Objective Predictors of Financial Toxicity in Oncology

ABSTRACT

BACKGROUND: Current financial toxicity (FT) screening tools rely on patient-reported risk factors. Underrepresented populations may not be forthcoming about FT fears due to cultural concerns of treatment withholding or migratory repercussions, if applicable. Identifying objective risk factors, such as social determinants of health (SDOH) and disease-specific factors (DSF), could reduce FT in patients with cancer and throughout health care systems.

METHODS: This was a multicenter retrospective study evaluating SDOH and DSF associated with FT (defined as ≥ $15,000 owed) related to cancer treatment. Inferential statistics were used to evaluate differences between the FT cohort and those who owed less than $15,000. Continuous data were compared with a Student t test or Mann-Whitney test, depending on distribution. Categorical outcomes were compared with a χ² test. A logistic regression model was used to evaluate multivariate associations with FT, using a P value of less than .05 to define significant results.

RESULTS: The sample comprised 162 records, 81 in each group. Univariate analyses demonstrated participants’ differences in age, relationship with a primary care provider, country of origin, insurance status, education level, need for an English interpreter, whether their disease was stage IV at diagnosis, recurrent or metastatic disease, use of immune checkpoint inhibitors, and use of targeted molecular therapy. Employment status and
marital status were not statistically different. The logistic regression model showed that lack of insurance and having stage IV disease at diagnosis were significantly associated with FT (P = .001 and P = .0495, respectively).

CONCLUSIONS: Objective FT screening can minimize response bias and incidence in those at increased risk. In our study, we found that individuals who are first-generation Hispanic immigrants and lack English proficiency faced significant barriers to receiving help for the high financial costs of
medical care. These findings identify specific subpopulations at risk for FT and will guide prospective interventions looking to minimize FT. Health care systems should analyze objective measures of FT while considering
loco-regional and subcultural SDOH/DSF to overcome response bias.

Financial toxicity (FT) refers to the negative effects associated with the cost of medical care for patients, their families, and health care systems.1-5 High prices for consultations, diagnostic or staging tests, treatments, and transportation, which may be only sporadically covered by insurance, and patients’ reduced productivity or inability to work during or after their treatment can lead to FT.6,7 FT can affect a patient’s and their family’s sense of well-being, ability to pay for basic needs, quality of life, health care access, and treatment adherence.6,7 In an attempt to mitigate FT, health care systems routinely offer financial navigation services, although many lack a standardized FT referral system or culturally appropriate screening procedures and tools. This results in variable accessibility and acceptability of FT navigation.

Unsurprisingly, FT remains prevalent for patients with cancer.8,9 Information about FT in certain underrepresented populations is scarce in part due to the aforementioned issues and patients’ hesitation to disclose financial vulnerability, due to a cultural belief that appropriate and timely health care may be withheld if they are unable to pay for it. It is crucial to prospectively identify those at higher risk for FT to minimize impact on their financial and general well-being and medical nonadherence while maintaining financial health of health care systems.

The aim of this study was to investigate social determinants of health (SDOH) and disease-specific factors (DSF) associated with FT in patients with cancer at our integrated health care system (Hartford Hospital, Hartford Healthcare, Connecticut). We hypothesized that by obtaining objective predictors of FT, we could identify and refer patients to financial navigation, resulting in lower debt for patients and the health care system.

Methodology
This was a single-center retrospective study looking to identify common SDOH and DSF of patients who had significant debt related to their cancer care. FT was defined as owing $15,000 or more without any account activity for 6 months or longer. The comparator group consisted of an unmatched cohort with less than $15,000 in debt owed to our cancer institute. To be included in either cohort, patients had to (1) be aged 18 to 89 years, (2) have a confirmed diagnosis of cancer and be receiving cancer care at our institution, (3) be of any sex, race, or ethnicity, and (4) have an active electronic medical record (EMR).

Following Institutional Review Board (Hartford Hospital) approval, the EMR was accessed to extract predetermined SDOH and DSF (Table 1). Inferential statistics were used to evaluate differences between the FT and comparator cohorts. Continuous data were compared with a Student t test or Mann-Whitney test, depending on distribution. Categorical outcomes were compared with a χ² test. A logistic regression model was used to evaluate multivariate associations with FT, using a P value of less than .05 to define statistically significant results using SPSS Statistics 29.0 (IBM; 2022).

Results
The sample comprised 162 patients, 81 in each group. There were statistically significant differences between cohorts in several SDOH variables, including age, lack of a primary care physician (PCP), non-US country of origin, lack of insurance, Spanish spoken at home, and the need for an English interpreter (Table 2). There was no association between FT and sex assigned at birth, gender identity, race, ethnicity, education, employment status, or marital status. Similarly, no association was found based on lifestyle habits (history or current use of alcohol, tobacco, or intravenous drug use).

