Emergency department visits and hospital readmissions in an Argentine health system

Abstract Background and goal of study The scope of health in the Sustainable Development Goals is much broader than the Millennium Development Goals, spanning functions such as health-system access and quality of care. Hospital readmission rate and ED-visits within 30 days from discharge are considered low-cost quality indicators. This work assesses an indicator of quality of care in a tertiary referral hospital in Argentina, using data available from clinical records. Purpose To estimate the rate of ED-visits and the hospital readmission rate (HRR) after a first hospitalization (First-H), and to identify associated factors. Methods This retrospective cohort included patients who had a First-H in Hospital Italiano de Buenos Aires between 2014-2015. Follow-up occurred from discharge until ED-visit, readmission, death, disaffiliation from health insurance, or 13 months. We present HRR at 30 days and ED-visits rate at 72 hours, using the Cox proportional-hazards regression model to explore associated factors, and reporting adjusted hazard ratios (HR) with their respective 95%CI. Results The study comprised 10,598 hospitalizations (median age was 68 years). Of these, 5,966 had at least one consultation to the ED during follow up, resulting in a 24 hour rate of consultations to ED of 1.51% (95%CI 1.29-1.72); at 48 hours 3.18% (95%CI 2.86-3.54); at 72 hours 4.71% (95%CI 4.32-5.13). In multivariable models, factors associated for 72 hours ED-visits were: age (aHR 1.06), male (aHR 1.14), Charlson Comorbidity Index (aHR 1.16), unscheduled hospitalization (aHR 1.39), prior consultation with the ED (aHR 1.08) and long hospital stay (aHR 1.39). Meanwhile, 2,345 patients had at least one hospital readmission (98% unscheduled), resulting a 24 hour rate of 0.5% (95%CI 0.42-0.71), at 48 hours 0.98% (95%CI 0.80-1.18), at 72 hours 1.4% (95%CI 1.2-1.6); at 30 days 7.7% (95%CI 7.2-8.2); at 90 days 13% (95%CI 12.4-13.8); and one-year 22.5% (95%CI 21.7-23.4). Associated factors for HRR at 30 days were: age (HR 1.16), male (HR 1.09), Charlson comorbidities score (HR 1.27), social service requirement during First-H (HR 1.37), unscheduled First-H (HR 1.16), previous ED-visits (HR 1.03) and length of stay (HR 1.08). Conclusion Priorities efforts to improve must include greater attention to patients' readiness prior discharge, to explore causes of preventable readmissions, and better support for patient self-management.