Medical Care Costs for Recurrent versus De Novo Stage IV Cancer by Age at Diagnosis.
Individuals diagnosed with advanced cancer comprise a sizeable proportion of cancer-related deaths and are responsible for substantial morbidity and spending in the United States (USA). There are two ways to develop advanced, noncurable cancer--be diagnosed with de novo metastatic stage IV cancer (i.e., the first occurrence of cancer) or develop recurrent metastatic cancer after previously having been treated for early-stage (i.e., stages I-III) disease. Given that most population-based datasets (including SEER and state-based tumor registries) cannot identify recurrent cancers (Warren and Yabroff 2015), little is known about whether there are cost differences between patients with de novo and recurrent metastatic cancers. In addition, many cost-effectiveness analyses associated with treatment trials for patients with metastatic cancer fail to differentiate between de novo versus recurrent metastatic cancer postdiagnosis cost trajectories (Takeda et al. 2007; Lange et al. 2014).There are several reasons why the medical care and costs of patients with different types of advanced cancer might differ significantly. In contrast to patients with newly diagnosed cancer, those with a metastatic recurrence have been previously treated with surgery, systemic therapy, and/or radiotherapy and may therefore be ineligible for standard, first-line treatment (Jassem et al. 2009). They may have "late effects" and/or comorbid conditions resulting from their prior disease and treatment (Carver et al. 2007); the treatments and costs for their initial, nonmetastatic cancer diagnosis may extend for many months (Eisen et al. 2015); and surveillance-related utilization and costs after the completion of curative treatment for their initial cancer diagnosis may be significant (Khatcheressian et al. 2006; Backhus et al. 2014; Steele et al. 2015; Hahn et al. 2016). In addition, the experience of having recurrent disease after an attempt of curative therapy may lead to a shift in patients' treatment goals and values (Weeks et al. 2012).
A number of studies have described the variation in utilization and costs associated with the initial treatment for cancer patients diagnosed at late stage versus early stage (Riley et al. 1995; Taplin et al. 1995; Warren et al. 2008; Yabroff et al. 2008, 2011). Several studies have described the high cost of care for breast cancer recurrence (Lamerato et al. 2006; Kamon et al. 2007; Lidgren et al. 2007; Engel-Nitz et al. 2015), and others have generated cost estimates for metastatic disease in patients 65 years and older (Yabroff et al. 2009; Bradley et al. 2017). However, in many of the databases used to generate these estimates, cancer recurrence cannot be identified reliably. This raises concern that cost estimates associated with the surveillance or survivorship phases of care, as well as those associated with the advanced disease phase of care, could be biased. For example, health care costs incurred because of recurrence after definitive treatment for early-stage disease could be attributed to the survivorship phase rather than the advanced disease phase of care. The extent of this potential bias is not well understood (Brown et al. 2002; Mariotto et al. 2011; Yabroff et al. 2011; Guy et al. 2013). In addition, few population-based datasets are available for the estimation of cancer care-related utilization and costs for patients less than 65 years of age. To address the lack of data related to the direct medical costs of cancer recurrence, we estimated average total annual and monthly medical care costs for patients diagnosed with recurrent breast, colorectal, or lung cancer, relative to patients diagnosed with de novo disease. We estimate costs not only among Medicare aged-eligible patients (age [greater than or equal to] 65 years) but also among those under age 65 at diagnosis, a group that we hypothesize will receive more aggressive (and more expensive) treatment for both recurrent and de novo metastatic cancers.
METHODS
Data Sources
Data for this study were obtained from the Cancer Research Network's (CRN) Virtual Data Warehouse (VDW) at the Colorado, Northwest, and Washington (formerly Group Health Cooperative) Kaiser Permanente regions. The CRN (http://cm.cancer.gov/) is a consortium of large health care systems affiliated with the Health Care Systems Research Network and the National Cancer Institute (Chubak et al. 2016). The VDW contains administrative, electronic health record (EHR) and other clinical data that have been extracted, processed, and maintained at each site (Ross et al. 2014). Cancer diagnoses were obtained from the Virtual Tumor Registry (VTR) component of the VDW, which adheres to standards of the National Cancer Institute's Surveillance, Epidemiology and End Results (SEER) program and the North American Association of Central Cancer Registries (NAACCR) (North American Association of Central Cancer Registries 2011). VTR data are derived from certified tumor registrars' manual review of patients' medical records and include demographics, diagnosis date, and tumor characteristics. VTR and VDW data are linked through a common, unique patient identifier (Ritzwoller et al. 2012, 2013, 2014; Hassett et al. 2014). Diagnosis and procedure-coded events are based on International Classification of Diseases, 9th and 10th revisions, Clinical Modification (ICD9 and ICD10), Healthcare Common Procedure Coding System (HCPCS), and the fourth edition of the Common Procedure Terminology codes (CPT-4). Oral and infused chemotherapy and immunotherapy were captured in the VDW infusion, pharmacy, and procedure files using methods previously described (Ritzwoller et al. 2013, 2014; Carroll et al. 2017). The VDW includes geocoded measures of socioeconomic status (e.g., median family income, and education) where patients' residential addresses are mapped to census block data using geocoding software. Deaths are derived from the tumor registries, membership data, state mortality files, and social security administration data. IRB approval was obtained from all three sites. All analyses were performed using SAS v 9.4 (SAS Inc., Cary, NC).
