Abstract
Background
While there is level I evidence demonstrating superiority of intravesical therapy in in patients with bladder cancer, surveillance strategies are primarily founded on expert opinion. We examined compliance with surveillance and treatment strategies and the pursuant impact on survival in patients with high-grade disease.
Methods
Using linked SEER-Medicare data, we identified subjects with a diagnosis of high-grade non-muscle-invasive disease in 1992–2002 who survived two years and did not undergo definitive treatment during that time. We used non-linear mixed-effects regression analyses to examine compliance with surveillance and treatment strategies. After adjusting for confounders using a propensity-score-weighted approach, we determined whether individual and comprehensive strategies during the initial two years after diagnosis were associated with survival after two years.
Results
Of 4,790 subjects, only one received all the recommended measures. While mean utilization for most measures significantly increased after 1997, only compliance with an induction course of BCG increased (13% to 20%, p<0.001). On multivariate analysis, compliance with ≥ 4 cystoscopies, ≥ 4 cytologies and BCG instillation was lower among octogenarians and higher among those with undifferentiated, Tis and T1 tumors, and among those diagnosed after 1997. Subjects compliant with these measures had a lower hazard of mortality (HR 0.41; 95% CI 0.18–0.93) than those who received <4 cystoscopies, <4 cytologies and no BCG.
Conclusion
There is a statistically significant survival advantage found among those who received at least half of the recommended care. Improving compliance with these process-of-care measures via systematic quality-improvement initiatives serve as the primary target to meliorate bladder cancer care.
Keywords (MeSH Terms): Urinary Bladder Neoplasms, Guideline Adherence, Quality of Healthcare, Survival
INTRODUCTION
In response to increased consumer and purchaser interest in ensuring high-quality care, the Agency for Health Care Policy and Research funded development of clinical guidelines for the management of patients with chronic conditions in the early 1990s. While there was growing skepticism regarding the quality of care delivered in America prior to publications by the Institute of Medicine (IOM),1, 2 it was not until McGlynn et al reported that only 55% of adults received the recommended care that the public became aware of the sizable gap between science and routine practice.3
While bladder cancer is not routinely referred to as a chronic condition, it shares many similar properties: It is common, necessitates multiple interventions, and costly. To address these concerns, clinical guidelines, like those set forth by the National Comprehensive Cancer Network (NCCN), the American Urological Association (AUA) and the European Association of Urology (EAU), were established in an attempt to the minimize morbidity and mortality associated with recurrence and progression of bladder cancer. The surveillance and treatment strategies embodied within these guidelines have been infused into education, specialty certification and reimbursement models. Despite this infusion, Schrag et al., using a relaxed definition for a single quality-of-care measure—endoscopic evaluation every six months instead of every three months—discovered that 40% patients underwent the recommended number endoscopic procedures.4 Using MEDSTAT claims data, Madeb et al. discovered that out of 14,677 subjects with non-muscle-invasive bladder cancer who underwent resection of their bladder tumor, only 49 (0.3%) received perioperative instillation of chemotherapy.5 Additionally, Huang et al., found that only 42% of patients with high-grade non-muscle-invasive bladder cancer received a single instillation of immune- or chemotherapy. 6
However, established clinical practice guidelines incorporate a comprehensive surveillance (cystoscopy, urinary cytology, and upper tract imaging) and treatment (perioperative mitomycin C and postoperative Bacillus Calmette-Guérin (BCG)) schedule and not just a single quality-of-care measure.7, 8 Additionally, the current surveillance strategies were derived from expert opinion and have yet to be associated with improved outcomes. Against this backdrop of low compliance with a single measure and an anticipated lower rate with multiple measures, we examined the overall compliance rate with established-clinical guidelines and its subsequent impact on survival.
METHODS
Data Source
We used the Surveillance, Epidemiology and End Results (SEER)-Medicare-linked database of the National Cancer Institute, which contains data on individuals aged 65 years and older, to identify bladder cancer patients who were diagnosed in 1992–2002. SEER data are summarized in the Patient Entitlement and Diagnosis Summary File (PEDSF) and contains data on patient demographics, tumor characteristics and follow-up information. The PEDSF was linked with 100% of Medicare claims from inpatient, outpatient, and national claims history files and was restricted to subjects who had Medicare Fee-for-Service coverage and for whom Medicare Parts A and B claims data were available for 12 months prior to and 24 months after diagnosis of bladder cancer.9
Study Population
The cohort consisted of patients at least 66 years of age with an incident diagnosis of high-grade (poorly, undifferentiated) urothelial (histology codes 8120, 8130) non-muscle-invasive (Ta, Tis, T1) bladder cancer (International Classification of Diseases, Ninth Revision (ICD-9) 188.0–188.9, 233.7) diagnosed between January 1, 1992 and December 31, 2002, for whom claims data were available through 2007. While beneficiaries are eligible for Medicare coverage at age 65, we limited our cohort to those 66 or older to allow at least one year of eligibility in Medicare before bladder cancer diagnosis to ascertain comorbidity data. We restricted our analysis to those with known grade, stage and histology and those who survived at least two years without undergoing definitive treatment (cystectomy, radiotherapy or systemic chemotherapy) during that timeframe.
