Abstract
Background
To examine temporal nationwide utilization patterns and predictors for use of positron emission tomography–computed tomography (PET/CT) in comparison to magnetic resonance imaging (MRI) and computed tomography (CT) among patients diagnosed with bladder cancer.
Materials and Methods
A total of 36,855 patients aged 66 years or older diagnosed with clinical stage TI-IV, N0M0 bladder cancer from 2004 to 2011 were analyzed. We used multivariable logistic regression analyses to discern factors associated with receipt of imaging within 12 months from diagnosis. The Cochran-Armitage test for trend was used to determine changes in the proportion of patients receiving imaging after cancer diagnosis.
Results
Independent of clinical stage, there was marked increase in use of PET/CT throughout the study period (2011 v 2004: OR 17.55, 95% CI = 10.14–30.38, P<0.001). While use of CT imaging remained stable during the study period, there was significantly decreased utilization of MRI (OR 0.60, 95% CI 0.49–0.75, P<0.001) in 2011 vs. 2004. The mean incremental cost of PET/CT versus CT and MRI was $1,040 and $612 (in 2016 dollars), respectively. Extrapolating these findings to the bladder cancer patients in U.S. results in excess spending of $11.6 million for PET/CT imaging.
Conclusion
We identified rapid adoption of PET/CT imaging independent of clinical stage, resulting in excess national spending of $11.6 million for this imaging modality alone. Further value-based research discerning the clinical versus economic benefits of advanced imaging among bladder cancer patients are needed.
Keywords: Imaging, positron emission tomography–computed tomography, PET/CT, bladder cancer
INTRODUCTION
There were an estimated 76,960 new cases and 16,390 deaths from bladder cancer in the United States in 20161. Clinical staging for bladder cancer commonly includes transurethral resection of the bladder tumor and upper tract imaging2–4. Imaging techniques such as positron emission tomography–computed tomography (better known as PET-CT or PET/CT), magnetic resonance imaging (MRI), and computed tomography (CT) can improve preoperative staging and follow-up surveillance.
Prior studies have explored the utility of PET/CT imaging in primary bladder cancer with limited evidence suggesting clinical superiority5–7. Moreover, meta-analyses have suggested PET/CT is ‘good’ in detecting metastatic disease but could not recommend this as the preferred imaging modality over other imaging due to limited studies and lack of comparative effectiveness research7. Taking the above into account, current guidelines recommend CT and/or MRI as the preferred abdominal imaging modality in staging bladder cancer patients2–4.
The American Board of Internal Medicine’s Choosing Wisely campaign and American Society of Clinical Oncology’s Value of Cancer Care Task Force have collaborated to encourage sustainable high-quality and high-value based cancer care8. Widespread adoption of costlier advanced imaging modalities such as PET/CT with lack of well-documented superiority over other imaging techniques can have a significant impact on the national health care system. Indeed, the Institute of Medicine recently conveyed a workshop aimed at controlling use of expensive advanced cancer care and treatments in the absence of comparative effectiveness research documenting superiority over less costly alternatives9. Utilization patterns regarding advanced imaging in bladder cancer remain largely unknown. Given this void in understanding we used a large population-based cancer registry to analyze utilization trends and costs associated with advanced imaging in bladder cancer patients.
METHODS
Database
We used the Surveillance, Epidemiology, and End Results (SEER) Medicare-linked database. The SEER registry, supported by the National Cancer Institute, contains patients’ demographic and cancer diagnosis information for approximately 30% of the U.S. population from 18 geographic regions, including Alaska, Arizona, Cherokee Nation, Connecticut, Detroit, Georgia, San Francisco-Oakland, San Jose-Monterey, Greater California, Hawaii, Iowa, Kentucky, Los Angeles, Louisiana, New Jersey, New Mexico, Seattle-Puget Sound, and Utah. The Medicare program contains health care claims and payments for 97% of US citizens age 65 and over. The SEER registry data is linked with Medicare claims data using a unique encrypted patient identifier.
Patient-Selection Criteria
The study population consisted of patients aged 66 years and older with an incident of bladder cancer diagnosed with clinical stage I to IV, N0, M0 transitional cell or urothelial carcinoma (American Joint Committee on Cancer Modified third edition; ICD-O-3 codes 8120 and 8130) from January 1, 2004 through December 31, 2011. We excluded the following patients: those with a bladder cancer diagnosis from a death certificate or autopsy, those without pathological confirmation, those without continuous Part A and Part B insurance coverage within 12 months of their cancer diagnosis, those without continuous Part A and Part B insurance coverage until death, and finally those that had health maintenance organization enrollment during the same period (Supplemental Material 1).
Identification of Imaging Modalities
The primary outcome of this study was receipt of imaging which included CT, MRI and/or PET/CT, for the purpose of diagnosis and surveillance. This was determined using Medicare claims data within one year after the date of bladder cancer diagnosis. We identified the three imaging modalities using the following HCPCS codes: PET/CT (78815 and 78816), MRI (74181–74183, 74185, 76498, and 72195–72197), and CT (codes 72191–72194, 74150–74170, 74176–74178, and 76497).
Patient Characteristics
Patient demographic information included age at cancer diagnosis (66–69, 70–74, 75–79, and 80 years or older), year of cancer diagnosis, race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, and other), marital status (single, married, and unknown), US census region (West, Northeast, Midwest and South), and neighborhood median household income (categorized into quartiles). Tumor characteristics, as reported by SEER data, included clinical stage, histologic grade and presence of hydronephrosis. We used the modification by Klabunde et al of the Charlson Comorbidity Index to quantify severity of preexisting comorbidities10, 11. Treatment within one year after bladder cancer diagnosis was determined from the Medicare claims using both ICD (ninth revision) procedure codes and level II Healthcare Common Procedure Coding System (HCPCS): Current Procedural Terminology (CPT) codes (Supplemental Material 2).