Study Sample
We identified patients diagnosed with de novo breast, colorectal, or lung cancers between January 1, 2000, and December 31, 2011, and patients who developed recurrent disease between January 1, 2000, and December 31, 2012, after having been initially diagnosed with stages I-III breast, colorectal, or lung cancer between January 1, 2000, and December 31, 2011. All de novo patients were identified from the VTR. Recurrent disease patients from KPCO and KPNW were identified through their respective VTRs, and those from KPWA were identified using our previously reported recurrence detection and timing algorithm (Hassett et al. 2014, 2017; Ritzwoller et al. 2017). Consistent with our previous analyses, eligibility for inclusion in the recurrence group was limited to patients who: (1) had no cancer prior to their incident breast, lung, or colorectal diagnosis; (2) completed definitive local-regional therapy for their incident cancer; and (3) survived and were followed for at least 30 days after definitive therapy. Censoring occurred if the VTR identified a second primary cancer. The date of the de novo or recurrent cancer diagnosis was considered the index date for this analysis. Patients were followed from 1 year prior to index date until death, disenrollment from the health plan, or one-year post-index date, whichever came first. Patients who disenrolled or were alive at the end of the study were censored as of those dates.
Utilization Measures
We examined inpatient, chemotherapy, and hospice use in the 12 months post-index date for de novo and recurrent cases across all three cancer cohorts. For recurrent cases, we also examined the proportion of cases who received chemotherapy and radiotherapy at the initial (i.e., incident stages I-III) diagnosis and in the twelve months after the index date (i.e., the recurrence date).
Costs
Medical care costs were estimated using the standard relative resource cost algorithm (SRRCA) (O'Keeffe-Rosetti et al. 2013). SRRCA applies uniform cost coefficients to standardized claims or EHR-derived utilization data to ensure that observed differences are not a result of differing pricing methods and billing rules (bundling of services). These estimates do not, however, account for variation in patient cost sharing or co-pays. Costs were reported in 2012 U.S. dollars. Average total monthly and annual costs were estimated from 1 year before the index date until disenrollment, death, or 1 year after the index date. For the monthly cost analysis, the calendar month associated with the index date was considered the index month.
Statistical Analysis
Patient characteristics were reported as means, medians, and standard deviations for interval-level variables (pre-index, index month, post-index) and percentages for categorical variables. Wilcoxon's rank-sum tests (for intervallevel variables) and chi-square tests of association (for categorical variables) were used to assess differences between the de novo and recurrent groups for all three cancer cohorts.
Differences in total annual and monthly costs between de novo versus recurrent cases were estimated separately for each cancer site across all age groups (and separately for patients aged <65 vs. [greater than or equal to]65) using generalized linear repeated-measures regression models with a gamma distribution and log link (Manning, Basu, and Mullahy 2005; Basu and Manning 2010; Guy et al. 2013). We also ran the generalized linear repeated-measures regression models where monthly costs were grouped into three-time periods: 12 months pre-index, index month, and 12 months post-index. The following covariates were included in the final adjusted model: age at index date, race/ethnicity, the Quan adaptation of the Charlson comorbidity index modified to exclude cancer diagnoses (Quan et al. 2005), census-based proxy for median family income, health plan (KPCO, KPNW, KPWA), index year, presence of an inpatient event within 12 months post-index date, receipt of chemotherapy within 12 months post-index date, and whether the individual died in the last month for which data were available. We address the potential bias in cost estimates due to differential follow-up among individuals that either disenrolled or died within 12 months following the index date using inverse proportional weights (IPWs), which account for key demographic and clinical factors related to differential follow-up. There were no significant differences in follow-up data among recurrent and de novo breast cancer patients, but there were differences in follow-up time among patients with lung and colorectal cancers.
RESULTS
Demographic and Clinical Characteristics
We identified 352 breast, 1,072 colorectal, and 4,041 lung cancer cases of de novo disease, along with 765 breast, 542 colorectal, and 340 lung cancer cases of recurrent disease after an early-stage initial diagnosis (see Figure SI: Consort Flow Diagram). As described in Table 1, relative to the recurrent cases, de novo colorectal and lung cancer patients were younger, and de novo lung cancer cases were more likely to be male. No clinically significant differences were found in race/ethnicity for de novo versus recurrent cases. Recurrent colorectal and lung cancer patients differed (higher) significandy (p < .054) in based proxies of income distribution. The year of the advanced cancer diagnosis, defined as the index year for this analysis, was relatively evenly distributed for de novo patients across all three cancer types. However, due to the study inclusion criteria (incident stages I-III diagnosis from 2000 to 2011 with a recurrent diagnosis through December 31, 2012), the recurrent cases were more heavily distributed during the latter years of the study. The comorbidity burden was statistically significandy higher only for the de novo lung cancer cases (relative to recurrent cases). A recurrence within 12 months of definitive therapy occurred in 18 percent, 31 percent, and 45 percent of the breast, colorectal, and lung cancer cases, respectively. In addition, relative to the recurrent cases, a larger proportion of the de novo colorectal (50 percent vs. 39 percent) and lung (71 percent vs. 52 percent) cancer cases died before the end of the study. The proportion of de novo cases who had at least one inpatient event after the index date was statistically significandy higher than the recurrent cases with 53 percent versus 42 percent for breast (p < .001), 76 percent versus 57 percent for colorectal (p < .001), and 60 percent versus 50 percent for lung cancers (p < .001), respectively. No statistically significant differences existed in the proportion of de novo versus recurrent cases who received chemotherapy or radiotherapy post-index date, across all three cancer sites. For recurrent cases, receipt of chemotherapy both at the initial (stages I-III) diagnosis and during the 12 months after the index date (recurrence) was higher for breast (51 percent) than colorectal (45 percent) or lung (22 percent). The use of radiotherapy in both time periods was also higher for breast (25 percent) than for colorectal (4 percent) or lung (12 percent). No statistically significant difference in hospice use was found for de novo versus recurrent colorectal or lung cancer cases; however, the hospice use for de novo breast cases was significantly lower than recurrent cases (6 percent vs. 12 percent, p-value .003).
