Introduction

The National Cancer Institute (NCI) defines a cancer survivor as any individual from the time of cancer diagnosis through the balance of life, including those who are receiving or have completed treatment, and who may have periods with and without active disease [15]. Collectively, there are over 18 million cancer survivors in the United States [3]. A growing population of survivors are people living with advanced or metastatic cancers; current estimates suggest that over 620,000 people in the United States are living with the most common types of metastatic cancers [9]. Advances in cancer therapies have transformed the survival trajectory of many metastatic cancers, with people living longer through periods with and without active disease, on and off treatment, and with evolving needs.

With survivors living longer and the cost of care and treatment rising, many experience substantial negative financial impact, even years after their initial cancer diagnosis [10]. There have been many terms used to describe financial impact, including financial hardship and financial toxicity. Financial hardship has been characterized as (1) material conditions (e.g., out-of-pocket costs, productivity loss, medical debt, bankruptcy) arising from cost of care and lower income due to employment effects during and after cancer treatment; (2) psychological responses to the increase in expenses or decreases in income; and (3) compensatory responses by patients and survivors to manage increased expenses or decreased income [2]. Financial toxicity describes the objective financial burden and subjective financial distress of patients with cancer as a result of treatment and health services [4].

People living with advanced or metastatic cancer may be particularly vulnerable to substantial financial impact [19]. Newer treatments including targeted and immunotherapies can be costly, leading to greater impact on individuals requiring these treatments, particularly as cost burdens continue to be shifted to patients [27]. Additionally, survivors and their caregivers may incur substantial out-of-pocket costs [27] and disruptions in employment related to medical care or treatment side effects [25]. Research in heterogeneous populations of cancer survivors suggests that financial impacts are associated with forgoing care, lack of adherence to prescription medications, and adverse health outcomes [2], including at least one study suggesting severe financial distress is associated with mortality [17]. Such impacts may be especially pronounced among underserved communities, where financial toxicity has been shown to perpetuate existing health disparities among those living with advanced cancer [22].

Despite the well-documented trend in rising cancer costs and disproportionate impact on vulnerable communities, there remains a dearth of evidence on people with advanced or metastatic disease, and its impact on adherence to treatment and healthcare utilization [19]. Of existing evidence, one study of 145 women with metastatic breast cancer showed that most experience financial toxicity [18], and another study found that those who were underinsured had significantly worse financial hardship [30]. Similarly, individuals with metastatic non-small cell lung cancer reported that over half reported two or more types of financial hardship (psychological, behavioral, material), and the most common out-of-pocket costs were medical, transportation, and lodging [12]. Limitations of the existing evidence for metastatic survivors are its focus on a single cancer type, care setting (e.g., academic medical center), and small sample sizes.

Given limitations of current evidence regarding financial impacts among survivors of advanced or metastatic cancer, the purpose of this study was to (1) describe financial toxicity reported by people with metastatic cancer; (2) identify clinical and sociodemographic characteristics associated with financial toxicity; and (3) examine associations between financial toxicity and compensatory strategies to offset costs, including postponing treatment, care, or supportive services.

Methods

Study population

Cancer Support Community’s (CSC’s) Cancer Experience Registry (CER) is an online, community-based research initiative designed to investigate the emotional, physical, practical, and financial impact of cancer. Adults (18 years or older) ever diagnosed with any cancer type are eligible to participate and invited to complete the open, web-based survey. Participants are recruited through CSC’s networks, including Cancer Support Community and Gilda’s Club partners, hospital and healthcare partners, advocacy partnerships, and social media. Eligibility criteria for study sample inclusion were: (1) participation in the CER survey from August 2017 to August 2021, (2) lived in the U.S., (3) completed the Functional Assessment of Chronic Illness Therapy COmprehensive Score for Financial Toxicity (FACIT-COST) measure, and (4) self-identified as having solid tumor metastatic cancer. Exclusion criteria included: (1) hematological malignancies; and (2) respondents who had missing data on the COST measure. Ethical and Independent Review Services (E&I, Independence, MO) served as the IRB of record (Study #16036). All procedures were in accordance with the ethical standards of the institutional research committee for studies involving human participants and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all participants.

Measures

Financial toxicity

Financial toxicity was measured using the FACIT-COST, a valid and reliable patient-reported measure of financial impact comprised of 11 items rated on a 5-point Likert scale (0 = Not at all; 4 = Very much) [6]. Responses were coded and summed (range: 0–44); lower scores indicate worse financial toxicity. This analysis utilized the COST measure as a dichotomous variable, where the cut-points are < 23 (mild, moderate, or severe financial toxicity) versus ≥ 23 (no financial toxicity), based on previous work where this approach has been documented to have high sensitivity (0.85) and specificity (0.82) to predict financial toxicity [8, 11, 23].