Quality-of-Care Measures
While there are slight variations between NCCN, AUA and EAU, we amalgamated the three published guidelines to generate compliance measures a priori. For the first two years after diagnosis, at a minimum, patients should undergo (1) lower urinary tract surveillance with cystoscopy and urine cytology every three months, (2) one upper tract image, (3) instillation of perioperative mitomycin C after transurethral urethral resection of bladder tumor (TURBT), and (4) an induction course of BCG. Translated into claims data, we anticipated that patients should undergo at least eight cystoscopies, eight cytologies, one upper tract image, one instillation of perioperative mitomycin C, and six instillations of BCG after the diagnostic TURBT. We relaxed the definition to count as compliant the use of BCG anytime during the first two years after diagnosis. We also relaxed the definition of perioperative mitomycin C to include a claim for instillation of any chemotherapeutic within three days of TURBT. While recent guidelines highlight compelling evidence for the utility of repeat TURBT in T1 disease, more frequent upper-tract imaging and maintenance BCG (an induction course, plus three weekly instillations at 3, 6, 9, 12, 18, 24, 30 and 36 months after diagnosis), we used an a fortiori argument with less stringent requirements and an exhaustive set of ICD-9 and Healthcare Common Procedure Coding Systems (HCPCS) codes from the Medicare claims record, to posit that if noncompliance with our measures were found to be high, then non-adherence with more stringent criteria would be far higher.
Study Variables
From the PEDSF, we determined patient age (66–69, 70–74, 75–79, ≥ 80), gender, race/ethnicity (White, Black, Hispanic, Other), marital status (married, other), tumor grade (poorly differentiated, undifferentiated), T-stage (Ta, Tis, T1), and year of diagnosis (categorical:1992–1997,1998–2002). We imputed subject socioeconomic status by utilizing 2000 US Census data to derive quartiles of ZIP code-level median household income (<$35,000, $35,000–$45,000, $45,001–$55,000, >$55,000) and percent of residents ≥ 25 years of age with at least four years of college education (categorical: <15.0%, 15.0%–25.0%, 25.1%–35.0%, >35.0%).10 We used the Klabunde et al modification of the Charlson Comorbidity index to quantify severity of preexisting comorbidities (0, 1, 2, ≥ 3).11, 12 For each patient, we noted the provider and institution where the initial bladder cancer was diagnosed utilizing the Unique Physician Identifier Number (UPIN) and the corresponding institution (Medicare provider number). The Medicare provider number was linked with the American Medical Association Masterfile to derive institution type—medical school affiliation (academic) as well as NCI designation as a Comprehensive Cancer Center (cancer center). Institution type was stratified into academic cancer center, non-academic cancer center, academic non-cancer center, non-academic non-cancer center and unknown. We discovered that only four patients were diagnosed at an NCI designated cancer center without medical school affiliation. Hence, they were included with those diagnosed at an academic cancer center. Cumulative volumes for surgeon and hospital were calculated after adjusting for inclusion of new providers and four new SEER registries in 2000. Caseload for transurethral resections was stratified into low, medium and high for each surgeon (<4, 4–11, ≥ 12) and hospital (≤10, 11–25, >25). We generated a region variable (West, Midwest, South, Northeast) from the SEER registry.
Statistical Analysis
We report differences in means and proportions in compliance using two-sample t test and Chi-square analyses. Since receipt of services and attendant compliance with quality-of-care measures may be clustered on the provider, we generated a mixed-effects logistic model to account for both fixed (covariates) and physician-level random effects. A post-estimation function from the mixed-effects model was utilized to generate propensity scores and inverse probability of treatment weights to adjust for measured variable bias. This propensity score model adjusted for patient age, gender, race/ethnicity, marital status, ZIP code-level income and education, comorbidity, tumor grade and stage, hospital type, and surgeon and hospital volume. We then examined the relative survival difference using competing-risks regression analysis stratified by individual and comprehensive quality-of-care measures. For individual measures, five separate mixed-effects, post-estimation propensity score analyses and competing-risks regression analyses were performed for cystoscopy (≥4 vs <4), cytology (≥4 vs <4), upper tract imaging (≥1 vs 0), perioperative instillation of intravesical mitomycin C (≥1 vs 0) and instillation of BCG within 90 days of diagnosis (≥1 instillation within 90 days vs ≥ 1 instillation after 90 days or no instillation). Similarly, six separate mixed-effects, post-estimation propensity score analyses and competing-risks regression analyses were performed to measure the association between compliance with comprehensive measures and survival: 1) <4 cystoscopy, <4 cytology and no BCG (referent); 2) <4 cystoscopy, <4 cytology and 1st instillation of BCG after 90 days; 3) <4 cystoscopy, <4 cytology and 1st instillation of BCG within 90 days; 4) ≥ 4 cystoscopy, ≥ 4 cytology and no BCG; 5) ≥ 4 cystoscopy, ≥ 4 cytology and 1st instillation of BCG after 90 days; and 6) ≥ 4 cystoscopy, ≥ 4 cytology and 1st instillation of BCG within 90 days. The most intense strategy that was clinically meaningful and statistically feasible was receipt of ≥ 4 cystoscopies, ≥ 4 cytologies and an induction course of BCG with the 1st instillation within 90 days. While this combined measure is not entirely comprehensive, it serves as essential minimum requirement urologists should follow—without dispute of overutilization—and based on the low compliance rate, would overcome computational limitations from rare events. The instillation within 90 days helped distinguish providers who instilled BCG based on the initial diagnosis (“preventers”) from those who utilized it in response to recurrences (“reactors”).
