Emergency Medicine

Refusal of emergency medical care: An analysis of patients who left without being seen, eloped, and left against medical advice

a b s t r a c t

Objective: Emergency department (ED) patients may elect to refuse any aspect of medical care. They may leave prior to physician evaluation, elope during treatment, or leave against medical advice during treat- ment. This study was undertaken to identify patient perspectives and Reasons for refusal of care.

Methods: This prospective study was conducted at an urban Level 1 Trauma Center. This study examined ED patients who Left without being seen , eloped during treatment, or left against medical advice during September to December 2018. This project included both chart review and a prospective patient survey.

Results: Among 298 participants, the majority were female (54%). Most participants were White (61%) or African American (36%). Thirty-eight percent of participants left against medical advice, 23% eloped, and 39% left without being seen by a provider. When compared to the general ED population, patients who refused care were significantly younger (p < 0.001). When comparing by groups, patients who left AMA were significantly older than those who eloped or left without being seen (p < 0.001). Among 68 patients interviewed by telephone, the most common stated reasons for refusal of care includED wait time (23%), unmet expectations (23%), and negative interactions with ED staff (15%).

Conclusion: ED patients who refused care were significantly younger than the general ED population. Common reasons cited by patients for refusal of care included wait time, unmet expectations, and neg- ative interactions with ED staff.

(C) 2019

  1. Introduction

patient autonomy is a valued patient right, to make individual choices regarding their medical care [1-8]. As a fundamental aspect of autonomy, patients may elect to refuse aspects of medical care. Patients may choose to leave prior to physician evaluation, refuse hospital admission, refuse certain specific tests or therapies, or refuse all treatments. Against medical advice discharges from the emergency department (ED) have been estimated at approxi- mately 3% of ED visits [9]. Patients who leave against medical advice have a higher rate of return ED visits and higher mortality [10,11].

Previous studies have identified risk factors for refusal of care that include alcohol use [12], substance abuse, trauma, younger age, male, mental health comorbidities, homelessness, and lack of medical insurance [13-15].

* Corresponding author at: Department of Emergency Medicine, Wright State University Boonshoft School of Medicine, 3525 Southern Boulevard, Kettering, OH 45429, United States.

E-mail address: [email protected] (C.A. Marco).

The process of refusal of care should include the determination of decisional capacity, delivery of information, including risks of refusing treatment and potential alternative treatments, and docu- mentation of these elements.

This study was undertaken to identify characteristics of patients who refuse Emergency medical care, compared to the ED popula- tion, and to identify cited reasons for refusal of care.

  1. Methods
    1. Study design

This study was approved by the Wright State University Institu- tional Review Board. In this prospective cohort study, descriptive characteristics of patients who left the ED without completing treatment were identified by medical record review. A prospective telephone survey was also administered to identify reasons for refusal of care. Patients who left the ED without completing treat- ment from September 2018 – December 2018 were categorized into 3 groups: (1) patients who left against medical advice


0735-6757/(C) 2019

(AMA), (2) patients who left before completing treatment (eloped), and (3) patients who left without being seen (LWBS).

    1. Study setting and population

This study was undertaken at Miami Valley Hospital, a level 1 trauma center in Dayton, Ohio, with approximately 90,000 patients annually. The ED utilizes the Emergency Severity Index for triaging patients, which ranks acuity using 5 levels, with 1 being most urgent and 5 being non-urgent. Multiple studies have demonstrated this 5 level ESI to be reliable and valid [16-18].

Inclusion criteria included ED patients aged 18 and over who left the ED AMA, eloped during treatment, or LWBS from Septem- ber 2018 – December 2018. Exclusion criteria for participation in this study were patients under age 18. Exclusion criteria for the prospective telephone survey also included those who declined to participate.

    1. Study protocol

Patients who left the ED without completing treatment (AMA, eloped, or LWBS) between September 2018 and December 2018 were identified from the daily EPIC ED census. Data were extracted from the electronic medical record, including patient demographic information (age, sex, and ethnicity) and patient health history including history of anxiety or other mental health problems, alco- hol use, and illicit drug use, day of the week, mode of arrival, arrival and departure times, chief complaint, triage score (ESI scale of 1- 5), vital signs, triage pain score, AMA category (left AMA, eloped during treatment, or LWBS), alcohol and toxicology screening results, and provider documentation of assessment of capacity, delivery of information, voluntary decision, and documented rea- son for refusal of care. Patients who refused care were compared to the general ED population during the study period using a uni- variate analysis.