Postponing care

Participants rated how frequently (Never, Rarely, Sometimes, Often, Always) they used compensatory strategies to reduce the cost of cancer treatment by: (1) postponing doctors’ appointments; (2) postponing follow-up screening and/or blood work; (3) postponing filling prescriptions; (4) skipping dosages of prescribed drugs; (5) delaying follow-up on recommendations for supportive care (e.g., physical therapy, occupational therapy, nutrition counseling); and (6) postponing seeking psychological counseling or support. Items were collapsed into dichotomous categories for each item (0 = Never; 1 = Rarely to Always). Conceptually similar strategies were grouped into three categories: postponing medical visits (items 1 and 2), postponing or skipping medication (items 3 and 4), and postponing allied health and/or psychosocial support care (items 5 and 6). The three categories were each recoded into dichotomous index variables, indicating whether a participant endorsed at least one of the two strategies (0 = neither strategy; 1 = one or both strategies).

Socio-demographics and clinical history

Participants provided information about age, gender identity, race, Hispanic ethnicity, annual household income, education, employment status, and insurance coverage. Rural–Urban Commuting Area (RUCA) codes were determined for all respondents who provided their zip code, using census data from March 2020. RUCA codes combine population density with commuting patterns to classify geographies from 1 (most urban) to 10 (most rural); geographical residence was categorized as rural (RUCA code ≥ 4), suburban (RUCA code 2 or 3), and urban (RUCA code 1) [28]. Participants also reported clinical history including primary cancer diagnosis, year first diagnosed with cancer, whether they were currently receiving treatment, and where they received treatment. Current comorbidities (yes/no) were assessed from a list of 12 common medical problems (back pain, diabetes, heart disease, high blood pressure, kidney disease, liver disease, lung disease, obesity, osteoarthritis, osteoporosis, rheumatoid arthritis, ulcer or stomach disease) modified from the Self-Administered Comorbidity Questionnaire [20]. The number of comorbidities was categorized as none, 1 or 2, and 3 or more.

Metastatic cancer status

Participants were classified as having metastatic cancer if they reported 1) having ever been diagnosed with metastatic (stage IV) cancer, 2) that their most recent cancer was stage IV at diagnosis, 3) that metastatic breast cancer (MBC) was their primary diagnosis, or 4) that their cancer was currently metastatic (stage IV).

Analysis

Data analysis was conducted using Stata, version 17.0 (StataCorp LLP). Descriptive statistics were calculated for the full sample and by presence of financial toxicity. The comparability between groups with and without financial toxicity was assessed using bivariate two-sample t-tests (means) and Wilcoxon rank-sum tests (median) for continuous variables and chi-square tests and Fisher’s Exact test (if cell size < 5) for categorical variables. Next, logistic regression was used to analyze the relationship between socio-demographic and clinical history variables and presence of financial toxicity (0 = no financial toxicity; 1 = mild, moderate, or severe financial toxicity), adjusting for all other variables. Finally, three logistic regression models assessed predictors of postponing care were conducted. In these models, financial toxicity (Yes/No) was the predictor variable and (1) postponing medical visits, (2) postponing or skipping medication, and (3) postponing supportive care and/or psychosocial support services were the response variables, respectively, adjusting for all previously mentioned socio-demographic and clinical history variables. Missing data were imputed using multivariate normal distribution such that the full sample was retained in regression models. The STATA suite of mi commands created 50 imputed data sets and averaged results across these data sets for final estimates. The variables with the highest proportion of missing information were income (19%), age (5%), and geographical residence (5%). All analytic variables were included in the imputation model. A p-value of < 0.05 was considered statistically significant.

Sensitivity analysis

To account for incongruous self-reported responses regarding history of metastatic disease, a sensitivity analysis was conducted excluding those individuals who answered ‘no’ or ‘don’t know’ to ever having been diagnosed with metastatic (Stage IV) cancer, or did not answer the question, but indicated most recent cancer was stage IV at diagnosis, primary diagnosis was MBC, or current metastatic cancer (stage IV). Inferential analyses were conducted on the smaller sample, who indicated ‘yes’ to ever having been diagnosed with metastatic cancer, to confirm consistency in magnitude and direction of observed associations.

Results

Participant characteristics

A total of 484 individuals met inclusion criteria (Table 1; Fig. 1). A third of respondents had MBC (31%); other highly represented cancers included lung (13%), gynecologic (10%), colorectal (9%), and prostate (7%). The mean age of respondents was 60 (range 21–88 years), and most respondents identified as female (71%) and non-Hispanic White (83%). Over half of the sample completed the survey two or more years after their initial cancer diagnosis (61%) and were currently receiving treatment for their cancer (78%).

