Introduction

Financial toxicity (FT) is now well recognized as an adverse outcome of cancer care that can impact personal well-being, health-related quality of life and health outcomes [1]. It is defined as the material or subjective financial distress experienced by patients as a result of their (cancer) diagnosis. Survey studies demonstrate that approximately 50% of patients with gynecologic cancer experience significant FT as a result of their diagnosis and treatment [2,3,4,5]. Broader population-based studies confirm that FT impacts 48–73% of patients with cancer [6, 7]. Younger age at diagnosis, female gender, non-white race, lower income and inadequate health insurance are significant predictors of FT in patients with cancer [2,3,4,5,6,7,8,9,10,11]. Many studies have demonstrated a strong association of increasing FT with diminished quality of life [2,3,4,5,6,7,8,9] and several have suggested increasing rates of non-adherence to recommended medical care [2, 4, 5, 7]. Moreover, one study found that in patients with cancer, filing for bankruptcy was a significant risk factor for mortality [11]. In another study, patients with high FT were more than 7 times more likely to report delaying or avoiding medical care as a result of their financial distress [2] than those without high FT.

Radiation treatment poses unique stressors on patients with cancer due to both the intensive treatment that may be associated with direct medical costs, as well as indirect expenses due to the unique structure of treatments necessitating higher transportation and parking costs or time away from work or lost wages. Further, the burden of side effects may limit an individual’s ability to work or care for others. In this study, we sought to characterize FT in patients receiving radiation treatment for gynecologic cancer. Additionally, notable financial hardships due to the COVID-19 pandemic disproportionately impacted low- and middle-income groups with higher proportions of income loss and unemployment [12]. Thus, we also sought to explore the impact of the pandemic on FT in this patient population.

Methods

Study population/recruitment

Following Institutional Review Board approval, all patients presenting to our gynecologic radiation oncology practice for a visit 1 month after completion of radiation were invited to complete the survey. The paper questionnaire was either completed at the visit or it was returned via mail in a pre-addressed envelope, and all results were entered into REDCap, a secure research database [13]. There were two periods of recruitment, August 2019-March 2020 and November 2020 to June 2021, resulting in pre-pandemic and pandemic cohorts.

Survey design

The survey included demographic questions, the COmprehensive Score for Financial Toxicity (COST) tool and the EQ-5D to measure quality of life (QOL) [14,15,16]. Due to the start of the COVID-19 pandemic in March of 2020, the study was paused for a period of time and once resumed, a set of questions related to the pandemic was added. High FT was defined as COST score ≤ 23 [2, 4]. Participants who answered fewer than 6 questions on the COST tool or who did not complete the survey correctly (as evidenced by selecting the same answer choice for the four reverse coded questions as the other seven questions) were excluded from analysis. A medical record review was performed to obtain additional demographic, disease and treatment characteristics.

Statistical analysis

To compare characteristics between the high and low FT groups, we used chi-square or Fisher’s exact test for categorical variables and Wilcoxon rank-sum test for continuous variables. We calculated the Spearman correlation coefficient to assess the association of FT with self-rated health and QOL. We used log-binomial regression to estimate risk ratios and 95% confidence intervals (CI) for associations between FT and cost-co** strategies, adjusting for age and insurance.

Results

We distributed surveys to 118 patients, and 112 responded, yielding a response rate of 95%. Eight surveys had incomplete data and 7 COST instruments were completed incorrectly. Of the remaining 97 evaluable respondents, 28% were categorized as having high FT (COST score ≤ 23). The median COST score was 15 (7–19) in the high FT group and 33 (28–36) in the low FT group (p < 0.01). Respondents with high FT were younger, with a median age of 61 (50–64), compared to those with low FT, who had a median age of 66 (60–72, p = 0.01). The distribution of race was similar between the high and low FT groups, with most patients being white (76%). Most patients were not employed (60%) and had insurance (97%). Employment status was not associated with high FT, but those in the high FT group were more likely to have private insurance (70%) compared to those in the low FT group (41%, p = 0.03). Partnership status, educational achievement and income were not significantly associated with high FT (all p30.24). There was no association with comorbidities measured by the Charlson Comorbidity Index, BMI, anxiety or depression (all p > 0.35) (Table 1).

