Abstract
Background:
Urothelial carcinoma (UC) is generally diagnosed early and may incur significant lifetime costs. This study estimated, from the payer’s perspective, the lifetime costs among patients diagnosed with UC according to stage at diagnosis.
Methods:
This retrospective analysis of the linked Surveillance, Epidemiology, and End Results (SEER)-Medicare database identified patients ≥66 years with newly diagnosed UC from 2004–2013. Patients were followed from UC diagnosis to death or last follow-up to estimate lifetime costs. Costs were allocated to 3 phases: diagnosis (≤3 months after diagnosis), terminal (≤3 months before death), and continuation (months between diagnosis and terminal phases). Survival-adjusted lifetime costs (total and major UC-related) were estimated for patients with UC based on stage at diagnosis (stages 0 through IV) and in a subgroup of patients receiving ≥1 systemic line of chemotherapy (LOC).
Results:
The sample included 15,588 patients: 3,446 stage 0 (8% ≥1 LOC; median [IQR] follow-up in months: 44 [23–71]); 3,902 stage I (12% ≥1 LOC; 33 [15–62]); 4,301 stage II (26% ≥1 LOC; 17 [7–39]); 1,612 stage III (25% ≥1 LOC; 17 [7–42]); and 2,327 stage IV (33% ≥1 LOC; 8 [3–18]). Median age was 78 years and 72% were male. Mean lifetime costs were lowest for stage IV patients (stage 0, $151,626; stage 1, $150,123; stage II, $149,728; stage III, $190,996; stage IV, $117,503). Hospitalizations not involving a cystectomy contributed about half of lifetime costs across all stages. Cystectomy contributed 2–13% of the total lifetime UC costs ($3,356 stage 0; $7,011 stage I; $11,855 stage II; $25,509 stage III; $11,693 stage IV). UC-related office visits contributed 8–15% of lifetime costs ($11,717 stage 0; $14,611 stage I; $19,882 stage II; $21,480 stage III; $17,820 stage IV).
Conclusion:
UC continues to be a costly cancer with stage III patients having highest lifetime costs. Hospitalizations drive most of the lifetime costs across all stages; most of these hospitalizations did not involve costs related to cystectomy. Treatment plans requiring shorter and fewer hospitalizations may lessen the economic burden of UC.
Keywords: Immunotherapy, SEER-Medicare, administrative claims, bladder cancer
Introduction
Urothelial carcinoma (UC) accounts for 90% of bladder cancer tumors in the US.1,2 Approximately 75% of patients present with non-muscle invasive bladder cancer (NMIBC) at diagnosis,3,4 of whom 30–70% will have same stage recurrence5 and 10–15 % will go on to develop invasive bladder cancer.4,6 Inclusive of all stages, the overall 5-year survival for bladder cancer is 77%, but is as low as 5% in patients with distant disease and is up to 95% for patients with in situ alone disease.7 Prognosis, management, and therapeutic aims are dependent on disease invasiveness at each stage. For NMIBC, treatment is directed at reducing recurrences and preventing disease progression; for muscle invasive bladder cancer (MIBC), definitive local therapy, without or with neoadjuvant or/and adjuvant chemotherapy are used for curative intent; for locally advanced (1M0) and metastatic disease (M1), the goal is to prolong and maintain quality of life, which can include use of systemic therapy (both chemotherapy and immune checkpoint inhibitors [ICIs]), surgery, and/or radiotherapy.8
Continuous treatment and surveillance makes bladder cancer one of the most expensive cancer types over the lifetime of a patient,9,10,11 and in fact have been shown to be the costliest cancer when compared to other malignancies.12–14 Previous studies have reported long term cost estimates for bladder cancer ranging from $102,700 to $125,541 in the US.11,13,15 In urothelial carcinoma (UC), similar to other cancers, the trajectory of cost accumulation is not linear but rather U-shaped, exhibiting high costs incurred around diagnosis and death, and relatively lower costs in-between.5,16 Lifetime costs of UC are expectedly stage-dependent since patients with early stage UC typically have longer survival and consequently higher costs during their disease journey from diagnosis, treatment, recovery/survivorship, and end of life. There could also be heterogeneity in lifetime costs within each stage, depending on the type and extent of care received. For example, in early stage bladder cancer, patients undergoing more treatment may incur three times as much cost compared to those in the lowest quartile, but the extent of care may not necessarily correlate with improved patient survival.17,18
Studies to date on the overall economic burden of bladder cancer,5,6,10,12,19,20 or more broadly UC, have reported costs per-patient-per month or cumulative costs over the course of patient follow-up rather than distinct phases of disease management, or have estimated costs for specific stages rather than evaluated across all stages of UC. Additionally, most of them covered time periods prior to 2011. With the advent of immunotherapy agents in the treatment landscape of advanced UC and ongoing efforts to define their potential role in earlier stages, it will become increasingly relevant to understand the relative cost-effectiveness of immunotherapy-based approaches compared to chemotherapy. To provide a framework for future comparative studies, as more real-world data on immuno-oncology drugs become available over time, the present research provides updated lifetime direct medical costs of managing and treating patients with UC.
