Understanding Financial Toxicity in Patients with Head and Neck Cancer: A Systematic Review

Background: Cancer treatment often results in financial burdens for patients including healthcare costs as well as treatment-induced disability leading to “financial toxicity” (FT) and decreased quality of life. The purpose of this review is to describe FT related to head and neck cancer (HNC) treatment, including quantifications of direct and indirect costs and descriptions of measurement tools. Methods: PubMed, Embase, Cochrane Library, and Web of Science databases were searched to identify articles published before April 2022. Full-text published studies were included if they assessed direct or indirect costs of HNC treatment; studies were excluded if they did not focus on HNC or financial burden. The risk of bias was assessed, and the results of the studies were synthesized. Results: Database searches yielded 530 unique studies, and 33 studies met the criteria for inclusion. Medical expenses for patients with HNC were higher than for patients with other cancers or controls in several studies. Major surgical procedures, neck dissection, free-flap reconstruction, and intensive care unit admission increased hospital costs. Trimodal therapy with surgery plus chemoradiation represented the most expensive treatment, and chemoradiation increased complication-related health care costs. In several studies, >50% of patients treated for HNC were disabled and did not return to work. One of the greatest contributors to the indirect cost of HNC treatment is the loss of lifetime wages. Patients with HNC are at risk for depression, anxiety, and social isolation, which are linked to a decreased quality of life and treatment non-adherence. The only tools used to assess FT in patients with HNC are the Comprehensive Score for financial Toxicity (COST) and the Financial Index of Toxicity (FIT). Conclusion: Financial toxicity is highly prevalent among patients with HNC. Further research is needed to validate the assessment tools for quantifying FT in HNC patients.


Introduction
Biomedical advances in HNC therapy, surgery, and technology have prolonged patient survival, slowed cancer progression, and reduced the use of toxic therapies. 1 However, treatments can still result in substantial physical, functional, and psychosocial morbidities that decrease the level at which patients can function. [2][3][4][5] In addition, the sizeable out-of-pocket costs for treatments further burden the estimated 15.5 million cancer survivors in the United States. 6,7 Cancer-related financial burden, or FT, encompasses the direct and indirect medical costs that patients incur throughout their treatment and surveillance ( Figure 1). Direct costs are those related to medical costs of cancer treatment, side effect management, and survivorship care, whereas indirect costs are those related to treatment-induced disability, lost productivity, and lost wages. The "toxicity" of these cancer-related financial burdens is reflected in patients' psychological distress and decreased quality of life (QOL), leading to increased rates of nonadherence to treatment and worse overall disease survival. [8][9][10] With approximately 550,000 new cases of HNC worldwide each year 11 and more than US$3.64 billion in total annual medical costs in the United States, 12 FT is an important and increasingly prevalent issue. A more thorough understanding of the sources of FT in patients with HNC is needed to design interventions and develop strategies to mitigate this problem.
The aims of this systematic review were to synthesize evidence on the components of financial burden faced by patients with HNC and address the following research questions and objectives: (1) Are patients with HNC more susceptible than patients with other cancers to FT? (2) What are the major factors contributing to the high costs of treatment in HNC? (3) Has FT been shown to adversely affect health-related QOL (HRQOL) in patients with HNC? and (4) To compare the metrics of validated tools that exist to quantitatively assess financial burden specifically in HNC. 2

Outcomes assessed and eligibility criteria
This study was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We identified articles published from the start of the database until the time of our search in April 2022. Primary outcomes included assessments of direct costs among HNC patients (including inpatient and outpatient services, surgical procedures, hospital costs, treatment modality, and treatmentrelated complications) and indirect costs (including adverse effects on employment and productivity). Consequences of FT associated with HNC treatment such as the negative impact on QOL and cost-coping strategies are included as a secondary outcome. The last outcome assessed was the use of quantitative assessments specifically addressing FT in patients with HNC. Studies assessing these outcomes during any timeframe of the HNC diagnosis, treatment, and survivorship periods were included in the review. Case reports, commentaries, abstract only, and non-English publications were excluded from fulltext review. Studies were excluded if they did not specifically address patients with HNC and financial burden or direct/ indirect costs as outlined above.

Search strategy and study selection
Independent searches of PubMed, Embase, Cochrane Library, and Web of Science databases were conducted to identify articles published from the start of the database until the time of our search in April 2022 describing FT in HNC. The keywords used were synonyms for FT (such as financial burden, burden of treatment, financial hardship, financial effects, financial impact, health care costs, cost of treatment, cost of illness, health expenditure, economic burden, and cost of illness) and synonyms for HNC (such as head and neck tumor, neoplasm or carcinoma, cancer of the head and neck, or head and neck surgery) in the title and abstract. These components were used to identify corresponding MeSH terms for search strings in PubMed. Search terms were modified on the basis of keyword availability for each database to ensure a broad scope of study inclusion. See Supplement 1 for a full list of search terms for each database. Additional references were obtained from a manual search of the bibliographies within the retrieved articles. Search results were uploaded to an EndNote database, and duplicates were removed. Subsequently, exclusion and inclusion criteria were applied, and eligible full-text studies were reviewed ( Figure 2).

