Provider Treatment Intensity and Outcomes For Patients With Early-Stage Bladder Cancer

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

ARTICLE

Provider Treatment Intensity and Outcomes for


Patients With Early-Stage Bladder Cancer
Brent K. Hollenbeck, Zaojun Ye, Rodney L. Dunn, James E. Montie, John D. Birkmeyer

Background Bladder cancer is among the most prevalent and expensive to treat cancers in the United States. In the
absence of high-level evidence to guide the optimal management of bladder cancer, urologists may vary
widely in how aggressively they treat early-stage disease. We examined associations between initial treat-
ment intensity and subsequent outcomes.

Methods We used the Surveillance, Epidemiology, and End Results–Medicare database to identify patients who
were diagnosed with early-stage bladder cancer from January 1, 1992, through December 31, 2002 (n = 20
713), and the physician primarily responsible for providing care to each patient (n = 940). We ranked the

Downloaded from http://jnci.oxfordjournals.org by on August 22, 2010


providers according to the intensity of treatment they delivered to their patients (as measured by their
average bladder cancer expenditures reported to Medicare in the first 2 years after a diagnosis) and then
grouped them into quartiles that contained approximately equal numbers of patients. We assessed asso-
ciations between treatment intensity and outcomes, including survival through December 31, 2005, and
the need for subsequent major interventions by using Cox proportional hazards models. All statistical
tests were two-sided.

Results The average Medicare expenditure per patient for providers in the highest quartile of treatment intensity
was more than twice that for providers in the lowest quartile of treatment intensity ($7131 vs $2830,
respectively). High–treatment intensity providers more commonly performed endoscopic surveillance and
used more intravesical therapy and imaging studies than low–treatment intensity providers. However, the
intensity of initial treatment was not associated with a lower risk of mortality (adjusted hazard ratio of
death from any cause for patients of low– vs high–treatment intensity providers = 1.03, 95% confidence
interval 0.97 to 1.09). Initial intensive management did not obviate the need for later interventions. In fact,
a higher proportion of patients treated by high–treatment intensity providers than by low–treatment inten-
sity providers subsequently underwent a major medical intervention (11.0% vs 6.4%, P = .02).

Conclusions Providers vary widely in how aggressively they manage early-stage bladder cancer. Patients treated by
high–treatment intensity providers do not appear to benefit in terms of survival or in avoidance of subse-
quent major medical interventions.

J Natl Cancer Inst 2009;101:571–580

Bladder cancer is the fifth most common new cancer diagnosis the optimal approaches to bladder cancer surveillance and treat-
and among the most expensive cancers to treat in the United States ment is largely lacking. Rather, current guidelines for the surveil-
(1). Nearly three-quarters of incident cases of bladder cancer are lance and treatment of early-stage bladder cancer are most
non–muscle-invasive (ie, early-stage) tumors (2), which are removed
endoscopically. In up to half of these patients, the disease will
progress to muscle-invasive cancer (3–6). Because mortality from Affiliations of authors: Division of Oncology, Department of Urology
muscle-invasive disease is common (7,8) and often requires a major (BKH, JEM), Division of Health Services Research, Department of Urology
(BKH, ZY, RLD, JEM), Department of Surgery (JDB), and Michigan Surgical
medical intervention (radical cystectomy, systemic chemotherapy, Collaborative for Outcomes Research and Evaluation (BKH, JDB), University
and/or radiation therapy), effective strategies for the prevention of Michigan Health System, Ann Arbor, MI.
and early detection of disease progression are of paramount Correspondence to: Brent K. Hollenbeck, MD, MS, Division of Oncology,
importance. Department of Urology, University of Michigan Health System, 1500 East
Medical Center Dr, TC 3875, Ann Arbor, MI 48109 (e-mail: bhollen@umich.
How best to achieve this goal remains unclear. Common strat- edu).
egies for bladder cancer surveillance and treatment that may be See “Funding” and “Notes” following “References.”
useful include intensive intravesical therapy, repeat endoscopic re- DOI: 10.1093/jnci/djp039
section after diagnosis, and frequent endoscopic surveillance (9,10). © The Author 2009. Published by Oxford University Press. All rights reserved.
However, evidence from randomized clinical trials establishing For Permissions, please e-mail: journals.permissions@oxfordjournals.org.

jnci.oxfordjournals.org JNCI | Articles 571


We first identified all Medicare patients aged 65–99 years who
CONTEXT AND CAVEATS
had an incident bladder cancer detected before death between
Prior knowledge January 1, 1992, and December 31, 2002, as documented by blad-
Little is known about how urologists vary in the aggressiveness der cancer codes 67.0–67.9 within the SEER–Medicare Patient
with which they treat patients during the first 2 years after a diag- Entitlement and Diagnosis Summary File. Next, we limited our
nosis of early-stage bladder cancer. study population to patients with early-stage bladder cancer [stage
Study design 0 or 1, defined according to the modified American Joint
Linked Surveillance, Epidemiology, and End Results (SEER)– Commission on Cancer (AJCC) (13)] by using specific codes pro-
Medicare data were used to identify patients who were diagnosed vided by SEER.
with early-stage bladder cancer and the physician primarily To ascertain the physician who had the primary responsibility
responsible for each patient’s care and to examine associations for each patient’s bladder cancer care, we first identified all early-
between initial treatment intensity and subsequent outcomes, stage bladder cancer–related procedures [as listed in the appendix
including survival.
of Schrag et al. (14)] that were performed within a 2-year period
Contribution following the patient’s diagnosis. Only claims for procedures that
Urologists who provided the most aggressive treatment had, on were performed for a primary diagnosis of bladder cancer were
average, more than double the Medicare expenditures per patient included. Next, we assigned each patient to the provider who had
compared with those who provided the least aggressive treatment, submitted the most of these Medicare procedure claims for the
but their patients did not appear to benefit in terms of survival or patient by using the Unique Physician Identifier Number. To