Methods
Data Source
This study was a retrospective database analysis using data from the linked Surveillance, Epidemiology, and End Results (SEER)-Medicare database. The SEER registry covers 28% of the US population and consists of 18 state and regional cancer registries,21,22 while the Medicare database contains administrative health care service claims and associated costs for all covered Medicare beneficiaries, from Medicare eligibility to death.22,23
Study Population
The study included Medicare beneficiaries 66 years or older who were newly diagnosed with histologically confirmed urothelial carcinoma from 2004 to 2013 (International Classification of Diseases for Oncology, Third Edition [ICD-O-3] codes 8120/3 – 8124/3) and who had available Medicare claims data from 2003 to 2014. Patients were followed from UC diagnosis until death, loss to follow-up, end of the study period (December 31, 2014), Health Maintenance Organization (HMO) enrollment, or Medicare disenrollment. Patients were excluded if they received a postmortem UC diagnosis, had an unknown diagnosis date, or had another cancer (except non-melanoma skin cancer) in the 5 years before UC diagnosis. Additional exclusion criteria included non-continuous enrollment in Medicare Parts A and B or enrollment in an HMO in the 12 months prior to diagnosis.
Our primary analysis focused on characterizing the lifetime economic burden in all UC patients who met the inclusion criteria by stage at diagnosis. For each stage, these patients will be referred to below as “average cohort patients” to distinguish them from the subgroup of patients who received a systemic line of chemotherapy (LOC) for UC within each stage, which will be referred to as “LOC patients.” This is an important subgroup since these patients are more likely to potentially benefit from the innovative therapies currently being examined in clinical trials that could replace or be used in conjunction with chemotherapy as the standard of care in the future. Of note, by our definition, if patients received neoadjuvant or adjuvant chemotherapy but not any subsequent systemic chemotherapy during their disease course were not considered as part of the “LOC patients” subgroup. Understanding the burden in LOC patients can help with future comparative cost analyses.
Baseline Characteristics
The SEER registries provided the baseline demographic information for the included UC patients. Each patient’s stage at diagnosis was also obtained from the SEER registries based on the tumor, node, metastasis (TNM) stage at time of UC diagnosis (Supplemental Table 2). Comorbidities that occurred up to 12 months prior to UC diagnosis were obtained from Medicare claims and used to calculate the Deyo adaptation of the Charlson Comorbidity Index (CCI).24 Using a previously validated claims-based prediction model, Medicare claims, specifically on wheelchair use, oxygen use, walking aid, hospital bed, hospice, skilled nursing facility, or hospitalization occurring 12 months prior to UC diagnosis, were used to calculate a proxy indicator (yes/no) variable for poor performance status.25
Lines of Chemotherapy
Receipt of chemotherapy was identified by Healthcare Common Procedure Coding System (HCPCS) and National Drug Codes (NDC) and included the following agents, either as monotherapy or in combination: carboplatin, cisplatin, docetaxel, doxorubicin, gemcitabine, ifosfamide, methotrexate, oxaliplatin, paclitaxel, pemetrexed, vincristine, and vinblastine. The LOC recipients may or may not have received prior neo-adjuvant or adjuvant chemotherapy (if they had undergone prior cystectomy). Among such patients, the identification of LOC was differentiated from neoadjuvant/adjuvant chemotherapy based on a 6-month gap post-cystectomy. Thus, if systemic chemotherapy was administered after 6 months of cystectomy, it was considered an LOC. Six months was chosen because adjuvant chemotherapy is generally given 4–6 weeks after cystectomy for up to 4 cycles that last over 3–4 months.
Outcomes
Primary outcomes were lifetime and phase-specific direct medical costs by stage at diagnosis for the average patients and the subgroup of LOC patients. Calculated all-cause costs represented actual Medicare payments rather than billed charges or “list price” established by the provider and thus the perspective of this analysis is the payer’s (Medicare) perspective.26,27 Costs were assessed in 2 ways: 1) by clinical setting of care as aligned with how Medicare aggregates its claims files, i.e., inpatient (MEDPAR), outpatient (Medicare Part B), physician and supplier (NCH), durable medical equipment (DME), home health agency (HHA), hospice, and prescription drug (Medicare Part D) and 2), by healthcare resource utilization categories. Healthcare resource utilization categories were identified using codes from the International Classification of Diseases 9th Edition Clinical Modifications Procedure Coding System (ICD-9-PCS), HCPCS, NDC, and Hospital Revenue Centers (Supplemental Table 1) and included hospitalization not involving a cystectomy, cystectomy, UC-related office visit, radiation, and systemic chemotherapy. Costs that did not fall into any of the above categories were lumped into “other” categorization. Costs were inflated to 2017 and presented in $US.
Costs over time were allocated into three clinically relevant disease phases: diagnosis phase (i.e. from diagnosis up to 3 months post diagnosis), terminal phase (covering the period 3 months before death), and continuation phase (i.e. the period between diagnosis and last follow-up date among patients who did not die or start of the terminal phase among patients who died) (Figure 1). Since this study evaluated lifetime costs across all stages of UC, it was important to select an interval period covering diagnosis and terminal phases (i.e., 3 months) that was not too long relative to the overall survival for patients with stage IV UC (~12 months).
Figure 1.