Cost Analysis
We measured the Medicare payments to these three imaging modalities within one year after bladder cancer diagnosis, and all reported costs were adjusted and normalized to 2016 U.S. dollars using the medical care component of the Consumer Price Index12. We also extrapolated the national excess medical spending on advanced imaging for bladder cancer care. Using the estimated nationwide new cases of bladder cancer in 2016 from the SEER registry and its stage distribution, we multiplied the number of patients in each stage group by the proportion expected to receive the imaging. Finally, the number of patients who received advanced imaging was multiplied by the mean differences of costs between advanced imaging (PET/CT) and the two imaging modalities (CT and MRI).
Statistical Analysis
We compared use of imaging modalities in bladder cancer patients stratified by demographic and clinical variables with Pearson χ2 tests. We performed a Cochran-Armitage test for trend to assess changes in the proportion of patients receiving imaging after cancer diagnosis, and also the various types of imaging modalities utilized from 2004 to 2011. Multivariable logistic regression models were used to determine adjusted odds ratios for use of the three imaging modalities. We assessed goodness-of-fit using the Hosmer and Lemeshow test. All statistical analyses were conducted using the SAS (version 9.4; SAS Institute, Cary, NC, USA) software suite. The criterion for statistical significance was a P value less than 0.05. Our study was exempted for approval by The University of Texas MD Anderson Cancer Center and The University of Texas Medical Branch Institutional Review Boards.
RESULTS
Patient demographics and clinical characteristics according to imaging modality are presented in Table 1. In total, 24,240 (65.8%) patients received one of these three imaging modalities within 12 months after bladder cancer diagnosis: 1,291 (3.5%) PET/CT, 1,495 (4.1%) MRI, and 21,454 (58.2%) CT. We also observed a greater use of PET/CT among female patients, residents in West region, patients diagnosed with hydronephrosis or high grade tumor, and patients who underwent surgery, chemotherapy, or radiation therapy.
Table 1.
Patient demographic and clinical characteristics by the type of imaging
| PET/CT (%) | MRI (%) | CT (%) | |||||||
|---|---|---|---|---|---|---|---|---|---|
|
| |||||||||
| Characteristic | Yes | No | p | Yes | No | p | Yes | No | p |
| Year of Diagnosis | <0.001 | <0.001 | <0.001 | ||||||
| 2004 | 14 (1.1) | 5138 (14.4) | 250 (15.1) | 4902 (13.9) | 2932 (12.3) | 2220 (17.0) | |||
| 2005 | 41 (3.2) | 4807 (13.5) | 244 (14.8) | 4604 (13.1) | 2948 (12.4) | 1900 (14.6) | |||
| 2006 | 160 (12.4) | 4452 (12.5) | 263 (15.9) | 4349 (12.4) | 2985 (12.5) | 1627 (12.5) | |||
| 2007 | 159 (12.3) | 4481 (12.6) | 220 (13.3) | 4420 (12.6) | 3040 (12.8) | 1600 (12.3) | |||
| 2008 | 226 (17.5) | 4358 (12.3) | 201 (12.2) | 4383 (12.5) | 3092 (13.0) | 1492 (11.4) | |||
| 2009 | 246 (19.1) | 4087 (11.5) | 139 (8.4) | 4194 (11.9) | 2996 (12.6) | 1337 (10.3) | |||
| 2010 | 259 (20.1) | 4202 (11.8) | 200 (12.1) | 4261 (12.1) | 3029 (12.7) | 1432 (11.0) | |||
| 2011 | 186 (14.4) | 4039 (11.4) | 134 (8.1) | 4091 (11.6) | 2801 (11.8) | 1424 (10.9) | |||
| Age Group | 0.002 | 0.001 | <0.001 | ||||||
| 66–69 | 219 (17.0) | 5464 (15.4) | 256 (15.5) | 5427 (15.4) | 3837 (16.1) | 1846 (14.2) | |||
| 70–74 | 288 (22.3) | 7986 (22.5) | 432 (26.2) | 7842 (22.3) | 5387 (22.6) | 2887 (22.2) | |||
| 75–79 | 356 (27.6) | 8668 (24.4) | 404 (24.5) | 8620 (24.5) | 5822 (24.4) | 3202 (24.6) | |||
| 80+ | 428 (33.2) | 13446 (37.8) | 559 (33.9) | 13315 (37.8) | 8777 (36.8) | 5097 (39.1) | |||
| Sex | <0.001 | <0.001 | <0.001 | ||||||
| Male | 865 (67.0) | 26424 (74.3) | 1088 (65.9) | 26201 (74.4) | 17066 (71.6) | 10223 (78.4) | |||
| Female | 426 (33.0) | 9140 (25.7) | 563 (34.1) | 9003 (25.6) | 6757 (28.4) | 2809 (21.6) | |||
| Race | 0.474 | <0.001 | 0.033 | ||||||
| Non-Hispanic White | 1151 (89.2) | 31576 (88.8) | 1405 (85.1) | 31322 (89.0) | 21119 (88.6) | 11608 (89.1) | |||
| Non-Hispanic Black | 53 (4.1) | 1440 (4.0) | 110 (6.7) | 1383 (3.9) | 984 (4.1) | 509 (3.9) | |||
| Hispanic | 28 (2.2) | 1026 (2.9) | 67 (4.1) | 987 (2.8) | 720 (3.0) | 334 (2.6) | |||
| Other | 59 (4.6) | 1522 (4.3) | 69 (4.2) | 1512 (4.3) | 1000 (4.2) | 581 (4.5) | |||