Costs
Adjusted total monthly medical care costs for 12 months before and after the index date for patients with de novo versus recurrent breast, lung, or colorectal cancers appear in Figures 1-3, respectively. Separate estimates are provided for individuals over and under 65. For all cancer types, monthly costs during the pre-index period were significantly higher (p < .0001) for recurrent cancer patients than for de novo patients, through approximately 1 month prior to the index month (Figures 1-3 and Table 2). During the index month, monthly costs were considerably higher for de novo than for recurrent cancer patients for both colorectal cancer age groups (p < .0001) and lung cancer patients aged 65 and older (p < .0001). Higher costs were observed for all de novo breast cancer patients and for lung cancer patients aged less than 65, relative to recurrent patients, although these results did not reach statistical significance. Monthly costs between post-index month 1 and month 12 were statistically significantly higher among de novo than among recurrent patients for all cancers and all age groups. Table S1 describes the final parameter estimates.
Differences in adjusted average total monthly costs were consistent with the adjusted average total annual costs--for both the 12-month period before and the 12-month period after the index diagnosis (Table 2). Recurrent cancer patients had statistically significantly (p < .005) higher annual costs relative to de novo patients in the pre-index period, for all age groups and all three cancer sites. For example, women less than 65 had an average total annual cost of $3,310 [95% CI $2,563-$4,27] in the year prior to a de novo breast cancer diagnosis, compared to $30,531 [95% CI $26,149-$35,649] in the year prior to a recurrence diagnosis, for an estimated annual cost difference of $27,221 (de novo vs. recurrence, p < .0001). For CRC and lung cancer patients less than 65, the average annual cost differential (de novo vs. recurrence) was estimated at $35,703 (p < .0001) and $35,465 (p < .0001), respectively. For patients aged 65 and older, the differentials were not as large, but remained statistically significant, with the exception of lung cancer patients. For the 12-month period starting with and extending beyond the index month, average annual costs were greater for de novo versus recurrent cancer patients for all cancer types and age groups (Figures 1-3); all differences were statistically significant (p < .006) for all groups, except for lung cancer patients aged 65 years and older (-$2,065, p = .33).
DISCUSSION
Our study illuminates the striking differences in overall medical care costs for de novo versus recurrent cancers. Most notable was the difference in annual total costs between the two groups in the months prior to the index advanced cancer diagnosis. For patients less than 65 years of age, estimated medical care costs during the year prior to a diagnosis of a recurrence were approximately five- to ninefold higher than during the year prior to a diagnosis of de novo metastatic disease. Much of this difference may be explained by residual treatment for the initial cancer diagnosis and ongoing surveillance costs (Loggers et al. 2014; Chang and Gould 2017; Merkow et al. 2017), but further analyses are warranted to determine the relative importance of these factors as contributors to this difference. Regardless, the high cost of medical care for patients previously treated for early-stage disease illustrates the need for accessible and assessable guideline-based surveillance care. This is particularly important for patients less than 65 years of age, who are the most vulnerable with respect to access to health insurance (Thorpe and Howard 2003; Davidoff et al. 2015).
Medical care costs during the index month of the advanced cancer diagnosis spiked. This jump was especially notable for de novo patients, relative to those with recurrent cancer. This spike in costs, and the corresponding spike in utilization (Figures 1-3), is consistent with our previous work, which noted a one- to threefold increase in monthly utilization of total medical care services during the months leading up to an initial cancer diagnosis (Hombrook et al. 2013). De novo patients were younger and in the lung and colorectal cohorts less likely to survive for 12 months after the index diagnosis (Table 1) than recurrent patients, but it is not clear that these factors were responsible for the de novo/recurrent cost differential. We identified higher inpatient use for de novo cases, suggesting that the cost differential could be driven by hospitalizations and tests/procedures performed to evaluate the new cancer diagnosis. Patients with recurrent cancer previously had diagnostic studies for their initial cancer diagnosis, so it would not be surprising to expect relatively lower "evaluative" costs at the time of their cancer recurrence diagnosis. While cost differences (de novo vs. recurrent) during the index month were lower for all cancer types, the differences were not statistically significant for breast cancer patients or for lung cancer patients less than 65.
