Inpatient Hemato-Oncology Service Use Before, During, and After COVID-19 Restrictions in a Romanian Tertiary Center

Introduction The coronavirus disease 2019 (COVID-19) pandemic disrupted cancer diagnosis and treatment pathways, but hospital-based evidence on inpatient hemato-oncology service use in Eastern Europe remains limited. This study evaluated how the pandemic period affected case capture, diagnostic confirmation, repeated inpatient representation, hospital admissions, and hospital days among adults with lymphoma, leukemia, or multiple myeloma treated in a Romanian tertiary center. Materials and methods A retrospective observational single-center analysis was performed at Sibiu County Emergency Clinical Hospital, Sibiu, Romania. Adult patients identified in hospital records between March 11, 2018, and March 8, 2024, were grouped into pre-pandemic, pandemic, and post-restriction periods. The analysis followed three cohort layers: initially identified patients, diagnostically confirmed cases, and final longitudinally eligible patients, defined as diagnostically confirmed patients with at least two standard inpatient admissions in the corresponding study period. Population-normalized hospital-based rates, rate ratios (RRs), 95% confidence intervals (CIs), proportion tests, chi-square tests, non-parametric comparisons, sensitivity analyses by major disease group, and adjusted patient-level models were used. Results The initially identified patient cohort included 1,314 patients, of whom 1,188 were diagnostically confirmed, and 749 met final longitudinal eligibility criteria. All patient counts decreased numerically during the pandemic period, but these changes did not reach statistical significance. In contrast, inpatient service-use indicators showed significant contraction. Hospitalizations declined from 1,450 before the pandemic to 1,325 during the pandemic period (RR 0.91, 95% CI 0.85-0.98, p = 0.015), while hospital days decreased from 8,123 to 6,553 (RR 0.81, 95% CI 0.78-0.83, p < 0.001). After restrictions, hospitalizations returned to baseline, but hospital days remained below the pre-pandemic level (RR 0.90, 95% CI 0.88-0.93, p < 0.001). Broad lymphoma, leukemia, and multiple myeloma distributions remained stable across periods. Disease-specific sensitivity analyses showed significant pandemic-period reductions in aggregate hospital days across all three major groups, whereas adjusted models showed comparable admissions and hospital days per patient retained in the final longitudinal cohort. Conclusions In this Romanian tertiary-center cohort, the pandemic period was associated mainly with aggregate inpatient service-use contraction rather than with broad disease-mix change. The findings indicate that stable cohort composition may coexist with reduced inpatient bed-day exposure. Combined monitoring of case capture, diagnostic confirmation, longitudinal inpatient representation, admissions, and hospital days may provide a descriptive hospital-based framework for distinguishing diagnostic-volume changes from service-intensity changes during healthcare disruptions.