Patients leaving against medical advice from the emergency department: How many of them actually return to continue their evaluation and what are the reasons the others do not?

Abstract

Listen

Translate article

Study objectives: There is little information about how many patients who leave against medical advice (AMA) return to the emergency department (ED) in the days immediately after their abrupt departure. We seek to determine the number of patients who leave AMA who actually return to the ED for resumption of further diagnostics and treatment and the reasons why they do not. Methods: A retrospective cohort analysis of patients leaving AMA during two 6-month periods in consecutive calendar years was performed at an urban academic ED. Patients leaving AMA were identified by treating staff, as well as daily medical record reviews. The study was conducted from May through October 2002 and July through December 2003. All patients were telephoned within 72 hours of their AMA departure and queried about their clinical status and their intent to return to the ED regardless of persistence, resolution, or worsening of symptoms. Data are reported as proportions and 95% confidence intervals (CIs). Approval was granted by the institutional review board. Results: A total of 199 patients were identified as having left the ED AMA. The predominant AMA discharge diagnoses, in order of predominance, included cardiovascular (acute coronary syndrome, rule out myocardial infarction, rule out pulmonary embolism, atrial fibrillation/flutter, and syncope), undifferentiated abdominal pain, respiratory (chronic obstructive pulmonary disease, asthma), and cellulitis. One hundred ninety-four patients could be reached by telephone within 72 hours. One hundred twenty-six of 194 patients (64.9%, 95% CI 57.6% to 71.5%) stated that their symptoms had either improved or were no longer present. Of these 126 patients, 109 (86.5%, 95% CI 78.9% to 91.7%) had original AMA discharge diagnoses referable to a potential cardiovascular pathology. Ninety-five patients (75.3%, 95% CI 66.7% to 82.4%) of the group whose symptoms had improved or abated stated they had no plans to seek further evaluation because their symptoms no longer existed. Of the remaining 31 patients in this group (24.6%, 95% CI 17.6% to 33.2%), 20 had returned in the intervening 72 hours, with 15 patients having significant clinical findings with further evaluation. Of the remaining 68 patients with continued symptoms, 36 (52.9%, 95% CI 40.5% to 64.9%) had returned for further evaluation. Nineteen of these 36 patients had significant findings on further evaluation. A total of 138 patients had not returned or sought follow-up care as instructed. In addition to the 95 patients expressing symptom improvement, 31 patients (22.4%, 95% CI 15.9% to 30.4%) opined that they had somewhat of a reluctance to return to the same ED for fear of possibly being made to feel embarrassed by staff about their original decision to disregard medical advice. The other 12 patients (8.7%, 95% CI 4.8% to 15.0%) who had not sought care, although still having symptoms, did not do so because of job and family commitments or because they would follow up with their personal physician. Conclusion: A number of patients who return to continue their care after having left AMA are found to have clinically significant pathology. A significant number of patients, particularly those who were believed to have cardiovascular maladies at their initial ED visit days earlier, do not plan to seek further evaluation and treatment as counseled because of the belief that the resolution or attenuation of their symptoms negates the need for further medical evaluation. Other patients express reticence about returning because they anticipate being met with derision by staff members and chastised for their original decision to leave.