Cost differentials persisted throughout the 12-month post-index period. These differences were reflected in the significantly lower average total annual costs for patients with recurrent versus de novo disease (except for breast cancer patients). Some have suggested that the use of aggressive and expensive initial treatments (Sagar, Lin, and Castel 2017) and/or end-of-life care could be driving costs for de novo cases. However, we found no significant differences between groups with respect to the proportion who received chemotherapy or radiotherapy in the 12 months following the index date. Across all cohorts, a minority of recurrent cases received chemotherapy in both the initial and recurrent settings (Yardley et al. 2014; Rier et al. 2017). Perhaps, providers are less likely to recommend aggressive, expensive therapy for recurrent versus de novo cases, because patients with recurrence may be relatively chemoresistant, whereas de novo cases are inherently chemotherapy naive and may be more likely to benefit from therapy (Malmgren et al. 2018). The average total annual cost estimates further illuminate the cost differences between groups (de novo vs. recurrent) by period (pre-/post-index). Also, we noted higher costs for patients in the younger age group --a finding previously noted by Taplin (Taplin et al. 1995) and Mariotto (Mariotto et al. 2011) in their estimates of initial care costs for patients with distant disease.
We are not aware of previous studies evaluating the medical care costs for patients with recurrent lung or colorectal cancer. For patients with recurrent breast cancer, our estimates of average total annual costs differed from those reported previously. Specifically, for 765 recurrent breast cancer cases (mean age 62.5), we estimated costs before and after recurrence to be $27,038 and $57,549 (2012$U.S., data not shown), respectively. Studying 62 patients with recurrent breast cancer who had at least 12 months of follow-up, Lamearto (Lamerato et al. 2006) reported billed charges of $12,344 before and $79,253 after recurrence (2003$U.S.). However, in addition to a much smaller cohort, this analysis included younger patients, used charges rather than costs, and did not include pharmacy charges. Karnon (Karnon et al. 2007) estimated costs in the year after a metastatic recurrence (2004[pounds sterling]) for patients who died/experienced a subsequent event (n = 56) and for patients who remained alive with no further event (n = 36), to be [pounds sterling]10,003 and [pounds sterling]9,409, respectively. Stokes (Stokes et al. 2008) used SEER/Medicare claims to estimate the 10-year costs of care for patients with recurrent breast cancer, finding that costs ranged from $11,000 to $19,183 (2004$U.S.). Even after adjusting for inflation and exchange rates, most of these other estimates are approximately 50 percent lower than our findings.
Our study begins to fill large gaps in the literature on the variation in medical care costs for patients with recurrent versus de novo metastatic disease, within and across three cancer types that comprise a sizeable proportion of the overall economic burden of cancer in the United States (Yabroff et al. 2011). The strengths of our study include the use of tumor registry data to identify cancer cases and recurrences--supplemented with a validated cancer recurrence detection and timing algorithm (Hassett et al. 2014; Ritzwoller et al. 2017). Second, we captured complete utilization and costs of medical care, both before and after the most important sentinel event in the cancer trajectory. We did not use health plan-specific charges, costs, or payments, given that they can vary for the same service, for the same diagnosis and/or treatment, and even across regions within the Kaiser Permanente program. Instead, we implemented a standardized costing algorithm that is based on Medicare fee schedules (O'Keeffe-Rosetti et al. 2013), thus allowing comparisons between our estimates and other published SEER-/Medicare-based estimates.
However, our study is not without limitations. First, we did not estimate "net" medical care costs relative to matched noncancer controls (Campbell and Ramsey 2009; Guy et al. 2013; Yabroff, Borowski, and Lipscomb 2013; Yabroff et al. 2013); however, findings from our prior work indicate that net costs (i.e., those attributable specifically to cancer) account for the majority of total costs in the year following diagnosis among de novo breast, colorectal, and lung cancer patients (Banegas et al. 2015). We did not decompose all the components driving variation in medical care costs within and between groups. The cost estimates reported in our study reflect utilization patterns of populations diagnosed with cancer between 2000 and 2011 (2012 for recurrent cancers), and thus, our estimates do not account for more recently approved, and often higher-cost, targeted therapies. Also, our study sample was not as racially, ethnically, and geographically diverse as the U.S. population. Further, we present cost estimates from the payer or health system perspective and do not capture patient cost liabilities, which may lead to substantial financial hardship for many patients and their families (Banegas and Yabroff 2013). Assessing the financial burden of patients with recurrent cancer is largely unknown and should be a focus of future research. Lasdy, our study is specific to the experience of patients receiving care from salary-based clinicians within three KP integrated health care systems. Consistent with our previous studies related to treatments and outcomes for patients with advanced cancer (Ritzwoller et al. 2012, 2014), we strongly believe that our findings generalize to other settings. However, given that financial incentives to provide more expensive treatments are mitigated within the KP setting, the cost estimates from this study may actually be less than what one might observe in a fee-forservice setting.
