Readmission After Myasthenia Gravis Discharge in a Nationally Representative Sample (P6.433)

Objective: To determine the 30-day readmission rate after myasthenia gravis (MG) related discharge in a nationally representative sample and examine factors associated with readmission. Background: Hospital readmission rates are evaluated to examine health care quality and outcomes. Readmissions have not been extensively studied for MG. Design/Methods: Retrospective cohort study of 30 and 90 day non-elective readmission after an index hospitalizations for MG. Index admissions were sampled from the Healthcare Cost and Utilization Project’s 2014 Nationwide Readmissions Database (NRD). Results: 4,305 admissions met inclusion criteria. Female sex (56.0%), age 55–84 (59.6%) and Medicare insurance (57.4%) were frequently represented among index hospitalizations. Most index hospitalizations occurred at metropolitan teaching hospitals (73.2%), and lasted 6–19 days (52.1%). Mechanical ventilation was common (16.0%). Sixteen percent of MG patients had a non-elective readmission within 30 days (n=694), 25.2% (n=1086) experienced non-elective readmission within 90 days. Medical comorbidity (Elixhauser score >=4 compared to 1: AOR 2.69, 1.57–4.62, Elixhauser score 3 versus 1: AOR 2.52, 1.40–4.58) independently predicted readmission at 30 days in regression models that included patient, clinical, hospital and index stay characteristics. This association was also significant at 90 days (Elixhauser score>4 versus 1: AOR 1.87, 1.16–3.01; Elixhauser score 3 versus 1: AOR 1.80, 1.09–2.96). Patient age, sex, primary payer, median zip code income, medical error occurrence and index disposition were not associated with a statistically significant increase or decrease in odds of readmission at 30 or 90 days. MG related symptoms were the most commonly documented reason for non-elective readmission within 30 (34.4%) and 90 days (36.0%). Conclusions: Patients hospitalized with MG are frequently readmitted. Medical complexity drives readmission among myasthenics and optimizing treatment during the index admission may help improve patient care. Disclosure: Dr. Thawani has nothing to disclose. Dr. Thibault has nothing to disclose. Dr. Crispo has nothing to disclose. Dr. Wright Willis has nothing to disclose.