Phases for Cost Allocation
Statistical Analysis
Descriptive statistics were used to summarize patient clinical characteristics and LOCs received. Mean, standard deviation (SD), median, and interquartile range (IQR) were reported for all continuous variables. Frequency and percentage were calculated for all categorical variables. The diagnosis and terminal phase-specific costs were reported as mean (and 95% confidence intervals) costs over the respective periods, while the continuation phase costs were reported as mean costs per month. Cumulative lifetime costs were survival-adjusted using the time to death event from Kaplan-Meier curves. This approach allows for the estimation of lifetime costs in the presence of heavy censoring and has been shown to be the most robust method used in this situation.28
Results
Patient Characteristics
A total of 15,588 patients met the inclusion criteria (Figure 2). At diagnosis, 22% of patients were diagnosed as stage 0, 25% were stage I, 28% were stage II, 10% were stage III, and 15% were stage IV. Median age at diagnosis was 78 years, 72% of patients were male, 87% were non-Hispanic white, and 56% were married (Table 1). Over half of the patients had CCI ≥1, and about a quarter had a proxy for poor performance status. Among stage IV patients, 53% were M1, 53% were N1 – N3, and 15% were T4/T4b at diagnosis (not mutually exclusive). As expected, median follow-up was lower for advanced stages of disease (44 months for stage 0, 33 months for stage I, 17 months for stage II, 17 months for stage III, 8 months for stage IV). The prevalence of LOC receipt increased with more advanced cancer (8% for stage 0, 12% for stage I, 26% for stage II, 25% for stage III, and 33% for stage IV patients). Very few patients received neoadjuvant or adjuvant chemotherapy only, without further receipt of line of systemic chemotherapy (1% stage 0, 6% stage II, 12% stage III, and 12% stage IV) (Table 1). Of note, more patients of lower stages were non-Hispanic White and married, which may highlight social support, access to care, and cultural beliefs as important determinants of surveillance and early diagnosis of UC.
Figure 2.
Patient Attrition
Table 1.
Patient demographic, clinical characteristics, and treatment patterns by cancer stage at UC diagnosis
| Characteristic | All (N= 15,588) |
Stage 0 (N= 3,446) |
Stage I (N= 3,902) |
Stage II (N= 4,301) |
Stage III (N= 1,612) |
Stage IV (N= 2,327) |
P value |
|---|---|---|---|---|---|---|---|
| Age at diagnosis, years, Median (IQR) | 78 (72 – 84) | 77 (72 – 83) | 78 (72 – 84) | 80 (73 – 85) | 77 (72 – 82) | 77 (71 – 82) | <0.001 |
| Male, N (%) | 11,155 (71.6) | 2,705 (78.5) | 2,898 (74.3) | 2,954 (68.7) | 1,115 (69.2) | 1,483 (63.7) | <0.001 |
| Race/Ethnicity, N (%) | <0.001 | ||||||
| Non-Hispanic White | 13,578 (87.1) | 3,060 (88.8) | 3,410 (87.4) | 3,768 (87.6) | 1,390 (86.2) | 1,950 (83.8) | |
| Non-Hispanic Black | 862 (5.5) | 142 (4.1) | 212 (5.4) | 240 (5.6) | 94 (5.8) | 174 (7.5) | |
| Hispanic | 603 (3.9) | 129 (3.7) | 147 (3.8) | 151 (3.5) | 60 (3.7) | 116 (5.0) | |
| Other | 545 (3.5) | 115 (3.3) | 133 (3.4) | 142 (3.3) | 68 (4.2) | 87 (3.7) | |
| Marital status at diagnosis, N (%) | <0.001 | ||||||
| Single (never married) | 1,102 (7.1) | 234 (6.8) | 254 (6.5) | 297 (6.9) | 113 (7.0) | 204 (8.8) | |
| Married | 8,771 (56.3) | 2,108 (61.2) | 2,239 (57.4) | 2,295 (53.4) | 891 (55.3) | 1,238 (53.2) | |
| Separated/Divorced/Widowed | 4,841 (31.1) | 795 (23.1) | 1,157 (29.7) | 1,519 (35.3) | 563 (34.9) | 807 (34.7) | |
| Unknown | 874 (5.6) | 309 (9.0) | 252 (6.5) | 190 (4.4) | 45 (2.8) | 78 (3.4) | |
| Census Location, N (%) | <0.001 | ||||||
| West | 6,130 (39.3) | 1,338 (38.8) | 1,412 (36.2) | 1,715 (39.9) | 661 (41.0) | 1,004 (43.1) | |
| South | 3,740 (24.0) | 800 (23.2) | 924 (23.7) | 1,088 (25.3) | 408 (25.3) | 520 (22.3) | |
| Northeast | 3,915 (25.1) | 929 (27.0) | 1,129 (28.9) | 986 (22.9) | 353 (21.9) | 518 (22.3) | |
| Midwest | 1,803 (11.6) | 379 (11.0) | 437 (11.2) | 512 (11.9) | 190 (11.8) | 285 (12.2) | |
| Charlson Comorbidity Index, N (%) | 0.002 | ||||||
| 0 | 7,078 (45.4) | 1,619 (47.0) | 1,759 (45.1) | 1,854 (43.1) | 749 (46.5) | 1,097 (47.1) | |
| 1 | 3,989 (25.6) | 895 (26.0) | 1,011 (25.9) | 1,112 (25.9) | 425 (26.4) | 546 (23.5) | |
| 2 | 2,098 (13.5) | 457 (13.3) | 526 (13.5) | 600 (14.0) | 204 (12.7) | 311 (13.4) | |
| 3+ | 2,423 (15.5) | 475 (13.8) | 606 (15.5) | 735 (17.1) | 234 (14.5) | 373 (16.0) | |
| Poor performance status, N (%) | 3,861 (24.8) | 777 (22.5) | 980 (25.1) | 1,129 (26.2) | 381 (23.6) | 594 (25.5) | 0.003 |
| M1, N (%) | 1,224 (7.9) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 1,224 (52.6) | - |