Cost differences between de novo and recurrent cancers reveal heterogeneity in care patterns that merits further investigation. Understanding differences in costs/utihzation for de novo and recurrent cancer patients is critical when developing treatment plans, evaluating factors associated with spending/quality (including assessing overuse and underuse), and studying episode-based reimbursement models. The costs reported by this study may serve as a benchmark for stage-specific phase-of-care oncology episode payment models and future cost-effectiveness studies of treatments for advanced cancer.
ACKNOWLEDGMENTS
Joint Acknowledgment/Disclosure Statement: This work was supported by a grant from the National Cancer Institute (R01 CA172143 DPR/MJH), with additional support from NCI (U2 C171524 to the Cancer Research Network, Lawrence Kushi, PI), R01 CA10527, U01 CA195565; The American Society of Clinical Oncology (Career Development Award) to MJH.
Disclosures: None.
Disclaimer: None.
REFERENCES
Backhus, L. M., F. Farjah, S. B. Zeliadt, T. K. Varghese, A. Cheng, L. Kessler, D. H. Au, and D. R. Hum. 2014. "Predictors of Imaging Surveillance for Surgically Treated Early-Stage Lung Cancer." Annals of Thoracic Surgery 98 (6): 1944--51; discussion 51-2.
Banegas, M. P., and K. R. Yabroff. 2013. "Out of Pocket, Out of Sight? An Unmeasured Component of the Burden of Cancer." Journal of the National Cancer Institute 105 (4): 252-3.
Banegas, M. P., K. R. Yabroff, M. O'Keefe Rosetti, D. R. Ritzwoller, R. A. Fishman, R. G. Salloum, J. Elston Lafata, and M. C. Hombrook. 2015. "Long-Term Medical Care Costs of Breast, Prostate, Lung and Colorectal Cancer for HMO Members." Journal of Patient-Centered Research and Reviews 2 (2): 80.
Basu, A., and W. G. Manning. 2010. "Estimating Lifetime or Episode-of-Illness Costs under Censoring." Health Economics 19 (9): 1010-28.
Bradley, C. J., K. R. Yabroff, A. B. Mariotto, C. Zeruto, Q. Tran, and J. L. Warren. 2017 "Antineoplastic Treatment of Advanced-Stage Non-Small-Cell Lung Cancer: Treatment, Survival, and Spending (2000 to 2011)." Journal of Clinical Oncology 35: 529-35.
Brown, M. L., G. F. Riley, N. Schussler, and R. Etzioni. 2002. "Estimating Health Care Costs Related to Cancer Treatment from SEER-Medicare Data." Medical Care 40 (8 Suppl):IV-17.
Campbell, J. D., and S. D. Ramsey. 2009. "The Costs of Treating Breast Cancer in the U.S.: A Synthesis of Published Evidence." Pharmacoeconomics 27 (3): 199-209.
Carroll, N. M., K. M. Burniece, J. Holzman, D. B. McQuillan, A. Plata, and D. P Ritzwoller. 2017. "Algorithm to Identify Systemic Cancer Therapy Treatment Using Structured Electronic Data. " JCO Clinical Cancer Informatics 1: 1-9.
Carver, J. R., C. L. Shapiro, A. Ng, L.Jacobs, C. Schwartz, K. S. Virgo, K. L. Hagerty, M. R. Somerfield, D. J. Vaughn, and Asco Cancer Survivorship Expert Panel. 2007. "American Society of Clinical Oncology Clinical Evidence Review on the Ongoing Care of Adult Cancer Survivors: Cardiac and Pulmonary Late Effects." Journal of Clinical Oncology 25 (25): 3991-4008.
Chang, C. R., and M. Gould. 2017. "Playing the Odds: Lung Cancer Surveillance after Curative Surgery." Current Opinion in Pulmonary Medicine 23 (4): 298-304.
Chubak, J., R. Ziebell, R. T. Greenlee, S. Honda, M. C. Hornbrook, M. Epstein, L. Nekhlyudov, P A. Pawloski, D. R. Ritzwoller, N. R. Ghai, H. S. Feigelson, H. A. Clancy, V. P. Doria-Rose, and L. H. Kushi. 2016. "The Cancer Research Network: A Platform for Epidemiologic and Health Services Research on Cancer Prevention, Care, and Outcomes in Large, Stable Populations." Cancer Causes & Control 27 (11): 1315-23.
Davidoff, A. J., S. C. Hill, D. Bernard, and K. R. Yabroff. 2015. "The Affordable Care Act and Expanded Insurance Eligibility among Nonelderly Adult Cancer Survivors." Journal of the National Cancer Institute 107 (9).
Eisen, A., G. G. Fletcher, S. Gandhi, M. Mates, O. C. Freedman, S. F. Dent, M. E. Trudeau, and Members of the Early Breast Cancer Systemic Therapy Consensus Panel. 2015. "Optimal Systemic Therapy for Early Breast Cancer in Women: A Clinical Practice Guideline." Current Oncology (Toronto, Ont.) 22 (Suppl 1): S67-81.
Engel-Nitz, N. M., Y. Hao, L. K. Becker, and R. Gerdes. 2015. "Costs and Mortality of Recurrent Versus de Novo Hormone Receptor-Positive/HER2(-) Metastatic Breast Cancer." Journal of Comparative Effectiveness Research 4 (4): 303-14.