Table 1 Sample characteristics of individuals with metastatic cancer in the Cancer Experience Registry by financial toxicity level, 2017–2021 (N = 484)
Full size table
Fig. 1
figure 1

Cohort selection

Full size image

Financial toxicity

Half of participants reported mild, moderate, or severe financial toxicity on the FACIT-COST measure (50%; Table 1). Factors significantly associated with financial toxicity in the bivariate analyses included age, gender identity, annual household income, education, employment status, insurance coverage, number of comorbidities, and receiving treatment at an academic medical center or comprehensive cancer center. Adjusting for age, gender identity, race/ethnicity, annual household income, urban/rural status, education, employment status, insurance coverage, currently receiving treatment, time since diagnosis, cancer type, number of comorbidities, and receiving all or some treatment at an academic or comprehensive cancer center, multivariable modeling revealed several factors that were significantly associated with financial toxicity, including age, race/ethnicity, annual household income, rural/urban residency, education, employment, time since diagnosis, and number of comorbidities. Specifically, older age was associated with lower odds of financial toxicity (OR 0.95; 95% CI 0.92–0.98) (Table 1). Conversely, compared to non-Hispanic Black respondents, individuals who identified as non-Hispanic White (OR 3.56; 95% CI 1.32–9.58), Hispanic (OR 4.42; 95% CI 1.19–16.43), or multiple races (OR 6.78; 95% CI 1.45; 31.75) had higher odds of financial toxicity. Respondents with an annual household income of < $40,000 per year (OR 4.35; 95% CI 2.18–8.70), whose educational attainment was less than a college degree (OR 2.09; 95% CI 1.12–3.89), and who were not employed (OR 2.74; 95% CI 1.40–5.35) also had higher odds of financial toxicity compared to those with higher income, greater education, and current employment, respectively. Cancer type was not associated with level of financial toxicity; however, being less than one year since cancer diagnosis (OR 2.64; 95% CI 1.29–5.42) was associated with higher odds of financial toxicity as compared with five or more years since diagnosis. Further, having 1–2 comorbidities (OR 2.29; 95% CI 1.22–4.28) or 3 + comorbidities (OR 2.36; 95% CI 1.19–4.70) were each associated with greater odds of financial toxicity than those without comorbidities.

Postponing care

Participants endorsed postponing or delaying care (Rarely to Always) to offset care costs, including postponing doctor’s appointments (21%); postponing follow-up screening and/or blood work (16%); postponing filling prescriptions (18%); skipping dosages of prescribed drugs (17%); delaying follow-up on recommendations for supportive services (26%); and postponing seeking psychological counseling or support (28%). Individuals with financial toxicity were significantly more likely to report postponing or delaying each type of care compared to those without financial toxicity (Table 2). After adjusting for age, gender identity, income, education, employment, insurance type, and the number of comorbidities, these associations remained statistically significant. When collapsing the six items into three different categories (postponing medical visits, suboptimal medication adherence, and postponing supportive services and psychosocial support), the odds of postponing care in each category were significantly higher for respondents with financial toxicity versus those without (postponing medical visits adjusted OR 2.58, 95% CI 1.45–4.58; suboptimal medication adherence adjusted OR 5.05, 95% CI 2.77–9.20; and postponing supportive services adjusted OR 4.16, 95% CI 2.53–6.85) (Table 3).

Table 2 Odds ratios and 95% confidence intervals for postponing care among those experiencing any financial toxicity compared to those not experiencing financial toxicity: individual postponing/delaying items
Full size table
Table 3 Odds ratios and 95% confidence intervals for postponing care among those experiencing any financial toxicity compared to those not experiencing financial toxicity: combined postponing/delaying items
Full size table

Sensitivity analysis

A total of 375 patients indicated ‘yes’ to ever having been diagnosed with metastatic (Stage IV) cancer; 95 patients answered ‘no’ and 14 responded ‘don’t know’, or skipped the question, but also indicated their most recent cancer was stage IV at diagnosis, primary diagnosis was MBC, or they currently had metastatic cancer (stage IV). Bivariate or multivariable analysis on the smaller sample (n = 375) did not alter the conclusions (results not shown).

Discussion

In this study of survivors with metastatic cancer, financial toxicity was common and significantly associated with postponement of medical and supportive care services. Self-reported lower income, being uninsured, being unemployed, and having less than a college degree were associated with greater odds of financial toxicity in this population. These findings build upon previous studies conducted among smaller samples of people living with metastatic breast and advanced lung cancers [12, 30] and are consistent with trends documented in survivors across the cancer continuum [31]. Vulnerabilities related to insurance and employment status, income, and educational attainment, coupled with length of treatment and treatment advances that can improve survival, may predispose people with advanced and metastatic disease to experience additional costly treatment and out-of-pocket costs throughout the trajectory of care. In addition, there may be great uncertainty in treatment plans and prognosis, making it difficult for individuals with advanced disease to prepare for the long-term financial burden of cancer and its treatment.