| Marital Status | 0.533 | 0.469 | 0.003 | ||||||
| Single | 182 (14.1) | 4638 (13.0) | 226 (13.7) | 4594 (13.0) | 3133 (13.2) | 1687 (12.9) | |||
| Married | 763 (59.1) | 21195 (59.6) | 960 (58.1) | 20998 (59.6) | 14048 (59.0) | 7910 (60.7) | |||
| Unknown | 346 (26.8) | 9731 (27.4) | 465 (28.2) | 9612 (27.3) | 6642 (27.9) | 3435 (26.4) | |||
| Census Region | <0.001 | <0.001 | <0.001 | ||||||
| West | 598 (46.3) | 13933 (39.2) | 622 (37.7) | 13909 (39.5) | 9171 (38.5) | 5360 (41.1) | |||
| Northeast | 257 (19.9) | 8984 (25.3) | 543 (32.9) | 8698 (24.7) | 5978 (25.1) | 3263 (25.0) | |||
| Midwest | 97 (7.5) | 4207 (11.8) | 163 (9.9) | 4141 (11.8) | 2725 (11.4) | 1579 (12.1) | |||
| South | 339 (26.3) | 8440 (23.7) | 323 (19.6) | 8456 (24.0) | 5949 (25.0) | 2830 (21.7) | |||
| Urban/Rural | 0.501 | 0.050 | 0.955 | ||||||
| Urban | 1261 (97.7) | 34834 (97.9) | 1628 (98.6) | 34467 (97.9) | 23331 (97.9) | 12764 (97.9) | |||
| Rural | 30 (2.3) | 730 (2.1) | 23 (1.4) | 737 (2.1) | 492 (2.1) | 268 (2.1) | |||
| Median Household Income, $ | 0.256 | <0.001 | 0.736 | ||||||
| <= 23,364 | 346 (26.8) | 9335 (26.2) | 359 (21.7) | 9322 (26.5) | 6276 (26.3) | 3405 (26.1) | |||
| 23,365 – 31,906 | 336 (26.0) | 8729 (24.5) | 403 (24.4) | 8662 (24.6) | 5834 (24.5) | 3231 (24.8) | |||
| 31,907 – 41,719 | 288 (22.3) | 8762 (24.6) | 453 (27.4) | 8597 (24.4) | 5825 (24.5) | 3225 (24.7) | |||
| >= 41,720 | 321 (24.9) | 8738 (24.6) | 436 (26.4) | 8623 (24.5) | 5888 (24.7) | 3171 (24.3) | |||
| Stage | <0.001 | <0.001 | <0.001 | ||||||
| I | 413 (32.0) | 29155 (82.0) | 912 (55.2) | 28656 (81.4) | 17326 (72.7) | 12242 (93.9) | |||
| II | 424 (32.8) | 3768 (10.6) | 349 (21.1) | 3843 (10.9) | 3648 (15.3) | 544 (4.2) | |||
| III | 125 (9.7) | 1103 (3.1) | 136 (8.2) | 1092 (3.1) | 1125 (4.7) | 103 (0.8) | |||
| IV | 329 (25.5) | 1538 (4.3) | 254 (15.4) | 1613 (4.6) | 1724 (7.2) | 143 (1.1) | |||
| Hydronephrosis | <0.001 | <0.001 | <0.001 | ||||||
| No | 1141 (88.4) | 34017 (95.7) | 1488 (90.1) | 33670 (95.6) | 22487 (94.4) | 12671 (97.2) | |||
| Yes | 150 (11.6) | 1547 (4.3) | 163 (9.9) | 1534 (4.4) | 1336 (5.6) | 361 (2.8) | |||
| Grade | <0.001 | <0.001 | <0.001 | ||||||
| Low | 167 (12.9) | 14341 (40.3) | 421 (25.5) | 14087 (40.0) | 8027 (33.7) | 6481 (49.7) | |||
| High | 1028 (79.6) | 15851 (44.6) | 1061 (64.3) | 15818 (44.9) | 12631 (53.0) | 4248 (32.6) | |||
| Unknown | 96 (7.4) | 5372 (15.1) | 169 (10.2) | 5299 (15.1) | 3165 (13.3) | 2303 (17.7) | |||
| Comorbidity Score | 0.293 | 0.006 | 0.000 | ||||||
| 0 | 671 (52.0) | 18505 (52.0) | 824 (49.9) | 18352 (52.1) | 12328 (51.7) | 6848 (52.5) | |||
| 1 | 336 (26.0) | 9150 (25.7) | 428 (25.9) | 9058 (25.7) | 6289 (26.4) | 3197 (24.5) | |||
| 2 | 132 (10.2) | 4134 (11.6) | 181 (11.0) | 4085 (11.6) | 2744 (11.5) | 1522 (11.7) | |||
| 3+ | 152 (11.8) | 3775 (10.6) | 218 (13.2) | 3709 (10.5) | 2462 (10.3) | 1465 (11.2) | |||
| Radical Cystectomy | <0.001 | <0.001 | <0.001 | ||||||
| No | 1009 (78.2) | 33485 (94.2) | 1372 (83.1) | 33122 (94.1) | 21527 (90.4) | 12967 (99.5) | |||
| Yes | 282 (21.8) | 2079 (5.8) | 279 (16.9) | 2082 (5.9) | 2296 (9.6) | 65 (0.5) | |||
| Chemotherapy | <0.001 | <0.001 | <0.001 | ||||||
| No | 712 (55.2) | 33194 (93.3) | 1278 (77.4) | 32628 (92.7) | 21120 (88.7) | 12786 (98.1) | |||
| Yes | 579 (44.8) | 2370 (6.7) | 373 (22.6) | 2576 (7.3) | 2703 (11.3) | 246 (1.9) | |||
| Radiation therapy | <0.001 | <0.001 | <0.001 | ||||||
| No | 1043 (80.8) | 34398 (96.7) | 1456 (88.2) | 33985 (96.5) | 22453 (94.2) | 12988 (99.7) | |||
| Yes | 248 (19.2) | 1166 (3.3) | 195 (11.8) | 1219 (3.5) | 1370 (5.8) | 44 (0.3) | |||
When assessing trends in receipt of imaging, the use of PET/CT significantly increased over the time period of study, from < 0.5% in 2004 to 4.4% in 2011 (P trend < 0.001). At the same time, the percentage of patients who received an MRI significantly decreased over the study period (P trend < 0.001) (Figure 1). We further assessed trends in receipt of imaging according to clinical stage (Figure 2). PET/CT increased from 2001 to 2011 across all clinical stages: I, 0.1% to 1.2%; II, 1.0% to 13.6%; III, 0.0% to 11.9%; and IV, 1.4% to 27.0% (All P trend < 0.001), respectively (Figure 2). In contrast, utilization decreased for MRI (P trend = 0.08) for clinical stage I, II and IV patients, while the use of CT imaging techniques remained essentially unchanged.