Guy Jr, G. P., D. U. Ekwueme, K. R. Yabroff, E. C. Dowling, C. Li, J. L. Rodriguez, J. S. de Moor, and K. S. Virgo. 2013. "Economic Burden of Cancer Survivorship among Adults in the United States." Journal of Clinical Oncology 31 (30): 3749-57.
Hahn, E. E., T. Tang, J. S. Lee, C. E. Munoz-Plaza, E. Shen, B. Rowley, J. L. Maeda, D. M. Mosen, J. C. Ruckdeschel, and M. K. Gould. 2016. "Use of Posttreatment Imaging and Biomarkers in Survivors of Early-Stage Breast Cancer: Inappropriate Surveillance or Necessary Care?" Cancer 122 (6): 908-16.
Hassett, M. J., D. P Ritzwoller, N. Taback, N. Carroll, A. M. Cronin, G. V. Ting, D. Schrag, J. L. Warren, M. C. Hornbrook, and J. C. Weeks. 2014. "Validating Billing/Encounter Codes as Indicators of Lung, Colorectal, Breast, and Prostate Cancer Recurrence Using 2 Large Contemporary Cohorts." Medical Care 52 (10):e65-73.
Hassett, M. J., H. Uno, A. M. Cronin, N. M. Carroll, M. C. Hombrook, and D. Ritzwoller. 2017. "Detecting Lung and Colorectal Cancer Recurrence Using Structured Clinical/Administrative Data to Enable Outcomes Research and Population Health Management." Medical Care 55 (12): e88-98.
Hombrook, M. C., R. A. Fishman, D. P. Ritzwoller, J. Elston-Lafata, M. C. O'Keeffe-Rosetti, and R. G. Salloum. 2013. "When Does an Episode of Care for Cancer Begin?" Medical Care 51 (4): 324-9.
Jassem, J., C. Carroll, S. E. Ward, E. Simpson, and D. Hind. 2009. "The Clinical Efficacy of Cytotoxic Agents in Locally Advanced or Metastatic Breast Cancer Patients Pretreated with an Anthracycline and a Taxane: A Systematic Review." European Journal of Cancer 45 (16): 2749-58.
Kamon, J., G. R. Kerr, W. Jack, N. L. Papo, and D. A. Cameron. 2007 "Health Care Costs for the Treatment of Breast Cancer Recurrent Events: Estimates from a UK-Based Patient-Level Analysis." British Journal of Cancer 97 (4): 479-85.
Khatcheressian, J. L., A. C. Wolff, T. J. Smith, E. Grunfeld, H. B. Muss, V. G. Vogel, F. Halberg, M. R. Somerfield, N. E. Davidson, and American Society of Clinical. 2006. "American Society of Clinical Oncology 2006 Update of the Breast Cancer Follow-up and Management Guidelines in the Adjuvant Setting." Journal of Clinical Oncology 24 (31): 5091-7
Lamerato, L., S. Havstad, S. Gandhi, D. Jones, and D. Nathanson. 2006. "Economic Burden Associated with Breast Cancer Recurrence: Findings from a Retrospective Analysis of Health System Data." Cancer 106 (9): 1875-82.
Lange, A., A. Prenzler, M. Frank, H. Golpon, T. Welte, and J. M. von der Schulenburg. 2014. "A Systematic Review of the Cost-Effectiveness of Targeted Therapies for Metastatic Non-Small Cell Lung Cancer (NSCLC)." BMC Pulmonary Medicine 14: 192.
Lidgren, M., N. Wilking, B. Jonsson, and C. Rehnberg. 2007. "Resource Use and Costs Associated with Different States of Breast Cancer." IntemationalJournal of Technology Assessment in Health Care23 (2): 223-31.
Loggers, E. T., P. A. Fishman, D. Peterson, M. O'Keeffe-Rosetti, C. Greenberg, M. C. Hombrook, L. H. Kushi, S. Lowry, A. Ramaprasan, E. H. Wagner, J. C. Weeks, and D. P. Ritzwoller. 2014. "Advanced Imaging among Health Maintenance Organization Enrollees with Cancer." Journal of Oncology Practice 10 (4): 231-8.
Malmgren, J. A., M. Mayer, M. K. Atwood, and H. G. Kaplan. 2018. "Differential Presentation and Survival of de Novo and Recurrent Metastatic Breast Cancer over Time: 1990-2010." Breast Cancer Research and Treatment 167 (2): 579-90.
Manning, W. G., A. Basu, and J. Mullahy. 2005. "Generalized Modeling Approaches to Risk Adjustment of Skewed Outcomes Data.." Journal of Health Economics 24 (3): 465-88.
Mariotto, A. B., K. R. Yabroff, Y. Shao, E. J. Feuer, and M. L. Brown. 2011. "Projections of the Cost of Cancer Care in the United States: 2010-2020." Journal of the National Cancer Institute 103 (2): 117-28.
Merkow, R. R., D. Korenstein, R. Yeahia, R. B. Bach, and S. S. Baxi. 2017 "Quality of Cancer Surveillance Clinical Practice Guidelines: Specificity and Consistency of Recommendations." JAMA Internal Medicine 177 (5): 701-9.
