ADULT UROLOGY

CLINICAL MODEL OF LIFETIME COST OF TREATING

BLADDER CANCER AND ASSOCIATED COMPLICATIONS
ELENIR B. C. AVRITSCHER, CATHERINE D. COOKSLEY, H. BARTON GROSSMAN,
ANITA L. SABICHI, LOIS HAMBLIN, COLIN P. DINNEY, AND LINDA S. ELTING

ABSTRACT
Objectives. To estimate the lifetime cost of bladder cancer and the contribution of complications to the
total costs.
Methods. We reviewed the medical records of a retrospective cohort of 208 patients with bladder cancer
who registered at our comprehensive cancer center from 1991 to 1999. We multiplied the number of
resources used during management of bladder cancer by their unit charges. We converted charges into costs
using the Medicare cost-to-charge ratio and inflated these to 2005 U.S. dollars. We estimated future costs
by creating two extreme hypothetical scenarios. In the best-case scenario, we assumed patients with
superficial disease developed recurrences at the cohort’s mean rate and that patients with muscle-invasive
disease were disease free after definitive therapy. Survival was based on the U.S. life expectancy in both
cases. In the worst-case scenario, we assumed patients with superficial disease developed muscle-invasive
disease and that all patients subsequently died of bladder cancer.
Results. The average cost of bladder cancer was $65,158 among the cohort patients. Sixty percent of this
cost ($39,393) was associated with surveillance and treatment of recurrences, and 30% ($19,811) was
attributable to complications. The lifetime cost of bladder cancer was lower for the worst-case scenario
($99,270) than for the best-case scenario ($120,684). However, a greater proportion of the costs were
attributable to complications with the worst-case scenario (43%, $42,290) compared with the best (28%,
$34,169).
Conclusions. The management of bladder cancer and its associated complications results in a major
economic burden. More cost-effective surveillance strategies and approaches for preventing complications
are crucial to minimizing the disease’s clinical and economic consequences. UROLOGY 68: 549–553, 2006.
© 2006 Elsevier Inc.

B ladder cancer is a common malignancy of the

urinary tract, accounting for considerable mor-
bidity and mortality. It is estimated that 356,557 new
2002, with 56,500 of these cases occurring in the
United States alone.1,2 Transitional cell carcinomas
are the most common form of bladder cancer, ac-
cases of bladder cancer were diagnosed worldwide in counting for nearly 90% of all such cases.3 The
disease is two to three times more common in men,
and its incidence is particularly high in industrial-
This study was supported, in part, by a grant from Pharmacia,
Incorporated (formerly Searle). ized countries, mainly because of its association
This study was presented, in part, at the 17th International with tobacco use and exposure to industrial chem-
Symposium of the Multinational Association of Supportive Care icals.4 In the United States, the high incidence of
in Cancer (MASCC), Geneva, Switzerland, July 2005. bladder cancer, coupled with the chronic nature of
From the Section of Health Services Research, Department of the disease, results in an estimate that, nationwide,
Biostatistics and Applied Mathematics, and Departments of Urol-
ogy, Clinical Cancer Prevention, and Radiation Oncology, Uni- more than 479,000 Americans live with bladder
versity of Texas M. D. Anderson Cancer Center, Houston, Texas cancer, making it the third most prevalent cancer
Reprint requests: Elenir B. C. Avritscher, M.D., Section of in men and the tenth most prevalent in women.5
Health Services Research, Department of Biostatistics and Ap- Because of its lengthy and resource-intensive
plied Mathematics, University of Texas M. D. Anderson Cancer course, often requiring costly lifetime surveillance
Center, 1515 Holcombe Boulevard, Unit 447, Houston, TX
77030. E-mail: ecaramel@mdanderson.org with periodic cystoscopic examinations, bladder
Submitted: December 15, 2005, accepted (with revisions): cancer management presents a particularly signif-
March 30, 2006 icant economic challenge. A better understanding

