Rural cancer financial toxicity screening

Abstract

Purpose

Cancer treatment expenses can lead to financial toxicity (FT), which reduces treatment compliance and impairs outcomes. Little is known regarding the FT among rural cancer populations, where added barriers impair accrual of survey data. To increase our understanding of FT experienced by these patients, we piloted a validated survey instrument and reported on the feasibility of administration.

Methods and Materials

Institutional approval was obtained to prospectively survey rural oncology patients undergoing radiation treatment. Baseline surveys were provided at simulation appointments; weekly surveys were captured during on-treatment visits. Respondents reported on demographics (including self-reported gender, race, education, income, insurance, employment) at baseline and on expenses, the COmprehensive Score for financial Toxicity (range, 0-44, modified such that higher score indicates worse toxicity), perception of providers’ financial empathy, and the minimum financially impactful amount of money at weekly visits. Completion rates and associations between demographic characteristics and FT were assessed with Mann–Whitney U test.

Results

Twenty-six participants were enrolled. Patients were elderly (mean 68.3 years old, SD 10.7), male (25 of 26), White (25 of 26). Forty-two percent were low-income (annual income < $48,000) and 50% had high school or less education. Most (n = 19, 73%) were insured through Medicare. Eighty-five percent of surveys were fully complete. The mean COmprehensive Score for financial Toxicity score at baseline was 14.0 (SD, 11.5; range, 0-38). The mean amount of money that would make a meaningful difference was $211 at baseline (interquartile range, $87.50-$350) and rose to $329 toward the end of the survey period (week 7).

Conclusions

FT screening of rural radiation oncology populations with a range of education is feasible with high fidelity of data collection. Future steps will identify patterns and predictors of severe FT and develop targeted interventions based on this feasibility study.

Introduction

Cancer is one of the most expensive conditions to manage.¹ In 2020, Medicare expenditures for cancer treatment totaled $200.7 billion and will grow to $245.6 billion by 2030.¹ Although the toxicity of chemotherapeutic and radiotherapeutic regimens is closely managed, damage to patients’ financial wellbeing remains underexamined. In 2013, Zafar et al² defined financial toxicity (FT) as “out of pocket expenses related to treatment…that can diminish quality of life and impede delivery of the highest quality care.” FT comprises material, psychological, and behavioral components. The material element refers to the decrease in net worth and the decreased ability to pay medical bills; the psychological element describes the anxiety secondary to these pressures; the behavioral element captures the tendency to avoid treatments, medications, and appointments to alleviate financial strain.³

These components of FT can significantly impact patient care. In 2019, 179 million patients reported some degree of FT, making them 3× more likely to skip visits and 2.5× more likely to avoid tests.⁴ Importantly, FT is distributed inequitably across cancer patients, with 50% of rural patients experiencing some degree of toxicity compared to 39% of urban patients.⁵ Thus, monitoring FT is especially important among rural patients undergoing treatment.

Studies have reported on the feasibility of FT screening in various settings.6, 7, 8 Although FT screenings are recommended for at-risk rural patients,⁹ implementing screening among this population has not been investigated.¹⁰ Thus, the objective of this study is to determine the feasibility of universal FT screening among rural cancer patients undergoing radiation treatment.

Methods and Materials

Institutional approval for this cross-sectional study was obtained to prospectively survey rural oncology patients undergoing radiation treatment. Baseline surveys were collected at simulation appointments (prior to radiation therapy treatment and incurred costs) and repeated at weekly treatments. Follow-up surveys were collected for up to 4 months after completion of therapy. Eligible patients included all patients undergoing radiation therapy at a single rural satellite site from September 2022 to January 2023.

At baseline, respondents completed an 8-page, 20-question survey (Appendix E1) that captured demographic data (sex, race, education, language, insurance, employment), weekly income, expenses, travel distance, COmprehensive Score for financial Toxicity (COST) questionnaire responses, perspectives regarding providers’ understanding and empathy of finances, and a participant-reported subjective estimate of a weekly sum of money that would meaningfully improve their finances. The COST questionnaire was developed by de Souza et al¹¹ in 2017 and has been used to measure financial toxicity in a range of cancer types.^12,13 It includes 11 questions relating to perceptions of cost burden graded on a 5-point Likert scale with total scores ranging from 0 to 44. We modified our questionnaire such that a higher sum of responses indicated worse FT. This version was completed by participants at weekly visits (Appendix E2).

Data were entered into a Research Electronic Data Capture (REDCap) database, cleaned in Stata and analyzed with Microsoft Excel. Processed data were described with mean, median, interquartile range, and standard deviation measures where appropriate. Associations between patient categories and FT scores were assessed with 1-tailed Mann–Whitney U tests (chosen for nonnormal distribution) with U values below the critical value for P < .05 considered statistically significant.

The primary outcome was the completion rate of all surveys based on total treatment weeks for each patient. The secondary outcome was the degree of FT as measured by COST scores.

