Abstract
This cross-sectional survey study examines financial distress and cost expectations among patients with cancer presenting for anticancer therapy.
The financial burden of cancer treatment is a well-established concern. Owing to cost sharing, even insured patients face financial burden and are at risk for worsened quality of life and increased mortality. Underinsured patients (those spending more than 10% of their income on health care costs) are a growing population, and are at risk given the looming heath policy and coverage changes on the horizon. In this setting, little is known about what expectations patients have regarding those costs and how those cost expectations might impact decision making.
Methods
After approval from the institutional review board at Duke University Medical Center, we conducted a cross-sectional survey study of financial distress and cost expectations among patients with cancer presenting for anticancer therapy. We enrolled a convenience sample of adult patients at a comprehensive cancer center and at 3 affiliated rural oncology clinics. Patients provided written informed consent and were compensated with $10 for completing the survey. Trained interviewers surveyed patients in person.
We abstracted the electronic health record for cancer diagnosis, stage, type of treatment, and duration of treatment at the time of enrollment. Demographics including race and income were obtained from the patient. Patient out-of-pocket expenses were based on patient’s best estimation of recent, averaged monthly costs. We surveyed patients about whether their actual costs met their expectations, and about how much they were willing to pay out-of-pocket for cancer treatment, not including insurance premiums. Financial distress was measured using a validated measure. We measured median relative cost of care, defined as monthly out-of-pocket costs divided by income. Expected financial burden, willingness to pay, and subjective financial distress were dichotomized to assess the impact of unexpected costs and high financial distress. We used hypothesis testing to examine variables associated with burden and distress. Multivariable logistic regression included specific variables of interest along with select variables found to be statistically significant in bivariate testing. Statistical analyses were performed using SAS software (version 9.4, SAS institute).
Results
Of 349 consecutive patients approached, 300 were eligible and agreed to participate, and 3 withdrew (86% response rate). Of the 300 patients, 157 (52%) were men. Patient characteristics, income, and costs are described in the Table along with unadjusted analyses. Forty-nine (16%) patients reported high or overwhelming financial distress (score >7).
Table. Patient Characteristics Associated With Unexpected Financial Burden and High or Overwhelming Financial Distress.
| Demographic | No. (%) | P Value | |||||
|---|---|---|---|---|---|---|---|
| (n = 300) | Unexpected Financial Burden (n = 118) | As or Lower Than Expected Financial Burden (n = 170)a | P Value | High or Overwhelming Financial Distress (n = 49) | No/Low/Average Financial Distress (n = 251) | ||
| Age, median (SD), y | 59.6 (11.8) | 57.0 (11.2) | 60.9 (11.9) | .005 | 54.3 (8.8) | 60.6 (12.0) | <.001 |
| Marriedb | 205 (68.3) | 73 (61.9) | 125 (73.5) | .04 | 21 (42.9) | 184 (73.3) | <.001 |
| Race | |||||||
| White | 226 (75.3) | 79 (66.9) | 136 (80.0) | .04 | 31 (63.3) | 195 (77.7) | .08 |
| Black/African American | 58 (19.3) | 30 (25.4) | 28 (16.5) | 15 (30.6) | 43 (17.1) | ||
| Other and/or unknown | 16 (5.3) | 9 (7.6) | 6 (3.5) | 3 (6.1) | 13 (5.2) | ||
| Primary insurance sourcec | |||||||
| Private insurance | 168 (56.0) | 74 (62.7) | 92 (54.1) | .13 | 33 (67.3) | 135 (53.8) | .01 |
| Medicare | 107 (35.7) | 35 (29.7) | 64 (37.6) | 9 (18.4) | 98 (39.0) | ||
| Medicaid | 22 (7.3) | 7 (5.9) | 14 (8.2) | 7 (14.3) | 15 (6.0) | ||
| Employment status | |||||||
| Retired | 110 (36.7) | 24 (20.3) | 78 (45.9) | <.001 | 4 (8.2) | 106 (42.2) | <.001 |
| Employed full time | 80 (26.7) | 29 (24.6) | 51 (30.0) | 8 (16.3) | 72 (28.7) | ||
| Unemployed and not seeking | 79 (26.3) | 48 (40.7) | 29 (17.1) | 28 (57.1) | 51 (20.3) | ||
| Other | 31 (10.3) | 17 (14.4) | 12 (7.1) | 9 (18.4) | 22 (8.8) | ||
| Annual household income range (includes those who gave a single value), $ | |||||||
| <20 000 | 41 (13.7) | 25 (21.2) | 14 (8.2) | .01 | 16 (32.7) | 25 (10.0) | <.001 |
| 20 000-39 999 | 53 (17.7) | 20 (16.9) | 33 (19.4) | 12 (24.5 | 41 (16.3) | ||
| 40 000-59 999 | 47 (15.7) | 21 (17.8) | 24 (14.1) | 6 (12.2) | 41 (16.3) | ||
| ≥60 000 | 137 (45.7) | 46 (39.0) | 84 (49.4) | 12 (24.5) | 125 (49.8) | ||
| Prefer not to say/unknown | 22 (7.3) | 6 (5.1) | 15 (8.8) | 3 (6.1) | 19 (7.6) | ||
| Monthly out-of-pocket costs, median (range), $ | 592 (3-47 250) | 703 (15-47 250) | 553 (3-26 756) | .03 | 728 (6-47 250) | 565 (3-26 756) | .07 |