| N1 – N3, N (%) | 1,258 (8.1) | 0 (0.0) | 0 (0.0) | NR | NR | 1,233 (53.0) | - |
| Follow-up, months, Median (IQR) | 23.2 (8.9 – 52.1) | 44.1 (23.1 – 71.1) | 33.3 (15.0 – 62.1) | 17 (6.9 – 38.9) | 17.1 (7.4 – 41.7) | 7.9 (3.3 – 17.7) | <0.001 |
| Treatment Patterns During Follow-up | <0.001 | ||||||
| (Neo-) Adjuvant Chemotherapy Only, N (%) | 850 (5.5) | 34 (1.0) | 80 (2.1) | 259 (6.0) | 191 (11.8) | 286 (12.3) | |
| No Systemic Chemotherapy, N (%) | 11,693 (75.0) | 3,133 (90.9) | 3,352 (85.9) | 2,913 (67.7) | 1,011 (62.7) | 1,284 (55.2) | |
| Received Systemic Chemotherapy, N (%) | 3,045 (19.5) | 279 (8.1) | 470 (12.0) | 1,129 (26.2) | 410 (25.4) | 757 (32.5) | |
| At least two LOC, N (%) | 959 (6.2) | 81 (2.4) | 155 (4.0) | 306 (7.1) | 125 (7.8) | 292 (12.5) | |
| At least three LOC, N (%) | 290 (1.9) | 24 (0.7) | 34 (0.9) | 88 (2.0) | 44 (2.7) | 100 (4.3) |
Abbreviation: IQR, interquartile range; LOC, line of systemic chemotherapy; ; NR – not reported as analyses of SEER-Medicare stipulate Ns<11 may not be directly reported.
Costs
Lifetime and phase-specific Costs by Stage
The average patient (as defined under the Methods Section, and includes patients irrespective of whether or not they received chemotherapy) diagnosed with earlier stages of UC (stage 0–II) had similar lifetime costs ($151,626 stage 0; $150,123 stage I; $149,728 stage II), while stage III patients had the highest ($190,996) and stage IV had the lowest lifetime cost burden at $117,503 (Table 2). The lifetime costs for the subgroup of LOC patients (as defined under the Methods section) were 6–24% higher than the average patient of the same stage. However, stage III LOC patients incurred lower lifetime costs ($166,851) than the average patient within the stage III group ($190,996). Stage IV LOC patients continued to have comparatively the lowest lifetime cost burden ($138,274) amongst the LOC patients across the four stages, whereas stage 0 LOC patients had the highest cost burden ($188,140) (Table 2).
Table 2.
Average lifetime and phase-specific costs per patient, by cancer stage and receipt of line of systemic chemotherapy (LOC).
| Phase specific costs (95% CI) | |||||
|---|---|---|---|---|---|
|
| |||||
| Diagnosis (3 months) | Continuation (monthly) | Terminal phase (3 months) | Lifetime Costs (95% CI) | ||
| Overall cohort | Stage 0 | $6,979 ($6,603 to $7,355) |
$984 ($914 to $1,054) |
$17,991 ($17,041 to $18,941) |
$151,626 (141,286 to 161,966) |
|
| |||||
| Stage I | $9,510 ($9,096 to $9,924) |
$939 ($895 to $983) |
$19,489 ($18,635 to $20,343) |
$150,123 (143,171 to 157,075) |
|
|
| |||||
| Stage II | $16,784 ($16,257 to $17,311) |
$862 ($825 to $899) |
$21,764 ($21,008 to $22,520) |
$149,728 (143,760 to 155,696) |
|
|
| |||||
| Stage III | $23,404 ($22,313 to $24,495) |
$1,110 ($946 to $1,274) |
$24,818 ($23,478 to $26,158) |
$190,996 (167,494 to 214,498) |
|
|
| |||||
| Stage IV | $19,290 ($18,441 to $20,139) |
$554 ($513 to $595) |
$26,571 ($25,500 to $27,642) |
$117,503 (110,326 to 124,680) |
|
|
| |||||
| Systemic Line of Chemotherapy Recipients | Stage 0 | $6,712 ($5,878 to $7,546) |
$1,208 ($1,047 to $1,369) |
$26,097 ($22,677 to $29,517) |
$188,140 (163,216 to 213,064) |
|
| |||||
| Stage I | $9,680 ($8,855 to $10,505) |
$1,067 ($992 to $1,142) |
$22,695 ($20,821 to $24,569) |
$169,561 (157,183 to 181,939) |
|
|
| |||||
| Stage II | $16,966 ($16,157 to $17,775) |
$911 ($863 to $959) |
$24,872 ($23,397 to $26,347) |
$159,219 (150,824 to 167,614) |
|
|
| |||||
| Stage III | $22,972 ($21,231 to $24,713) |
$930 ($854 to $1,006) |
$24,011 ($21,911 to $26,111) |
$166,851 (153,136 to 180,566) |
|
|
| |||||
| Stage IV | $22,346 ($20,925 to $23,767) |
$690 ($610 to $770) |
$26,942 ($25,178 to $28,706) |
$138,274 (124,790 to 151,758) |
|
Abbreviation: LOC, line of systemic chemotherapy
Stage III patients had the highest diagnosis phase ($23,404) and continuation phase ($1,110 per month) costs, and the second highest terminal phase ($24,818) costs. Stage IV patients had the second highest diagnosis ($19,290) and the highest terminal phase costs ($26,571), but the lowest continuation phase costs ($554 per month). In the subgroup of LOC patients, stage 0 patients had the highest continuation phase ($1,208 per month) costs, while stage IV patients continued to have the lowest continuation phase ($690 per month) costs (Table 2). Patients diagnosed with earlier stages of UC spent more time in the continuation phase compared to those diagnosed with advanced stages (45.2 months for stage 0, 36.6 months for stage I, 23.3 months for stage II, 24.4 months for stage III, 10.7 months for stage IV) (Supplemental Table 3).
