Endocrine stress phenotypes as drivers of extreme inpatient severity and resource utilization without increased mortality in cancer hospitalizations: National Inpatient Sample, 2018–2022.

Abstract

e23201 Background: Endocrine complications during cancer hospitalization may reflect acute physiologic stress rather than chronic disease. Whether stress hyperglycemia and adrenal insufficiency function as high-acuity, resource-intensive inpatient phenotypes that amplify physiologic severity without increasing in-hospital mortality is incompletely characterized nationally. Methods: A serial cross-sectional, hospitalization-level analysis of the 2018–2022 Healthcare Cost and Utilization Project National Inpatient Sample (NIS) was performed. Adult hospitalizations with a principal diagnosis of malignancy were identified using ICD-10-CM codes C00–C97 and D45–D47. Stress hyperglycemia was defined as hyperglycemia without diabetes, and adrenal insufficiency was identified via inpatient diagnosis proxies. Primary outcomes were All Patient Refined Diagnosis Related Group (APR-DRG) severity of illness subclass, an ordinal measure, and extreme severity defined as APR-DRG level 4. Secondary outcomes included mechanical ventilation as an ICU-level care proxy, length of stay (LOS), total hospitalization cost, and in-hospital mortality. Results: Stress hyperglycemia occurred in 2.09% of cancer hospitalizations and was associated with higher mean inpatient severity than the reference group (APR-DRG 3.02 vs 2.50). Extreme inpatient severity occurred in 30.97% of stress hyperglycemia hospitalizations compared with 15.64% in the reference group. After adjustment, stress hyperglycemia remained independently associated with higher inpatient severity (adjusted β 0.45, 95% CI 0.42–0.48) and increased odds of extreme severity (adjusted OR 2.25, 95% CI 2.06–2.45), and was associated with longer LOS (10.32 vs 6.64 days) and higher mean hospitalization cost ($48,257 vs $29,765). Adrenal insufficiency was similarly associated with increased inpatient severity (mean 2.98 vs 2.56) and greater odds of extreme severity (adjusted OR 2.04, 95% CI 1.47–2.85), along with increased mechanical ventilation utilization. Neither stress hyperglycemia (adjusted OR 1.05, p = 0.223) nor adrenal insufficiency (adjusted OR 1.23, p = 0.123) was independently associated with increased in-hospital mortality after adjustment. Conclusions: In U.S. cancer hospitalizations, endocrine stress phenotypes amplify physiologic severity and drive disproportionate resource use without increasing adjusted in-hospital mortality. This suggests stress hyperglycemia and adrenal insufficiency serve as administrative markers of acute inpatient vulnerability, with relevance to ICU capacity, LOS, and cost. Recognizing these phenotypes may improve inpatient risk stratification and guide earlier resource planning in oncology care.