There was a significant association between FT and several DSF variables: stage IV disease at diagnosis, whether they had recurrent or metastatic disease, use of immune checkpoint inhibitors (ICI), and use of targeted molecular therapy (Table 3). There were no statistically significant differences between FT and the comparator based on site of disease, use of chemotherapy, radiation therapy, or surgery and/or hormonal therapy. When all variables that showed a univariate difference were included in a logistic regression model, lack of insurance and stage IV at diagnosis were significantly associated with FT (IV = .001 and P = .0495, respectively). Patients covered by Medicare and private insurance had a statistically lower likelihood of experiencing FT than the referent uninsured patients (OR, 0.027; 95% CI, 0.004-0.172). Patients covered by Medicaid constituted too small a sample (n = 4) to be evaluated. Patients with earlier stage cancer at diagnosis (compared with the referent, stage IV) also were associated with a statistically lower likelihood of experiencing FT (stage I: OR, 0.032; 95% CI, 0.003-0.383; P = .007; stage II: 0.076, 0.007-0.797; P = .032). Statistics for patients with stage III disease did not differ from those with stage IV disease.

Discussion
FT can negatively impact patients’ and their families’ sense of well-being, quality of life, and medical adherence1-4, and also can jeopardize the financial stability of health care systems.5 Many institutions lack a systematic FT screening procedure and depend on patients or families to express concerns regarding FT before being referred to financial navigators. Further, current screening tools likely underreport FT in underrepresented populations, given linguistic and cultural incongruences.10-12 It has been hypothesized that certain at-risk underrepresented populations, such as first-generation Hispanic patients and their families or caregivers, are hesitant to disclose financial concerns due to cultural beliefs that their health care may be altered, withheld, or refused if they were unable to pay, either out-of-pocket or through insurance. In a first step toward creating a standardized financial navigation model at our integrated cancer system, we decided to identify SDOH and DSF associated with FT in our patient population. We hypothesized that this would allow us to more efficiently utilize current FT resources, which are frequently outnumbered by patient demand, given the vulnerable populations commonly served by our integrated health care network.

Our study identified that patients aged 50 to 69 years and those lacking a PCP and insurance were at increased risk for FT. In addition, we demonstrated that non-US citizens, non-English speakers, and those requiring an interpreter to receive their cancer care were statistically associated with FT. This confirms our hypothesis that underrepresented communities, especially first-generation immigrants and those for whom English is a second language, are at increased risk not only for FT but also for medical jeopardy, as noted by their lack of a PCP.

We also found that those patients with higher burden of disease (de novo stage IV, recurrent or metastatic disease) and those requiring newer oncologic treatments (eg, immunotherapy or targeted molecular therapy) were at increased risk for FT. High prices for newer cancer therapies are the result of complex models that take into account research and development costs, duration of patents, and arrangements with payers, pharmacy benefit managers, and health care systems. Although some drug companies may offer patient assistance programs (PAPs) for uninsured or at-risk patients, these programs often require information that non-US citizens may not have available or that underserved populations may be hesitant or unable to provide. Examples include all income sources, bank account balances, salary stubs, proof of immigration status, and insurance information.

Educating patients on the safety of PAPs, clarifying misconceptions that the purpose of a PAP is not to identify patients with financial or migratory troubles, and helping them complete required paperwork when linguistic or reading abilities are limited may increase the number of underserved patients benefiting from PAPs. In addition, confirming linguistic appropriateness of paperwork and making certain questions optional, especially those questions that underserved patients may perceive as intrusive, could also minimize patient avoidance. In this study, considering that FT was not associated with early-stage disease, preventive and early detection interventions should continue to be prioritized.

The American Society of Clinical Oncology (ASCO) Patient Centered Cancer Care Certification Program showed disparities in the robustness of FT screening procedures in academic vs community-based cancer centers, with the latter appearing to prioritize the matter further.13 The program recommends adequate patient engagement in and empowerment to share in decision-making about their cancer care and access to cancer care at the right time, in the right setting and helps guide improved financial navigation in those centers looking to join the program.13 The program also requests that all patients be routinely provided with a best estimate of out-of-pocket expenses for any new therapy that is offered and that “patient financial counseling services, including assistance programs that are available, are routinely provided to all patients.”14

Certain institutions may not be able to increase their financial navigation resources due to preexisting FT impacting their health care system. Securing insurance reimbursement for these services, which has been lacking, would likely increase FT resource availability in many institutions. In addition, we believe it is necessary to guarantee cultural and linguistic sensitivity and congruence of intake and financial navigation forms to minimize patient
avoidance while increasing acceptability and retention.

Until all of this has been accomplished, identifying and prioritizing those with objective SDOH and DSF predictors of FT at their time of entry into the health care system will allow early identification of at-risk populations and increase efficiency of available resources while securing financial well-being of patients, their families, and health care systems.