Patients were then contacted by telephone to obtain verbal con- sent to conduct a brief interview over the phone consisting of the following questions: (1) ”Why did you leave the ED prior to com- pleting treatment?”, (2) ”Was there anything that might have con- vinced you to agree to treatment?”, and (3) ”Do you have any comments about treatment in the emergency department?” Patients were attempted to be contacted by telephone a maximum of 3 times.

    1. Outcome measures

The primary outcomes were demographic and characteristics of patients who left the ED AMA, eloped during treatment, or LWBS, compared to the ED population, and stated reasons for refusal of care.

    1. Data analysis

Data were analyzed with SAS v9.4 (Cary, NC). Categorical vari- ables were presented as N (%) and a Chi-square or Fisher’s Exact test was used to test for an overall association between groups (left AMA, eloped, LWBS) and the variable characteristic. Continuous variables were presented as median and interquartile range. Krus- kal Wallis two-tailed test was used to test for an overall difference in the variable between groups. For multiple comparisons, if the overall test was significant (p < 0.05), groups were then compared two at a time.

  1. Results
    1. Descriptive characteristics

Two hundred ninety-eight eligible participants were identified (1% of ED patients during the study period). The median age for all three AMA categories was 36 (IQR [27.0, 49.0]). Most were female (N = 160; 54%). The majority of patients were White (N = 181, 61%), followed by African American (N = 105, 36%), His-

panic (N = 3, 1%), Asian (N = 1, 0.3%), and other (N = 5, 2%). The most common category of refusal of care was left without being seen (N = 115; 39%), left against medical advice (N – 113; 38%),

and elopement (N = 70; 24%) (Table 1).

    1. Comparison of patient groups: Left AMA, eloped, or left without being seen

The three groups of patients who left the ED prior to completing treatment were compared (Table 2). The median age of patients who left AMA (40, IQR [32, 56]) was significantly higher than the eloped (35, IQR [28,45]) and left without being Seen (33, IQR [26,

Table 1

Descriptive statistics for 298 patients who left the ED prior to completing treatment.

n (%) or median [IQR]

No. Patients 298

Age (median, IQR) 36.0 [27.0, 49.0]



138 (46.3%)


160 (53.7%)


African American

105 (35.6%)


1 (0.3%)


181 (61.4%)


3 (1.0%)


0 (0%)


5 (1.7%)

Triage Level


1 (0.4%)


29 (10.1%)


168 (58.5%)


75 (26.1%)


14 (4.9%)

Triage Vital Signs

BP (systolic, median, IQR)

136 [122, 150]

HR (median, IQR)

90 [77.5, 104.5]

RR (median, IQR)

18 [16, 18]

Temp (median, IQR)

98.2 [97.9, 98.6]

Pain Score (median, IQR)

7 [3, 10]

AMA Category

Left AMA

113 (37.9%)


70 (23.5%)

Left without being seen

115 (38.6%)


109 (36.8%)

cocaine use

40 (13.5%)

opiate use

65 (22.0%)

Benzodiazepine Use

10 (3.4%)

Methamphetamine Use

36 (12.2%)

marijuana use

87 (29.4%)

Other Drugs Used

14 (4.7%)

ED Documentation Assessment of Capacity

92 (31.0%)

Delivery of Information (risks, benefits, alternatives)

97 (32.7%)

Voluntary Decision

116 (38.9%)

ED Documentation for 113 patients who left AMA

Assessment of capacity

70 (62%)

Delivery of information

78 (69%)

Voluntary decision

74 (66%)

Table 2

Comparison of patient groups: Left AMA, eloped or left without being seen.

Left AMA


Left without being Seen

Overall P-value Group Comparison

No. Patients




Age (median, IQR)

40 [32, 56]

35 [28, 45]

33 [26, 42]


Gender Male

61 (54.0%)

31 (44.3%)

46 (40.0%)



52 (46.0%)

39 (55.7%)

69 (60.0%)



African American

33 (29.5%)

26 (37.7%)

46 (40.4%)


77 (68.8%)

40 (58.0%)

64 (56.1%)


2 (1.8%)

3 (4.4%)

4 (3.5%)

Mode of ED Arrival Walk-in

69 (63.3%)

54 (79.4%)

92 (86.8%)



40 (36.7%)

14 (20.6%)

14 (13.2%)

Triage Level 1 & 2

21 (19.3%)

6 (8.8%)

3 (2.7%)



68 (62.4%)

36 (52.9%)

64 (58.2%)

4 & 5

20 (18.4%)

26 (38.3%)

43 (39.1%)

Triage BP (median, IQR)

138 [125, 151]

135 [122, 151]

135 [121, 150]


Triage RR (median, IQR)

18 [16, 18]

18 [16, 18]

18 [16, 18]