Lifetime Costs by the Setting of Care Provided
Regardless of the stage at diagnosis, in the average patients, hospitalizations were the primary cost drivers of lifetime costs, comprising more than 50–65% of all costs, followed by outpatient costs and physician Part B costs (Figure 3). Hospice accounted for only 2–5% of lifetime costs; increased disease stage did not correlate with a gradient of higher lifetime cost attribution for hospice (Figure 3).
Figure 3.
Lifetime costs by stage of urothelial carcinoma at diagnosis, stratified by setting of care
Lifetime Costs by Healthcare Resource Utilization Categories
Hospitalizations and Office visits
Hospitalizations that did not involve a cystectomy accounted for 49–53% of lifetime costs in the average patients. The cost burden as a percentage of total lifetime costs was highest for stage III average patients ($100,356; 53%). In the subgroup of LOC patients, hospitalizations that did not involve a cystectomy contributed to 42–47% of lifetime costs, with stage IV LOC patients having the highest burden ($64,642; 47%).
UC-related office visits accounted for 8–15% of lifetime costs in the average patients. The cost burden as a percentage of total lifetime costs was highest for stage IV average patients ($17,820; 15%). In the subgroup of LOC patients, UC-related office visits contributed to 13–22% of lifetime costs, with stage II LOC patients having the highest burden ($34,386; 22%).
Cystectomy, Chemotherapy, and Radiation
Cystectomy accounted for 2–13% of lifetime costs in the average patients. The cost burden as a percentage of total lifetime costs was highest for stage III average patients ($25,509; 13%). In the subgroup of LOC patients, cystectomy contributed to 2–5% of lifetime costs, with stage III LOC patients having the highest burden ($8,959; 5%).
The cost burden of systemic chemotherapy was minimal, contributing only 1–4% in the average patients. Stage IV patients accumulated the highest cost of systemic chemotherapy ($4,831 in the average patients and $10,238 in LOC patients).
Radiation costs accounted for 1–5% of lifetime costs in the average patients. The cost burden as a percentage of total lifetime costs was highest for stage II average patients ($7,932; 5%). In the subgroup of LOC patients, radiation contributed to 3–9% of lifetime costs, with stage II LOC patients having the highest burden ($14,527; 9%) (Table 3).
Table 3.
