Using G8 and carg toxicity score to predict emergency room (ER) visits, hospitalizations, and mortality in older patients with newly diagnosed cancer.

Authors

null

Amit Arora

Kaiser Permanente, Fremont, CA

Amit Arora , Hongxin Sun , John L. Shaia , Tatjana Kolevska , Dinesh Kotak , Kate Belohlav , Grant Richard Williams , Raymond Liu

Organizations

Kaiser Permanente, Fremont, CA, The Permanente Medical Group Consulting Services, Oakland, CA, Permanente Medical Group, San Francisco, CA, Kaiser Permanente Northern California, Vallejo, CA, Thomas Jefferson Univ Hosp, Bensalem, PA, Kaiser Permanente, Oakland, CA, The University of Alabama at Birmingham, O'Neal Comprehensive Cancer Center, Birmingham, AL, The Permanente Medical Group, Department of Hematology Oncology, San Francisco, CA

Research Funding

No funding received

Background: ASCO and NCCN guidelines recommends Geriatric Screening (G8) and CARG chemotherapy toxicity tool assessment for all older patient before receiving chemotherapy as high risk G8 (< 14) and CARG (≥10) are associated with increased chemotherapy toxicities. We conducted a pilot to understand predictors of high risk G8/CARG and if high risk G8/CARG can predict ER/hospitalization and mortality in community-based Oncology clinics in Kaiser Permanente Northern California. Methods: G8 and CARG were administered to all patients ≥65 years with newly diagnosed cancer from 5/1/21 to 12/31/21. Patients were followed for at least 30 days after assessment for ER/hospitalization and mortality. The median follow-up days from referral to ER/hospitalization was 96 days (range 0-273 days). Chi-Square tests were applied for G8/CARG risk category with demographic and utilization variables. Cox proportional-hazards models were performed to see the association between G8/CARG score and days from referral to ER/hospitalization, and days from referral to death, adjusted for age, sex, race, and cancer type. Results: During this pilot 1082 patients (52% female) completed G8, and 516 patients (57% female) completed CARG. Percentage of patients with high risk G8/CARG increased with each decade (G8: < 70 yrs (58%), 70-79 (63%), 80-89 (90%), ≥ 90 (100%); p < 0.001); (CARG: < 70 yrs (19%), 70-79 (43%), 80-89 (65%), 90 and above (81%); p < 0.001). More men than women had high risk CARG (48% vs. 39%, p = 0.012). Ethnicity was not associated with high risk G8 / CARG. Upper GI cancers (UGI) were associated with highest proportion of patients with high risk G8 (88%) and CARG (58%) whereas breast cancer (BC) had the lowest proportion of patients with high risk G8 (46%) and CARG (14%); p < 0.001. In the adjusted G8 model for ER/hospitalization, high risk G8 vs low risk (HR 1.58, CI 1.23-2.03, p = 0.0003) was related to ER/hospitalization. In the adjusted CARG model for ER/ hospitalization, high risk CARG vs low risk (HR 2.42, CI 1.37-4.29, p = 0.0024) and medium risk CARG vs low risk (HR 2.17, CI 1.23-3.83, p = 0.0074) were related to ER/hospitalization. In the adjusted G8 model for mortality, high risk G8 vs low risk (HR 4.52, CI 2.28-8.97, p < 0.0001) were related to mortality. In the adjusted CARG model for mortality, high risk CARG vs low risk (HR 3.92, CI 1.21-12.74, p = 0.023) and medium risk CARG vs low risk (HR 1.59, CI 0.48-5.33, p = 0.45) were related to mortality. Conclusions: This community-based pilot shows that increasing age is associated with high risk G8 / CARG. G8 and CARG assessment at the time of initial cancer diagnosis can predict early ER/hospitalization and mortality in older adults with cancer and should be included as a part of initial assessment.

Disclaimer

This material on this page is ©2024 American Society of Clinical Oncology, all rights reserved. Licensing available upon request. For more information, please contact licensing@asco.org

Abstract Details

Meeting

2022 ASCO Annual Meeting

Session Type

Poster Session

Session Title

Symptoms and Survivorship

Track

Symptom Science and Palliative Care

Sub Track

Geriatric Models of Care

Citation

J Clin Oncol 40, 2022 (suppl 16; abstr 12055)

DOI

10.1200/JCO.2022.40.16_suppl.12055

Abstract #

12055

Poster Bd #

301

Abstract Disclosures