Table 1 Participant demographics, stratified by high and low financial toxicity
Full size table

The most common type of cancer was uterine cancer (64%), and most patients had stage I disease (54%). Sixty percent of patients had external beam radiation therapy with or without brachytherapy, while 40% of patients had brachytherapy alone. There was no significant association of cancer type, cancer stage, or type or duration of radiation therapy with extent of FT (all p > 0.28). A majority (63%) of respondents also received chemotherapy, which was not significantly associated with FT (Table 2). A higher proportion of patients received all of their chemotherapy and radiation at our urban tertiary care center 87% in the low FT group as compared to 70% in the high FT group (p = 0.07).

Table 2 Cancer and treatment, stratified by high and low financial toxicity
Full size table

Respondents in the low FT group were more likely to report no changes to their employment (61%) as compared to those in the high FT cohort (26%, p = 0.002). A similar pattern was noted in employment status of the primary caregiver, with the majority of the low FT group reporting no changes to caregiver employment (79%) and nearly half of the high FT cohort reporting some change to caregiver employment (48%, p = 0.003). Patients with increased FT were more likely to report needing to move from full-time to part-time employment (p = 0.002) and needing to take unpaid time off (p = 0.07). Difficulty paying for transportation or parking was a consideration in making medical decisions in 22% of respondents with high FT compared to 7% with low FT (p = 0.06). Medical bills, decreased ability to work and money needed for transportation or parking more often contributed to financial stress in respondents with high FT as compared to low FT (all p < 0.049, Table 3). Greater FT correlated with worse QOL (r = 0.37, p < 0.01) and poorer self-rated health (r = 0.27, p = 0.01).

Table 3 Factors contributing to financial stress
Full size table

In a multivariable analysis adjusting for age and type of insurance, high FT was significantly associated with cost-co** strategies compared to those with low FT (Fig. 1). Respondents with high FT were 6.0 times (95% CI: 1.0–35.9) more likely to delay or avoid medical care, 13.6 times (95% CI: 2.9–64.3) more likely to borrow money, and 6.9 times (95% CI: 1.7–27.2) more likely to reduce spending on basic goods.

Fig. 1
figure 1

Risk of cost-co** strategies in respondents with high compared to low financial toxicity

Full size image

Forty-eight respondents (49%) completed the survey prior to the beginning of the COVID-19 pandemic, and 49 (51%) completed the survey during the pandemic. Prior to the pandemic, the median cost score was 27 (19–34), which improved to 32 (25–35) during the pandemic (p = 0.07). In the pre-pandemic cohort, 56% had household incomes < $100,000, which increased to 78% in the pandemic cohort (p = 0.08). There were no significant changes in cost-co** strategies or factors that contributed to financial stress in the pre-pandemic cohort compared to the pandemic cohort. Examining the financial impacts of COVID-19 only among the 49 respondents in the pandemic cohort, those with high FT were more likely to have lost their job (20% vs. 0%, p = 0.04), lost income (30% vs. 5%, p = 0.05), applied for financial assistance (30% vs. 0%, p = 0.01), reported more financial stress compared to before the pandemic (50% vs. 5%, p = 0.002), and were more likely to delay or avoid medical care for any reason (60% vs. 20%, p = 0.01).

Discussion

This study is the first to report on FT in a group of patients that has undergone radiation therapy for gynecologic malignancies and to evaluate the impact of the COVID19 pandemic on FT in this population. Compared to our prior work at the same institution and other studies utilizing the COST tool, the proportion of patients with FT is slightly lower at approximately 28%, but we similarly found younger age and private insurance to be associated with higher FT, and treatment characteristics, such as cancer type and stage, were not associated with degree of FT. In contrast to prior studies, income and receipt of chemotherapy were not associated with FT [2, 7], which may be due to insufficient power as a result of the sample size or the timing of administration of the survey. Unlike other studies that surveyed patients months to years after diagnosis, our survey was administered only one month after completion of treatment, and it is possible that patients have yet to experience the full financial impact of their disease.