We utilized a maximum likelihood, competing-risks regression model as described by Fine and Gray. 13 Competing-risks regression analysis was performed to characterize the risks of bladder cancer-specific mortality across intensity of surveillance and treatment. We defined failure as bladder cancer mortality and the competing event as non-bladder cancer mortality. Estimates are reported as sub-hazard ratios (HR) with corresponding 95% confidence intervals. Additionally, since patients treated by the same provider may have similar outcomes, we accounted for this potential clustering by utilizing the Huber/White/sandwich variance estimator to the competing-risks regression analysis to yield more conservative confidence intervals. While competing-risks regression analysis does not have a goodness-of-fit statistic, we utilized the Cox model as a proxy. We confirmed non-violation of the proportional hazards assumption using “log-log” plots and adequate model fit using Hosmer and Lemeshow analysis.14. We conducted all analyses with STATA software (College Station, Texas). All statistical tests were two-tailed, and the probability of a type I error was set at 0.05. The institutional review board at UCLA approved the study protocol.
RESULTS
We identified 4,790 subjects and the plurality was octogenarian, male, white, married and without any comorbid conditions. The majority was diagnosed in 1998–2002 with a poorly differentiated T1 tumor by a medium-volume surgeon at a non-academic non-cancer center (Table 1).
Table 1.
Cohort characteristics
| Variables | Number | % |
|---|---|---|
| Age-group
| ||
| 66–69 | 701 | 14.6% |
| 70–74 | 1246 | 26.0% |
| 75–79 | 1230 | 25.7% |
| ≥80 | 1613 | 33.7% |
|
| ||
| Gender
| ||
| Male | 3695 | 77.1% |
| Female | 1095 | 22.9% |
|
| ||
| Race
| ||
| White | 4377 | 91.4% |
| Black | 115 | 2.4% |
| Hispanic | 126 | 2.6% |
| Other | 172 | 3.6% |
|
| ||
| Marital
| ||
| Married | 3107 | 64.9% |
| Other | 1683 | 35.1% |
|
| ||
| Charlson Score
| ||
| 0 | 3354 | 70.0% |
| 1 | 984 | 20.5% |
| 2 | 315 | 6.6% |
| ≥3 | 137 | 2.9% |
|
| ||
| % Of Subjects in ZIP code ≥ 25 years of age with ≥ 4 years of college education
| ||
| <15% | 1009 | 21.1% |
| 15–25% | 1231 | 25.7% |
| 25% 35% | 992 | 20.7% |
| >35% | 1558 | 32.5% |
|
| ||
| Median ZIP code household income
| ||
| <$35,000 | 801 | 16.7% |
| $35,000–$45,000 | 1146 | 23.9% |
| $45,000–$55,000 | 1232 | 25.7% |
| >$55,000 | 1611 | 33.6% |
|
| ||
| Region
| ||
| West | 2448 | 51.1% |
| Midwest | 951 | 19.9% |
| South | 448 | 9.4% |
| Northeast | 943 | 19.7% |
|
| ||
| Year
| ||
| 1992–1997 | 1809 | 37.8% |
| 1998–2002 | 2981 | 62.3% |
|
| ||
| Surgeon Volume
| ||
| Low (<4) | 963 | 20.1% |
| Medium (4–11) | 2373 | 49.5% |
| High (≥12) | 1454 | 30.4% |
|
| ||
| Hospital Volume
| ||
| Low (≤10) | 1563 | 32.6% |
| Medium (11–25) | 1613 | 33.7% |
| High (>25) | 1614 | 33.7% |
|
| ||
| Institution Type
| ||
| Non-Academic Non-Cancer Center | 3362 | 70.2% |
| Academic Non-Cancer Center | 1042 | 21.8% |
| Academic Cancer Center | 96 | 2.0% |
| Unknown | 290 | 6.0% |
|
| ||
| Grade
| ||
| Poorly Differentiated | 3814 | 79.6% |
| Undifferentiated | 976 | 20.4% |
|
| ||
| Stage
| ||
| Ta | 1822 | 38.0% |
| Tis | 488 | 10.2% |
| T1 | 2480 | 51.8% |
On univariate analysis, while the mean utilization of cystoscopy, upper-tract imaging, and intravesical instillation of BCG significantly increased after the establishment of established-clinical guidelines, the proportion compliant with each measure did not significantly increase after 1997, with the exception of receipt of an induction course of BCG (13.0% to 19.7%, p <0.001) (Table 2).
Table 2.
