Financial distress in patients with advanced cancer

Purpose We examined the frequency and severity of financial distress (FD) and its association with quality of life (QOL) and symptoms among patients with advanced cancer in France. Design In this cross-sectional study, 143 patients with advanced cancer were enrolled. QOL was assessed using the Functional Assessment of Cancer General (FACT-G) and symptoms assessed using Edmonton Assessment System (ESAS) and Hospital Anxiety and Depression Scale (HADS). FD was assessed using a self-rated numeric scale from 0 to 10. Results Seventy-three (51%) patients reported having FD. Patients reported having FD were most likely to be younger (53.8 (16,7SD) versus 62 (10.5SD), p<0.001), single (33 (62%) versus 40(44%), p = 0.03) and had a breast cancer (26 (36%), p = 0.024). Patients with FD had a lower FACT-G score (59 versus 70, p = 0.005). FD decreased physical (14 versus 18, p = 0.008), emotional (14 versus 16, p = 0.008), social wellbeing (17 versus 19, p = 0.04). Patients with FD had higher HADS-D (8 versus 6 p = 0.007) and HADS-A (9 versus 7, p = 0.009) scores. FD was linked to increased ESAS score (59 (18SD) versus 67 (18SD), p = 0.005) and spiritual suffering (22(29SD) versus 13(23SD), p = 0.045). Conclusion The high rate of patient-reported FD was unexpected in our studied population, as the French National Health Insurance covers specific cancer treatments. The FD was associated with a poorer quality of life. Having a systematic assessment, with a simple tool, should lead to future research on interventions that will increase patients’ QOL.

Introduction The French healthcare system covers the total cost of specific treatments for cancer patients, but does not include the cost of expenses such as the daily charge for the hospital room, or certain low-risk drugs that are not considered an essential part of cancer treatment. Private insurance provides reimbursement of any such remaining costs incurred by the patient. Persons who cannot afford private insurance due to economic problems are entitled to free 'top-up' universal state medical insurance. As specific cancer treatment is free of charge, financial difficulty is considered to be rare in France, and is not systematically assessed. However, there may be other consequences of cancer that affect the family income, such as loss of salary, which can lead to increased financial burden. A study in 2007 [1] in the United States of America found a correlation between the increase of financial difficulties and the decrease of quality of life (QOL) in patients with advanced cancer. In 2016 Perrone et al [2] highlighted the association between financial difficulties with clinical outcomes in a country with public health system. Literature [3,4] highlighted the need to take into account the subjective distress and the consequences of the objective financial burden. Financial distress (FD) is a subjective social class identification that can lead to lower self-esteem and a lower sense of control over life. It has been said that patients withdrawing themselves from healthcare treatment for economic reasons is not uncommon in France [5][6][7]. Financial difficulties have recently been found to be the most frequent source of distress by cancer patients in a community cancer center in the USA [8]. Despite these results, one of the less explored effects of cancer is its association on finances, the part of FD in the overall suffering and the consequences of FD on QOL. The goal of this study was to evaluate the association of FD on QOL in patients with advanced cancer in France. The primary objective was to determine the association between FD and QOL. Secondary objectives were to characterize the frequency and intensity of FD in advanced cancer patients in France, to evaluate the association between FD and cancer-related symptoms, to determine the association between FD and subjective financial burden, financial worries and concerns and financial difficulties. We hypothesized that FD is associated with poorer QOL and worse psychological and physical symptoms in patients with advanced cancer.

Method and methods
This cross-sectional survey was approved by the French Institutional Review Board (Centre de Protection des Personnes Lyon Sud-Est; CPP number 12/116, A 13-465; 2012-A01405-38) and registered as a clinical trial NCT01868100. No financial compensation was provided to participate in the study. Written informed consent has been obtained from each patient.

Definition and assessment
According to literature there are many terms used for describing patients feeling concerning financial issues [9][10][11][12]. In this study we used the term FD which defined a subjective experience of distress attributed by the patient to financial burden. Objective financial burden is defined as the percentage of total family income used for direct out-of-pocket expenditure on health care during the last month, which has not been covered by private insurance [13,14] Subjective financial burden is defined as the impact of FD on patients' wellbeing.

Population
From the 19 th of March 2013 to 30rd of June 2014, 143 patients were enrolled in the Teaching Hospital (TH) Centre Hospitalo-Universitaire de Lyon Sud (CHLS) and a Comprehensive Cancer Centre, Institut de Cancérologie de la Loire (ICL) in Rhône-Alpes. Outpatients or hospitalized patients above 18 years old with advanced lung or breast or colorectal or prostate cancer were included. Advanced cancer was determined by the presence of a relapse, or a metastasis or a second-line-treatment chemotherapy, or a local advanced disease. Non-inclusion criteria were applied to patients presenting with cognitive disorders diagnosed by medical staff. Informed consent was obtained for the patients before the inclusion.

Data collection and survey instruments
Patient demographics including cancer diagnosis and staging and ECOG (Eastern Cooperative Oncology Group) [15] were collected on chart report. We used the following validated instruments. ESAS (Edmonton Symptom Assessment System) measures the intensity of the most common symptoms within the last 24 hours of cancer patients: pain, fatigue, nausea, depression, anxiety, drowsiness, shortness of breath, appetite, sleep, and wellbeing [16][17][18]. For this study we added an item assessing spiritual pain. The HADS (Hospital Anxiety and Depression Scale) is a self-assessment tool with 14 items, with two separate subscales for anxiety and depression. A score of 16 or higher is suggestive of a severe case of anxiety or depression [19][20][21]. FACT-G (Functional Assessment of Cancer Therapy-General) is a 27-question tool validated in cancer patients and has four subscales to assess wellbeing. From these subscales a QOL score is obtained. Both the total score and the individual scores have good internal reliability (alpha = 0.72-0,85) [22,23]. No specific scales exist to assess Financial Burden and FD in France. In this study, financial burden was assessed by questions used in previous studies concerning patients with cancer [24,25]. High burden is defined as more than 10% of total family income spent on out-of-pocket expenses [26,27]. The patients were asked to rate from a "strongly agree to strongly disagree" questions concerning their physical, emotional, social and familial distress. The statement was considered positive when they answered "strongly agree, or agree or somewhat agree". Subjective financial burden and FD, financial concerns and worries are measured by an exploratory questionnaire. Patients were asked to score a numeric rating scale from 0 to 10 with 0 being best and 10 being worst. Patients were considered having FD if they quoted 1 or above. High FD was defined quoting 4 and above.