Downloaded from http://jnci.oxfordjournals.org by on August 22, 2010


in avoidance of subsequent major medical interventions.
ensure adequate reliability in our profiles of individual physician
Implications practice styles, we included only those providers who had treated
It may be possible to eliminate unnecessary procedures and thus at least 10 patients diagnosed with bladder cancer between January
reduce the costs of caring for patients with early-stage bladder 1, 1992, and December 31, 2002. Using this method, our final
cancer. study population consisted of 20 713 patients who were treated by
Limitations 940 providers, 99.4% of whom were urologists.
The use of observational data did not allow the authors to account
for unmeasured differences between patients in different treatment Characterization of Treatment Intensity
intensity groups. The use of SEER–Medicare data limits the gener- Treatment intensity was defined in terms of early-stage bladder
alizabity of the finding to patients older than 65 years. cancer expenditures, which were measured at the patient level and
From the Editors included inpatient and outpatient Medicare payments incurred
within the first 2 years after bladder cancer diagnosis. We included
only those expenditures that were associated with a primary dia-
gnosis code for bladder cancer [ie, International Classification of
commonly based on the opinions of experts or on observational Diseases, 9th Revision (ICD-9) (15) codes 188.x (bladder cancer),
data and generally favor more intensive regimens (10). Con- 233.7 (carcinoma in situ of the bladder), and V105.4 (personal his-
sequently, urologists vary widely in how they approach early-stage tory of bladder cancer)]. Payments were standardized to account
bladder cancer. for the regional variation in Medicare reimbursement (16).
In this context, we used linked Surveillance, Epidemiology, and Expenditures related to systemic chemotherapy, radiation therapy,
End Results (SEER)–Medicare data to identify differences in the and those incurred after these interventions were not included.
practice styles of US urologists during the first 2 years after an Because radical cystectomy is generally considered an effective
early-stage bladder cancer diagnosis. We were particularly inter- treatment for patients with high-risk early-stage bladder cancer
ested in the extent to which the initial treatment intensity of urolo- (17,18), expenditures related to cystectomy were included. All pay-
gists was associated with patients’ outcomes. ments were price adjusted to 2005 dollars by using the Medicare
Economic Index (16), and the sum of the price-adjusted payments
was attributed to the primary bladder cancer care provider. The
Methods providers were first ranked according to their average expenditures
Study Population for the 2-year period after the patients’ bladder cancer diagnosis
We used 1992–2005 SEER–Medicare linked data to identify a and then sorted into four treatment intensity groups (quartiles)
cohort of patients with early-stage bladder cancer. As detailed that contained approximately the same number of patients per
elsewhere (11), the files in this database provide a rich source of quartile.
information on Medicare patients included in SEER, a nationally To explore the practice patterns underlying treatment inten-
representative collection of population-based registries that collect sity, we characterized processes of care by using the ICD-9 and
information about all incident cancer patients from diverse geo- Healthcare Common Procedure Coding System (HCPCS) codes
graphic areas in the United States. By December 31, 2005, the in the Medicare files. The HCPCS codes are composed primarily
SEER registries included approximately 26% of the US popula- of Common Procedure Terminology codes (19) in addition to
tion (12). For each Medicare patient in SEER, the SEER- codes used exclusively by Medicare. For this study, we focused on
Medicare–linked files contain 100% of the Medicare claims from processes of care that were plausibly relevant to surveillance and
the inpatient, outpatient, and national claims history files. survival. As shown in the Appendix, we divided our process-of-care

572 Articles | JNCI Vol. 101, Issue 8 | April 15, 2009


measures into three categories: surveillance related (including correlations in mortality were used to derive variance–covariance
endoscopic examination of the bladder, upper urinary tract evalu- estimators. These sandwich estimators were then incorporated
ation, urinary studies, and imaging studies), treatment related into the Cox proportional hazards models that measured the asso-
(including intravesical therapy and repeat endoscopic resection ciations between treatment intensity and outcomes. For all Cox
within 60 days of the initial resection), and medical services models, we confirmed the assumption of proportionality by visual
(including visits to the urologist and visits to other physicians). inspection of the hazard plots and by goodness-of-fit testing (31).
For the secondary outcomes (use of cystectomy, systemic che-
Outcomes motherapy, and/or radiation therapy), we fit logistic models to
For all outcome measures, we used the patient as the unit of analy- estimate the association between provider treatment intensity and
sis. The primary outcome was all-cause mortality, which avoided patient-level outcome, by adjusting for patient age (5-year age
potential problems with misclassification of the cause of death groups), sex, race (white, black, or other), comorbidity (0, 1, 2, or ≥3),
(20–24) and was measured from January 1, 1992, through December socioeconomic status (low, medium, or high), the ICD-O-2
31, 2005, by using explicit vital status fields in SEER. However, tumor grade (low, high, or unknown), and the AJCC stage (Ta,
because the vast majority of patients with early-stage bladder can- Tis, T1, or Ta or T1 not otherwise specified). We computed
cer are likely to die from competing causes (6), we also assessed adjusted percentages for each outcome by back-transforming the
bladder cancer–specific mortality as a secondary outcome by using predicted use of the intervention from the logistic model. To
the cause-of-death field available in SEER. Finally, we assessed the examine the association between the use of a therapy and provider
patient’s need for a subsequent major medical intervention as evi- treatment intensity, Cox proportional hazards models were used to

