A comprehensive analysis of temporal trends of between-hospital variation in mortality, readmission and length of stay using logistic regression

Abstract

Despite the benefits of studying multiple patient outcomes together, research on between-hospital variation has often focused on single outcomes or disease-specific study populations. In this study we examined nationwide temporal trends and between-hospital variation in in-hospital mortality, 30-day readmissions and length-of-stay above the All-Patient-Refined Diagnoses-Related-Group (APR-DRG)-specific 90th percentile (pLOS). We modelled 13,660,187 admissions derived from an administrative database occurring between 2008 and 2018 in 90 (89%) Belgian acute-care hospitals. We applied an APR-DRG-specific logistic regression to study temporal trends in outcomes, hospital-level associations between outcomes, associations of outcomes with hospitals characteristics, and to evaluate how many and which APR-DRGs explained between-hospital variation. Our proposed analytical model managed to achieve novel insights into healthcare quality of care, illustrating the high potential administrative databases can provide. It was revealed that between-hospital variation in outcomes is likely due to systemic hospital factors. This is illustrated by the fact that baseline bottom-performing hospitals remained underperforming throughout the study period and vice versa. APR-DRG-specific between-hospital variation assessments further confirmed this. When hospitals have overall outcome ratios that significantly deviate from the benchmark, this seems to be driven by a considerable number of APR-DRGs, comprising a diverse set of pathologies. This urges a healthcare policy reform wherein longitudinal follow-up and benchmarking of patient outcomes should become the starting point towards targeted quality improvement interventions.