Results

Recruitment and completeness of data

Twenty-seven patients were consented, with 1 withdrawing consent from the study. Twenty-two out of 26 patients (85%) completed all surveys based on the number of treatments delivered (Fig. E1).

Participant characteristics

At baseline, our sample population was elderly (mean age, 68.3), mostly male (96%), and mostly White (96%) (Table 1). Forty-two percent (n = 11) had incomes lower than $48,000 (cutoff point drawn from the inherent characteristics of the baseline data) and 50% (n = 13) had attained an education level of high school or less (Fig. 1).

Table 1.

Patient demographics

	No.	%
Age	26
<50	1	4%
50-59	4	15%
60-69	9	35%
70-79	8	31%
>80	2	8%
Gender	26
Male	25	96%
Female	1	4%
Race	26
White	25	96%
Other	1	4%
Insurance type
Employer provided	5	19%
Self-insured	6	23%
Medicare + Medicaid	4	15%
Medicare	19	73%
Tricare	2	8%
Uninsured	0	0%
Unsure	0	0%
Prefer not to say	0	0%
Income level
<$48,000	11	42%
>$48,000	15	58%
Level of education
High school or less	13	50%
Beyond high school	13	50%

Figure 1.

Median responses to “minimum amount of money that would make a meaningful difference to financial status.”

FT and meaningful money

The median COST score indicating the degree of subjective FT was 14.0 at baseline and rose to 18.0 at treatment end (Fig. E2), with higher scores indicating greater FT. Scores began increasing around week 6. Although the median “amount of money that would make a big difference” was $239 overall, which slightly varied over time (Fig. 1). There was a negative association between income level and reported median COST scores (Fig. 2). When comparing patients with incomes greater versus < $48,000 per year we found a statistically significant difference (23.0 vs 6.20, effect size=16.8) in reported median COST scores (Fig. E4, N_{low income} = 11, N_{high income} = 15, U = 8.5, p = .0019).

Figure 2.

Association between income level and COST scores.

Abbreviation: COST = COmprehensive Score for financial Toxicity.

Mean COST scores also varied according to educational attainment. Patients without any college attendance reported notable differences in mean COST scores (15.6 vs 9.33, effect size = 6.27) than their peers with some degree of college attendance but did not achieve statistical significance (Fig. E5, N_{low education} = 13, N_{high education} = 13, U = 45, P = .24). Over time, patients' reported perceptions of provider awareness of their financial situation increased (Fig. 3).

Figure 3.

Proportions of patients responding to “How much do you think you team understands your financial situation?”

Foregone expenses

Nine participants avoided expenses to save money for cancer related treatment. Gas, food, and entertainment were the most frequently foregone expenses, with an average reduction at baseline of $36.50, $175, and $50, respectively (Fig. 4) These values stayed stable through the duration of the survey period as patients continued to respond “yes” in these categories each week of their treatment period.

Figure 4.

Proportions of patients responding to “Which categories did you reduce expenses in?”

Discussion

Our study found that FT screening was feasible among rural cancer patients as evidenced by the high weekly completion rate (85%). The timeline of COST scores and requested supplemental funds (Fig. E2, E3) suggest periods of acute FT for each patient undergoing treatment around week 2 and week 6. These insights, along with the greater reported toxicity among low-income patients and those with lower educational attainment, indicate opportunities for targeted intervention for these at-risk groups and time periods.

Rural cancer centers face unique challenges from their reimbursement stream, patient population, staffing, and geography. In our study, patients traveled an average of 30 to 40 miles each way to receive treatment. These distances cause treatment delays that can increase mortality for rural patients—1 study estimated 2-year survival rates at 76% for nonrural patients without a treatment delay versus 27% for rural patients who experienced treatment delays attributed to travel distances.¹⁴ Policy changes may help alleviate this burden. The American Society for Radiation Oncology has proposed a novel Center for Medicare and Medicaid payment model of up to $500 per patient per episode to “support transportation to radiation oncology treatment services for beneficiaries who are at risk because of transportation barriers.” These indirect cash transfers can help overcome structural disparities and reduce the inequities imposed by geography.

Limitations

Qualitative interviews may capture additional nuance of subjective financial pressures; we plan to perform these in subsequent investigations. The study was conducted at a single rural site amid a racially and demographically homogenous patient catchment area, thus limiting generalizability. However, because 2020 census data shows similar proportions older than 65 among the surveyed counties and rural counties across the nation,¹⁵ these results are applicable to other rural American counties.

Conclusion

The results indicate that weekly screening for FT among rural patients undergoing radiation therapy is feasible and sustainable over the treatment period. These screenings can normalize discussions of cost concerns, target interventions, and alleviate financial anguish.

Disclosures

Nirav Kapadia reports financial support was provided by Northern New England Clinical Oncology Society (NNECOS). If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.