Overall and compliant use of services stratified by year the NCCN guidelines for non-muscle-invasive bladder cancer were established.
| Mean (SD) | Number Compliant (%) | |||||
|---|---|---|---|---|---|---|
| Quality-of-Care Measures | 1992–1997 | 1998–2002 | p-value | 1992–1997 | 1998–2002 | p-value |
| Cystoscopy | 4.73 (2.07) | 4.86 (2.02) | 0.04 | 84 (4.6%) | 143 (4.8%) | 0.81 |
| Cytology | 2.26 (2.75) | 2.24 (2.65) | 0.80 | 98 (5.4%) | 143 (4.8%) | 0.34 |
| Upper tract imaging | 1.81 (1.50) | 1.97 (1.64) | <0.001 | 1582 (87.4%) | 2610 (87.5%) | 0.92 |
| Mitomycin C | 1.63 (3.50) | 1.08 (3.00) | <0.001 | 50 (2.8%) | 95 (3.2%) | 0.41 |
| BCG | 3.76 (5.16) | 4.89 (5.38) | <0.001 | 235 (13.0%) | 586 (19.7%) | <0.001 |
With regard to comprehensive care, out of the 4,790 subjects, only one case was compliant with all the quality-of-care measures (Table 3); hence the title of this report. Relaxing the definition to mandate ≥ 8 cystoscopies, ≥ 8 cytologies and an induction course of BCG, yields 21 cases fulfilling this requirement (0.4%). In fact, over 65% did not have receipt of at least ≥ 1 cystoscopy, ≥ 1 cytology and a single instillation of intravesical BCG.
Table 3.
Number compliant with progressive relaxation of guidelines
| Compliance criteria | N | % |
|---|---|---|
| ≥8 Cystoscopies and ≥ 8 cytologies and ≥ 6 instillations of BCG | 21 | 0.4% |
| ≥8 Cystoscopies and ≥ 8 cytologies and ≥ 1 instillation of BCG | 26 | 0.5% |
| ≥4 Cystoscopies and ≥ 4 cytologies and ≥ 6 instillations of BCG | 631 | 13.2% |
| ≥4 Cystoscopies and ≥ 4 cytologies and ≥ 1 instillation of BCG | 846 | 17.7% |
| ≥4 Cystoscopies and ≥ 1 cytology and ≥ 1 instillation of BCG | 1603 | 33.5% |
| ≥1 Cystoscopy and ≥ 1 cytology and ≥ 1 instillation of BCG | 1790 | 37.4% |
| ≥1 Cystoscopy and ≥ 1 instillation of BCG | 2558 | 53.4% |
| ≥1 Cystoscopy | 4602 | 96.1% |
Table 4 presents a mixed-effects model assessing receipt of ≥ 4 cystoscopies, ≥ 4 cytologies and an induction course of BCG with the first dose within 90 days of diagnosis. A higher odds of compliance with the aforementioned measures was found among subjects diagnosed with an undifferentiated grade (OR 1.74; 95% CI 1.30–2.32), Tis (OR 2.12; 95% CI 1.41–3.20) T1 (OR 1.56; 95% CI 1.19–2.04) tumor and diagnosed after 1997 (OR 1.55; 95% CI 1.17–2.05) and lower odds among octogenarians (OR 0.44; 95% CI 0.30–0.66). When year of diagnosis was represented as a continuous variable, we discovered a 10% increase in odds of compliance with time (OR 1.10; 95% CI 1.05–1.15). A post-estimation prediction of compliance over time was performed and at the current rate of progress, we would predict 25% compliance by 2024.
Table 4.
Mixed-effects model predicting compliance with ≥4 cystoscopy, ≥4 cytology and ≥6 instillations of BCG with the 1st instillation within 90 days of diagnosis.
| Variables | Odds Ratio | 95% CI |
|---|---|---|
| Age-group
| ||
| 66–69 (referent) | 1.00 | … |
| 70–74 | 1.34 | (0.93–1.94) |
| 75–79 | 0.92 | (0.63–1.34) |
| ≥80 | 0.44 | (0.30–0.66) |
|
| ||
| Gender
| ||
| Male (referent) | 1.00 | … |
| Female | 1.24 | (0.91–1.68) |
|
| ||
| Race
| ||
| White (referent) | 1.00 | … |
| Black | 0.57 | (0.23–1.45) |
| Hispanic | 0.73 | (0.33–1.61) |
| Other | 0.34 | (0.14–0.85) |
|
| ||
| Marital
| ||
| Married (referent) | 1.00 | … |
| Other | 0.82 | (0.62–1.08) |
|
| ||
| Charlson Score
| ||
| 0 (referent) | 1.00 | … |
| 1 | 1.16 | (0.87–1.55) |
| 2 | 1.04 | (0.63–1.70) |
| ≥3 | 0.64 | (0.29–1.41) |
|
| ||
| % Of Subjects in ZIP code ≥ 25 years of age with ≥ 4 years of college education
| ||
| <15% (referent) | 1.00 | … |
| 15–25% | 0.95 | (0.63–1.43) |
| 25%–35% | 0.97 | (0.60–1.56) |
| >35% | 1.08 | (0.66–1.79) |
|
| ||
| Median ZIP code household income
| ||
| <$35,000 (referent) | 1.00 | … |
| $35,000–$45,000 | 0.82 | (0.52–1.30) |
| $45,000–$55,000 | 1.03 | (0.64–1.66) |
| >$55,000 | 1.15 | (0.68–1.95) |
|
| ||
| Region
| ||
| West (referent) | 1.00 | … |
| Midwest | 1.28 | (0.81–2.02) |
| South | 0.87 | (0.52–1.47) |