North American Association of Central Cancer Registries. 2011. "NAACCR Strategic Management Plan: 2011-2016" [accessed on January 8, 2011]. Available at https://www.naaccr.org/
O'Keeffe-Rosetti, M. C., M. C. Hombrook, R. A. Fishman, D. R. Ritzwoller, E. M. Keast, J. Staab, J. E. Lafata, and R. Salloum. 2013. "A Standardized Relative Resource Cost Model for Medical Care: Application to Cancer Control Programs." Journal of the National Cancer Institute. Monographs 2013 (46): 106-16.
Quan, H., V. Sundararajan, P. Halfon, A. Fong, B. Bumand, J. C. Luthi, L. D. Saunders, C. A. Beck, T. E. Feasby, and W. A. Ghali. 2005. "Coding Algorithms for Defining Comorbidities in ICD-9-CM and ICD-10 Administrative Data." Medical Care 43(11): 1130-9.
Rier, H. N., M. D. Levin, J. van Rosmalen, M. Bos, J. C. Drooger, P. deJong, J. E. A. Portielje, E. M. R. Elsten, A. J. Ten Tije, S. Sleijfer, and A. Jager. 2017. "First-Line Palliative HER2-Targeted Therapy in HER2-Positive Metastatic Breast Cancer Is Less Effective after Previous Adjuvant Trastuzumab-Based Therapy." Oncologist 22 (8): 901-9.
Riley, G. F., A. L. Potosky, J. D. Lubitz, and L. G. Kessler. 1995. "Medicare Payments from Diagnosis to Death for Elderly Cancer Patients by Stage at Diagnosis." Medical Care 33 (8): 828-41.
Ritzwoller, D. R., N. M. Carroll, T. Delate, M. C. Hombrook, L. Kushi, E. J. Aiello Bowles, J. M. Freml, K. Huang, and E. T. Loggers. 2012. "Patterns and Predictors of First-Line Chemotherapy Use among Adults with Advanced Non-Small Cell Lung Cancer in the Cancer Research Network." Lung Cancer 78 (3): 245-52.
Ritzwoller, D. R., N. Carroll, T. Delate, M. O'Keeffe-Rossetti, R. A. Fishman, E. T. Loggers, E. J. Aiello Bowles, J. Elston-Lafata, and M. C. Hombrook. 2013. "Validation of Electronic Data on Chemotherapy and Hormone Therapy Use in HMOs." Medical Care 51 (10): e67-73.
Ritzwoller, D. R., N. M. Carroll, T. Delate, M. C. Hombrook, L. Kushi, E. J. Bowles, E. T. Loggers, and A. Menter. 2014. "Comparative Effectiveness of Adjunctive Bevacizumab for Advanced Lung Cancer: The Cancer Research Network Experience." Journal of Thoracic Oncology 9 (5): 692-701.
Ritzwoller, D. P., M. J. Hassett, H. Uno, A. M. Cronin, N. M. Carroll, M. C. Hombrook, and L. C. Kushi. 2017 "Development, Validation, and Dissemination of a Breast Cancer Recurrence Detection and Timing Informatics Algorithm." Journal of the National Cancer Institute 110 (3): djx200.
Ross, T. R., D. J. Ng, J. S. Brown, R. Pardee, M. D. Hombrook, G. Hart, and J. F. Steiner. 2014. "The HMO Research Network Virtual Data Warehouse: A Public Data Model to Support Collaboration." eGEMS2 (1): 2.
Sagar, B., Y. S. Lin, and L. D. Castel. 2017. "Cost Drivers for Breast, Lung, and Colorectal Cancer Care in a Commercially Insured Population over a 6-Month Episode: An Economic Analysis from a Health Plan Perspective." Journal of Medical Economics 20: 1018-23.
Steele, S. R., G. J. Chang, S. Hendren, M. Weiser, J. Irani, W. D. Buie, J. F Rafferty, and C. P. G. C. o. t. A. S. o. C. and Rectal Surgeons. 2015. "Practice Guideline for the Surveillance of Patients after Curative Treatment of Colon and Rectal Cancer." Diseases of the Colon and Rectum 58 (8): 713-25.
Stokes, M. E., D. Thompson, E. L. Montoya, M. C. Weinstein, E. P. Winer, and C. C. Earle. 2008. "Ten-Year Survival and Cost Following Breast Cancer Recurrence: Estimates from SEER-Medicare Data." Value Health 11 (2): 213-20.
Takeda, A. L., J. Jones, E. Loveman, S. C. Tan, and A. J. Clegg. 2007. "The Clinical Effectiveness and Cost-Effectiveness of Gemcitabine for Metastatic Breast Cancer: A Systematic Review and Economic Evaluation." Health Technology Assessment 11 (19): iii, ix-xi, 1-62.
Taplin, S. H., W. Barlow, N. Urban, M. T. Mandelson, D. J. Timlin, L. Ichikawa, and P. Nefcy. 1995. "Stage, Age, Comorbidity, and Direct Costs of Colon, Prostate, and Breast Cancer Care." Journal of the National Cancer Institute 87 (6): 417-26.