Downloaded from http://jnci.oxfordjournals.org by on August 22, 2010


denced by their treatment with radical cystectomy, systemic che- take into account the timing of the therapy while adjusting for the
motherapy, and radiation therapy. These major interventions were covariates.
identified using the appropriate ICD-9 and HCPCS codes within To account for potential unmeasured confounding by disease
the inpatient, national claims history, and outpatient files. severity (ie, patients who received more intensive treatment may
have had more aggressive disease), we conducted a sensitivity
Statistical Analysis analysis in which initial treatment intensity and patient survival
For all of the analyses, the exposure was provider treatment inten- were measured in separate populations. For this analysis, treatment
sity, which was categorized as quartiles of patients. We first sought intensity was assessed by profiling the providers’ practice patterns
to understand differences in patient demographics and disease using data from January 1, 1992, through December 31, 1998, and
characteristics according to provider treatment intensity. Next, we survival was then assessed among the same providers’ patients who
characterized the practice styles of these providers by exploring were diagnosed with early-stage bladder cancer between January 1,
associations between treatment intensity and processes of care. 1999, and December 31, 2002.
For all of these comparisons, statistical inference was made using All analyses were carried out with SAS software (version 9.1;
chi-square and Kruskal–Wallis tests for categorical and continuous Cary, NC). All statistical tests were two-tailed, and the probability
data, respectively. of a type I error was set at .05. The study protocol was approved
To examine the association between treatment intensity and by the institutional review board of the University of Michigan.
survival, we fit a Cox proportional hazards model by adjusting for
patient and disease characteristics, including patient age (in 5-year
age groups), sex, race (white, black, or other), the International Results
Classification of Diseases for Oncology, 2nd edition (ICD-O-2) Medicare expenditures for the initial management of early-stage
(25) tumor grade (low, high, or unknown), and the AJCC (13) bladder cancer varied by more than twofold among quartiles of
tumor stage (Ta, Tis, T1, or Ta or T1 not otherwise specified). In provider treatment intensity and ranged from mean per-patient
addition, we adjusted for socioeconomic status by using a compos- expenditures of $2830 for low–treatment intensity providers to
ite measure that was assessed at the level of the patient’s ZIP code, $7131 for high–treatment intensity providers. Table 1 presents
as described by Diez Roux et al. (26). Patients were separated into clinical and disease characteristics as the average percentage of
three equally sized groups of socioeconomic status according to patients treated by providers within each quartile of provider treat-
the summary score for this composite measure: low (score ment intensity. Patient age at diagnosis, sex, comorbidity, and
range = ⫺12.23 to 1.19), medium (score range = 1.20 to 5.89), and tumor grade did not vary according to the initial treatment inten-
high (score range = 5.90 to 20.76). Patient comorbidities were iden- sity by the provider. Compared with providers in the lowest
tified by using ICD-9 diagnosis codes (15) in Medicare inpatient quartile of treatment intensity, those in the highest quartile of
and outpatient claims for health care encounters that had occurred treatment intensity treated patients with slightly more severe blad-
during the 12-month period preceding the bladder cancer diagnosis. der cancers, as evidenced by the higher proportion of their patients
We used the Klabunde et al. (27) adaptation of the Charlson with high-grade (29.1% vs 28.5%, P = .02) and stage T1 (28.6% vs
comorbidity index (28) to assess comorbidity. Patients were classi- 24.3%, P < .001) disease.
fied according to their comorbidity index score (0, 1, 2, or ≥3), As shown in Table 2, high–treatment intensity providers (ie,
which was treated as a categorical variable. Because patients who those in the highest quartile of treatment intensity) had higher
were treated by the same provider may have similar outcomes (29), rates of all surveillance- and treatment-related processes of care
we adjusted the models to account for this potential clustering by during the initial management of patients with early-stage blad-
using more robust standard errors (30). Briefly, within-cluster der cancer than low–treatment intensity providers (ie, those in

jnci.oxfordjournals.org JNCI | Articles 573


Table 1. Patient and disease characteristics by provider treatment intensity*

Quartiles of provider treatment intensity


Characteristic 1 (low) 2 3 4 (high) P
No. of patients 5198 5154 5177 5184 —
No. of providers 254 204 224 258 —
Mean Medicare expenditures per 2830 3962 4956 7131 —
patient in 2005 US dollars
Patient age at diagnosis (%), y .32
65–69 17.5 17.9 17.0 18.6
70–74 25.4 25.9 26.7 26.5
75–79 25.6 25.9 25.3 25.3
80–84 17.9 17.4 18.5 17.0
≥85 13.6 12.9 12.5 12.6
Female sex (%) 25.2 24.9 26.2 24.9 .42
Race (%) <.001
White 92.9 94.4 91.8 92.4
Black 1.8 1.8 2.4 3.5
Other 5.3 3.8 5.8 4.1
Socioeconomic status† (%) <.001

Downloaded from http://jnci.oxfordjournals.org by on August 22, 2010


Low 30.4 31.3 33.7 38.0
Medium 32.3 36.8 33.2 31.0
High 37.3 31.9 33.1 31.0
Comorbidity index (%) .19
0 45.1 44.0 43.3 42.5
1 29.6 30.1 29.3 30.5
2 14.1 14.7 15.5 14.8
≥3 11.2 11.2 11.9 12.2
Tumor grade‡ (%) .02
Low 65.1 64.1 63.8 62.6
High 28.5 28.7 28.9 29.1
Unknown 6.4 7.2 7.3 8.3
Tumor stage§ (%) <.001
Ta 58.4 58.9 54.1 53.0
Tis 6.5 6.5 7.4 7.5
T1 24.3 23.8 26.1 28.6
Ta or T1, not otherwise specified 10.8 10.8 12.4 10.9

* All P values are two-sided (chi-square test). — = not applicable.