| Northeast | 1.13 | (0.75–1.71) |
|
| ||
| Year
| ||
| 1992–1997 (referent) | 1.00 | … |
| 1998–2002 | 1.55 | (1.17–2.05) |
|
| ||
| Surgeon Volume
| ||
| Low (<4) (referent) | 1.00 | … |
| Medium (4–11) | 0.93 | (0.64–1.36) |
| High (≥12) | 0.78 | (0.49–1.24) |
|
| ||
| Hospital Volume
| ||
| Low (≤10) (referent) | 1.00 | … |
| Medium (11–25) | 0.94 | (0.65–1.35) |
| High (>25) | 1.06 | (0.70–1.61) |
|
| ||
| Institution Type
| ||
| Non-Academic Non-Cancer Center (referent) | 1.00 | … |
| Academic Non-Cancer Center | 1.39 | (0.95–2.03) |
| Academic Cancer Center | 2.04 | (0.87–4.79) |
| Unknown | 0.78 | (0.43–1.44) |
|
| ||
| Grade
| ||
| Poorly Differentiated (referent) | 1.00 | … |
| Undifferentiated | 1.74 | (1.30–2.32) |
|
| ||
| Stage
| ||
| Ta (referent) | 1.00 | … |
| Tis | 2.12 | (1.41–3.20) |
| T1 | 1.56 | (1.19–2.04) |
With regard to individual measures, a statistically significant lower hazard of mortality was discovered among subjects with receipt of ≥ 4 cystoscopies (HR 0.61; 95% CI 0.47–0.79), ≥ 4 cytologies (HR 0.55; 95% CI 0.39–0.80) and instillation of BCG within 90 days of diagnosis (HR 0.71; 95% CI 0.51–0.98). Upper tract imaging (HR 1.14; 95% CI 0.77–1.73) and instillation of perioperative mitomycin C (HR 0.74; 95% CI 0.34–1.65) were not associated with a survival advantage (Table 5). With regard to comprehensive measures, only those who had ≥ 4 cystoscopies, ≥ 4 cytologies and initiation of BCG within 90 days of diagnosis was associated with a statistically significant lower hazard for mortality than those who underwent <4 cystoscopies, <4 cytologies and no instillation of BCG (HR 0.41; 95% CI 0.18–0.93). There was a trend towards statistical significance among men who underwent ≥ 4 cystoscopies, ≥ 4 cytologies and initiation of BCG after 90 days of diagnosis (HR 0.43; 95% CI 0.16–1.16). Less frequent surveillance or delayed initiation (or no use) of BCG was not associated with a lower hazard for mortality (Table 6).
Table 5.
Propensity score-adjusted competing-risks regression analysis measuring the association between individual compliance measures and survival.
| Variable | HR | 95% CI |
|---|---|---|
| Cystoscopy
| ||
| <4 Cystoscopies (referent) | 1.00 | … |
| ≥4 Cystoscopies | 0.61 | (0.47–0.79) |
|
| ||
| Cytology
| ||
| <4 Cytologies (referent) | 1.00 | … |
| ≥4 Cytologies | 0.55 | (0.39–0.80) |
|
| ||
| Upper Tract Imaging
| ||
| No upper tract image (referent) | 1.00 | … |
| ≥1 Upper tract image | 1.15 | (0.77–1.73) |
|
| ||
| Perioperative Instillation of Mitomycin C
| ||
| No instillation of mitomycin C after any TURBT (referent) | 1.00 | … |
| At least one instillation of mitomycin C after any TURBT | 0.74 | (0.34–1.65) |
|
| ||
| Postoperative Instillation of BCG
| ||
| No instillations, or ≥ 1 instillation with 1st dose after 90 days (referent) | 1.00 | … |
| ≥1 instillation with 1st dose within 90 days | 0.71 | (0.51–0.98) |
Table 6.
Propensity score-adjusted competing-risks regression analysis measuring the association between comprehensive compliance measures and survival.
| Surveillance and Treatment Strategy | HR | 95% CI |
|---|---|---|
| <4 cystoscopy, <4 cytology and no BCG (referent) | 1.00 | *** |
| <4 cystoscopy, <4 cytology and 1st instillation of BCG after 90 days | 1.62 | 0.67–3.94 |
| <4 cystoscopy, <4 cytology and 1st instillation of BCG within 90 days | 0.88 | 0.36–2.11 |
| ≥4 cystoscopy, ≥ 4 cytology and no BCG | 1.11 | 0.34–3.60 |
| ≥4 cystoscopy, ≥ 4 cytology and 1st instillation of BCG after 90 days | 0.43 | 0.16–1.16 |
| ≥4 cystoscopy, ≥ 4 cytology and 1st instillation of BCG within 90 days | 0.41 | 0.18–0.93 |
DISCUSSION
We discovered a single case of comprehensive compliance out of 4,790 eligible patients. We did not anticipate that the considerable discordance between established guidelines and routine practice would yield a single case to report. Moreover, those who received at minimum ≥ 4 cystoscopies, ≥ 4 cytologies and initiation of BCG within 90 days of diagnosis had improved survival when compared with those who did not.