Contrary to other studies [31], the findings showed that people with metastatic cancer who identified as non-Hispanic White, Hispanic, or more than one race had greater odds of financial toxicity compared to those identifying as non-Hispanic Black. These results may be influenced by sampling artifacts or other unmeasured contextual factors. For example, a recent survey on racial bias in oncology care delivery showed that, while Black respondents more often reported negative care experiences compared to White respondents, they also reported the highest rates of oncologist conversations about cost of treatment and insurance coverage as compared to other races [21]. The intersection of race/ethnicity, financial toxicity, and other contextual factors remains an important area of study, and future research may center on those who would most benefit from tailored support to prevent, identify, and mitigate financial toxicity.

Given the role of financial toxicity in perpetuating health disparities [22], the consistent associations between financial toxicity and postponement of care by people with metastatic cancer is notable. Those who reported financial toxicity were more than five times more likely to postpone filling prescriptions or skipping doses of their medications, and more than twice as likely to postpone medical care. Prior research has suggested that associations between financial toxicity and mortality for cancer survivors may be due in part to treatment non-adherence [17]. In addition, higher out-of-pocket costs have been associated with greater odds of non-adherence to oral anticancer medications [29] and general prescription utilization [5]. These results demonstrate the critical impact of financial toxicity on accessing healthcare services and medications and underscore the importance of identifying and addressing financial hardship early.

Additionally, those reporting financial toxicity had four times greater odds of delaying follow-up recommendations for receiving supportive care and psychosocial services. Comprehensive whole-person care, including distress management [7] and rehabilitation services [24], have been recommended for all people with cancer, and there is growing recognition of the importance of this care for people with advanced and metastatic cancer in particular [26]. By postponing these services, people may experience worse health-related quality of life and severe physical or psychosocial impairments. Future research should examine the care decisions that individuals make after a cancer diagnosis, both during treatment and in long-term follow-up care, and the impact of financial toxicity on those decisions. Prospective longitudinal research is also needed to clarify what predicts improved or worsening financial toxicity and coping behaviors over time in people living longer with advanced and metastatic cancer, to inform future interventions.

Collectively, the results demonstrating a link between financial toxicity and care postponement among people with metastatic cancer underscore the need for cancer care delivery and healthcare policy solutions. Indeed, there is growing consensus that routine screening for financial toxicity should be embedded into oncology care [1, 2, 4, 31] and that high quality care should include patient-provider communication about cost of treatment as part of informed decision making [1, 13]. Ideally, a multi-level approach to prevent and mitigate financial hardship at the patient, provider, practice, employer, and policy levels should be implemented to address the growing cost of care for people with cancer [31]. In the near term, research is needed that identifies effective navigation for financial burden, including how to best support health care teams in discussing care costs in a manner that takes into account metastatic patients’ perspectives and preferences. It is important to note two recent efforts to further research focused on people living with metastatic cancer. NCI recently released a request for application (RFA): Research to Understand and Address the Survivorship Needs of Individuals Living with Advanced Cancer, to support research that includes strategies to understand and/or address financial concerns among this population [16]. Additionally, CancerX is a public–private partnership to boost innovation in the fight against cancer. Its efforts include the development of evidence, best practices, and implementation projects to reduce the burden of cancer [14]. Future research on financial toxicity for people living with metastatic cancer may leverage these opportunities.

The strengths of the current analysis include participation by a broad population of metastatic cancer survivors, representing over 25 diagnostic cancer types from across the United States. This expands upon previous literature that recruited people living with advanced and metastatic cancer from a single care setting and/or one cancer type. The results, however, should be viewed in light of certain limitations. Selection bias may be present as the web-based survey was shared widely and survey participation required internet access. Survey respondents may be more engaged with healthcare and supportive services than those who did not participate. While the current sample includes geographically diverse participants, the sample is not nationally representative of all people living with metastatic cancer in the U.S. as study participants were predominantly female and identified as non-Hispanic White. Additionally, the use of the COST measure as a dichotomous outcome (no financial toxicity versus mild, moderate, or severe financial toxicity) may overestimate risk for mild financial toxicity and underestimate risk for moderate-severe financial toxicity particularly for individuals who may be financially vulnerable due to low income or lack of employment. We also acknowledge limitations in potential errors of self-reported data in terms of cancer history and stage.

In summary, this study revealed that financial toxicity is common for many people with metastatic cancer and is associated with postponing medical visits, suboptimal medication adherence, and postponing supportive care services. With increases in the number of people living longer with metastatic cancer, the growing evidence that financial toxicity perpetuates health disparities, and the rising costs of cancer-directed therapies, there will continue to be a need to systematically screen and intervene to prevent, identify, and mitigate financial impact for this population.