Predictors of discharge against medical advice in adult trauma patients

Abstract

Background: Patients who leave against medical advice (AMA) have higher readmission rates and mortality. However, little is known about the characteristics of trauma patients that leave AMA. The purpose of this study was to identify predictors for leaving AMA in adult trauma patients.

Methods: The Trauma Quality Improvement Program database was queried between 2010 and 2016 for patients ≥18 years of age presenting after trauma. Two groups were compared: those who left AMA and those that did not. Bivariate analysis using Chi-squared and Mann-Whitney U tests was performed. A multivariable logistic regression analysis was performed to identify predictors for leaving AMA.

Results: Of 1,403,466 trauma patients identified, 10,659 (0.76%) left AMA. Patients that left AMA were younger (median age, 48 vs. 53 years-old, p < 0.001), more often male (82.1% vs. 62.8%, p < 0.001), more likely to be black (23.6% vs. 14.9%, p < 0.001), and more likely to be uninsured (27.0% vs. 12.3%, p < 0.001). Patients leaving AMA were more likely to test positive for alcohol (36.1% vs. 17.4%, p < 0.001) or drug use (36.0% vs. 17.2%, p < 0.001) at time of admission. On multivariable logistic regression, the strongest predictors for leaving AMA were: no insurance (OR 2.00, CI 1.88-2.14, p < 0.001), alcohol use (OR 1.85, CI 1.74-1.96, p < 0.001) or drug use (OR 1.83, CI 1.72-1.94, p < 0.001), male gender (OR 1.83, CI 1.71-1.97, p < 0.001), and stab mechanism of injury (OR 1.58, CI 1.43-1.73, p < 0.001).

Conclusion: In adult trauma patients, male gender, stab mechanism of injury, being uninsured, and alcohol/drug use were strong predictors of leaving AMA. The risk factors identified may help in developing strategies aimed at preventing trauma patients from leaving AMA.

Keywords: Discharge against medical advice; Leave against medical advice; Predictors; Trauma.