Triage Pain Score (median, IQR)

7 [2, 10]

8 [5, 10]

7 [3, 9]



44 (39.3%)

27 (38.6%)

38 (33.3%)


Cocaine Use

14 (12.4%)

7 (10.0%)

19 (16.8%)


Opiate Use

29 (25.7%)

15 (21.4%)

21 (18.6%)


Benzodiazepines Use

3 (2.7%)

2 (2.9%)

5 (4.4%)


Methamphetamine Use

17 (15.0%)

6 (8.6%)

13 (11.5%)


Marijuana Use

33 (29.2%)

21 (30.0%)

33 (29.2%)


Categorical variables tested with Chi-square or *Fisher’s Exact test. Continuous or interval scaled variables tested with Kruskal-Wallis 2-tailed test.

42]) categories (P < 0.001). Patients who left AMA had significantly more ambulance arrivals (40, 36.7%) than patients who left with- out being seen (14, 13.2%) (P < 0.001). Patients who left AMA were more likely to have lower triage Acuity levels (Fig. 1). There were no differences between groups by gender, ethnicity, triage BP, triage RR, triage pain score, ETOH use, cocaine use, opiate use, ben- zodiazepine use, methamphetamine use, or marijuana use.

    1. Comparing demographics of all three AMA group patients to control ED patient population

When compared to the general ED population (patients who did not leave AMA, LWBS, or eloped during the study period), patients who left the ED prior to completing treatment were significantly

younger (mean 39, compared to 46.3, p < 0.001) There were no dif- ferences by sex, ethnicity, or mode of ED arrival (Table 3).

    1. Patient interview results

A total of 68 patients (23%) in the dataset responded to the tele- phone survey after their ED visit. To assess the potential for selec- tion bias, we compared demographic and other patient characteristics between those who did and did not complete the telephone survey. Univariate Chi-square, Fisher’s Exact, or Mann Whitney Wilcoxon p-values < 0.05 were used to indicate signifi- cance differences. The survey participants were less likely to have tested positive for opiates or meth, more likely to be African Amer- ican, and were older. While 4% of the patients who completed the

Percent Distrubution of Patient Triage Levels among AMA Patient Groups


















Left AMA Eloped Left Without being Seen Triage Level 1-2 Triage Level 3 Triage Level 4-5

Fig. 1. Percent distribution of patient triage levels among AMA patient groups.

Table 3

Comparing demographics of all AMA-eloped patients to controls (all ED patients Sept 2018 to Dec 2018 who were not AMA-eloped).

n (%) or mean (n, SD) p-




N — 298

N = 31,294

Age (years)

39.0 (n = 298, SD

46.3 (n = 30543, SD






138 (46.3%)

13,605 (43.5%)



160 (53.7%)

17,689 (56.5%)



105 (35.6%)

12,573 (40.8%)




181 (61.4%)

17,694 (57.4%)


9 (3.1%)

562 (1.8%)

Mode of ED



215 (76.0%)

21,476 (71.7%)



68 (24.0%)

8476 (28.3%)

survey had tested positive for opiates, 27% of those who did not complete the survey had tested positive for opiates, (p < 0.001). While 3% of the patients who completed the survey had tested pos- itive for methamphetamine, 15% of those who did not complete the survey had tested positive for meth, p = 0.01. More participants were African American (51%), compared to 32% of those who did not complete the survey (p = 0.004). The median age of the patients who completed the survey was 42 years compared to 35 years among non-completers (p = 0.003). No differences were detected with respect to sex, mode of arrival, triage level, ETOH, cocaine, marijuana, benzodiazepine use, BP nor pain.

The median number of days following the ED visit to telephone contact of the patient was 30 (IQR [27, 35]). Many patients stated at least one reason for refusal of care. The most reported reasons included wait time (23%), unmet expectations (23%), and negative interactions with ED staff (15%) (Table 4). Sample verbatim com- ments illustrated several patient perspectives (Table 5).

  1. Discussion

Patients who refused ED care cited three common reasons, including wait time, expectations not met, and negative interaction with ED staff. These findings confirm previous studies which found that a long wait time is the top reason patients LWBS [19-21]. ED wait times vary significantly and are related to triage acuity. Because many patients in this study had lower triage acuity, they may have experienced higher wait times. One patient stated, ”I know next time if I want to be seen quickly, I should just call an ambu- lance because that’s the only way they won’t make you wait.”

Table 4

Survey of 68 participants: ”Why did you leave the ED prior to completing treatment?”