Figure 1.
Percent of patients receiving any imaging, CT, MRI, and PET/CT, after bladder cancer diagnosis from 2004–2011 (Any imaging: Ptrend, <0.001; CT: Ptrend, P <0.001; MRI: Ptrend, P <0.001; PET/CT: Ptrend, <0.001).
Figure 2.
Percent of patients receiving PET/CT, MRI, or CT imaging after a bladder cancer diagnosis. A). PET/CT (Stage I: Cochrane Armitage test of trend, P < 0.001; Stage II: Cochrane Armitage test of trend, P = < 0.001; Stage III: Ptrend, < 0.001; Stage IV: Ptrend, P < 0.001) B). MRI (Stage I: Ptrend, P = 0.001; Stage II: Ptrend, P <0.001; Stage III: Ptrend, P = 0.083; Stage IV: Ptrend, P = 0.031). C). CT (Stage I: Ptrend, P <0.001; Stage II: Ptrend, P = 0.379; Stage III: Ptrend, P = 0.658; Stage IV: Ptrend, P = 0.480).
We used multivariable logistic regression models to evaluate factors associated with utilization of each the three imaging modalities for patients diagnosed with bladder cancer. We noted a marked increase in use of PET/CT during the study period (2011 v 2004: OR 17.55, 95% CI = 10.14–30.38, P<0.001) (Table 2). Predictors associated with an increased likelihood of receiving PET/CT included female vs. male gender (OR 1.28, 95% CI 1.12–1.46, p=0.001), White vs. non-White (non-Hispanic Black: OR 0.74, 95% CI 0.55–0.99, p= 0.047; Hispanic: OR 0.54, 95% CI 0.36–0.81, p =0.003), married v single marital status (OR 1.21, 95% CI 1.01–1.45, p=0.034), being diagnosed with high vs. low grade tumors (OR 1.89, 95% CI 1.56–2.28, p<0.001), clinical stage higher than I (Stage I: OR 6.17, 95% CI 5.25–7.24, p<0.001; Stage II: OR 5.86, 95% CI 4.67–7.35, p<0.001, and Stage III: OR 11.20, 95% CI 9.39–13.35, p<0.001), and the presence of hydronephrosis (yes vs. no OR 1.40, 95% CI 1.15–1.70, p<0.001).
Table 2.
Multivariable model discerning predictors for receipt of positron emission tomography–computed tomography, magnetic resonance imaging and computed tomography
| PET/CT | MRI | CT | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
| ||||||||||||
| Characteristic | OR | 95% CI | P-value | OR | 95% CI | P-value | OR | 95% CI | P-value | |||
| Year of Diagnosis | ||||||||||||
| 2004 | 1.00 | 1.00 | 1.00 | |||||||||
| 2005 | 3.00 | 1.63 | 5.52 | <.001 | 1.02 | 0.84 | 1.22 | 0.877 | 1.16 | 1.06 | 1.26 | <.001 |
| 2006 | 13.30 | 7.67 | 23.07 | <.001 | 1.15 | 0.96 | 1.38 | 0.136 | 1.39 | 1.28 | 1.51 | <.001 |
| 2007 | 13.25 | 7.64 | 22.98 | <.001 | 0.95 | 0.79 | 1.15 | 0.591 | 1.46 | 1.34 | 1.59 | <.001 |
| 2008 | 19.89 | 11.54 | 34.29 | <.001 | 0.86 | 0.71 | 1.04 | 0.119 | 1.60 | 1.47 | 1.75 | <.001 |
| 2009 | 23.94 | 13.90 | 41.22 | <.001 | 0.62 | 0.50 | 0.77 | <.001 | 1.73 | 1.59 | 1.89 | <.001 |
| 2010 | 25.21 | 14.65 | 43.39 | <.001 | 0.89 | 0.74 | 1.09 | 0.261 | 1.69 | 1.55 | 1.85 | <.001 |
| 2011 | 17.55 | 10.14 | 30.38 | <.001 | 0.60 | 0.49 | 0.75 | <.001 | 1.55 | 1.42 | 1.70 | <.001 |
| Age Group | ||||||||||||
| 66–69 | 1.00 | 1.00 | 1.00 | |||||||||
| 70–74 | 0.91 | 0.76 | 1.10 | 0.348 | 1.16 | 0.98 | 1.36 | 0.08 | 0.89 | 0.82 | 0.96 | 0.002 |
| 75–79 | 1.04 | 0.86 | 1.25 | 0.694 | 0.92 | 0.78 | 1.08 | 0.31 | 0.84 | 0.78 | 0.90 | <.001 |
| 80+ | 0.66 | 0.55 | 0.79 | <.001 | 0.76 | 0.65 | 0.89 | <.001 | 0.72 | 0.67 | 0.78 | <.001 |
| Sex | ||||||||||||
| Male | 1.00 | 1.00 | 1.00 | |||||||||
| Female | 1.28 | 1.12 | 1.46 | <.001 | 1.35 | 1.21 | 1.50 | <.001 | 1.40 | 1.32 | 1.48 | <.001 |
| Race | ||||||||||||
| Non-Hispanic White | 1.00 | 1.00 | 1.00 | |||||||||
| Non-Hispanic Black | 0.74 | 0.55 | 0.99 | 0.047 | 1.50 | 1.21 | 1.86 | <.001 | 0.86 | 0.76 | 0.97 | 0.012 |
| Hispanic | 0.54 | 0.36 | 0.81 | 0.003 | 1.49 | 1.15 | 1.95 | 0.003 | 1.19 | 1.03 | 1.37 | 0.015 |
| Other | 0.86 | 0.64 | 1.14 | 0.287 | 0.94 | 0.73 | 1.21 | 0.615 | 0.95 | 0.85 | 1.06 | 0.327 |
| Marital Status | ||||||||||||
| Single | 1.00 | - | 1.00 | |||||||||
| Married | 1.21 | 1.01 | 1.45 | 0.034 | - | 1.09 | 1.01 | 1.17 | 0.021 | |||