† Based on summary scores determined according to Diez Roux et al. (26); low (score range = ⫺12.23 to 1.19), medium (score range = 1.20 to 5.89), and high
(score range = 5.90 to 20.76).
‡ International Classification of Diseases for Oncology, 2nd edition (25).
§ Modified American Joint Commission on Cancer (13).

the lowest quartile of treatment intensity). Compared with died from any cause between January 1, 1992, and December 31,
patients who were treated by low–treatment intensity providers, 2005. Patients treated by low–treatment intensity providers had a
those treated by high–treatment intensity providers were, on similar risk of death as those who were treated by high–treatment
average, followed up more rigorously with bladder endoscopy intensity providers (adjusted hazard ratio [HR] of death = 1.03,
(8.3 vs 7.3 procedures, P < .001), urine cytology (2.3 vs 1.3 tests, 95% confidence interval [CI] = 0.97 to 1.09) after adjusting for
P < .001), and radiographic imaging (6.0 vs 5.1 studies, P < .001). differences in demographics and cancer severity (ie, tumor grade
Treatment-related processes of care followed similar trends. and stage). When the patients were stratified by tumor grade and
Patients who were treated by high–treatment intensity providers stage, we observed the anticipated effects of these markers of
received statistically significantly more instillations (5.0 vs 2.6, P disease severity on survival, that is, patients with high-grade or T1
< .001) and induction courses (0.6 vs 0.5, P < .001) of intravesical disease had generally lower survival than their counterparts with
therapy than patients who were treated by low–treatment inten- low-grade or Ta disease, respectively, at all levels of provider treat-
sity providers. ment intensity. However, as with the primary analysis, we observed
Despite these differences in provider practice style, the median no survival benefit associated with more intensive care. For exam-
survival of patients was similar across all four quartiles of provider ple, among patients with T1 disease—a population with the high-
treatment intensity (Table 3) and ranged from 76.5 months for est risk of disease progression—those treated by low–treatment
those whose providers were in the second highest quartile to 79.8 intensity providers had a similar risk of death as those treated by
months for those whose providers were in the second lowest high–treatment intensity providers (adjusted HR of death = 0.98,
quartile (P = .50). Overall, 11 485 (55.4%) of the 20 713 patients 95% CI = 0.88 to 1.09).

574 Articles | JNCI Vol. 101, Issue 8 | April 15, 2009


Table 2. Provider practice styles according to average treatment intensity

Quartiles of provider treatment intensity


Processes of care 1 (low) 2 3 4 (high) P*
Surveillance related
Endoscopic surveillance, mean number of procedures 7.3 7.7 7.9 8.3 <.001
Upper urinary tract evaluation, mean number of tests 0.9 1.0 1.0 1.2 <.001
Radiographic imaging, mean number of studies 5.1 5.1 5.4 6.0 <.001
Urinary cytology, mean number of tests 1.3 1.7 1.9 2.3 <.001
Urine cytology, mean % of patients 41.3 47.8 50.4 56.8 <.001†
Urinalysis, mean number of tests 7.3 7.3 8.4 8.5 <.001
Treatment related
Intravesical therapy, mean number of instillations 2.6 3.4 4.1 5.0 <.001
Induction courses‡ of intravesical therapy, mean number 0.5 0.5 0.6 0.6 <.001
Induction intravesical therapy, mean % of patients 21.7 25.8 30.2 36.0 <.001†
Repeat endoscopic resection, mean % of patients 4.7 5.6 6.4 9.6 <.001†
Medical services
Visits to the urologist, mean number 3.4 3.5 4.1 4.2 <.001
Visits to the other physicians, mean number 20.2 19.4 19.8 19.9 .12

Downloaded from http://jnci.oxfordjournals.org by on August 22, 2010


* P values based on two-sided Kruskal–Wallis test except where indicated.
† Two-sided chi-square test.
‡ Five or more treatments of intravesical therapy within a 45-day period.

Overall, 1613 (7.8%) patients died from bladder cancer Discussion


between January 1, 1992, and December 31, 2005. However, as
We found that urologists vary widely in the intensity of treatment
with the primary outcome, we observed no benefit of treatment
they provide during the first 2 years after a diagnosis of early-stage
intensity to bladder cancer–specific survival (Table 3). In fact,
bladder cancer. On average, providers in the highest quartile of
patients who were treated by low–treatment intensity providers
treatment intensity had more than double the Medicare expendi-
had a 30% lower risk of death compared with those treated by tures per patient compared with those in the lowest quartile. The
high–treatment intensity providers (adjusted HR of death from high–treatment intensity style of practice was characterized by a
bladder cancer = 0.70, 95% CI = 0.59 to 0.83). Similar relation- greater use of all measured health services, including intravesical
ships between treatment intensity and bladder cancer–specific therapy, endoscopy, urinary studies, and imaging. However, this
survival were evident after stratifying patients by tumor grade aggressive early treatment approach did not improve survival or
and stage. prevent patients from having to undergo major medical interven-
As shown in Figure 1, patients treated by high–treatment tions in subsequent years. In fact, compared with patients treated by
intensity providers were not less likely to require a subsequent low–treatment intensity urologists, those treated by high–treatment
major medical intervention than those treated by low–treatment intensity urologists were nearly two and one-half times more likely
intensity providers (11.0% vs 6.4%, P = .02). Indeed, patients to undergo radical cystectomy and nearly twice as likely to receive
treated by high–treatment intensity providers were more likely any major medical intervention, even after accounting for patient
than patients treated by low–treatment intensity providers to differences.
undergo radical cystectomy, even after adjustment for differ- These findings highlight the lack of clinical consensus in
ences between the two groups of patients (3.9% vs 1.6%, P < how best to manage patients with early-stage bladder cancer.
.001). Current guidelines for the management of non–muscle-inva-
In a sensitivity analysis, we repeated the primary analysis by sive bladder cancer generally favor the more intensive regi-
assessing treatment intensity and survival in separate patient mens of endoscopy and intravesical therapy (9). However,
populations. Briefly, treatment intensity was measured by use of neither of these regimens has convincingly demonstrated the
providers’ practice patterns for their patients diagnosed between ability to prevent disease progression or to prolong patient
January 1, 1992, and December 31, 1998. Overall mortality was survival (32–34). Indeed, only one randomized trial (32) to our
then assessed for the same providers among their patients diag- knowledge has explored the question of optimal endoscopic
nosed with bladder cancer between January 1, 1999, and surveillance care. Because that study included only 97 patients,
December 31, 2002. As with the primary analysis, we observed the findings were inconclusive. In light of the limited
no differences in overall mortality according to treatment high-level evidence to guide clinical practice, the considerable
intensity (eg, adjusted HR of death for low vs high treatment variation in the early treatment of bladder cancer is not
intensity = 1.00, 95% CI = 0.89 to 1.13). A similar null relation- surprising.
ship was observed when we used bladder cancer–specific sur- One potential limitation of our analysis relates to unmeasured
vival as the outcome. differences in patients among the physician treatment intensity