Reason N %

Wait Time 26 23

Expectations not met 25 23

Negative interaction with ED staff



Did not want to be admitted



Did not want recommended treatment/tests 11 10

Family obligations 11 10

Improved symptoms 9 8

Other obligations 2 2

Other 15 14

Table 5

Sample verbatim comments from patients who refused ED care.*

I was treated like a drug addict. I would never return. I felt better. I just wanted to lie down.

They wouldn’t give me water so I left. I was scared.

People with drug related issues should not be treated differently. I would have stayed but I had things to do.

All of the staff were on their phones; I did not like it.

The Emergency Department is slow and they do not care about patients. Just needed to leave due to another emergency.

* Including patients who eloped, left without being seen, and left against medical advice.

The second most frequently stated reason patients left was patient expectations not being met. This commonly included patients feeling like their pain was not being adequately treated, with multiple instances of patients being denied narcotic pain medication after requesting it. Pain the most frequent ED chief complaint [22,23]. Patients have high expectations when it comes to pain treatment. One study of ED patients found that patients reported an average expectation for pain relief of 72% while 18% of patients expected 100% relief of their pain [24].

Negative interactions with ED staff was another common rea- son for refusal of care. The negative interactions with ED staff included rude staff, including nurses and a resident physician, or wanting the staff to be more caring and concerned about their Medical issues. Previous studies have demonstrated that negative and patronizing attitudes, impatience, a lack of understanding from medical staff, and insufficient explanations to patients con- tribute to the development of negative attitudes in patients [25]. These negative attitudes developed by patients, in turn, contribute to their decision to leave without completing treatment.

Another commonly cited reason was not wanting to be admit- ted to the hospital. These results confirm previous studies which note higher short-term recidivism rates (within 30 days) than those who complete their care in the ED [26].

Family obligations included issues such as family transporta- tion, family emergencies, and unspecified prior family commit- ments. The observed rate of patients leaving due to family obligations is similar to a previous study about hospitalized patients with asthma who left AMA [27].

Patients not wanting the recommendED treatments/tests included not wanting or being afraid of having surgery performed, not wanting to have imaging performed, and not seeing the benefit of certain procedures (e.g. receiving IV fluids). Improvement of symptoms was another frequent reason with a rate similar those found in previous studies, which also speculated that this could be due to these patients presenting to the ED with self-limited or relatively Benign conditions [21].

Appropriate documentation of refusal of medical care is impor- tant to ensure the appropriate assessment of capacity, delivery of information, and voluntary decision by the patient. Documenting proper elements of informed consent and informed refusal of care provides some liability protection for ensuing litigation as well as aiding ethically to ensure physician’s respect their patient’s auton- omy [28-32]. In this study, we found that documentation was vari- able. Sixty two percent of charts documented assessment of capacity, 69% documented delivery of information, and 66% docu- mented the voluntary decision. It is uncertain whether lack of doc- umentation reflects inadequate communication, or inadequate documentation of the encounter. Because patients were not quer- ied about assessment of capacity, information delivery or a volun- tary decision during the phone interview, it is uncertain whether the medical record accurately reflects the elements of assessment of capacity.

  1. Conclusions

ED patients who refused care were significantly younger than the general ED population. Common reasons cited by patients for refusal of care included wait time, unmet expectations, and nega- tive interactions with ED staff.


The authors wish to thank Nancy Buderer, MS, for her statistical expertise in the data analysis for this project.


  1. Prentice ED. Informed consent: the most important protector. Acad Emerg Med 1999;6:774-5.
  2. Moskop JC. Informed consent in the emergency department. Em Med Clin N Am 1999;17:327-40.
  3. Hansson MO. Balancing the quality of consent. J Med Ethics 1998;24:182-7.
  4. Young Jr WF. Informed consent. Ann Emerg Med 1997;30:350-1.
  5. Etchells E, Sharpe G, Walsh P, et al. Bioethics for clinicians: consent. CMAJ 1996;155:177-80.
  6. Faden RR, Beauchamp TL, King NMP. A history and theory of informed consent. New York: Oxford University Press; 1998.
  7. Marco CA, Brenner J, Krauss C, McGrath N, Derse A. Refusal of emergency medical treatment: case studies and ethical foundations. Ann Emerg Med 2017 Nov;70(5):696-703. PMID:28559033.
  8. Simon JR. Against medical advice, refusal of care, and informed consent. In: Marco CA, Schears RM, editors. Ethical dilemmas in emergency medicine. Cambridge University Press; 2015.
  9. Choung AL, Cho JP, Choi SC, et al. Patients who leave the Emergency Department against medical advice. Clin Exp Emerg Med 2016;3:88-94.
  10. Gabayan GZ, Asch SM, Hsia RY, Zingmond D, Liang LJ, Han W, et al. Factors associated with short-term bounce-back admissions after emergency department discharge. Ann Emerg Med 2013;62(2):136-144.e1.
  11. Geirsson OP, Gunnarsdottir OS, Baldursson J, Hrafnkelsson B, Rafnsson V. Risk of repeat visits, hospitalisation and death after uncompleted and completed visits to the emergency department: a prospective observation study. Emerg Med J 2013;30(8):662-8.
  12. Joo Jeong, Song KJ, Kim YJ, et al. The association between acute alcohol intoxication and discharge against medical advice of injured patients in the ED. Am J Emerg Med 2016;34:464-8.
  13. Yong TY, Fok JS, Hakendorf P, et al. Characteristics and outcomes of discharges against medical advice among hospitalized patients. Int Med J 2013:798-802.
  14. de Guise E, LeBlanc J, Dagher J, Tinawi S, Lamoureux J, Marcoux J, et al. Characteristics of patients with acute traumatic brain injury discharged

