The inpatient cancer diagnostic service: An academic cancer center experience.

Abstract

e23015 Background: Rapid inpatient cancer diagnosis may reduce time to treatment and improve outcomes, particularly among medically complex patients. The Ohio State University (OSU) Comprehensive Cancer Center established an Inpatient Cancer Diagnostic Service (ICDS) for hospitalized patients with suspected new malignancy to expedite diagnosis and treatment. We describe clinical outcomes associated with this inpatient diagnostic care model. Methods: We retrospectively reviewed hospitalized patients evaluated by ICDS between January 1, 2018 and June 30, 2021. Data included pathology, cancer type, time to oncology follow-up, time to treatment initiation, and overall survival (OS). OS was calculated from admission to death, censoring patients alive at last follow-up. Kaplan–Meier and log-rank tests were used. Results: A total of 1,281 patients met inclusion criteria; 252 did not undergo inpatient tissue sampling (192 deferred to outpatient evaluation, 56 declined further workup and enrolled in hospice, and 4 were diagnosed with hepatocellular carcinoma based on imaging and included in the analysis). Inpatient diagnostic evaluation was completed in 1,033 patients (median age 64 years; 57.6% male). Benign pathology was identified in 17.3%. Among patients with cancer (n = 854), gastrointestinal (30.0%) and thoracic (26.6%) malignancies were most common (Table). Hematologic malignancies comprised 7.4%, most commonly lymphoma (41.6%). A total of 552 patients (64.6%) established oncology follow-up at OSU; 474 (55.5%) received cancer-directed therapy at OSU and 201 received treatment outside OSU. Mean time from admission to diagnosis was 6.4 days; among those followed at OSU, mean time to first oncology visit and treatment initiation was 18.7 and 29.3 days. With median follow-up of 4.2 years, patients treated at OSU had improved OS compared with those treated elsewhere (P < 0.001), likely reflecting referral and selection bias at a tertiary cancer center. One-year survival was 59.1% for patients treated at OSU, 31.3% for those treated elsewhere, and 3.1% for patients referred to hospice. Median OS was 1.48, 0.33, and 0.06 years, respectively. Conclusions: The ICDS model facilitated timely diagnosis and oncologic follow-up and was associated with improved survival among patients receiving longitudinal care at OSU. This ICDS model may be scalable to other comprehensive cancer centers. Limitations include retrospective design and potential selection bias among patients receiving care at OSU. Cancer diagnoses. Cancer Type Number, (%) Median OS in Years, (95% CI) Gastrointestinal 260 (30%) 0.44 (0.31-0.60) Thoracic 228 (26.6%) 0.55 (0.40-0.77) Genitourinary 70 (8.2%) 1.04 (0.55-1.71) Head and neck 65 (7.6%) 1.00 (0.54-2.02) Hematologic 64 (7.4%) 3.82 (2.41-NR) Other 61 (7%) 0.87 (0.72-1.44) Unknown primary 53 (6%) 0.13 (0.09-0.21) Breast 37 (4.3%) 1.86 (0.81-3.06) Gynecologic 16 (1.8%) 2.02 (0.50-NR)