Radiation-specific factors, such as type of radiation and duration of radiation treatment, were not associated with high FT. However, several factors, including some that are unique burdens of radiation treatment, were noted to be associated with FT. Specifically, respondents with high FT more often reported changes to employment (i.e., moving from full-time to part-time employment, taking unpaid time off or losing their job) than those with low FT. A recent analysis of patients with breast cancer receiving radiation treatment demonstrated that employment and marital status were associated with financial hardship, but that duration of treatment, while associated with measures of decreased employment, did not independently predict higher FT [17]. Further, our study demonstrated that more patients in the high FT group reported caregiver employment changes, notably caregivers taking more paid and unpaid time off from work. Other studies assessing caregiver FT have shown that FT in caregivers is associated with higher FT in patients and more advanced disease [18]. Additionally, over 60% of patients and caregivers reported lost wages due to underemployment or unemployment [18]. These employment changes can be particularly challenging for the younger patients who rely on employer-based insurance, have yet to accumulate large savings and also may have young dependents or elderly parents whom they are caring for.

When respondents were asked what contributed most to their financial stress, decreased ability to work, medical bills, and cost of transportation and parking were significantly more stressful for respondents with high FT. Finally, the risks of using cost-co** strategies, such as decreasing spending, borrowing money and delaying or avoiding medical care, among respondents with high FT were similar to our prior work [2, 4].

We noted some surprising results when comparing the cohorts surveyed before and during the COVID-19 pandemic. While not statistically significant, respondents reported lower FT, as evidenced by lower COST scores, during the pandemic. This was in spite of participants reporting lower income in the pandemic cohort, which is consistent with the national trends in income [12]. There were no other significant differences in the two cohorts related to demographic, disease or treatment variables. Perhaps the natural experiment of the dramatic shift in healthcare and individual behaviors during the COVID-19 pandemic may generate some hypotheses about strategies to reduce FT. For example, significant government funding and assistance during the pandemic may have alleviated financial stress for respondents. The shift to remote work may also have allowed for more flexible work schedules in patients and caregivers and therefore less disruption to work schedules and more transportation options. Or perhaps, the reduced total household expenses for individuals in the setting of significantly reduced travel and activities freed up income to pay for care, thereby decreasing financial stress despite lower household income. However, we also must consider that our pandemic cohort does not include respondents who were not present to care due to concerns over COVID-19 exposure and infection or possibly due to extreme financial hardship.

Our findings should be considered in the context of interventions that may buffer against FT. Such interventions span from the individual to the institutional, health system and governmental levels. At the individual level, interventions include screening for FT and patient navigation programs, which have shown promise in pilot studies [19,20,21]. At institutional and governmental levels, work-leave policies that incorporate protections for both patients and caregivers and flexible working schedules may minimize FT. Finally, health systems and governments must work to expand insurance coverage and reduce out-of-pocket costs (i.e., through contracts negotiated with pharmaceutical and insurance companies to reduce out-of-pocket costs related to drug costs, co-pays and deductibles) [22].

This study has several strengths. To our knowledge, it is the first study to assess FT in patients with gynecologic cancer undergoing radiation treatment, and the survey had a very high response rate of 95%. Additionally, due to the timing of the study overlap** with the onset of the COVID-19 pandemic, we were able to compare FT before and during the pandemic and investigate pandemic-specific effects on financial stress. This study is limited by the small sample size, especially in our ability to compare respondents before and during the pandemic. Our findings may not be generalizable to a broader population, given that our respondents received care at a single urban, academic institution and had insurance. Even so, our study sample was heterogeneous in terms of income, insurance type, education, and employment status.

In conclusion, FT is a substantial concern for patients with gynecologic cancer undergoing radiation treatment. Privately insured, younger patients are at higher risk of FT regardless of stage of disease, type of cancer or radiation treatment. Multivariable models highlighted the increased odds of delaying or avoiding care and need to borrow money or decrease spending on necessities for participants with high FT. More work is needed in a larger population to further investigate the lower level of FT during the pandemic and if confirmed, to consider what societal, policy and healthcare delivery changes may have led to subtle improvement in financial hardship for patients with cancer.