Against Medical Advice but Not Infection: The Readmission Ramifications of Leaving AMA after Trauma Care.

Authors:
Fabiana C Sanchez, Grace M Niziolek, Grant V Bochicchio, Jason A Snyder, Leonardo J Diaz Gandica, Lindsay M Kranker, Marco J Henriquez, Marguerite W Spruce, Melissa Canas, Michael W Alchaer, Ricardo A Fonseca

Body of Abstract:
Background:
Trauma patients who leave the hospital against medical advice (AMA) are at higher risk of adverse outcomes, including hospital readmission. Infection-related readmissions may be driven by incomplete treatment, missed follow-up care, or limited access to outpatient resources. This study compares infection-related readmission rates in trauma patients that left AMA versus those discharged per standard hospital protocol.
Methods:
We conducted a retrospective cohort study using data from an urban Level 1 trauma registry from 2018 to 2022. The study population included trauma patients with an Injury Severity Score (ISS) greater than or equal to 9 who resided within the four counties surrounding our institution. Patients who died during the index hospitalization or were discharged to hospice were excluded. Patients were categorized into those who left against medical advice (AMA) and those who underwent standard discharge (non-AMA). The primary outcome was infection-related readmission (IRR). Data collected included patient demographics, mechanism of injury (MOI), ISS, culture results, antibiotic usage, and length of stay. Statistical analyses conducted included t-tests, chi-square tests, and binary logistic regression.
Results:
Among the 5,845 trauma patients, 130 (2.2%) left against medical advice (AMA), and 5,715 (97.8%) were discharged under standard care (non-AMA). AMA patients were more likely to be male (84.6% vs 61.7%, p < 0.001), younger (40.9 years vs 53.0 years, p < 0.001), Black non-Hispanic (66.2% vs 47.6%, p < 0.001), and injured by a penetrating mechanism (33.8% vs 21%, p < 0.001). A total of 213 patients (3.6%) were readmitted within 30 days; patients who left AMA had an overall 30-day readmission rate that was fivefold higher than the non-AMA trauma population (16.9% vs. 3.3%, p < 0.001). A total of 90 patients were readmitted due to infection (1.5%, 9 AMA vs 81 non-AMA). The infection-related readmission rate was significantly higher in the AMA group compared to those discharged non-AMA (6.9% [9/130] vs 1.4% [81/5715], p <0.001). Logistic regression showed that leaving AMA was a significant predictor of readmission due to infection (OR = 4.592, 95% CI: 2.202–9.576, p < 0.001). Conclusions: Leaving the hospital AMA is an independent predictor of readmission due to infection after controlling for sex, age, race/ethnicity, and social vulnerability. This underscores the need for targeted interventions and closer follow-up for this high-risk group to reduce infection-related readmissions, especially among those with a penetrating mechanism of injury. Further research is needed to stratify the patient’s reason for leaving AMA to reduce the associated risk of adverse outcomes incurred by patients.