmun WE. Discharges against medical advice: a community hospital’s experience. Can J Rural Med. 2004;9(4):265. Fall.Google Scholar
- 17.Crabtree BF, Miller WL. Doing qualitative research, 2nd ed. Thousand Oaks, Calif.: Sage Publications; 1999.Google Scholar
- 18.Census 2000 Demographic Profiles. Available at http://www.ctnow.com/extras/census/0600900980070.pdf. Accessed May 16, 2008.
- 20.Kleinman A. The Illness Narratives: Suffering, Healing, and the Human Condition. New York: Basic Books, Inc; 1988.Google Scholar
- 23.Accreditation Council for Graduate Medical Education (ACGME). Outcome project: enhancing residency education through outcomes assessment. Available from: http://www.acgme.org/Outcome/. Accessed May 16, 2008.
- 24.Association of American Medical Colleges (AAMC). Learning objectives for medical student education—guidelines for medical schools: report 1 of the medical school objectives project. January 1998. Available from: http://www.aamc.org/meded/msop/. Accessed May 16, 2008.
- 25.Liaison Committee on Medical Education (LCME). Functions and structures of a medical school: standards for accreditation of medical education programs leading to the M.D. degree. July 2003. Available from: http://www.lcme.org/. Accessed May 16, 2008.
- 32.Hass LJ, Leiser JP, Magill MK, Sanyer ON. Management of the difficult patient. Am Fam Physician. 2005;72(10):2063–8. Nov 15.Google Scholar
- 33.AMA Principles of Medical Ethics, American Medical Association, Chicago, Illinois, 2001. http://www.cirp.org/library/statements/ama/. Accessed May 16, 2008.