Health Care Resource Use Specific Lifetime Costs by Stage at Diagnosis
| Hospitalization (not involving a cystectomy) | Cystectomy | UC- related office visit | Radiation | Systemic Chemotherapy | Other | ||
|---|---|---|---|---|---|---|---|
| Stage 0 | All | $73,903 (64,692 to 83,112) |
$3,356 (2,507 to 4,205) |
$11,717 (10,983 to 12,451) |
$1,603 (1,267 to 1,939) |
$837 (580 to 1,095) |
$60,210 (55,667 to 64,753) |
| LoC | $84,573 (62,686 to 106,484) |
$2,893 (1,261 to 4,525) |
$23,800 (19,496 to 28,105) |
$5,664 (3,890 to 7,437) |
$7,096 (4,921 to 9,272) |
$64,114 (53,479 to 74,749) |
|
| Stage I | All | $73,249 (68,641 to 77,853) |
$7,011 (5,041 to 8,982) |
$14,611 (13,631 to 15,591) |
$2,988 (2,224 to 3,752) |
$1,392 (1,092 to 1,692) |
$50,872 (46,226 to 55,518) |
| LoC | $71,659 (64,158 to 79,208) |
$4,729 (1,485 to 7,972) |
$28,390 (25,142 to 31,638) |
$7,711 (6,014 to 9,408) |
$7,383 (5,736 to 9,030) |
$49,689 (41,305 to 58,073) |
|
| Stage II | All | $72,709 (69,620 to 75,802) |
$11,855 (9,321 to 14,391) |
$19,882 (18,627 to 21,137) |
$7,932 (6,970 to 8,894) |
$2,443 (2,110 to 2,776) |
$34,907 (30,779 to 39,035) |
| LoC | $69,864 (63,488 to 76,278) |
$2,986 (1,945 to 4,026) |
$34,386 (31,366 to 37,402) |
$14,527 (12,271 to 16,783) |
$6,579 (5,627 to 7,530) |
$30,877 (27,186 to 34,568) |
|
| Stage III | All | $100,356 (83,362 to 117,385) |
$25,509 (19,171 to 31,847) |
$21,480 (19,513 to 23,449) |
$4,566 (3,559 to 5,573) |
$3,653 (3,026 to 4,279) |
$35,432 (20,664 to 50,200) |
| LoC | $74,557 (65,025 to 84,092) |
$8,959 (5,964 to 11,951) |
$25,924 (22,284 to 29,564) |
$7,917 (5,407 to 10,426) |
$7,973 (6,446 to 9,501) |
$41,521 (33,373 to 49,669) |
|
| Stage IV | All | $59,494 (54,978 to 64,023) |
$11,693 (9,274 to 14,113) |
$17,820 (16,568 to 19,075) |
$3,118 (2,571 to 3,665) |
$4,831 (4,315 to 5,347) |
$20,547 (15,738 to 25,356) |
| LoC | $64,642 (53,281 to 76,018) |
$4,691 (3,045 to 6,338) |
$27,781 (25,096 to 30,474) |
$4,321 (3,414 to 5,228) |
$10,238 (9,052 to 11,424) |
$26,601 (20,229 to 32,973) |
|
Abbreviation: UC, urothelial carcinoma; LOC – at least one line of chemotherapy
Other costs
Costs that are unlikely related to UC accounted for 17–40% of lifetime costs in the average patients, with earlier stages having higher costs in that category. The cost burden as a percentage of total lifetime costs was highest for stage 0 average patients ($60,210, 40%) and lowest for stage IV ($20,547, 17%). In the LOC subgroups, costs that are unlikely related to UC accounted for 19–34% of lifetime costs.
Discussion
Communicating the cost of cancer care to patients is becoming increasingly incorporated by clinical oncology organizations as one of the core principles to improve shared decision-making, patient satisfaction, and optimize out-of-pocket spending and outcomes.29,30 Our study provides a contemporary detailed update of the lifetime economic burden of UC by adopting a phase-based, stage-dependent approach. The study results provide relevant benchmarks and context to further understand and communicate the economic burden of UC as the treatment landscape for UC continues to evolve.18–21 Additionally, this study generalizes to both the average UC patient by stage as well as the subgroup of patients who receive a systemic line of chemotherapy and therefore assumed to have progressed to advanced disease where systemic chemotherapy is recommended.
Our study found that the trajectory of cost accumulation does not necessarily increase linearly with stage of disease but rather depends on the extent of care received during the diagnosis, continuation, and terminal phases of UC. A recent study by Sloan et al15 also using the linked SEER–Medicare dataset (1998–2013) demonstrated that the cumulative costs for bladder cancer patients over a 16-year period were higher for patients with lower stages at first diagnosis ($124,320 for localized; $104,161 for regional; and $64,922 for distant disease), which was mostly due to the longer (16-year) survival of patients presenting with lower stage disease (13.3% for localized; 4.4% for regional; and 0.0% for distant disease), possibly reflecting long-term routine surveillance, including imaging, scans, lab tests, and/or other diagnostic procedures in inpatient, outpatient, and office settings.31 Interestingly, and somewhat similar to the estimates by Sloan et al for localized bladder cancer, stage 0 patients in our study treated with LOC generated the greatest lifetime burden ($188,140) as well as continuation costs ($1,208 per month), compared to LOC recipients of higher disease stage. This could possibly be related to greater extent of care, surveillance, and continuous imaging in stage 0 patients who also subsequently progress to more advanced disease over time and get treated with LOC. Similar to the Sloan et al report, costs in our study were high in the first few months of diagnosis, after which they declined over the continuation phase.
Our study found that stage III patients had the highest lifetime costs ($190,996) and stage IV had the lowest lifetime costs ($117,503). Stage III patients had the highest diagnosis, continuation, and second highest terminal costs (second only to stage IV terminal phase costs). One reason why stage III patients may have the highest lifetime costs is the intensity of care received in a potentially curative setting where treatment plans often involve a multi-disciplinary approach and multi-modal care that may also include cystectomy and/or (neo)-adjuvant therapy.32 At the same time, however, stage III LOC recipients had lower lifetime costs than the average stage III patients, mainly attributable to lower costs during the continuation phase. Future studies should causally examine if lower lifetime costs among stage III patients are due to treatment with LOC.