Data collection
Three independent authors (M.R.-S., S.P., and C.P.) participated in the review process and screened and selected studies for full-text review and inclusion. Discrepancies regarding article inclusion were resolved via discussion among the authors to achieve consensus with input from the senior author (N.G.). For full articles included in the final review, data regarding the study population, study method, tools and outcomes used, figures regarding direct cost of treatment, and objective and subjective indirect cost statistics were collected and tabulated.

Assessment of bias
Studies were evaluated for risk of bias using the Newcastle-Ottawa Quality Assessment Scale (NOS) by three authors (M.R.-S., S.P., and C.P.) independently, with discrepancies resolved after joint article review and discussion with input

3
from the senior author (N.G.) ( Table 1). The NOS includes three categories for quality assessment: selection, comparability, and outcome. Selection was graded on a 4-point scale with 1 point each rewarded if (1) the study cohort was representative of the average HNC population, (2) the selection of the comparison group was from the same community as the exposed cohort, (3) the ascertainment of data was drawn from medical records or structured interview, and (4) the demonstration that the outcome of interest was not present at the start of the study. Comparability was graded on a 2-point scale with 1 point each rewarded if (1) the study provided a control group for non-HNC patients or (2) a group was selected to control for additional relevant factors such as socioeconomic factors, cancer subsite, insurance, or treatment modality. Outcome was graded on a 3-point scale with 1 point each rewarded if (1) the assessment of the study outcome was independent, blinded, or based on objective data from records; (2) follow-up was long enough to assess outcomes; and (3) there was a limited loss of patient data due to attrition.

Results
The database searches and manual cross-reference search identified a total of 675 articles. After applying the PRISMA protocol criteria, a total of 33 studies were included in the literature review. There was heterogeneity in the reporting of financial burden for patients with HNC. Studies varied on reporting total healthcare costs over varying time frames, out-of-pocket costs, annual costs, median versus mean costs, or costs based on insurance status, etc, and thus, a qualitative synthesis of the studies was used to integrate study results by analyzing common themes present in the available literature. It was not possible to perform a meta-analysis on study results.

Clinical Medicine Insights: Oncology
Jones et al 17 reported that the cost of a hospital bed accounted for 36.4% of total cost related to HNC treatment, while 24.5% of the total cost was for operating room fees. Postoperative ICU admission was an additional source of added cost. 17,20 Four studies reported on the cost of treatment related to treatment modality. 2,18,19,21 Surgery alone had the lowest cost, followed by radiation alone, surgery plus radiation, and chemoradiation, whereas trimodal therapy (including surgery and chemoradiation) was the most expensive. 2,18,19,21 Costs associated with multimodality treatment were as high as US$153,892 in one study (twice the cost of single-mode therapy). 2 Chemotherapy and radiation therapy increased the respective median overall costs 42% to 43% (P < .001) in Tom et al 18 and five-fold and two-fold in Moore et al. 21 Radiotherapy for early glottic cancer was 15.5 times more costly than endoscopic laser surgery in Brandenburg et al. 22 Lang et al 19 reported that radiotherapy alone (US$56,900) was less costly than multimodal therapy with chemoradiotherapy (US$91,564) and had lower 6-month complication-related health care costs (US$6,223) than chemoradiotherapy (US$15,825).

Clinical Medicine Insights: Oncology
Ten studies reported on complications related to HNC treatment and the resultant increase in health care costs, often the results of prolonged hospitalization stays. [14][15][16][17]19,20,[23][24][25][26] Chemotherapy-related complications were specifically associated with significant increases in health care costs in three studies. 16,19,25 For example, Lang et al 19 reported that chemotherapy complications cost an extra US$14,714, primarily for inpatient costs: treatment-related complications occurred in 86% of patients treated with chemoradiotherapy versus 51% in those who were treated with radiotherapy (P < .001); oral complications, dehydration, and electrolyte imbalances were the most expensive complications and were more common (P < .001) among patients who received chemoradiotherapy. Nonzee et al 23