against medical advice in a level 1 urban trauma centre. Brain Inj 2014;28 (10):1288-94.

  1. Choi Namkee G, DiNitto Diana M, Nathan Marti C, Choi Bryan Y. Substance use disorders with presenting problems and outcomes in older adults’ emergency department visits. Acad Emerg Med 2015;22:1316-26.
  2. Eitel DR, Travers DA, Rosenau AM, Gilboy N, Wuerz RC. The Emergency Severity Index triage algorithm version 2 is reliable and valid. Acad Emerg Med 2003;10(10):1070-80.
  3. Tanabe P, Gimbel R, Yarnold PR, Kyriacou DN, Adams JG. Reliability and validity of scores on the emergency severity index version 3. Acad Emerg Med 2004;11(1):59-65.
  4. Hocker M, Gerardo C, Theiling J, et al. NHAMCS validation of emergency severity index as an indicator of emergency department resource utilization. West J Emerg Med 2018;19(5):855-62.
  5. Fraser J, Atkinson P, Gedmintas A, Howlett M, McCloskey R, French J. A comparative study of patient characteristics, opinions, and outcomes, for patients who leave the emergency department before medical assessment. Can J Emerg Med 2017;19(5):347-54.
  6. Clarey AJ, Cooke MW. Patients who leave emergency departments without being seen: literature review and English data analysis. Emerg Med J 2012;29:617-21.
  7. Johnson M, Myers S, Wineholt J, Pollack M, Kusmiesz AL. Patients who leave the emergency department without being seen. J Emerg Nurs 2009;35:105-8.
  8. Hawkins SC, Smeeks F, Hamel J. Emergency management of chronic pain and drug-seeking behavior: an alternate perspective. J Emerg Med 2008;34:125-9.
  9. Ducharme J. emergency pain management: a Canadian Association of emergency Physicians (CAEP) consensus document. J Emerg Med 1994;12:855-66.
  10. Fosnocht DE, Heaps ND, Swanson ER. Patient expectations for pain relief in the ED. Am J Emerg Med 2004;22(4):286-8.
  11. Landau SF, Bendalak J, Amitay G, Marcus O. Factors related to negative feelings experienced by emergency department patients and accompanying persons: an Israeli study. Isr J Health Policy Res 2018;7:6.
  12. Ding R, Jung JJ, Kirsch TD, Levy F, McCarthy ML. Uncompleted emergency department care: patients who leave against medical advice. Acad Emerg Med 2007;14:870-6.
  13. Baptist AP, Warrier I, Arora R, Ager J, Massanari RM. Hospitalized patients with asthma who leave against medical advice: characteristics, reasons, and outcomes. J Allergy Clin Immunol 2007;119:924-9.
  14. Yu KT, Green RA. Critical aspects of emergency department documentation and communication. Emerg Med Clin North Am. 2009;27(4):641-54 [ix].
  15. Levy F, Mareineiss DP, Lacovelli C. The importance of a proper against- medical-advice (AMA) discharge: how signing out AMA may create significant liability protection for providers. J Emerg Med 2012 Sep;43(3):516-20.
  16. Devitt PJ, Devitt AC, Dewan M. An examination of whether discharging patients against medical advice protects physicians from malpractice charges. Psych Serv 2000 Jul;51(7):899-902.
  17. Devitt PJ, Devitt AC, Dewan M. Does identifying a discharge as ”against medical advice” confer legal protection?. J Fam Pract 2000;49(3):224-7.
  18. Berger JT. Discharge against medical advice: ethical considerations and professional obligations. J Hosp Med 2008;3(5):403-8.

Leave a Reply

Your email address will not be published. Required fields are marked *