Factors that may explain the inadequacy of compliance with guideline-recommended care include: 1) the dearth of evidence-based medicine for surveillance strategies, 2) the surveillance and treatment strategies are too stringent, 3) patient preferences and treatment-related toxicities, 4) publication of practice guidelines after 1997, and 5) meta-analyses documenting the benefits of perioperative chemotherapy in 2004. With regard to the first three factors, a surveillance schedule that is too stringent with the attendant patient discomfort that is based on expert opinion will undoubtedly yield low compliance. In the absence of validation of surveillance strategies, we anticipate that compliance will remain low. Our study is especially timely, since it is among the first in bladder cancer to examine the relationship between process-of-care surveillance measures and an outcome that matters to cancer patients—survival. We hope our findings demonstrating a process-outcomes link will improve utilization of surveillance and treatment services. With regard to point 4, although the guidelines were first published in 1997, by limiting our cohort to those with high-risk disease, we expected preference-sensitive variation in this high-risk group to err on overutilization, not underutilization. Moreover, the benefits of BCG were well known prior to 1992. In fact, Southwest Oncology Group 8795 was closed prematurely in 1991 after accruing half of the projected number of subjects (349 in lieu of 720) as evidenced by the superiority of BCG over mitomycin C in recurrence-free survival (let alone placebo). With regard to the point 5—meta-analysis demonstrating an advantage of a single perioperative intravesical dose of chemotherapy was first reported in 2004—multiple studies published in the 1980s and 1990s demonstrated the benefits of perioperative intravesical chemotherapy. 7, 15–21 Some contend that with time, compliance with practice guidelines will be sufficient. Notwithstanding the statistically significant increase of compliance with at ≥ 4 cystoscopies, ≥ 4 cytologies and an induction course of BCG, we anticipate only 25% compliance by the year 2024. This anticipated 25% rate of compliance with recommended care (in 2024) is congruent with the deficit seen in patients with atrial fibrillation and hip fractures in 2003.3 Not coincidentally, atrial fibrillation and hip fractures have since been prioritized in the Initial National Priorities for Comparative Effectiveness Research.22
Our findings are commensurate with others depicting the underutilization of effective care in patients with bladder cancer, spanning extent of disease, from the underuse of definitive treatment and neoadjuvant chemotherapy for patients with at least stage II disease, 37% and 1%, respectively,23 to endoscopic surveillance and perioperative instillation of mitomycin in patients with non-muscle-invasive disease, 40% and 0.3%, respectively.4–6 Our findings may appear at odds with those of Strope et al., who have queried a similar cohort and discovered increased utilization of intravesical therapy and physician office visits.24 While office visits was not a process metric in our analysis, our findings echoed theirs with regard to intravesical therapy (post-operative immunotherapy). Moreover, while we also demonstrate increased mean utilization of most services over time we note that the compliance rate did not increase; hence the increased utilization may be as a result of non-coordinated care. With regard to treatment intensity and survival, our findings are at odds with 1) Hollingsworth et al., who demonstrated that while increasing treatment intensity was associated with major interventions (radical cystectomy, radiotherapy, and systemic chemotherapy), it was not significantly associated with median cancer-specific survival (albeit there was a trend with p=0.07); 25 and 2) the Hollenbeck et al. analysis that demonstrated that increasing treatment intensity was actually associated with a significantly higher hazard for cause-specific mortality (HR 1.43). 26 In fact, based on Hollingsworth et al. analysis, if provider treatment intensity was modified to reflect the 25th percentile, the authors anticipate Medicare savings of $18.7 million without detriment to oncologic outcomes. Our analysis differed from the Michigan’s group in that we excluded individuals who 1) underwent definitive treatment (cystectomy, radiotherapy or systemic chemotherapy) within the initial two-year period, or 2) did not survive during the initial two years, since a significant proportion of individuals with non-muscle-invasive bladder cancer on the initial transurethral resection may have been understaged. Therefore, this may in part explain why in their analyses, they discovered that the upper quartile of treatment intensity may not benefit from more services, 25, 26 since more intense surveillance and intravesical treatment strategies are unlikely to rescue a patient who has progressed to myoinvasion.27–30
While our sample size is large, our study has methodological limitations. As with any observational study, omitted-variable bias may impact adherence rates with established-clinical guidelines and the propensity score model. Patient preferences for surveillance and treatment strategies may have also confounded our findings of significant non-compliance. While the discomfort associated with endoscopic evaluation and intravesical therapy, may have contributed to noncompliance, we attempted to minimize confounding by limiting our cohort to those with high-grade tumors—a disease state that warrants higher intensity of surveillance and treatment—and to those who survived at least two years without undergoing definitive treatment (thus excluding those with competing causes of death and those who were understaged and quickly progressed). Last, while the surveillance strategies were statistically significant predictors of lower mortality, they may be a proxy for improved access to care. However, in the absence of an association between education and median income with the quality measures, this is not likely to be a significant cause.
Despite these limitations, our findings serve to alert patients and providers to the wide gap between guideline-recommended care and routine practice. So, how do we bridge the chasm between clinical practice guidelines and routine care? One approach relies on restructuring payment policies through performance-based incentive programs to explicitly promote improvements in quality of care. Pay-for-performance incentives and alternative payment models (e.g., PROMETHIUS® and accountable care organizations) have been integrated into most commercial and public health plans (e.g., the Centers for Medicare & Medicaid Services sponsors the Hospital Quality Improvement Demonstration Project).31 By linking incentives with physician adherence to clinically effective measures, facilitating positive patient outcomes and avoiding complications, we hope to bend the arc of the care trajectory for bladder cancer by implementing adequate cystoscopies and cytologies and initiation of BCG. These measures are feasible and effective in minimizing the morbidity and mortality associated with bladder cancer. Alternative strategies such as clinical reminders, electronic health records and development of scorecards not tied to financial incentives, may improve quality of care without the unintended consequences—gaming the system for financial incentives, stifling innovation and hindering development of creative solutions to improving quality and the potential to paradoxically penalize providers caring for underserved, non-compliant patients who are at high risk for progression.
