JCO Oncol Pract. 2026 Apr 9:OP2501192. doi: 10.1200/OP-25-01192. Online ahead of print.
ABSTRACT
PURPOSE: Approximately three quarters of patients with sarcoma report financial toxicity, defined as the psychosocial and economic burden of cancer care. The Comprehensive Score for Financial Toxicity (COST) is a screening questionnaire validated in patients with cancer. However, its use is not widespread in sarcoma care, likely in part due to its complexity. This study aimed to characterize financial toxicity in patients with sarcoma and evaluate whether shorter screening tools can identify high-risk patients.
METHODS: We conducted a survey and retrospective cohort study of patients treated for sarcoma at a large, academic sarcoma center in 2016-2024. Patients completed the COST questionnaire (11 graded questions, each scored on a five-point Likert scale from 0 to 4) and the shorter, four-question Worry about Affording Healthcare Scale (WAHS). We obtained clinical information from the electronic health record. We used descriptive statistics to summarize responses to COST, Pearson’s correlation coefficient to assess the association between WAHS and COST, and stepwise Spearman correlations to evaluate whether an abbreviated (1-4 question) version of COST correlated highly with total COST score.
RESULTS: Among 205 participants, the mean COST score was 20/44 (standard deviation = 6; lower scores indicate greater financial toxicity). Respondents worried most about job loss (mean COST: 0.99) and family finances (mean COST: 1.11). WAHS was weakly correlated with COST (r = 0.47, P < .01). A four-item abbreviated COST tool was strongly correlated with the total COST (ρ = 0.906, P < .0001) and demonstrated high sensitivity (96%) and moderate specificity (75%).
CONCLUSION: An abbreviated four-item COST tool could act as a clinically feasible, accurate screening tool after external validation. Routine financial toxicity screening could facilitate timely detection and referral to essential financial navigation resources.
PMID:41955544 | DOI:10.1200/OP-25-01192