| Unknown | 1.06 | 0.87 | 1.30 | 0.558 | - | 1.06 | 0.98 | 1.15 | 0.128 | |||
| Census Region | ||||||||||||
| West | 1.00 | 1.00 | 1.00 | |||||||||
| Northeast | 0.71 | 0.61 | 0.83 | <.001 | 1.41 | 1.24 | 1.60 | <.001 | 1.11 | 1.04 | 1.17 | 0.001 |
| Midwest | 0.50 | 0.40 | 0.63 | <.001 | 0.89 | 0.74 | 1.07 | 0.215 | 1.05 | 0.98 | 1.14 | 0.179 |
| South | 0.94 | 0.81 | 1.08 | 0.37 | 0.89 | 0.77 | 1.03 | 0.128 | 1.30 | 1.22 | 1.39 | <.001 |
| Median Household Income, $ | ||||||||||||
| <= 23,364 | - | 1.00 | 1.00 | |||||||||
| 23,365 – 31,906 | - | 1.18 | 1.01 | 1.37 | 0.038 | 1.04 | 0.98 | 1.11 | 0.207 | |||
| 31,907 – 41,719 | - | 1.25 | 1.07 | 1.47 | 0.005 | 1.07 | 1.00 | 1.14 | 0.065 | |||
| >= 41,720 | - | 1.21 | 1.03 | 1.43 | 0.019 | 1.12 | 1.04 | 1.20 | 0.002 | |||
| Grade | ||||||||||||
| Low | 1.00 | 1.00 | 1.00 | |||||||||
| High | 1.89 | 1.56 | 2.28 | <.001 | 1.41 | 1.23 | 1.61 | <.001 | 1.61 | 1.52 | 1.69 | <.001 |
| Unknown | 0.94 | 0.72 | 1.22 | 0.641 | 1.07 | 0.88 | 1.29 | 0.506 | 0.97 | 0.91 | 1.04 | 0.44 |
| Clinical Stage | ||||||||||||
| I | 1.00 | 1.00 | 1.00 | |||||||||
| II | 6.17 | 5.25 | 7.24 | <.001 | 2.41 | 2.09 | 2.78 | <.001 | 3.74 | 3.39 | 4.13 | <.001 |
| III | 5.86 | 4.67 | 7.35 | <.001 | 3.14 | 2.57 | 3.85 | <.001 | 5.91 | 4.81 | 7.26 | <.001 |
| IV | 11.20 | 9.39 | 13.35 | <.001 | 3.88 | 3.29 | 4.56 | <.001 | 6.46 | 5.42 | 7.70 | <.001 |
| Hydronephrosis | ||||||||||||
| No | 1.00 | 1.00 | 1.00 | |||||||||
| Yes | 1.40 | 1.15 | 1.70 | <.001 | 1.57 | 1.31 | 1.87 | <.001 | 1.42 | 1.25 | 1.61 | <.001 |
| Comorbidity Score | ||||||||||||
| 0 | - | 1.00 | 1.00 | |||||||||
| 1 | - | 1.08 | 0.96 | 1.22 | 0.209 | 1.12 | 1.07 | 1.19 | <.001 | |||
| 2 | - | 1.04 | 0.88 | 1.23 | 0.64 | 1.03 | 0.96 | 1.11 | 0.456 | |||
| 3+ | - | 1.38 | 1.18 | 1.62 | <.001 | 0.93 | 0.86 | 1.00 | 0.059 | |||
In our multivariable analysis, there was significantly decreased utilization of MRI (OR 0.60, 95% CI 0.49–0.75, P<0.001) in 2011 vs. 2004, respectively. Predictors associated with increased likelihood of receiving MRI included female vs male gender (OR 1.35, 95% CI 1.21–1.50, p=0<.001), non-White (non-Hispanic black: OR 1.50, 95% CI 1.21–1.86, p<.001; Hispanic: OR 1.49, 95% CI 1.15–1.95, p=0.003), Northwest vs. West region (OR 1.41, 95% CI 1.24–1.60, p<0.001), being diagnosed with high vs. low grade tumors (OR 1.41, 95% CI 1.23–1.61, p<0.001), clinical stage II, III, and IV vs. I (Stage II: OR 6.17, 95% CI 5.25–7.24, p<0.001; Stage III: OR 5.86, 95% CI 4.67–7.35, p<0.001, and Stage IV: OR 11.20, 95% CI 9.39–13.35, p<0.001), hydronephrosis yes vs. no (OR 1.57, 95% CI 1.31–1.87, p<0.001), and comorbidity score 3 or more (OR 1.38, 95% CI 1.18–1.62, P<0.001).
Predictors associated with increased likelihood of receiving CT were younger age (70–74, OR 0.89, 95% CI 0.82–0.96, P=0.002; 75–79, OR 0.84, 95% CI 0.78–0.90, P<0.001; 80+, OR 0.72, 95% CI 0.67–0.78, P<0.001; versus patients age 66–69 years old), female vs male gender (OR 1.40, 95% CI 1.32–1.48, p<0.001), Hispanic vs. non-Hispanic White and non-Hispanic Black race/ethnicity (Hispanic, OR 1.19, 95% CI 1.03–1.037, p=0.015), married vs. single marital status (OR 1.09, 95% CI 1.01–1.17, p=0.021), highest median household income quartile (OR 1.12, 95% CI 1.04–1.20, p=0.002), being diagnosed with high vs. low grade tumors (OR 1.61, 95% CI 1.52–1.69, p<0.001), clinical stage II, III, and IV vs. I (Stage II: OR 3.74, 95% CI 3.39–4.13, p<0.001; Stage III: OR 5.91, 95% CI 4.81–7.26, p<0.001, and Stage IV: OR 6.46, 95% CI 5.42–7.70, p<0.001), and hydronephrosis yes vs. no (OR 1.42, 95% CI 1.25–1.61, p<0.001),
The mean incremental cost of PET/CT vs. CT and MRI was $1,040 and $612 (in 2016 dollars), respectively. The estimated national excess in health care costs for PET/CT imaging compared to less costlier CT and MRI techniques was $11.6 million (Supplemental Material 3).
