Skip to main content


Predictive Model for at Home Care Among Patients With Solid-Tumor Malignancies

February 09, 2021

Researchers developed a predictive model for identifying admissions of patients with cancer who were potentially suitable for in-patient level care at home, or hospital at home (JCO Oncol Pract. 2021;OP2000663. doi:10.1200/OP.20.00663).

“Hospital at home is a means of providing inpatient-level care at home. Selection of admissions potentially suitable for [hospital at home] in oncology is not well studied,” wrote Kevin Chen, MD, MHS, Section of General Internal Medicine, Yale School of Medicine, and colleagues.

 Thus Dr Chen and colleagues aimed to create a predictive model to identify admissions of patients with solid tumors potentially suitable for hospital at home care.

In order to be considered potentially suitable, admissions could not involve escalation of care, rapid response evaluation, in-hospital death, telemetry, surgical procedure, consultation to a procedural service, advanced imaging, transfusion, restraints, and nasogastric tube placement.

A multivariable logistic regression model was used to explore baseline variables that could potentially predict potential suitability for hospital at home.

A total of 3322 admissions between January 1, 2015, and June 12, 2019 were analyzed. These patients had solid-tumor malignancies and unplanned admissions. Of the patients analyzed, 905 patients were potentially suitable for hospital at home.

Overall 13 factors predicted potential suitability for hospital at home in the deviation cohort (n = 1,097). These included admission source; temperature and respiratory rate at presentation; hemoglobin; breast cancer, GI cancer, or malignancy of secondary or ill-defined origin; admission for genitourinary, musculoskeletal, or neurologic symptoms, intestinal obstruction or ileus, or evaluation of secondary malignancy; and emergency department visit in prior 90 days.

“Hospital admissions of patients potentially suitable for [hospital at home] may be identifiable using data available at admission,” Dr Chen and colleagues concluded.—Marta Rybczynski

Back to Top