© 2006 ELSEVIER INC. 0090-4295/06/$32.00

ALL RIGHTS RESERVED doi:10.1016/j.urology.2006.03.062 549
 TABLE I. Frequency of episodes of complications and patients affected by disease stage
Invasive Disease Superficial Disease
Episodes Patients Episodes Patients
Complication Type (n ⴝ 679) (n ⴝ 97) (n ⴝ 137) (n ⴝ 137)
Systemic chemotherapy toxicity 323 (47) 57 (59) 0 (0) 0 (0)
Urinary tract related* 131 (19) 52 (54) 101 (74) 57 (42)
Comorbidity related 81 (12) 40 (41) 17 (12) 15 (11)
Complex infection 60 (9) 36 (37) 5 (4) 5 (4)
Severe pain 20 (3) 15 (15) 0 (0) 0 (0)
Gastrointestinal complications 19 (3) 17 (18) 2 (1) 2 (1)
Thromboembolism 14 (2) 9 (9) 2 (1) 1 (1)
Bleeding (nonurinary) 12 (2) 10 (10) 0 (0) 0 (0)
Cystectomy/neobladder related† 10 (2) 4 (4) 5 (4) 5 (4)
Drug reaction 5 (1) 5 (5) 5 (4) 4 (3)
Metastasis related 3 (1) 3 (3) 0 (0) 0 (0)
None NA 15 (15) NA 70 (51)
KEY: NA ⫽ Not applicable.
Data presented as numbers, with percentages in parentheses.
* Urinary tract-related complications encompassed urinary tract infection and inflammation (including irritative voiding symptoms and urinary bleeding), urinary inconti-
nence, urinary obstruction, renal insufficiency, and bladder perforation.
†
Cystectomy/neobladder-related complications included wound dehiscence, fistula requiring surgical intervention, and erectile dysfunction.

study has examined the differential economic bur-

TABLE II. Distribution of costs by type
den of superficial and invasive disease or the con-
of expenses
tribution of complications to the total costs. There-
Mean Cost/Patient fore, we developed a clinical model of the lifetime
Cost Component (% of Total Cost)
cost of treating bladder cancer from the provider’s
Admission $16,778 (26) perspective, including patients younger than 65
Surgical procedures $15,781 (24) years of age, and describe the contribution of com-
Office visits $7,067 (11)
plications to the total costs.
Laboratory tests $6,517 (10)
Chemotherapy $5,420 (8)
Diagnostic imaging $4,864 (7)
MATERIAL AND METHODS
Ostomy care $3,496 (5)
Prescription drugs $3,022 (5) We obtained clinical and resource use data by reviewing the
Blood transfusions $1,748 (3) medical records of a retrospective cohort of 306 consecutive
Radiotherapy $465 (1) Houston-area patients who presented to The University of
Total $65,158 (100) Texas M.D. Anderson Cancer Center (MDACC) with transi-
tional cell carcinoma of the bladder from January 1, 1991 to
December 31, 1999. The cohort was limited to patients who
resided in the Houston area (ie, Harris County and its six
of the exact resources used during the treatment of contiguous counties) to ensure complete ascertainment of all
costs of cancer care. Of the initial 306 patients, 98 were ex-
bladder cancer and the factors associated with cluded from the analysis. Of these 98 patients, 45 had received
greater costs is important to guide policy makers a portion of their cancer care after registration from other
and providers in planning and allocating scarce providers, 31 were seen for a second opinion only, and 22 had
healthcare resources. However, scant information a history of other primary malignancies. As a result, 208 pa-
is available on which to base such decisions, par- tients were eligible for the analysis. The institutional review
board approved the study with a waiver of informed consent.
ticularly for patients younger than 65 years of age We first developed a clinical model of bladder cancer care
with bladder cancer. A review of the published reflecting care intervals for diagnosis, initial therapy, surveil-
studies on the economic burden of bladder cancer lance, management of recurrences, and terminal care. Through
conducted by Botteman et al.6 revealed that the medical record review, we then collected the total number and
lifetime cost per patient from the diagnosis of blad- type of resources used and subcategorized them into the care
intervals, as well as into whether they were disease or compli-
der cancer to death ranges from $96,000 to cation related. Complications were defined as any adverse
$187,000 (in 2001 U.S. dollars); the greatest per- event directly or indirectly related to bladder cancer and/or its
patient lifetime cost of all cancers among Medicare treatment. For patients who were diagnosed before registra-
patients.6 However, these cost estimates were tion at MDACC, we also collected the date and stage of the
based solely on Medicare patients, which excludes initial diagnosis and of all previous recurrences. We multi-
plied the number of resources used during care at MDACC by
a substantial number of patients younger than 65 their unit charges, obtained from our institutional accounting
years of age with bladder cancer.7–9 In addition, to system. To obtain cost estimates generalizable to other provid-
the best of our knowledge, no previously published ers nationwide, the charges were transformed into costs using