jnci.oxfordjournals.org JNCI | Articles 575


Table 3. All-cause and bladder cancer–specific mortality risks according to average provider treatment intensity*

Median survival in months by quartiles of Adjusted HR of death (95% CI) for patients of
provider treatment intensity low– vs high–treatment intensity providers
Bladder cancer–
Patient stratification 1 (low) 2 3 4 (high) All-cause mortality specific mortality
All patients 76.8 79.8 76.5 78.0 1.03 (0.97 to 1.09)† 0.70 (0.59 to 0.83)†
Stratified by grade
Low 85.1 86.6 83.4 85.9 1.03 (0.96 to 1.11)‡ 0.66 (0.52 to 0.85)‡
High 62.0 64.0 62.5 63.3 1.03 (0.92 to 1.15)‡ 0.81 (0.63 to 1.03)‡
Unknown 71.1 80.7 72.8 71.3 0.96 (0.78 to 1.19)‡ 0.39 (0.20 to 0.75)‡
Stratified by stage
Ta 85.3 87.4 84.3 89.9 1.06 (0.98 to 1.14)§ 0.74 (0.57 to 0.97)§
Tis 76.8 72.0 74.9 72.4 0.91 (0.74 to 1.11)§ 0.49 (0.27 to 0.89)§
T1 75.8 76.6 76.7 78.9 0.98 (0.88 to 1.09)§ 0.68 (0.54 to 0.86)§
Ta or T1 64.4 64.8 63.7 61.7 1.06 (0.89 to 1.27)§ 0.72 (0.44 to 1.18)§

* HR = hazard ratio; CI = confidence interval.


† Adjusted for age, sex, race, socioeconomic status, comorbidity, tumor grade, and tumor stage.
‡ Adjusted for age, sex, race, socioeconomic status, comorbidity, and tumor stage.

Downloaded from http://jnci.oxfordjournals.org by on August 22, 2010


§ Adjusted for age, sex, race, socioeconomic status, comorbidity, and tumor grade.

groups. In particular, patients treated by high–treatment inten- would require systematic variation in unmeasured risk factors
sity urologists might have more aggressive disease than those across providers, which is probably less likely than variation in
treated by low–treatment intensity urologists, which could such risks across patients. Finally, our sensitivity analysis to assess
explain the apparent lack of benefit associated with treatment treatment intensity and survival in separate patient populations
intensity. We addressed this well-described limitation of observa- also failed to demonstrate a survival advantage of more aggressive
tional data (35,36) in several ways. First, we used a clinical regis- treatment.
try to ascertain patients’ bladder cancer stage and grade, which Although more aggressive early treatment intensity was not
are, arguably, the most important determinants of death in the associated with survival, it was associated with higher rates of
bladder cancer patient population (7,37). Patients in the different major medical interventions, including radical cystectomy, sys-
treatment intensity groups were similar with respect to age, sex, temic chemotherapy, and radiation therapy. There are several
and comorbidity. Second, we assessed treatment intensity at potential explanations for this finding. First, as discussed above,
the level of the provider. Relative to a patient-level analysis, this high–treatment intensity providers may have been treating sicker
approach is less susceptible to selection bias to the extent that it patients who, ultimately, required such interventions. However,
unmeasured confounding seems unlikely given our approach of
measuring treatment intensity at the level of the provider and the
large differences in the rates of major intervention. Second, it is
possible that more intensive therapy could, paradoxically, increase
the risk of disease progression and thus the need for major medi-
cal interventions. However, we know of no biological mechanism
to support this possibility. Third and perhaps most likely, pro-
vider practice styles with regard to the management of early-
stage bladder cancer, as measured by their initial treatment
intensity, may be consistent with those for more advanced dis-
ease. Simply put, urologists who treat aggressively early are likely
to provide aggressive treatment in all aspects of bladder cancer
care, and vice versa.
A second limitation of our findings relates to their applica-
bility to the broader population of patients with early-stage
Figure 1. Use of major medical interventions among patients with bladder cancer. Because we relied on SEER–Medicare data, our
early-stage bladder cancer by quartiles of provider treatment intensity, findings may not be generalizable to patients younger than 65 years.
expressed as the percentage of patients adjusted for age, sex, race, However, it is important to note that nearly three-quarters of
comorbidity, socioeconomic status, and tumor grade and stage. The
adjusted percentages were obtained by back-transforming data from bladder cancer cases in the United States occur annually within
the logistic models. Statistical inference was based on the Wald chi- the Medicare population (12). In early-stage bladder cancer,
square statistic obtained from the Cox proportional hazards models for unlike in prostate cancer, treatment decisions generally are
comparisons between the highest and lowest treatment intensity
quartiles within interventions, adjusting for the above covariates. All not made on the basis of the patient’s age. Thus, extrapolation
P values are two-sided. of our findings to the broader cohort (ie, all patients with