Indirect costs associated with HNC treatment
Eleven studies included in the systematic review reported on indirect costs associated with HNC treatment (Table 3). 2,6,9,[28][29][30][31][32][33][34]40 Eight studies reported on issues surrounding employment and disability induced by their cancer treatment. 2 28 For many patients that were able to continue working, there was lost productivity from missed work for clinic visits, reduced work hours, and use of unpaid time off and sick leave. 27,29 Massa et al, 29 reported that patients' greatest financial and stress burdens (13% and 18%, respectively) were from lost income as a result of absence from work. Mongelli et al 28 found that lost productivity at work was significantly (P < .01) associated with worse fatigue (β = 5.99) and social functioning (β = 4.07). Issues with social eating and anxiety from oral dysfunction in some HNC survivors were identified as barriers to employment. 19,23,32 In one study, patients with oral, oropharyngeal, and salivary gland cancers who continued to work used 44.9 more short-term disability days than other employees (P < .01). 2 In a study by Cohen et al, 31

Consequences of FT
Ten studies included in the systematic review reported on consequences of FT associated with HNC treatment (Table 4). 3,8,9,27,28,[35][36][37][38][39] Four studies reported that the HRQOL for HNC patients significantly declined because of their health status and treatment-related social function, appetite, sleep, pain, mucositis, and dysphagia. 8,[35][36][37] Four studies reported on depression, anxiety, or stress related to cancer treatment. 3,28,29,35 Chen et al 3 found that 17% (36/211) of patients treated with radiation for SCC HNC had depression 1 year later (related to smoking, feeding tube dependence, and presence of tracheostomy or laryngeal stoma). Mongelli et al 28 reported that financial difficulties for 43% of thyroid cancer survivors were associated with worse anxiety (β = 5.07; P < .01) and depression (β = 5.47; P < .01). Nilsen et al 35 reported that of 228 survivors of HNC, 132 (56%) reported that treatment-related symptoms of depression and anxiety impacted their daily life. In the study by Massa et al, 29 38.9% of the stress reported by patients with HNC was attributable to medical expenses, while transportation concerns were the greatest source of stress for 27 patients and surgery with adjuvant chemoradiation was reported as more stressful than surgery alone. Four studies    Patients who reported their mood as "somewhat depressed" and "extremely depressed" at 1 year after completion of RT was 12% and 5%, respectively; at 3 years, 8% and 7%; at 5 years, 9% and 4%. The proportion of patients using antidepressant medications was 6% (2 of 36), 11% (2 of 18), and 0% (0 of 7), at 1, 3, and 5 years. Self-reported data on demographics, income, wealth, cost-coping strategies, out-ofpocket costs, medication adherence, and social isolation Medicaid patients were more likely to use cost-coping-strategies, associated with decreased wealth and higher total out-of-pocket costs (P < .01). Patients with high perceived social isolation were more likely to use cost-coping strategies (OR, 11.5), to report nonadherance to supportive medications (21.4 vs 5.45 days over 6 months), and to miss appointments ( Patients who continued working used sick leave, took unpaid time off, or arranged to work fewer hours; patients with lower COST scores were more likely to report worry and financial burden (P < .0001) and to use savings or borrow money, skip clinic visits, or not fill prescriptions because of the cost Mongelli et al 28

Assessment of FT in HNC
Several instruments have been developed to assess a patient's financial burden in relation to their cancer treatment; however, we only identified two instruments that were used to assess financial burden specifically in patients with HNC. Table 5 summarizes comparative metrics for these two tools: the COST and the FIT. The COST survey was developed and validated by de Souza et al 41 in 2014 in a group of 155 patients and was studied again in 2017 in 233 patients; 39 both groups had advanced-stage IV solid tumor cancers and were receiving chemotherapy. The authors developed an 11-item survey, with each item graded on a 5-point Likert scale (0 to 4): total scores range from 0 to 44, and lower scores indicate worse FT. The tool is meant to assess patientreported financial distress and how this impacts their lives.
The COST survey has been used in two published studies in patients with HNC. 27,38 In a study of 64 patients with HNC treated with radiation therapy by Beeler et al, 27 patients with greater FT (indicated by lower COST scores) exhibited costcoping strategies: 54% decreased spending on food and clothing, 43% used savings, and 13% borrowed money to pay for treatment. Patients with lower COST scores and greater FT were significantly more likely to refuse recommended tests, miss clinic visits, and not take medications as prescribed. 27 Patients who required a feeding tube and supportive infusions were statistically more likely to have a lower COST score and greater FT, suggesting FT is associated with a higher risk for morbidity. 27 Patients with lower COST scores were also significantly more likely to report worry and financial burden. 27 Mady et al 38 found that 30 of 104 (28.8%) survivors of HNC had high FT and that COST was associated with HRQOL; 60% of the survivors reported using cost-coping mechanisms. In these two studies, lower COST scores and worse FT were associated with younger age, lower income, single individuals, lower education levels, and the larynx as the primary site of cancer. 27,38 The FIT was developed and validated by Hueniken et al 40 in 2020 in 430 patients with HNC (of multiple stages and who were treated with multiple treatment modalities) 12 to 24 months after treatment at a single institution in Canada. The FIT incorporates a nine-item survey to measure three subdomains of FT: financial stress, financial strain, and lost productivity. Each item is graded on a scale from 0 to 1, and the total score is calculated from the mean scores of all responses and multiplied by 100; total scores range from 0 to 100, and higher scores represent worse FT. Their study demonstrated that worse FT and higher FIT scores correlated with a lower baseline household income and higher amount of lost wages. 40 This initial study shows promise for the use of the FIT survey in the evaluation of patients with HNC; however, it has not been validated in further published studies.