550 UROLOGY 68 (3), 2006

 TABLE III. Cost per care interval by disease stage
Mean Cost/Patient Mean Cost/Interval (95% CI)
for All Patients
Care Interval (% of Total Costs) Superficial Disease Muscle-Invasive Disease P Value*
Initial workup at M.D. Anderson $3,700 (6) $2,401 (1,942–2,859) $4,430 (3,006–5,854) ⬍0.001
Initial disease treatment $11,444 (18) $6,548 (5,379–7,717) $42,035 (32,080–51,990) ⬍0.001
Surveillance $13,457 (21) $4,525 (3,783–5,259)§ $9,768 (6,782–12,761)§ ⬍0.001
Treatment of recurrences $25,936 (39) $6,401 (5,173–7,630) $36,101 (29,820–42,381) ⬍0.001
Terminal care† $10,621 (16) NA $59,161 (39,505–78,816) NA
KEY: CI ⫽ confidence interval; NA ⫽ not applicable.
* Mann-Whitney test.
†
Terminal care only included costs for intervals in which patients died of bladder cancer.
§
Mean cost/yr.

state-specific Medicare cost-to-charge ratio.10 Pharmaceutical at the initial diagnosis. Of the 208 patients, 186
costs were based on the pharmaceutical industry’s average (89%) were diagnosed before registration at
wholesale price as listed in the Red Book.11 The costs were
then inflated to 2005 U.S. dollars using the Consumer Price
MDACC, with 58 (28%) having a previous history
Index for medical care.12 of recurrent disease. The mean follow-up time was
For patients who were diagnosed before registration at 5.3 years from the initial diagnosis (95% confi-
MDACC, we applied the cohort’s average cost per care interval dence interval [CI] 4.7 to 5.9) and 3.7 years (95%
and proportions attributed to complications by stage of dis- CI 3.3 to 4.0) from registration at MDACC. The
ease to previous treatment and surveillance periods and added
these values to the cohort’s observed costs to obtain the co-
patients experienced a mean of 0.7 recurrences per
hort’s total costs. We then estimated the range of future costs year (95% CI 0.5 to 0.8). By the end of the study
by creating two extreme hypothetical clinical scenarios. In the period, 97 patients (47%) had developed muscle-
best-case scenario, we assumed patients with superficial dis- invasive disease and 80 patients (38%) had
ease developed recurrences at the cohort’s average rate and died—31 (15%) of bladder cancer and 49 (23%) of
that all patients with muscle-invasive disease were disease free
after definitive therapy. In both cases, survival was determined
other causes.
according to U.S. life expectancy by sex and age, because we Complications requiring treatment affected 144
assumed the cohort’s probability of survival and mortality to patients (69%). A total of 816 episodes of compli-
not be affected by bladder cancer and, thus, to be the same as cations were observed during care at our institu-
that for the general population. For each patient with superfi- tion; 137 (17%) complicated the course of superfi-
cial disease, we applied the cohort’s rate of superficial recur-
rence to the predicted years of survival to obtain the number of
cial disease and 679 (83%) of muscle-invasive
future recurrences. We then multiplied the number of recur- disease (Table I). Urinary tract-related complica-
rences by the cohort’s mean cost and duration of superficial tions accounted for most of the episodes of super-
recurrences. Subsequently, we applied the cohort’s annual ficial disease complications (74%), and chemo-
cost of superficial disease surveillance to the patient’s pre- therapy and urinary tract-related complications
dicted lifespan (subtracted by the estimated duration of the
future recurrences) and added these costs to those of the esti-
were the most frequent muscle-invasive disease
mated recurrences to obtain the future costs. For patients with complications, accounting for 47% and 19% of all
muscle-invasive disease, we applied the cohort’s annual cost such complications, respectively (Table I).
for muscle-invasive disease surveillance to the estimated fu-
ture years of survival. In the worst-case scenario, we assumed
patients with superficial disease developed muscle-invasive RESOURCE USE AND COSTS
disease and that the entire cohort received terminal care (at Management of bladder cancer consumed con-
the cohort’s mean cost of terminal care) and subsequently died siderable healthcare resources. A mean of 2.9 cys-
of bladder cancer, with survival based on the cohort’s average toscopies per patient per year (95% CI 2.5 to 3.4)
length of terminal care interval.
and 49 cystectomies were performed during care at
MDACC. All but 25 patients required hospitaliza-
RESULTS
tion for a mean of 18 days/yr (95% CI 12 to 23): 1.3
PATIENT CHARACTERISTICS AND OUTCOMES days/yr (95% CI 1.0 to 1.7) for patients with super-
A total of 208 patients with bladder cancer were ficial disease and 36 days/yr (95% CI 26 to 47) for
eligible for the analysis. Of these, 97 (47%) were patients with muscle-invasive disease.
younger than 65 years of age at registration, 156 The mean cost of bladder cancer was $65,158 (95%
(75%) were men, and 177 (85%) were white. Most, CI 57,922 to 72,394) for the cohort patients (Table
147 patients (71%), were initially diagnosed with II). Admissions and surgical procedures together
superficial disease (American Joint Committee on accounted for 50% of the total costs ($32,559), and
Cancer Stage 1 or lower), and the remaining 61 surveillance for, and treatment of, tumor recur-
(29%) already had muscle-invasive disease (Amer- rences accounted for 60% ($39,393; Tables II and
ican Joint Committee on Cancer Stage 2 or greater) III). In addition, 30% of the total cost ($19,811)