576 Articles | JNCI Vol. 101, Issue 8 | April 15, 2009


early-stage bladder cancer) would appear to be reasonable. Given the lack of association between treatment intensity and
Although overall treatment intensity was not associated with survival, our findings suggest the opportunity for reducing costs
better outcomes, it is possible that greater use of individual by eliminating unnecessary procedures and thus reducing waste-
aspects of early-stage bladder cancer care (eg, endoscopic ful spending for the care of patients with early-stage bladder
surveillance) could afford a benefit for some patients. Using cancer, which is already among the most expensive cancers in the
observational data to identify such components of care may United States (1). In light of the small but nontrivial risks associ-
provide better and more efficient care in patients with early- ated with early-stage bladder cancer surveillance and treatment,
stage bladder cancer. Finally, the lack of an association between the overuse of a high–treatment intensity practice style is worri-
treatment intensity and all-cause mortality among patients tra- some given its lack of association with any benefit for the patients.
ditionally felt to be at high risk of disease progression (ie, those Identifying best practices of care for patients diagnosed with
with stage T1 and/or high-grade tumors) does not preclude the early-stage bladder cancer must ultimately await the findings
possibility that some groups of patients may benefit from from future well-designed randomized clinical trials. In the
greater intensity of care; rather, it suggests that such patient meantime, urologists should not assume that more aggressive
populations are not readily identifiable by the grade and stage management of early-stage bladder cancer will translate into
information captured in SEER. better outcomes for their patients.

Appendix Table Codes used to identify processes of care*

Downloaded from http://jnci.oxfordjournals.org by on August 22, 2010


HCPCS codes (National
Process Variable ICD-9 diagnosis Claims History and
Category of care specification codes (MEDPAR) Outpatient files) Descriptor
Surveillance Endoscopic Discrete variable: 57.32, 57.39 52 000, 52 001, 52 005, 52 007 Cystoscopy: with irrigation and
related surveil- use for each 1-month evacuation of blood clots;
lance interval for 2 years with ureteral catheter; with
after diagnosis brush biopsy of the ureter
(range = 0–24 months)
57.33 52 204, 52214 Cystoscopy with biopsy,
fulguration
52 250 Cystoscopy with insertion of
radioactive substance, with or
without biopsy or fulguration
52 260, 52 265 Cystoscopy with bladder
dilation
57.91 52 270, 52 275, 52 276, Cystoscopy with urethral
52 277, 52 281, 52 282, dilation or urethrotomy
52 283, 52 285
52 290, 52 300, 52 301, Cystoscopy with ureteral
52 305 meatotomy
57.0 52 310, 52 315 Cystoscopy with removal of
foreign body
52 320, 52 325, 52 327, Cystoscopy for ureteral
52 330, 52 332, 52 334 calculus
52 317, 52 318 Cystoscopy with lithalopaxy
52 341, 52 342, 52 343 Cystoscopy for ureteral
stricture
57.92, 60.2 52 347, 52 400, 52 450, Cystoscopy with transurethral
52 500, 52 510, 52 601, prostate surgery
52 606, 52 612, 52 614,
52 620, 52 630, 52 640,
52 647, 52 648, 52 700
56.31, 56.33 52 344, 52 345, 52 346, Cystoscopy with ureteroscopy
52 351, 52 352, 52 353,
52 354, 52 355
52 224 Cystoscopy, with fulguration or
treatment of minor (<0.5 cm)
bladder lesions, with or
without biopsy
57.49 52 234, 52 235, 52 240 Cystoscopy, with fulguration
and/or resection of small (0.5
up to 2 cm), medium (2 up
to 5 cm), large (≥5 cm)
bladder tumors

(Table continues)

jnci.oxfordjournals.org JNCI | Articles 577


Appendix Table (continued).

HCPCS codes (National


Process Variable ICD-9 diagnosis Claims History and
Category of care specification codes (MEDPAR) Outpatient files) Descriptor
Upper Discrete variable: 74 400, 74 410, 74 415 Intravenous urography
urinary use for each 1-month
tract eval- interval for 2 years
uation after diagnosis (range
0–24)
74 420 Retrograde urography
74 425 Antegrade urography
74 150, 74 160, 74 170 Abdominal CT
74 181, 74 182, 74 183, 74 185 Abdominal MRI
Urinary Discrete variables: 81 000, 81 001, 81 002, Urinalysis
studies use (uniquely for 81 003, 81 005
urinalysis and cytol-
ogy) for each 1-month
interval for 2 years after
diagnosis (range 0–24)

Downloaded from http://jnci.oxfordjournals.org by on August 22, 2010


Dichotomous variable:
use of urine cytology
at least once during
2-year interval
88 104, 88 106, 88 107, Urine cytology and
88 108, 88 112, 88 160, 88 161, cytogenetics
88 162, 88 271, 88 272,
88 273, 88 274, 88 275, 88 291
Imaging Continuous variable: 71 010, 71 015, 71 020, Chest radiography
studies number of unique 71 021, 71 022, 71 023,
claims over 2-year 71 030, 71 034, 71 035
interval
71 250, 71 260, 71 270, 71 275 Chest CT
71 550, 71 551, 71 552, 71 555 Chest MRI
74 150, 74 160, 74 170 Abdominal CT
74 181, 74 182, 74 183, 74 185 Abdominal MRI
72 191, 72 192, 72 193, 72 194 Pelvis CT
72 195, 72 196, 72 197 Pelvis MRI
76 497 Unlisted CT
76 498 Unlisted MRI
78 810 PET scan
74 400, 74 410, 74 415 Intravenous urography
74 420 Retrograde urography
74 425 Antegrade urography
76 700, 76 770, 76 705, Abdominal or renal ultrasound
76 775, 76 778
78 700, 78 701, 78 704, 78 707, Nuclear medicine renal scan
78 708, 78 709, 78 715
78 300, 78 305, 78 306, Nuclear medicine bone scan
78 315, 78 320
78 800, 78 801, 78 802, Radiopharmaceutical
78 990 localization of tumor or distri-
bution of radiopharmaceuti-
cal agents
Treatment Intravesical Continuous variables: 51 720 Bladder instillation of
related therapy number of unique anticarcinogenic agent
claims over 2-year
interval; and, measure
number of induction
courses† over 2-year
interval
Dichotomous variable:
use of induction che-
motherapy at least
once during 2-year
interval