Discussion
The reviewed evidence indicates that medical expenses are higher for patients with HNC than for patients with other cancers. 2,6,13 The added costs were attributable to major surgical procedures, multimodal treatment (trimodal therapy was the most expensive), treatment-related complications (particularly for chemotherapy and radiation therapy), and longer hospital stays. 2,[14][15][16][17][18][19]21 The indirect costs contributing to FT in patients with HNC reflect the long-term side effects of treatment, including lost employment/productivity as well as impacts on important daily and social functions such as swallowing, eating, speech, and communication. Additional burdens of depression, anxiety, and social isolation related to HNC treatment reduce treatment adherence and HRQOL. Two tools were reported for the quantitative assessment of FT in patients with HNC: the FIT and the COST. The FIT survey was used in one Canadian study of patients with HNC, and the COST survey was validated in several studies in the United States in patients with advanced-stage IV solid tumor cancers and employed in two original research studies of HNC patients. The COST survey demonstrated internal consistency, test-retest reliability, and concurrent validity for 233 patients via three indices, whereas the FIT test demonstrated concurrent validity for 49 patients via one inventory.
The current models of care will be increasingly inadequate to meet the needs of the growing population of cancer survivors. Thus, novel strategies are necessary to better address these needs. Cancer centers, physicians, health insurance providers, and patients all play important roles in mitigating the effects of FT. Healthcare systems should be more transparent about the cost of treatment and provide better access to designated cancer centers. Physicians can engage in patient-centered cancer care and follow professional cancer society guidelines to avoid low-value interventions. Physicians can also make a point to discuss patient goals and lifestyle changes as well as designate a member of the health care team to help with early referrals to supportive care services. Recently, the use of telemedicine rather than in-person clinic appointments has shown utility in reducing costs as well. 42,43 Insurance providers can restructure cost-sharing plans and offer more payment options and financial assistance. Patients should ideally be fully engaged in the formation of their treatment goals and be self-motivated in seeking further education and counseling.
Early interventions can lead to long-term reduction in health care costs and improve patient outcomes. A study by Yarlagadda et al 44 aimed to better prepare patients for lifestyle changes during the treatment of HNC through preadmission social work counseling, which improved patient-reported feelings of support during hospitalization and treatment. Social work counseling that continued throughout treatment and discharge reduced patient anxiety and stress, improved perceived communication, and facilitated earlier access to resources. 44 Patients who received this counseling were better prepared for instructions from nurses, more motivated, and had a more timely discharge. 44 Patients that have better access to information regarding support groups, financial advice, and the long-term effects of treatment on the ability to work, physical functioning, and QOL have shown reduced feelings of isolation, anxiety, and depression, leading to better outcomes. [44][45][46][47][48][49][50] A study evaluating the perspectives of patients with HNC on the cost of care revealed that patients preferred having a conversation, particularly with insurance representatives and hospital-based financial advisors, about the costs before initiating treatment. 49 Multidisciplinary HNC clinics have been shown to help coordinate care by allowing patients to see multiple providers during the same visit, reducing the overall time off from work for follow-up care while also facilitating treatment adherence. 45,51 Multidisciplinary cancer care also emphasizes rehabilitation interventions with the inclusion of physical and occupational therapy, speech therapy, and vocational rehabilitation, helping survivors return to work and thus ease economic burden. [52][53][54] Multidisciplinary cancer treatment teams may also include a specialized patient navigator or financial advisor who can help patients and caregivers cope with financial burdens. 45

Limitations
Although the literature search was comprehensive, this review is subject to low-quality evidence of available literature and publication bias. Interpretations of FT reflect patients' subjective experience and may be dependent on the cancer type, treatment regimen, and variability in provider and patient preferences, which may lead to confounding bias and decreased homogeneity of results. Only two tools have been reported to quantify FT in cancer patients, which need further validation. Finally, because of the heterogeneity among studies, we were unable to conduct a meta-analysis of the data.

Conclusions
Financial toxicity is an important consideration for the wellbeing of patients with HNC. More widespread use of screening and measurement tools, such as the COST and FIT, will raise awareness of the FT endured by these patients. Clinics should have referral resources and strategies available to address these issues to mitigate the economic burden of cancer care, which will improve the QOL for cancer survivors.