Irrespective of the approach, in the absence of a broad quality-improvement initiative, the diffusion of clinically effective care will be slow, and many more unnecessary recurrences, procedures and deaths will be realized. This is an especially critical point since progress in preventing bladder cancer-related mortality lags behind other diseases—a 5% relative reduction in mortality rate over the last 15 years.32 This contribution is slight when one considers that bladder cancer is the fifth most common cancer contrasts starkly with prostate, breast, lung and colon cancers where we have seen greater reductions in mortality during the same period.33
Acknowledgments
FUNDING SOURCE: This work was supported by the American Cancer Society (117496-PF-09-147-01-CPHPS (Principal Investigator: KC)); Ruth L. Kirschstein National Research Service Award Extramural (1 F32 CA144461-01 (Principal Investigator: KC)); Jonsson Comprehensive Cancer Center Seed Grant (Principal Investigator: MSL)); and National Institute of Diabetesand Digestive and Kidney Diseases (N01-DK-1-2460 (Principal investigator:M.S.L.))
Role of the Sponsor: The sponsor was not involved with the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript.
Footnotes
Author Contributions: Dr. Chamie had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Financial Disclosures: No financial disclosures to report.
Disclaimer: This study used the linked SEER-Medicare database. The interpretation and reporting of these data are the sole responsibility of the authors.
Additional Contributions: We acknowledge the efforts of the Applied Research Program, National Cancer Institute; the Office of Research, Development, and Information, Centers for Medicare & Medicaid Services; Information Management Services Inc; and the SEER program tumor registries in the creation of the SEER-Medicare database.
References
- 1.Institute of Medicine. Crossing the Quality Chasm. National Academy Press; 2001. [Google Scholar]
- 2.Kohn LT, Corrigan J, Richardson WC, Donaldson MS. To err is human: Building a safer health system. Washington, DC: National Academy Press; 2000. [PubMed] [Google Scholar]
- 3.McGlynn EA, Asch SM, Adams J, et al. The quality of health care delivered to adults in the United States. N Engl J Med. 2003;348(26):2635–45. doi: 10.1056/NEJMsa022615. [DOI] [PubMed] [Google Scholar]
- 4.Schrag D, Hsieh LJ, Rabbani F, Bach PB, Herr H, Begg CB. Adherence to surveillance among patients with superficial bladder cancer. J Natl Cancer Inst. 2003;95(8):588–97. doi: 10.1093/jnci/95.8.588. [DOI] [PubMed] [Google Scholar]
- 5.Madeb R, Golijanin D, Noyes K, et al. Treatment of nonmuscle invading bladder cancer: do physicians in the United States practice evidence based medicine? The use and economic implications of intravesical chemotherapy after transurethral resection of bladder tumors. Cancer. 2009;115(12):2660–70. doi: 10.1002/cncr.24311. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Huang GJ, Hamilton AS, Lo M, Stein JP, Penson DF. Predictors of intravesical therapy for nonmuscle invasive bladder cancer: results from the surveillance, epidemiology and end results program 2003 patterns of care project. J Urol. 2008;180(2):520–4. doi: 10.1016/j.juro.2008.04.016. discussion 24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Tolley DA, Parmar MK, Grigor KM, et al. The effect of intravesical mitomycin C on recurrence of newly diagnosed superficial bladder cancer: a further report with 7 years of follow up. J Urol. 1996;155(4):1233–8. [PubMed] [Google Scholar]
- 8.Lamm DL, Blumenstein BA, Crawford ED, et al. A randomized trial of intravesical doxorubicin and immunotherapy with bacille Calmette-Guerin for transitional-cell carcinoma of the bladder. N Engl J Med. 1991;325(17):1205–9. doi: 10.1056/NEJM199110243251703. [DOI] [PubMed] [Google Scholar]
- 9.Warren JL, Klabunde CN, Schrag D, Bach PB, Riley GF. Overview of the SEER-Medicare data: content, research applications, and generalizability to the United States elderly population. Med Care. 2002;40(8 Suppl):IV-3–18. doi: 10.1097/01.MLR.0000020942.47004.03. [DOI] [PubMed] [Google Scholar]
- 10.Bach PB, Guadagnoli E, Schrag D, Schussler N, Warren JL. Patient demographic and socioeconomic characteristics in the SEER-Medicare database applications and limitations. Med Care. 2002;40(8 Suppl):IV-19–25. doi: 10.1097/00005650-200208001-00003. [DOI] [PubMed] [Google Scholar]
- 11.Klabunde CN, Potosky AL, Legler JM, Warren JL. Development of a comorbidity index using physician claims data. J Clin Epidemiol. 2000;53(12):1258–67. doi: 10.1016/s0895-4356(00)00256-0. [DOI] [PubMed] [Google Scholar]
- 12.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373–83. doi: 10.1016/0021-9681(87)90171-8. [DOI] [PubMed] [Google Scholar]
- 13.Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. Journal of the American Statistical Association. 1999;94(446):496–509. [Google Scholar]