Hospitalizations not related to cystectomy were a major cost driver for all patients in our study, accounting for nearly half of all costs, regardless of stage of disease at diagnosis or phase of treatment (diagnosis, continuation, terminal). This is to be expected as the study cohort represented elderly patients with a median age of 78 years, close to a quarter of whom with poor performance status, and close to 30% with CCI score of 2 and above. Future studies are needed to understand why patients with UC are being admitted to the hospital for reasons other than cystectomy. Some hospitalization costs may have been a result of complications from procedures, which occur especially among older patients and patients with multiple comorbidities.5,10,11,33 Such complications can result in extended hospital stays and increased costs. As a result, treatment plans that require fewer and shorter hospitalizations can lead the way to lessening the economic burden of UC. Office visit costs were another significant source of economic burden in UC patients. UC has a high risk of recurrence, and as a result continuous follow-up care and/or surveillance is needed. Depending upon their disease burden and disease course and whether they undergo cystectomy, a bladder sparing procedure, or receive chemotherapy, patients are often followed up frequently with lab tests and imaging. For patients with early stage disease and at high risk for disease recurrence more frequent office visits, cystoscopies, imaging and possible TURBT are often required.8,11,31,34
Our study is not without limitations. First, our study population consisted of Medicare patients with an average age of 78 years; therefore, our results are not generalizable to a younger non-Medicare population. Second, clinical characteristics such as M stage, performance status, and Charlson comorbidity index were measured at initial UC diagnosis rather than continuously updated over the follow-up period where changes can impact treatment decisions. For example, a patient could be diagnosed with M0 disease but develop metastasis over time. Third, the use of ICD-9, HCPCS, and Revenue Center codes to identify healthcare resources can be subject to potential miscoding. Fourth, it was not possible to identify whether radiation was given for locoregional treatment versus palliative care. Lastly, while our study provides an update to literature estimates, it covers a time period (2003–2014) prior to the approval of immune checkpoint inhibitors in bladder cancer.2
The present study highlights the significant financial burden incurred by UC patients and the healthcare system alike.30,35 Our study provides a needed update to the literature on the economic burden incurred by patients with stages 0 – IV urothelial carcinoma and comprehensively reports costs by disease phase, the clinical context of care, and by receipt of systemic chemotherapy. Estimates reported here provide a benchmark on cost of care prior to the approval of immuno-oncology therapies and could be helpful in informing patients, payers and stakeholders alike about the economic burden of UC during shared-decision making discussions with treating physicians.30,35
Supplementary Material
Acknowledgements:
This study used the linked SEER-Medicare database. The interpretation and reporting of these data are the sole responsibility of the authors. The authors acknowledge the efforts of the National Cancer Institute; the Office of Research, Development and Information, CMS; Information Management Services (IMS), Inc; and the SEER program tumor registries in the creation of the SEER-Medicare database. Editorial assistance was provided by Sophie Roberts, an employee of Pharmerit International.
Disclosures:
This study was funded by AstraZeneca. AA was an employee of Pharmerit International when the work was conducted but is a current employee of AstraZeneca and has stock options. YD and RS are employees of AstraZeneca and have stock options. VC and MB are current employees of Pharmerit International. MB is also a shareholder of Pharmerit International. CJ was an employee of Pharmerit International at the time of study conduct. Part of AH’s research time was supported by a Merit Review Award (I01 BX000545), Medical Research Service, Department of Veterans Affairs. The authors on this manuscript have no other relevant financial or other relationships to disclose.
References
- 1.Cancer Stat Facts: Bladder Cancer. 2018; https://seer.cancer.gov/statfacts/html/urinb.html. Accessed January 30, 2019.
- 2.Dietrich B, Srinivas S. Urothelial carcinoma: the evolving landscape of immunotherapy for patients with advanced disease. Research and reports in urology. 2018;10:7–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Anastasiadis A, de Reijke TM. Best practice in the treatment of nonmuscle invasive bladder cancer. Ther Adv Urol. 2012;4(1):13–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Park JC, Citrin DE, Agarwal PK, Apolo AB. Multimodal management of muscle-invasive bladder cancer. Current problems in cancer. 2014;38(3):80–108. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Svatek RS, Hollenbeck BK, Holmäng S, et al. The Economics of Bladder Cancer: Costs and Considerations of Caring for This Disease. European Urology. 2014;66(2):253–262. [DOI] [PubMed] [Google Scholar]
- 6.Sievert KD, Amend B, Nagele U, et al. Economic aspects of bladder cancer: what are the benefits and costs? World journal of urology. 2009;27(3):295–300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Survival Rates for Bladder Cancer. 2019; https://www.cancer.org/cancer/bladder-cancer/detection-diagnosis-staging/survival-rates.html. Accessed July 22, 2019.