UROLOGY 68 (3), 2006 551

was attributable to the treatment of complications:

P Value*

0.001
0.102
⬍0.001

⬍0.001
26% ($16,934) for muscle-invasive disease compli-
cations and 4% ($2,312) for superficial disease
complications. The care intervals with complica-
tions were consistently costlier than those without

$16,252 (4,453–28,052)

$12,101 (5,588–18,613)
$2,826 (2,167–3,485)

$7,164 (6,029–8,299)
complications for both superficial disease and

No Complications
muscle-invasive disease intervals (Table IV). Ap-
plying these data to the hypothetical scenarios, the

TABLE IV. Cost per care interval with complication versus periods with no complications by disease stage
Muscle-Invasive Disease
lifetime cost of bladder cancer was lower for the
worst-case scenario ($99,270) than for the best-
case scenario ($120,684) because of the longer sur-
vival in the latter group. However, a greater pro-
portion of the costs was attributable to
complications with the worst-case scenario
(43%, $42,290) compared with the best-case sce-

$49,520 (39,846–59,195)

$43,265 (36,157–50,373)
$13,558 (6,389–20,726)
$8,172 (3,867–12,477)
nario (28%, $34,169).

Complications
COMMENT
Our estimates of the lifetime cost for treating blad-
der cancer ($65,158 for the cohort, $99,270 for the
worst-case scenario and $120,684 for the best-case
scenario) have indicated that the economic burden
caused by the disease in patients of all ages treated
at a comprehensive cancer center can be substan-

P Value*
⬍0.001
0.008
0.007
⬍0.001
tial. Most importantly, our study has shown that
complications significantly contribute to the life-
time costs of managing bladder cancer. Almost one
third of the total costs were attributed to the treat-

$1,986 (1,640–2,332)
$5,792 (4,558–7,026)
$4,062 (3,332–4,792)
$4,992 (4,054–5,930)
ment of complications. We have also shown that No Complications
surveillance for bladder tumor recurrence and
treatment of the eventual recurrences accounted
Superficial Disease

for 60% of the total costs of managing bladder

cancer.
These data suggest that to minimize the disease’s
clinical and economic consequences, it is crucial to
identify more cost-effective surveillance strategies
and approaches for preventing bladder cancer-re-
$9,574 (7,888–11,259)

$11,688 (7,238–16,138)
$6,354 (3,275–9,432)

$6,369 (5,166–7,572)

lated complications. During the past years, a num-

ber of tumor markers have been identified for the
Complications

detection of bladder cancer tumors.13 The use of

Data presented as mean cost, with 95% confidence interval in parentheses.

proteomic and genomic-based markers holds great

promise for bladder cancer surveillance. However,
because these markers still need prospective vali-
dation and some have only shown improved tumor
detection when used adjunctly with cystoscopy, it
is essential that economic evaluations be con-
Initial workup at M.D. Anderson

ducted alongside their clinical trials.14 –17 Given the

heavy burden that surveillance and treatment of
Surveillance (mean cost/yr)

recurrences currently has on total costs, the assess-

Treatment of recurrences
Initial disease treatment

ment of the relative merits of new surveillance and

treatment strategies in light of their cost-effective-
ness profile will provide critical information for
* Mann-Whitney test.

guiding providers and policy-makers in evaluating

Care Interval

the cost impact of alternative strategies before dis-

semination into urologic practice.
However, our study had several limitations.
First, our estimates were derived exclusively from