| Relative cost of care, median (range), %d | 11 (0-7150) | 17 (1-7150) | 10 (0-796) | .003 | 31 (3-7150) | 10 (0-796) | <.001 |
| Primary cancer diagnosis | |||||||
| Colorectal cancer | 81 (27.0) | 41 (34.7) | 37 (21.8) | .004 | 17 (34.7) | 64 (25.5) | .02 |
| Breast cancer | 53 (17.7) | 26 (22.0) | 24 (14.1) | 14 (28.6) | 39 (15.5) | ||
| Lung cancer | 52 (17.3) | 20 (16.9) | 30 (17.6) | 9 (18.4) | 43 (17.1) | ||
| Pancreas or biliary cancer | 39 (13.0) | 8 (6.8) | 30 (17.6) | 2 (4.1) | 37 (14.7) | ||
| Other | 75 (25.0) | 23 (19.5) | 49 (28.8) | 7 (14.3) | 68 (27.1) | ||
| Stage at enrollment | |||||||
| Stage IV | 162 (54.0) | 61 (51.7) | 91 (53.5) | .52 | 18 (36.7) | 144 (57.4) | .02 |
| Metastatic recurrence | 73 (24.3) | 27 (22.9) | 45 (26.5) | 15 (30.6) | 58 (23.1) | ||
| Localized/stage I-III | 65 (21.7) | 30 (25.4) | 34 (20.0) | 16 (32.7) | 49 (19.5) | ||
| Quality of life, 0-10, median (range) | 8 (1-10) | 7 (1-10) | 8 (3-10) | .002 | 6 (1-10) | 8 (3-10) | <.001 |
| Financial distress, 1-10, median (range)e | 3.6 (1-10) | 5.8 (1.2-10) | 2.6 (1-9.1) | <.001 | 8.1 (7.1-10) | 3.3 (1-7.0) | NA |
Abbreviation: NA, not applicable.
Twelve participants were unknown or refused to answer the question regarding expected burden.
Married also includes 6 patients who answered “married-like relationship.”
Three patients had “other” insurance, not shown in the table were included in the testing under “other.”
Relative cost of care calculated as monthly costs divided by monthly income, calculated for patients who provided (1) single, nonzero value for income and (2) cost information (n = 213); this number is reported as a percent.
Financial distress score is reported as an inverted score from 1 to 10. Interpretation as follows: no distress (1-1.9), low distress (2-4), average distress (4.1-7), high distress (7.1-9), overwhelming distress (>9).
The median relative cost of care was 11%. The relative cost of care for patients with high or overwhelming distress was 31% vs 10% for those with no, low, or average financial distress. One hundred eighteen (39%) participants endorsed higher than expected financial burden from cancer care. In unadjusted analysis, unexpected burden was associated with being younger, unmarried, nonwhite, unemployed/not retired, having lower household income, higher costs, colorectal/breast cancer diagnosis, lower quality of life and higher financial distress (Table). In adjusted analysis, experiencing higher than expected financial burden was associated with high or overwhelming financial distress (OR, 4.78; 95% CI, 2.02-11.32; P < .01) and with decreased willingness to pay for cancer care (OR, 0.48; 95% CI, 0.25-0.95; P = .03).
Discussion
More than one-third of insured cancer patients receiving anticancer therapy faced out-of-pocket costs that were greater than expected, and patients with the most distress were underinsured, paying almost one-third of their income in health care-related costs. Patients at risk for unexpected costs had less household income and faced higher out-of-pocket costs.
Facing unexpected treatment costs was associated with lower willingness to pay for care, even when adjusting for financial burden. This suggests that unpreparedness for treatment-related expenses may impact future cost-conscious decision making. Interventions to improve patient health care cost literacy might impact decision making. Indeed, the Institute of Medicine has listed cancer cost-related health literacy as a high priority for future research, and this priority has been included in the Center for Medicare and Medicaid’s Oncology Care Model. Future studies should test interventions for cost mitigation through shared decision making.
References
- 1.Ubel PA, Abernethy AP, Zafar SY. Full disclosure—out-of-pocket costs as side effects. N Engl J Med. 2013;369(16):1484-1486. [DOI] [PubMed] [Google Scholar]
- 2.Zafar SY, Peppercorn JM, Schrag D, et al. The financial toxicity of cancer treatment: a pilot study assessing out-of-pocket expenses and the insured cancer patient’s experience. Oncologist. 2013;18(4):381-390. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Fenn KM, Evans SB, McCorkle R, et al. Impact of financial burden of cancer on survivors’ quality of life. J Oncol Pract. 2014;10(5):332-338. [DOI] [PubMed] [Google Scholar]
- 4.Ramsey SD, Bansal A, Fedorenko CR, et al. Financial insolvency as a risk factor for early mortality among patients with cancer. J Clin Oncol. 2016;34(9):980-986. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.The Commonwealth Fund. Biennial Health Insurance Survey, 2003-2016. http://www.commonwealthfund.org/interactives-and-data/surveys/biennial-health-insurance-surveys/2017/biennial-explorer. Accessed April 19, 2017.
- 6.Institute of Medicine Delivering High-Quality Cancer Care: Charting a New Course for a System in Crisis. Washington, DC: National Academies Press; 2013. [PubMed] [Google Scholar]