- 8.Flaig T, Spiess P, Agarwal N, et al. NCCN Guidelines Insights: Bladder Cancer, Version 5.2018. Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw. 2018;16(9):1041–1053. [DOI] [PubMed] [Google Scholar]
- 9.Avritscher EB, Cooksley CD, Grossman HB, et al. Clinical model of lifetime cost of treating bladder cancer and associated complications. Urology. 2006;68(3):549–553. [DOI] [PubMed] [Google Scholar]
- 10.Mossanen M, Gore JL. The burden of bladder cancer care: direct and indirect costs. Current Opinion in Urology. 2014;24(5):487–491. [DOI] [PubMed] [Google Scholar]
- 11.Yeung C, Dinh T, Lee J. The health economics of bladder cancer: an updated review of the published literature. Pharmacoeconomics. 2014;32(11):1093–1104. [DOI] [PubMed] [Google Scholar]
- 12.Botteman MF, Pashos CL, Redaelli A, Laskin B, Hauser R. The health economics of bladder cancer. Pharmacoeconomics. 2003;21(18):1315–1330. [DOI] [PubMed] [Google Scholar]
- 13.Riley GF, Potosky AL, Lubitz JD, Kessler LG. Medicare payments from diagnosis to death for elderly cancer patients by stage at diagnosis. Medical care. 1995;33(8):828–841. [DOI] [PubMed] [Google Scholar]
- 14.James AC, Gore JL. The costs of non-muscle invasive bladder cancer. Urologic Clinics. 2013;40(2):261–269. [DOI] [PubMed] [Google Scholar]
- 15.Sloan FA, Yashkin AP, Akushevich I, Inman BA. The Cost to Medicare of Bladder Cancer Care. European Urology Oncology. 2019. [DOI] [PubMed]
- 16.de Oliveira C, Pataky R, Bremner KE, et al. Phase-specific and lifetime costs of cancer care in Ontario, Canada. BMC cancer. 2016;16(1):809. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hollingsworth JM, Zhang Y, Krein SL, Ye Z, Hollenbeck BK. Understanding the variation in treatment intensity among patients with early stage bladder cancer. Cancer. 2010;116(15):3587–3594. [DOI] [PubMed] [Google Scholar]
- 18.Hollenbeck BK, Ye Z, Dunn RL, Montie JE, Birkmeyer JD. Provider treatment intensity and outcomes for patients with early-stage bladder cancer. Journal of the National Cancer Institute. 2009;101(8):571–580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Cooksley CD, Avritscher EBC, Grossman HB, et al. Clinical Model of Cost of Bladder Cancer in the Elderly. Urology. 2008;71(3):519–525. [DOI] [PubMed] [Google Scholar]
- 20.James AC, Gore JL. The Costs of Non-Muscle Invasive Bladder Cancer. Urologic Clinics of North America. 2013;40(2):261–269. [DOI] [PubMed] [Google Scholar]
- 21.Noone A-M, Lund JL, Mariotto A, et al. Comparison of SEER Treatment Data With Medicare Claims. Medical care. 2016;54(9):e55–e64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.SEER. Registry Groupings in SEER Data and Statistics. SEER Registries https://seer.cancer.gov/registries/terms.html, 2018.
- 23.Potosky AL, Riley GF, Lubitz JD, Mentnech RM, Kessler LG. Potential for Cancer Related Health Services Research Using a Linked Medicare-Tumor Registry Database. Medical Care. 1993;31(8):732–748. [PubMed] [Google Scholar]
- 24.Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. Journal of clinical epidemiology. 1992;45(6):613–619. [DOI] [PubMed] [Google Scholar]
- 25.Davidoff AJ, Gardner LD, Zuckerman IH, Hendrick F, Ke X, Edelman MJ. Validation of disability status, a claims-based measure of functional status for cancer treatment and outcomes studies. Medical care. 2014;52(6):500. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Riley GF. Administrative and Claims Records as Sources of Health Care Cost Data. Medical Care. 2009;47(7_Supplement_1):S51–S55. [DOI] [PubMed] [Google Scholar]
- 27.Yabroff KR, Lamont EB, Mariotto A, et al. Cost of Care for Elderly Cancer Patients in the United States. JNCI: Journal of the National Cancer Institute. 2008;100(9):630–641. [DOI] [PubMed] [Google Scholar]
- 28.Wijeysundera HC, Wang X, Tomlinson G, Ko DT, Krahn MD. Techniques for estimating health care costs with censored data: an overview for the health services researcher. Clinicoecon Outcomes Res. 2012;4:145–155. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Meropol NJ, Schrag D, Smith TJ, et al. American Society of Clinical Oncology guidance statement: the cost of cancer care. Journal of Clinical Oncology. 2009;27(23):3868–3874. [DOI] [PubMed] [Google Scholar]
- 30.Casilla-Lennon MM, Choi SK, Deal AM, et al. Financial toxicity among patients with bladder cancer: reasons for delay in care and effect on quality of life. The Journal of urology. 2018;199(5):1166–1173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Spiess PE, Agarwal N, Bangs R, et al. Bladder Cancer, Version 5.2017, NCCN Clinical Practice Guidelines in Oncology. 2017;15(10):1240. [DOI] [PubMed] [Google Scholar]
- 32.Blakely T, Atkinson J, Kvizhinadze G, Wilson N, Davies A, Clarke P. Patterns of cancer care costs in a country with detailed individual data. Medical care. 2015;53(4):302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Noyes K, Singer EA, Messing EM. Healthcare economics of bladder cancer: cost-enhancing and cost-reducing factors. Current opinion in urology. 2008;18(5):533–539. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Woldu SL, Bagrodia A, Lotan Y. Guideline of guidelines: non-muscle-invasive bladder cancer. BJU Int. 2017;119(3):371–380. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Messing EM. Financial Toxicity of Having Bladder Cancer. Bladder Cancer (Amsterdam, Netherlands). 2018;4(3):351. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Bladder Cancer Stages. 2018; https://www.cancer.org/cancer/bladder-cancer/detection-diagnosis-staging/staging.html. Accessed January 30, 2019.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.