DISCUSSION
Guidelines recommend use of CT and MRI as the principle imaging in the staging and management of bladder cancer2–4. While there is uncertainty regarding use in metastatic patients, there are no evidence to suggest its clinical value in the non-metastatic setting7. In the present study, we assessed trends in use of PET/CT, MRI, and CT among bladder cancer patients. Our study revealed a significant shift in the type of imaging modality performed during the study period. Specifically, we observed marked 16-fold increased use of PET/CT regardless of clinical stage. This rapid adoption of PET/CT translated into excess national spending of approximately $11 million.
PET/CT has become the standard of care for other malignancies due to improved sensitivity and specificity over CT or MRI.13–15 However, whether or not PET/CT improves the accuracy in bladder cancer staging is a matter of debate. When PET/CT was used in detecting the primary tumor, the reported sensitivity ranged from 54% to 86.7% with a specificity from 25% to 100%5, 18–20. Conversely, the sensitivity dropped for PET/CT in preoperative staging, ranging from 46% to 60%16, 19, 21. There have been only a few studies that have reported PET/CT to be more sensitive than CT for preoperative staging19, 22. Although PET/CT may offer the ability to detect additional lesions and more frequently upstage patients, the final clinical impact on actual treatment changes may be relatively low and not adequately quantified in these studies7, 23–25. Due to the limited number of comparative effectiveness studies available, as well as the relatively small number of patients included in these studies, current guidelines do not recommend the use of PET/CT imaging for bladder cancer staging2–4. One major challenge of using PET in bladder cancer patients is that the fluorodeoxyglucose (FDG) is excreted into the urinary system, some have refuted this benefit as an acceptable initial imaging modality for staging bladder cancer patients.16, 17 The recent studies have found that use of other tracers than FDG, such as C11-methionine and C11-choline, may improve the visualization of PET imaging in bladder cancer.26, 27 However, further research is needed to support the application of these new tracers into clinical practice.
In the present study, predictors for receipt of advanced imaging largely included clinical and pathologic determinants. Patients with high grade tumors, >T2 or greater clinical stage, and those with increased comorbidities were the most likely to receive advanced imaging. Our finding that patients who underwent chemotherapy were more likely to receive advanced imaging may reflect a higher index of suspicion for more advanced disease in this population. Moreover, we also observed geographic variation in receipt of advanced imaging. Patients residing in the western US regions were significantly more likely to receive advanced imaging. This variability in practice patterns may be a reflection of a larger number of PET/CT scanners in the western region compared to other SEER regions with inherent improved access to this imaging and/or market influences28. A prior study have associated increased use of advanced imaging with physician self-referral arrangements as a major driver of health care costs29. Interestingly, that study was derived from a large private insurer in California.
The cost difference between PET/CT and other imaging were substantial at a national level. Our study estimated that the excess spending on advanced imaging will impose about $12 million in cost-expenditures. This may be an underestimate as our analysis used Medicare reimbursement rates for imaging which are historically lower than private insurance payers. The substantial economic costs of adopting advanced technology from diagnosis to treatment is an important issue of current health care reform30, 31. The cost of PET was only 1.5% of the Medicare spending on cancer care, however the contribution of PET to cancer care spending will continue to increase due to the higher growth rate of imaging cost than the cost of cancer care.32 Our data highlights the need for health policy measures to limit utilization and the associated costs of advanced imaging which are not guideline-recommended over less costly imaging modalities9.
Our findings must be interpreted in the context of the study design. First, the SEER-Medicare database provides a national representative sample of elderly patients which findings may not be generalizable to younger populations. Second, Medicare claims data do not collect information on glomerular filtration rate and urine creatinine clearance. Both of these are determinants are often used as surrogates regarding appropriateness of using CT or MRI in patients with poor renal function. Third, claims data do not contain information regarding patient and physician preference which are important determinants in the decision-making process on imaging selection33, 34. Fourth, we did not require a corresponding diagnosis code for bladder cancer when an imaging was identified since we found in the sensitivity analysis that only 20% bladder cancer patients received imaging billed with this diagnosis code which would have largely underestimated utilization of imaging. Finally, with limited clinical information available from claims data to determine the intent of advanced imaging, our study merely focused on the national trends in advanced imaging adoption. We made no attempt to discern trends in the appropriateness of the various imaging modalities used in bladder cancer. The appropriateness of various imaging modalities remains to be determined given recent guideline panel recommendations on appropriateness of use of the varying imaging modalities in bladder cancer4, 35, 36.
CONCLUSIONS
We identified rapid adoption of PET/CT imaging without comparative effectiveness research documenting clinical superiority over less costlier guideline-recommended imaging. These findings have important implications regarding health policy decision-making and the need for improved value-based bladder cancer care.
Supplementary Material
CLINICAL PRACTICE POINTS.
Current European and United States Guidelines on bladder cancer recommend CT and/or MRI as the preferred abdominal imaging modality over the PET/CT in preoperative staging and follow-up surveillance.
The American Board of Internal Medicine’s Choosing Wisely campaign and American Society of Clinical Oncology’s Value of Cancer Care Task Force have collaborated to encourage sustainable high-quality and high-value based cancer care.
Data from this large population-based cancer registry analysis of utilization patterns and economic impact regarding advanced imaging in bladder cancer showed a sharp increase in the use of advanced PET/CT imaging during the study period, accompanied with an excess national spending of approximately $11 million.
These findings suggested that value-based bladder cancer care is needed in community practice. Researches on comparative effectiveness of PET/CT imaging over less costlier imaging techniques are lacking to support contemporary trend of PET/CT imaging.