- 14.Messing EM, Young TB, Hunt VB, et al. Hematuria home screening: repeat testing results. J Urol. 1995;154(1):57–61. doi: 10.1016/s0022-5347(01)67224-0. [DOI] [PubMed] [Google Scholar]
- 15.Devonec M, Bouvier R, Sarkissian J, Bendimerad O, Gelet A, Dubernard JM. Intravesical instillation of mitomycin C in the prophylactic treatment of recurring superficial transitional cell carcinoma of the bladder. British journal of urology. 1983;55(4):382–5. doi: 10.1111/j.1464-410x.1983.tb03327.x. [DOI] [PubMed] [Google Scholar]
- 16.Huland H, Otto U. Mitomycin instillation to prevent recurrence of superficial bladder carcinoma. Results of a controlled, prospective study in 58 patients. European urology. 1983;9(2):84–6. doi: 10.1159/000474054. [DOI] [PubMed] [Google Scholar]
- 17.Zincke H, Benson RC, Jr, Hilton JF, Taylor WF. Intravesical thiotepa and mitomycin C treatment immediately after transurethral resection and later for superficial (stages Ta and Tis) bladder cancer: a prospective, randomized, stratified study with crossover design. The Journal of urology. 1985;134(6):1110–4. doi: 10.1016/s0022-5347(17)47647-6. [DOI] [PubMed] [Google Scholar]
- 18.Tolley DA, Hargreave TB, Smith PH, et al. Effect of intravesical mitomycin C on recurrence of newly diagnosed superficial bladder cancer: interim report from the Medical Research Council Subgroup on Superficial Bladder Cancer (Urological Cancer Working Party) Br Med J (Clin Res Ed) 1988;296(6639):1759–61. doi: 10.1136/bmj.296.6639.1759. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Giesbers AA, Van Helsdingen PJ, Kramer AE. Recurrence of superficial bladder carcinoma after intravesical instillation of mitomycin-C. Comparison of exposure times. British journal of urology. 1989;63(2):176–9. doi: 10.1111/j.1464-410x.1989.tb05159.x. [DOI] [PubMed] [Google Scholar]
- 20.Kim HH, Lee C. Intravesical mitomycin C instillation as a prophylactic treatment of superficial bladder tumor. The Journal of urology. 1989;141(6):1337–9. doi: 10.1016/s0022-5347(17)41300-0. discussion 39–40. [DOI] [PubMed] [Google Scholar]
- 21.Maier U, Baumgartner G. Metaphylactic effect of mitomycin C with and without hyaluronidase after transurethral resection of bladder cancer: randomized trial. The Journal of urology. 1989;141(3):529–30. doi: 10.1016/s0022-5347(17)40881-0. [DOI] [PubMed] [Google Scholar]
- 22.IOM (Institute of Medicine) Initial National Priorities for Comparative Effectiveness Research. Washington, DC: National Academy Press; 2009. [Accessed: August 21, 2010]. Available: compliance qoc propensity cancer revised.doc. [Google Scholar]
- 23.Schrag D, Mitra N, Xu F, et al. Cystectomy for muscle-invasive bladder cancer: patterns and outcomes of care in the Medicare population. Urology. 2005;65(6):1118–25. doi: 10.1016/j.urology.2004.12.029. [DOI] [PubMed] [Google Scholar]
- 24.Strope SA, Ye Z, Hollingsworth JM, Hollenbeck BK. Patterns of care for early stage bladder cancer. Cancer. 2010;116(11):2604–11. doi: 10.1002/cncr.25007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.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):3587–94. doi: 10.1002/cncr.25221. [DOI] [PubMed] [Google Scholar]
- 26.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):571–80. doi: 10.1093/jnci/djp039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lambert EH, Pierorazio PM, Olsson CA, Benson MC, McKiernan JM, Poon S. The increasing use of intravesical therapies for stage T1 bladder cancer coincides with decreasing survival after cystectomy. BJU Int. 2007;100(1):33–6. doi: 10.1111/j.1464-410X.2007.06912.x. [DOI] [PubMed] [Google Scholar]
- 28.Herr HW, Sogani PC. Does early cystectomy improve the survival of patients with high risk superficial bladder tumors? J Urol. 2001;166(4):1296–9. [PubMed] [Google Scholar]
- 29.Lee CT, Dunn RL, Ingold C, Montie JE, Wood DP., Jr Early-stage bladder cancer surveillance does not improve survival if high-risk patients are permitted to progress to muscle invasion. Urology. 2007;69(6):1068–72. doi: 10.1016/j.urology.2007.02.064. [DOI] [PubMed] [Google Scholar]
- 30.Schrier BP, Hollander MP, van Rhijn BW, Kiemeney LA, Witjes JA. Prognosis of muscle-invasive bladder cancer: difference between primary and progressive tumours and implications for therapy. Eur Urol. 2004;45(3):292–6. doi: 10.1016/j.eururo.2003.10.006. [DOI] [PubMed] [Google Scholar]
- 31.Rosenthal MB, Landon BE, Normand SL, Frank RG, Epstein AM. Pay for performance in commercial HMOs. N Engl J Med. 2006;355(18):1895–902. doi: 10.1056/NEJMsa063682. [DOI] [PubMed] [Google Scholar]
- 32.American Cancer Society Surveillance Research. Cancer Statistics 1984–2009. [Google Scholar]
- 33.Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA Cancer J Clin. 2010;60(5):277–300. doi: 10.3322/caac.20073. [DOI] [PubMed] [Google Scholar]