Frailty phenotypes and acute care escalation in hospitalized patients with solid tumors: A National Inpatient Sample analysis, 2016–2023.

Abstract

e23202 Background: Frailty-aware oncology is increasingly emphasized, yet inpatient frailty is inconsistently captured and often reduced to single markers. Whether cumulative multi-domain frailty burden predicts inpatient outcomes among hospitalized patients with solid tumors remains unclear. Methods: A survey-weighted analysis of the National Inpatient Sample (NIS), 2016–2023, was performed. Adult hospitalizations with solid tumor malignancies (ICD-10-CM C00–C80) were included. Hematologic malignancies (C81–C96) and palliative care encounters (Z51.5) were excluded. Frailty proxies included malnutrition (E43/E44/E46), cachexia or weight loss (R64), sarcopenia (M62.84), dysphagia (R13*), and pressure ulcer (L89*). Frailty burden was categorized as 0, 1, 2, or ≥3 domains. The primary outcome was in-hospital mortality. Secondary outcomes included any airway escalation (mechanical ventilation or tracheostomy), shock (R57*), length of stay (LOS), and hospitalization cost. Survey-weighted multivariable logistic regression estimated adjusted odds ratios (aORs) with 95% confidence intervals (CIs). LOS and cost were modeled using survey-weighted linear and gamma regression. Results: Among 3,046,500 unweighted solid tumor hospitalizations (weighted subpopulation size 15,232,496), frailty burden distribution was: 0 domains 79.46% (95% CI 79.32–79.61), 1 domain 16.46% (95% CI 16.34–16.57), 2 domains 3.61% (95% CI 3.57–3.65), and ≥3 domains 0.48%(95% CI 0.47–0.49). Unadjusted mortality increased stepwise with frailty burden from 2.02% (95% CI 1.99–2.05) to 9.81% (95% CI 9.31–10.33), and airway escalation from 2.60% (95% CI 2.57–2.63) to 11.50% (95% CI 10.98–12.05). LOS increased from 5.00 days (95% CI 4.99–5.02)to 11.52 days (95% CI 11.29–11.75), and mean cost from $20,943.82 (95% CI $20,532.55–$21,355.09) to $35,440.96 (95% CI $34,009.64–$36,872.28). After adjustment, frailty burden remained strongly associated with mortality (1 domain aOR 2.08, 95% CI 2.04–2.12; 2 domains aOR 3.14, 95% CI 3.04–3.25; ≥3 domains aOR 4.18, 95% CI 3.91–4.47) and airway escalation (1 domain aOR 2.58, 95% CI 2.53–2.63; 2 domains aOR 3.77, 95% CI 3.66–3.88; ≥3 domains aOR 4.69, 95% CI 4.41–4.98). In phenotype-specific models, malnutrition (aOR 1.86, 95% CI 1.82–1.90) and pressure ulcer (aOR 2.25, 95% CI 2.17–2.34) were the strongest predictors of mortality, while dysphagia was most strongly associated with airway escalation (aOR 2.98, 95% CI 2.91–3.06). Conclusions: In nationally representative solid tumor hospitalizations, cumulative multi-domain frailty burden identifies stepwise risk for inpatient mortality, acute care escalation, LOS, and cost. Frailty phenotyping using routinely coded inpatient markers may support earlier multidisciplinary assessment and inform risk-adjusted inpatient oncology care pathways.