(Table continues)

578 Articles | JNCI Vol. 101, Issue 8 | April 15, 2009


Appendix Table (continued).

HCPCS codes (National


Process Variable ICD-9 diagnosis Claims History and
Category of care specification codes (MEDPAR) Outpatient files) Descriptor
51 700 Bladder irrigation, with or
without instillation
51 701 Insertion of a nonindwelling
catheter
51 702, 51 703 Insertion of a temporary
indwelling catheter
J8520, J8521 Capecitabine
J9000, J9001 Doxorubicin
J9031 Intravesical Bacille
Calmette-Guérin
J9201 Gemcitabine
J9280, J9290, J9291 Mitomycin
J9340 Thiotepa
Repeat Dichotomous variable: 57.49 52 234, 52 235, 52 240 Cystoscopy, with fulguration
endo- use of repeat endo- and/or resection of small
scopic scopic resection within (0.5 up to 2 cm); medium

Downloaded from http://jnci.oxfordjournals.org by on August 22, 2010


resection 60 days of initial resec- (2 up to 5 cm); large (≥5 cm)
tion or biopsy tumors
Medical Visits to the Continuous variable: 99 201–99 205 New patient, outpatient
services urologist number of unique
claims over 2-year
interval
99 211–99 215 Established patient, outpatient
99 218–99 220, 99 234–99 236 Initial, established, observation
unit
99 221–99 223 Initial, inpatient
99 231–99 233 Subsequent, inpatient
99 241–99 245 Consult, outpatient
99 251–99 255 Consult, inpatient, new
99 261–99 263 Consult, inpatient, established
99 271–99 275 Confirmatory consult
Visits to Continuous variable: 99 201–99 205 New patient, outpatient
other phy- number of unique
sicians claims over 2-year
interval
99 211–99 215 Established patient, outpatient
99 218–99 220, 99 234–99 236 Initial, established, observation
unit
99 221–99 223 Initial, inpatient
99 231–99 233 Subsequent, inpatient
99 241–99 245 Consult, outpatient
99 251–99 255 Consult, inpatient, new
99 261–99 263 Consult, inpatient, established
99 271–99 275 Confirmatory consult

* ICD-9 = International Classification of Diseases, 9th Revision; MEDPAR = Medicare Provider Analysis and Review file; HCPCS = Healthcare Common
Procedure Coding System; CT = computed tomography; MRI = magnetic resonance imaging; PET = positron emission tomography.
† Induction intravesical therapy: at least five unique claims for instillation of any anticarcinogenic agent within a 45-day period.

References 4. Holmang S, Hedelin H, Anderstrom C, Holmberg E, Busch C, Johansson


1. Riley GF, Potosky AL, Lubitz JD, Kessler LG. Medicare payments from SL. Recurrence and progression in low grade papillary urothelial tumors.
diagnosis to death for elderly cancer patients by stage at diagnosis. Med J Urol. 1999;162(3 pt 1):702–707.
Care. 1995;33(8):828–841. 5. Holmang S, Hedelin H, Anderstrom C, Johansson SL. The relationship
2. Snyder C, Harlan L, Knopf K, Potosky A, Kaplan R. Patterns of among multiple recurrences, progression and prognosis of patients
care for the treatment of bladder cancer . J Urol . 2003 ; 169 ( 5 ): with stages Ta and T1 transitional cell cancer of the bladder followed
1697–1701. for at least 20 years. J Urol. 1995;153(6):1823–1826; discussion
3. Haukaas S, Daehlin L, Maartmann-Moe H, Ulvik NM. The long-term 1826–1827.
outcome in patients with superficial transitional cell carcinoma of 6. Herr HW. Tumor progression and survival of patients with high grade,
the bladder: a single-institutional experience. BJU Int. 1999;83(9): noninvasive papillary (TaG3) bladder tumors: 15-year outcome. J Urol.
957–963. 2000;163(1):60–61.