Statistical analysis
For the descriptive analysis, categorical variables were expressed with average and standard deviation (SD), continuous variables with proportion and percentage. The Student T test was used to compare the categorical variables and the chi-square test for the continuous variables. When Student T test or chi-square test could not be used the Mann-Whitney test or Fisher test was applied. One hundred and twenty-five patients were necessary to allow us to detect a Pearson correlation between FD and quality of life as small as 0.25 with 80% power at a 5% type I error rate. A 10% dropout rate was expected so we recruited 143 patients. A cancer-related type stratification was used. We estimated a Pearson and Spearman correlation between FD and quality of life. We compared FD with socioeconomic status using ANOVA or Wilcoxon rank sum tests as appropriate. P values less than 0.05 were considered statistically significant. We used a Bonferroni correction for all statistical analyses. For the statistical analysis SPPS software (Chicago, Inc) 19.0 was used.

Results
Three hundred and twenty-five patients were screened; of these 124 were not included due to their type of cancer. Twenty-one patients refused to participate. Thirty-seven could not be enrolled due to non-inclusion criteria.
Baseline characteristics of the patients and patients' FD (Table 1) The TH recruited 94 patients (66%) and the CCC 49 patients (34%). Of these, 67 patients (47%) were considered ECOG 2. Seventy (49%) patients were female of whom 42 (29%) reported having FD (p = 0.03). The average age was 58 years old. Patients who reported having FD had a lower average age (53.8 (16,7SD)) than patients with no FD (62 (10.5SD), p<0.001). Fifty-three (37) patients were single. Single patients had more FD (33 (62%)) than those who were married or living with a partner (28 (44%)) (p = 0.003). Eighty-five (59%) of the population did not have Subjective financial burden is shown in Table 2 The average financial concerns and difficulties rates were 3 (SD 3). Patients having FD had more financial difficulty Association of financial distress on patients' quality of life and symptoms are shown in Table 3 The average score on the FACT-G was 63 (SD 18). Patients with FD have a lower FACT-G score   Association between patient-reported financial distress and other subjective terms Table 4 shows the association between the financial distress reported by patients and other subjective terms such as subjective financial burden (r = 0.843, p<0.001), financial concerns (r = 0.791, p<0.001), financial worries (r = 0.816, p<0.001) and financial difficulties (r = 0.88, p<0.001).  . This study showed that loss of employment was linked to: the younger and older, active population, those with a low level of education, and those engaged in manual work, or work of a precarious nature, and in small businesses. Prognosis influenced work activity. Patients with good prognoses had maintained job activity at 89% (office work) and 74% (physical work), whereas this was 48% and 28% respectively for patients with poor prognoses [29]. We also acknowledge that cancer type influences our results. Except for lung cancer, other types of cancers chosen in our study do not have the poorer prognoses and previous studies showed that poorer prognosis is associated with a higher Financial Difficulty [32][33][34][35]. Further study should involve other types of cancers that are more closely linked to difficult socioeconomic status and poorer prognoses. In 2012, the poverty rate in the Rhône-Alpes region was 12.1% (the national rate is 14%) [36,37]. The unemployment rate is above the national one (9.3%). Despite the fact that Rhône-Alpes is one of the richest regions in France our study shows that patients with advanced cancer are prone to have FD. In order to be more representative of the French population, a major study involving all regions of France should be conducted. In recent literature [28], 130 patients out of 147 in the USA had FD. Of 91% patients with  [38]. Both the American results and ours highlight that FD is not completely explained by the healthcare system but the French healthcare system seems to offer moderate protection against FD. That may be due to the 100% coverage by French healthcare. This study has some limitations linked to the design of this cross-sectional study. The link between the financial distress and the others factors is not a cause effect relationship. The strong association we observed between FD and the subjective report needs to be confirmed in other studies with larger sample sizes. As only two centers were involved we cannot generalize our results and thus more, largerscale studies, based on a mix of public and private hospitals, are needed. Furthermore our study used non validated items due to a lack of existing ones so future studies should focus on the validation of items assessing financial distress and financial difficulty. Even if our FD assessment was exploratory, it is an extremely simple way of expressing FD that should trigger a social worker visit. Validation of tools should be made in order to assess FD among patients with cancer and identify interventions to help them. A new tool, Comprehensive Score for Financial Toxicity [39], may be a new lead but needs further investigation. Despite the fact that cancer treatment costs are largely covered by the French healthcare system, this study shows that FD is frequent and is associated with poorer QOL, and more intense psychological, physical and spiritual symptoms. Thus, FD and Financial Difficulty assessment should be systematic.

This study shows a correlation between FD and QOL and between FD and psychological symptoms (HADS-A and HADS-D). It shows correlation between FD and spiritual suffering and between FD with the ESAS in patients
Supporting information S1 Appendix. This questiondoinaire was used to interview the patients. (DOCX)

Author Contributions
Conceptualization: WR MF EB MD MR CB.