Conclusions
Sufficient financial navigators to serve patient populations in a culturally and linguistically adequate manner are crucial to increase recruitment and retention into those services. Many institutions may be unable to immediately increase financial navigation resources because they remain nonreimbursable. Identifying objective predictors of FT allows for early identification and prioritization of those at higher risk and should facilitate better use of already available financial navigation resources. Further, we believe that taking into account loco-regional SDOH/DSF, subcultural differences, and patient misconceptions on financial navigation and PAPs would allow response bias to be overcome. In our patient population, non-US country of origin and lack of English proficiency supported that first-generation Hispanics are inappropriately navigated for FT. These findings identify specific subpopulations at risk for FT and can guide prospective interventions looking to increase efficiency of available services. Additional studies are needed to determine if an opting-out financial navigation system can minimize FT in underrepresented populations.

Funding: No funding sources.

Corresponding author: Alvaro Mendendez, MD, is a breast medical oncologist at Hartford HealthCare Cancer Institute, Hartford, Connecticut. He is an assistant professor of medicine, Department of Internal Medicine, University of Connecticut, Farmington, Connecticut. alvaro.mendendez@hhchealth.org

About the authors: Aleksei Bazhenov, MD, PhD, is affiliated with Yale New Haven Hospital, New Haven, Connecticut; he is formerly affiliated with University of Connecticut School of Medicine, internal medicine residency. Luiza Doro, MD, and Alvaro Menendez, MD, are affiliated with the University of Connecticut School of Medicine, internal medicine residency, in Farmington. Menendez is also affiliated Hartford HealthCare Cancer Institute and is medical director, cancer disparities and health equity, Hartford HealthCare Cancer Institute in Connecticut. David M. O’Sullivan, PhD, is affiliated with Hartford Hospital, Hartford HealthCare.

References
1. Halpern MT, Thamm C, Knowles R, Chan RJ. Cancer survivors’ experience of care and financial toxicity: results from a national survey. JCO Oncol Pract. 2025;21(2):226-234. doi:10.1200/OP.24.00370
2. Kadambi S, Wang Y, Job A, et al. Financial toxicity in older adults with cancer and their caregivers. JCO Oncol Pract. 2025;21(1):92-99. doi:10.1200/OP.24.00105
3. Pisu M, Martin MY. Financial toxicity: a common problem affecting patient care and health. Nat Rev Dis Primers. 2022;8(1):7. doi:10.1038/s41572-022-00341-1
4. Sadigh G, Switchenko J, Weaver KE, et al. Correlates of financial toxicity in adult cancer patients and their informal caregivers. Support Care Cancer. 2022;30(1):217-225. doi:10.1007/s00520-021-06424-1
5. Mariotto AB, Enewold L, Zhao J, Zeruto CA, Yabroff KR. Medical care costs associated with cancer survivorship in the United States. Cancer Epidemiol Biomarkers Prev. 2020;29(7):1304-1312. doi:10.1158/1055-9965.EPI-19-1534
6. Abrams HR, Durbin S, Huang CX, et al. Financial toxicity in cancer care: origins, impact, and solutions. Transl Behav Med. 2021;11(11):2043-2054. doi:10.1093/tbm/ibab091
7. Zafar SY, McNeil RB, Thomas CM, Lathan CS, Ayanian JZ, Provenzale D. Population-based assessment of cancer survivors’ financial burden and quality of life: a prospective cohort study. J Oncol Pract. 2015;11(2):145-150. doi:10.1200/JOP.2014.001542
8. Khan HM, Ramsey S, Shankaran V. Financial toxicity in cancer care: implications for clinical care and potential practice solutions. J Clin Oncol. 2023;41(16):3051-3058. doi:10.1200/JCO.22.01799
9. Shankaran V, Li L, Fedorenko C, et al. Risk of adverse financial events in patients with cancer: evidence from a novel linkage between cancer registry and credit records. J Clin Oncol. 2022;40(8):884-891. doi:10.1200/JCO.21.01636
10. Witte J, Mehlis K, Surmann B, et al. Methods for measuring financial toxicity after cancer diagnosis and treatment: a systematic review and its implications. Ann Oncol. 2019;30(7):1061-1070. doi:10.1093/annonc/mdz140
11. de Souza JA, Yap BJ, Wroblewski K, et al. Measuring financial toxicity as a clinically relevant patient-reported outcome: the validation of the Comprehensive Score for financial Toxicity (COST). Cancer. 2017;123(3):476-484. doi:10.1002/CNCR.30369
12. Peppercorn J, Gelin M, Masteralexis TE, Zafar SY, Nipp RD. Screening for financial toxicity in oncology research and practice: a narrative review. JCO Oncol Pract. 2025;21(1):5-11. doi:10.1200/OP-24-00989
13. ASCO certified: patient-centered cancer care standards. Accessed February 20, 2025. https://cdn.bfldr.com/KOIHB2Q3/as/j9kr2bht6k2vc5bc6728bq/Oncology-Medical-Home-Standards-Manual
14. Makhoul I, Power S, Smith C, Niece T, O’Mahony D. ASCO domestic quality programs. JCO Glob Oncol. 2024;10:e2400058. doi:10.1200/GO.24.0005