The development of a financial toxicity patient-reported outcome in cancer: The COST measure
Version of Record online: 20 JUN 2014
© 2014 American Cancer Society
Volume 120, Issue 20, pages 3245–3253, October 15, 2014
How to Cite
de Souza, J. A., Yap, B. J., Hlubocky, F. J., Wroblewski, K., Ratain, M. J., Cella, D. and Daugherty, C. K. (2014), The development of a financial toxicity patient-reported outcome in cancer: The COST measure. Cancer, 120: 3245–3253. doi: 10.1002/cncr.28814
- Issue online: 3 OCT 2014
- Version of Record online: 20 JUN 2014
- Manuscript Accepted: 17 APR 2014
- Manuscript Revised: 29 MAR 2014
- Manuscript Received: 18 DEC 2013
- cost of cancer care;
- quality of life;
- comparative effectiveness;
- patient-reported outcome;
- outcomes research
Considering patients' experience is essential for optimal decision-making. However, despite increasing recognition of the impact of costs on oncology care, there is no patient-reported outcome measure (PROM) that specifically describes the financial distress experienced by patients.
The content for a comprehensive score for financial toxicity (COST) was developed with a stepwise approach: step 1) a literature review and semistructured, qualitative interviews with patients for content generation; step 2) patients' assessment of the items for importance to their quality of life; step 3) pilot testing assessing interitem (IIC) and item–total (ITC) correlations to identify redundancy (Spearman rho, >0.7) and statistically unrelated content (P > .05); and step 4) exploratory factor analysis. Sociodemographic data were collected.
In total, 155 patients with advanced cancer who were receiving treatment (20 patients in step 1, 35 patients in step 2, and 100 patients in steps 3 and 4) participated in the PROM development. In step 1, the literature was reviewed, and 20 patients generated 147 items, which were reduced to 58 items because of redundancy. In step 2, 35 patients rated the 58 items on importance, and 30 items were retained. In step 3, 46 patients assessed the 30 items, 14 items were excluded because of high IIC, and 3 were excluded because of nonsignificant ITC. In step 4, 2 items were discarded because of poor loadings in a factor analysis of 100 patients, resulting in an 11-item PROM.
The content for a financial toxicity PROM was developed in 155 patients. The provisional COST measure demonstrated face and content validity as well as internal consistency and should be validated in larger samples. Cancer 2014;120:3245–3253. © 2014 American Cancer Society.