Acknowledgments
This study was conducted with the support of the Institute for Translational Sciences at the University of Texas Medical Branch, supported in part by a Clinical and Translational Science Award Mentored Career Development (KL2) Award (KL2TR001441) from the National Center for Advancing Translational Sciences, National Institutes of Health (NIH), Comparative Effectiveness Research on Cancer in Texas (CERCIT) (RP140020) and the National Cancer Institute (NCI) (K05 CA134923) (SBW). This study was funded in part by the NIH Bladder SPORE (5P50CA091846-03) (AMK). This work was supported in part by the Duncan Family Institute and a fellowship from The University of Texas MD Anderson Cancer Center's Halliburton Employees Foundation (Huo). We thank Dr. Gary Deyter from the Department of Health Services Research at The University of Texas MD Anderson Cancer Center for reviewing and editing the manuscript. This study used the linked Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database. The interpretation and reporting of these data are the sole responsibility of the authors. The authors acknowledge the efforts of the Applied Research Program, NCI; the Office of Research, Development and Information, CMS; Information Management Services (IMS), Inc.; and the SEER program tumor registries in the creation of the SEER database.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflict of Interest: None
References
- 1.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA: a cancer journal for clinicians. 2016;66:7–30. doi: 10.3322/caac.21332. [DOI] [PubMed] [Google Scholar]
- 2.Clark PE, Agarwal N, Biagioli MC, et al. Bladder cancer. J Natl Compr Canc Netw. 2013;11:446–475. doi: 10.6004/jnccn.2013.0059. [DOI] [PubMed] [Google Scholar]
- 3.Babjuk M, Bohle A, Burger M, et al. EAU Guidelines on Non-Muscle-invasive Urothelial Carcinoma of the Bladder: Update 2016. European urology. 2017;71:447–461. doi: 10.1016/j.eururo.2016.05.041. [DOI] [PubMed] [Google Scholar]
- 4.Witjes JA, Compérat E, Cowan NC, et al. EAU Guidelines on Muscle-invasive and Metastatic Bladder Cancer: Summary of the 2013 Guidelines. European urology. 2014;65:778–792. doi: 10.1016/j.eururo.2013.11.046. [DOI] [PubMed] [Google Scholar]
- 5.Anjos DA, Etchebehere ECSC, Ramos CD, Santos AO, Albertotti C, Camargo EE. 18F-FDG PET/CT Delayed Images After Diuretic for Restaging Invasive Bladder Cancer. Journal of Nuclear Medicine. 2007;48:764–770. doi: 10.2967/jnumed.106.036350. [DOI] [PubMed] [Google Scholar]
- 6.Goodfellow H, Viney Z, Hughes P, et al. Role of fluorodeoxyglucose positron emission tomography (FDG PET)-computed tomography (CT) in the staging of bladder cancer. BJU international. 2014;114:389–395. doi: 10.1111/bju.12608. [DOI] [PubMed] [Google Scholar]
- 7.Lu YY, Chen JH, Liang JA, et al. Clinical value of FDG PET or PET/CT in urinary bladder cancer: a systemic review and meta-analysis. European journal of radiology. 2012;81:2411–2416. doi: 10.1016/j.ejrad.2011.07.018. [DOI] [PubMed] [Google Scholar]
- 8.Rocque GB, Williams CP, Jackson BE, et al. Choosing Wisely: Opportunities for Improving Value in Cancer Care Delivery? Journal of oncology practice. 2017;13:e11–e21. doi: 10.1200/JOP.2016.015396. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.National Cancer Policy F, Board on Health Care S, Institute of M, National Academies of Sciences E, Medicine. Appropriate Use of Advanced Technologies for Radiation Therapy and Surgery in Oncology: Workshop Summary. Washington (DC): National Academies Press (US); 2016. The National Academies Collection: Reports funded by National Institutes of Health. Copyright 2016 by the National Academy of Sciences. All rights reserved. [PubMed] [Google Scholar]
- 10.Klabunde CN, Potosky AL, Legler JM, Warren JL. Development of a comorbidity index using physician claims data. Journal of Clinical Epidemiology. 2000;53:1258–1267. doi: 10.1016/s0895-4356(00)00256-0. [DOI] [PubMed] [Google Scholar]
- 11.Charlson ME, Sax FL, MacKenzie CR, Fields SD, Braham RL, Douglas RG. Assessing illness severity: does clinical judgment work? J Chronic Dis. 1986;39:439–452. doi: 10.1016/0021-9681(86)90111-6. [DOI] [PubMed] [Google Scholar]
- 12.Mauchley DC, Mitchell JD. Current estimate of costs of lung cancer screening in the United States. Thoracic Surgery Clinics. 2015;25:205–215. doi: 10.1016/j.thorsurg.2014.12.005. [DOI] [PubMed] [Google Scholar]
- 13.Agrawal A, Rangarajan V. Appropriateness criteria of FDG PET/CT in oncology. The Indian Journal of Radiology & Imaging. 2015;25:88–101. doi: 10.4103/0971-3026.155823. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.International Atomic Energy Agency. A Guide to Clinical PET in Oncology: Improving Clinical Management of Cancer Patients. 2017 IAEA-TECDOC-1605 ed2008. [Google Scholar]
- 15.Delbeke D, Coleman RE, Guiberteau MJ, et al. Procedure guideline for tumor imaging with 18F-FDG PET/CT 1.0. Journal of nuclear Medicine. 2006;47:885–895. [PubMed] [Google Scholar]
- 16.Swinnen G, Maes A, Pottel H, et al. FDG-PET/CT for the Preoperative Lymph Node Staging of Invasive Bladder Cancer. European urology. 2010;57:641–647. doi: 10.1016/j.eururo.2009.05.014. [DOI] [PubMed] [Google Scholar]