552 UROLOGY 68 (3), 2006

a single comprehensive cancer center, which may 2. American Cancer Society: Cancer facts and fig-
have led to an overestimation of the costs and im- ures 2002. Available at: www.cancer.org/downloads/STT/
CancerFacts&Figures2002TM.pdf. Accessed April 8, 2005.
pact of the complications. In contrast, it is possible 3. Izawa JI, and Grossman HB: Localized bladder cancer.
that care is more efficiently delivered in such cen- Curr Treat Options Oncol 1: 423– 432, 2000.
ters, which would have led to a corresponding un- 4. Johansson SL, and Cohen SM: Epidemiology and eti-
derestimation of the costs and complications. The ology of bladder cancer. Semin Surg Oncol 13: 291–298,
balance of these opposing phenomena, however, 1997.
5. National Cancer Institute: NCI Fact Book. Available at:
can only be gauged in a larger and more represen- www3.cancer.gov/admin/fmb/03Factbk.pdf. Accessed April
tative cohort. Second, the retrospective nature of 12, 2005.
the data collection could have resulted in incom- 6. Botteman MF, Pashos CL, Redaelli A, et al: The health
plete ascertainment of resource use for treatment economics of bladder cancer: a comprehensive review of the
published literature. Pharmacoeconomics 21: 1315–1330,
of complications and a corresponding underesti- 2003.
mation of their contribution to the total costs. 7. Riley GF, Potosky AL, Lubitz JD, et al: Medicare pay-
Third, by applying the cohort’s proportions attrib- ments from diagnosis to death for elderly cancer patients by
uted to complications to the estimated future costs, stage at diagnosis. Med Care 33: 828 – 841, 1995.
we may have underestimated the economic burden 8. Baker S, Kessler LG, and Smucker RC: Site-Specific
Treatment Costs for Cancer: An Analysis of the Medicare Contin-
of complications in the hypothetical scenarios, be- uous History Sample File. Cancer Care and Costs: DRG’s and
cause comorbidity-related complications triggered Beyond 1989. Ann Arbor, Health Administration Press Per-
by bladder cancer or its treatment are expected to spectives, 1989.
increase with age. Finally, we only analyzed direct 9. U.S. Environmental Protection Agency: Cost of Illness
medical costs. By excluding indirect costs (ie, the Handbook. Cost of Bladder Cancer. Available at: www.epa.gov/
oppt/coi/docs/II_8.pdf. Accessed April 11, 2005.
costs associated with productivity loss), we under- 10. Department of Health and Human Services, Centers for
estimated the true economic burden of bladder Medicare and Medicaid Services: Changes to the hospital in-
cancer, which can be particularly germane to the patient prospective payment systems and fiscal year 2003
working-age population of patients with bladder rates. Federal Register 67: 49982–50289, 2003.
cancer.6,18 11. Medical Economics: Drug Topics Red Book. Montvale,
NJ, Medical Economics Company, 2003.
12. U.S. Department of Labor, Bureau of Labor Statistics:
Consumer price indexes. Available at: www.bls.gov/cpi/
CONCLUSIONS home.htm. Accessed August 9, 2004.
13. Lokeshwar VB, Habuchi T, Grossman HB, et al: Bladder
The management of bladder cancer, particularly tumor markers beyond cytology: International Consensus
surveillance for recurrence and the treatment of Panel on Bladder Tumor Markers. Urology 66(6 suppl 1):
the eventual recurrences, results in major clinical 35– 63, 2005.
14. Grossman HB, Soloway M, Messing E, et al: Surveil-
and economic burdens. Complications associated lance for recurrent bladder cancer using a point-of-care pro-
with bladder cancer and its treatment contribute teomic assay. JAMA 295: 299 –305, 2006.
significantly to the disease’s lifetime costs. Identi- 15. Van Le TS, Miller R, Barder T, et al: Highly specific urine-
fying more cost-effective surveillance strategies based marker of bladder cancer. Urology 66: 1256 –1260, 2005.
and approaches for preventing bladder cancer 16. Hautmann SH, Toma M, Lorenzo Gomez MF, et al:
Immunocyte and the HA-HAase urine tests for the detection of
complications is crucial to minimizing the dis- bladder cancer: a side by side comparison. Eur Urol 46: 466 –
ease’s clinical and economic consequences. 471, 2004.
17. van Rhijn BW, Lurkin I, Kirkels WJ, et al: Microsatellite
analysis—DNA test in urine competes with cystoscopy in fol-
REFERENCES low-up of superficial bladder carcinoma: a phase II trial. Can-
1. International Agency for Research on Cancer: cer 92: 768 –775, 2001.
GLOBOCAN 2002. Cancer incidence, mortality and preva- 18. Ruttenberg R, and Powers M: Economics of notification
lence worldwide. IARC Cancer Base, No. 5, version 2.0. and medical screening for high-risk workers. J Occup Med 28:
Available at: http:www.dep.iarc.fr. Accessed April 7, 2005. 757–764, 1986.

UROLOGY 68 (3), 2006 553