jnci.oxfordjournals.org JNCI | Articles 579


7. Stein JP, Lieskovsky G, Cote R, et al. Radical cystectomy in the treatment 25. World Health Organization. International Classification of Diseases for
of invasive bladder cancer: long-term results in 1,054 patients. J Clin Oncology, Second Edition. Geneva: World Health Organization; 1990.
Oncol. 2001;19(3):666–675. 26. Diez Roux AV, Merkin SS, Arnett D, et al. Neighborhood of residence
8. Dimopoulos MA, Moulopoulos LA. Role of adjuvant chemotherapy in the and incidence of coronary heart disease [see comment]. N Engl J Med.
treatment of invasive carcinoma of the urinary bladder. J Clin Oncol. 2001;345(2):99–106.
1998;16(4):1601–1612. 27. Klabunde CN, Potosky AL, Legler JM, Warren JL. Development of a
9. Hall MC, Chang SS, Dalbagni G, et al. Guideline for the management of comorbidity index using physician claims data. J Clin Epidemiol.
nonmuscle invasive bladder cancer (stages Ta, T1, and Tis): 2007 update. 2000;53(12):1258–1267.
J Urol. 2007;178(6):2314–2330. 28. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of clas-
10. The National Comprehensive Cancer Network. Bladder Cancer V.2.2008. sifying prognostic comorbidity in longitudinal studies: development and
http://www.nccn.org/professionals/physician_gls/PDF/bladder.pdf . validation. J Chronic Dis. 1987;40(5):373–383.
Accessed June 4, 2008. 29. Panageas KS, Schrag D, Riedel E, Bach PB, Begg CB. The effect of clus-
11. Warren JL, Klabunde CN, Schrag D, Bach PB, Riley GF. Overview of the tering of outcomes on the association of procedure volume and surgical
SEER-Medicare data: content, research applications, and generalizability to outcomes [see comment]. Ann Intern Med. 2003;139(8):658–665.
the United States elderly population. Med Care. 2002;40(8 suppl):IV-3-18. 30. Lin DY. Cox regression analysis of multivariate failure time data: the
12. Surveillance, Epidemiology, and End Results (SEER) Program. http:// marginal approach. Stat Med. 1994;13(21):2233–2247.
seer.cancer.gov/. Accessed December 12, 2006. 31. Kleinbaum D. Evaluating the proportional hazards assumption. In: Diez
13. Greene FL, Page EL, Fleming ID. AJCC Cancer Staging Manual. K, Gail M, Krickeberg K, Singer B, eds. Survival Analysis. New York:
New York: Springer-Verlag; 2002. Springer; 1996:129–166.
14. Schrag D, Hsieh LJ, Rabbani F, Bach PB, Herr H, Begg CB. Adherence 32. Olsen LH, Genster HG. Prolonging follow-up intervals for non-invasive
to surveillance among patients with superficial bladder cancer. J Natl bladder tumors: a randomized controlled trial. Scand J Urol Nephrol

Downloaded from http://jnci.oxfordjournals.org by on August 22, 2010


Cancer Inst. 2003;95(8):588–597. Suppl. 1995;172:33–36.
15. US Public Health Service. International Classification of Diseases, 9th revi- 33. Soloway MS, Sofer M, Vaidya A. Contemporary management of stage T1
sion. (PHS) 92-1260. Washington, DC: US GPO; 1992. transitional cell carcinoma of the bladder. J Urol. 2002;167(4):1573–1583.
16. Brown ML, Riley GF, Schussler N, Etzioni R. Estimating health care 34. Shelley MD, Court JB, Kynaston H, Wilt TJ, Coles B, Mason M.
costs related to cancer treatment from SEER-Medicare data. Med Care. Intravesical bacillus Calmette-Guerin versus mitomycin C for Ta and T1
2002;40(8 suppl):IV-104-17. bladder cancer. Cochrane Database Syst Rev. 2003(3):CD003231.
17. Herr HW, Sogani PC. Does early cystectomy improve the survival of 35. Stukel TA, Fisher ES, Wennberg DE, Alter DA, Gottlieb DJ, Vermeulen MJ.
patients with high risk superficial bladder tumors? J Urol. 2001;166(4): Analysis of observational studies in the presence of treatment selection bias:
1296–1299. effects of invasive cardiac management on AMI survival using propensity
18. Stein JP. Indications for early cystectomy. Semin Urol Oncol. 2000;18(4): score and instrumental variable methods. JAMA. 2007;97(3):278–285.
289–295. 36. Radford MJ, Foody JM. How do observational studies expand the evi-
19. American Medical Association. Current Procedure Terminology. CPT 2005, dence base for therapy? [comment]. JAMA. 2001;286(10):1228–1230.
Standard Edition. Chicago, IL: AMA Press; 2004. 37. Heney NM, Nocks BN, Daly JJ, et al. Ta and T1 bladder cancer: location,
20. Bach PB, Guadagnoli E, Schrag D, Schussler N, Warren JL. Patient demo- recurrence and progression. Br J Urol. 1982;54(2):152–157.
graphic and socioeconomic characteristics in the SEER-Medicare database
applications and limitations. Med Care. 2002;40(8 suppl):IV-19-25. Funding
21. Hoel DG, Ron E, Carter R, Mabuchi K. Influence of death certificate
American Cancer Society Pennsylvania Division—Dr. William and Rita
errors on cancer mortality trends. J Natl Cancer Inst. 1063;85(13):
Conrady Mentored Research Scholar Grant (MSRG-07-006-01-CPHPS to
1063–1068.
B.H.); American Urological Association Foundation (to B.H.); Astellas Pharma
22. Feuer EJ, Merrill RM, Hankey BF. Cancer surveillance series: interpret-
US, Inc. (to B.H.); National Cancer Institute (R01 CA098481-01A1, K05
ing trends in prostate cancer—part II: cause of death misclassification and
CA115571-01 A2 to J.B.).
the recent rise and fall in prostate cancer mortality. J Natl Cancer Inst.
1999;91(12):1025–1032.
23. Penson DF, Albertsen PC, Nelson PS, Barry M, Stanford JL. Determining Notes
cause of death in prostate cancer: are death certificates valid? J Natl The study sponsors had no role in the design of the study; the collection, analy-
Cancer Inst. 2001;93(23):1822–1823. sis, or interpretation of the data; the writing of the manuscript; or the decision
24. Weinstock MA, Reynes JF. Validation of cause-of-death certification for to submit the manuscript for publication.
outpatient cancers: the contrasting cases of melanoma and mycosis fungoi- Manuscript received November 5, 2008; revised January 9, 2009; accepted
des. Am J Epidemiol. 1998;148(12):1184–1186. February 2, 2009.

580 Articles | JNCI Vol. 101, Issue 8 | April 15, 2009

You might also like