- 17.Schöder H, Larson SM. Seminars in nuclear medicine. Vol. 34. Elsevier; 2004. Positron emission tomography for prostate, bladder, and renal cancer; pp. 274–292. [DOI] [PubMed] [Google Scholar]
- 18.Harkirat S, Anand S, Jacob M. Forced diuresis and dual-phase <sup>18</sup>F-fluorodeoxyglucose-PET/CT scan for restaging of urinary bladder cancers. Indian Journal of Radiology and Imaging. 2010;20:13–19. doi: 10.4103/0971-3026.59746. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lodde M, Lacombe L, Friede J, Morin F, Saourine A, Fradet Y. Evaluation of fluorodeoxyglucose positron-emission tomography with computed tomography for staging of urothelial carcinoma. BJU international. 2010;106:658–663. doi: 10.1111/j.1464-410X.2010.09212.x. [DOI] [PubMed] [Google Scholar]
- 20.Lu Y-Y, Chen J-H, Liang J-A, et al. Clinical value of FDG PET or PET/CT in urinary bladder cancer: A systemic review and meta-analysis. European journal of radiology. 2012;81:2411–2416. doi: 10.1016/j.ejrad.2011.07.018. [DOI] [PubMed] [Google Scholar]
- 21.Drieskens O, Oyen R, Van Poppel H, Vankan Y, Flamen P, Mortelmans L. FDG-PET for preoperative staging of bladder cancer. European Journal of Nuclear Medicine and Molecular Imaging. 2005;32:1412–1417. doi: 10.1007/s00259-005-1886-9. [DOI] [PubMed] [Google Scholar]
- 22.Apolo AB, Riches J, Schöder H, et al. Clinical Value of Fluorine-18 2-Fluoro-2-Deoxy-D-Glucose Positron Emission Tomography/Computed Tomography in Bladder Cancer. Journal of Clinical Oncology. 2010;28:3973–3978. doi: 10.1200/JCO.2010.28.7052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Mertens LS, Fioole-Bruining A, Vegt E, Vogel WV, van Rhijn BW, Horenblas S. Impact of 18F-fluorodeoxyglucose (FDG)-positron-emission tomography/computed tomography (PET/CT) on management of patients with carcinoma invading bladder muscle. BJU international. 2013;112:729–734. doi: 10.1111/bju.12109. [DOI] [PubMed] [Google Scholar]
- 24.Goodfellow H, Viney Z, Hughes P, et al. Role of fluorodeoxyglucose positron emission tomography (FDG PET)-computed tomography (CT) in the staging of bladder cancer. BJU International. 2014;114:389–395. doi: 10.1111/bju.12608. [DOI] [PubMed] [Google Scholar]
- 25.Kibel AS, Dehdashti F, Katz MD, et al. Prospective Study of [(18)F]Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography for Staging of Muscle-Invasive Bladder Carcinoma. Journal of Clinical Oncology. 2009;27:4314–4320. doi: 10.1200/JCO.2008.20.6722. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.de Jong IJ, Pruim J, Elsinga PH, Jongen MM, Mensink HJ, Vaalburg W. Visualisation of bladder cancer using 11C-choline PET: first clinical experience. European Journal of Nuclear Medicine and Molecular Imaging. 2002;29:1283–1288. doi: 10.1007/s00259-002-0881-7. [DOI] [PubMed] [Google Scholar]
- 27.Schöder H, Larson SM. Positron emission tomography for prostate, bladder, and renal cancer. Seminars in Nuclear Medicine. 2004;34:274–292. doi: 10.1053/j.semnuclmed.2004.06.004. [DOI] [PubMed] [Google Scholar]
- 28.Buck AK, Herrmann K, Stargardt T, Dechow T, Krause BJ, Schreyögg J. Economic Evaluation of PET and PET/CT in Oncology: Evidence and Methodologic Approaches. Journal of Nuclear Medicine Technology. 2010;38:6–17. doi: 10.2967/jnmt.108.059584. [DOI] [PubMed] [Google Scholar]
- 29.Mitchell JM. Utilization trends for advanced imaging procedures: evidence from individuals with private insurance coverage in California. Medical care. 2008;46:460–466. doi: 10.1097/MLR.0b013e31815dc5ae. [DOI] [PubMed] [Google Scholar]
- 30.Dinan MA, Curtis LH, Hammill BG, et al. Changes in the use and costs of diagnostic imaging among medicare beneficiaries with cancer, 1999–2006. JAMA. 2010;303:1625–1631. doi: 10.1001/jama.2010.460. [DOI] [PubMed] [Google Scholar]
- 31.Nguyen PL, Gu X, Lipsitz SR, et al. Cost Implications of the Rapid Adoption of Newer Technologies for Treating Prostate Cancer. Journal of Clinical Oncology. 2011;29:1517–1524. doi: 10.1200/JCO.2010.31.1217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Yang Y, Czernin J. Contribution of Imaging to Cancer Care Costs. Journal of Nuclear Medicine. 2011;52:86S–92S. doi: 10.2967/jnumed.110.085621. [DOI] [PubMed] [Google Scholar]
- 33.Barry MJ, Edgman-Levitan S. Shared Decision Making — The Pinnacle of Patient-Centered Care. New England Journal of Medicine. 2012;366:780–781. doi: 10.1056/NEJMp1109283. [DOI] [PubMed] [Google Scholar]
- 34.Gogineni K, Shuman KL, Chinn D, Gabler NB, Emanuel EJ. Patient demands and requests for cancer tests and treatments. JAMA Oncology. 2015;1:33–39. doi: 10.1001/jamaoncol.2014.197. [DOI] [PubMed] [Google Scholar]
- 35.NCCN Bladder Cancer Practice Guidelines (Version 2.2016) Williston Park, N.Y: The National Comprehensive Cancer Network; 2016. [Google Scholar]
- 36.Milowsky MI, Rumble RB, Booth CM, et al. Guideline on Muscle-Invasive and Metastatic Bladder Cancer (European Association of Urology guideline): American Society of Clinical Oncology Clinical Practice Guideline Endorsement. Journal of Clinical Oncology. 2016 doi: 10.1200/JCO.2015.65.9797. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.