Thorpe, K. E., and D. Howard. 2003. "Health Insurance and Spending among Cancer Patients." Health Affairs Suppl Web Exclusives: W3-189-98.
Warren, J. L., and K. R. Yabroff. 2015. "Challenges and Opportunities in Measuring Cancer Recurrence in the United States." Journal of the National Cancer Institute 107 (8).
Warren, J. L., K. R. Yabroff, A. Meekins, M. Topor, E. B. Lamont, and M. L. Brown. 2008. "Evaluation of Trends in the Cost of Initial Cancer Treatment." Journal of the National Cancer Institute 100 (12): 888-97
Weeks, J. C., P. J. Catalano, A. Cronin, M. D. Finkelman, J. W. Mack, N. L. Keating, and D. Schrag. 2012. "Patients' Expectations about Effects of Chemotherapy for Advanced Cancer." New England Journal of Medicine 367 (17): 1616-25.
Yabroff, K. R., L. Borowski, and J. Lipscomb. 2013. "Economic Studies in Colorectal Cancer: Challenges in Measuring and Comparing Costs." Journal of the National Cancer Institute Monographs 2013 (46): 62-78.
Yabroff, K. R., E. B. Lamont, A. Mariotto, J. L. Warren, M. Topor, A. Meekins, and M. L. Brown. 2008. "Cost of Care for Elderly Cancer Patients in the United States." Journal of the National Cancer Institute 100 (9): 630-41.
Yabroff, K. R., J. L. Warren, D. Schrag, A. Mariotto, A. Meekins, M. Topor, and M. L. Brown. 2009. "Comparison of Approaches for Estimating Incidence Costs of Care for Colorectal Cancer Patients." Medical Care 47 (7 Suppl 1): S56-63.
Yabroff, K. R., J. Lund, D. Kepka, and A. Mariotto. 2011. "Economic Burden of Cancer in the United States: Estimates, Projections, and Future Research." Cancer Epidemiology, Biomarkers &Prevention 20 (10): 2006-14.
Yabroff, K. R., S. Francisci, A. Mariotto, M. Mezzetti, A. Gigli, and J. Lipscomb. 2013. "Advancing Comparative Studies of Patterns of Care and Economic Outcomes in Cancer: Challenges and Opportunities." Journal of the National Cancer Institute. Monographs 2013 (46): 1-6.
Yardley, D. A., R. A. Kaufman, A. Brufsky, M. U. Yood, H. Rugo, M. Mayer, C. Quah, B. Yoo, and D. Tripathy. 2014. "Treatment Patterns and Clinical Outcomes for Patients with de Novo versus Recurrent HER2-Positive Metastatic Breast Cancer." Breast Cancer Research and Treatment 145 (3): 725-34.
SUPPORTING INFORMATION
Additional supporting information may be found online in the Supporting Information section at the end of the article.
Appendix SA1: Author Matrix.
Figure S1: Qualifying Patients for Recurrence Cost Analysis.
Table S1: Average Total Monthly Cost Model Parameter Estimates by Cancer Site.
Debra P. Ritzwoller [iD], Paul A. Fishman, Matthew P. Banegas, Nikki M. Carroll, Maureen O'Keeffe-Rosetti, Angel M. Cronin, Hajime Uno, Mark C. Hombrook, and Michael J. Hassett
Address correspondence to Debra P. Ritzwoller, Ph.D., Institute for Health Research, Kaiser Permanente Colorado, Waterpark III, 2550 S. Parker Rd., Suite 200, Aurora, CO 80014; e-mail: debra.ritzwoller@kp.org. Paul A. Fishman, Ph.D., is with the Department of Health Services, University of Washington, Seatde, WA; and Kaiser Permanente Washington Health Research Institute, Seattle, WA. Matthew P. Banegas, Ph.D., Maureen O'Keeffe-Rosetti, M.S., and Mark C. Hombrook, Ph.D., are with the Kaiser Permanente Center for Health Research, Portland, OR. Nikki M. Carroll, M.S., is with the Institute for Health Research, Kaiser Permanente Colorado, Aurora, CO. Angel M. Cronin, M.S., is with the Dana-Farber Cancer Institute, Boston, MA. Hajime Uno, Ph.D., and Michael J. Hassett, M.D., M.P.H., are with the Dana-Farber Cancer Institute, Boston, MA; and Harvard Medical School, Boston, MA.
DOI: 10.1111/1475-6773.13014
Table 1: Demographics of De Novo and Recurrent Cohorts by Cancer Site
Breast Cancer
De Novo Recurrent
N(%) N(%)
Total cases 352 765
Demographics
Race/ethnicity (*)
White 297 (84) 638 (83)
Nonwhite 38 (11) 79 (10)
Unknown 17 (5) 48 (6)
Female 352 (100) 765 (100)
Age on index date
21-54 years 113 (32) 238 (31)
55-69 years 126 (36) 267 (35)
70+ years 113 (32) 260 (34)
<65 years 214 (61) 421 (55)
Mean age at index (SD) 61.6 (13.5) 62.5 (14.7)
Income (census block level based)
<$40,000 22 (6) 79 (10)
$40,000-$60,000 125 (36) 252 (33)
$60,000+ 205 (58) 434 (57)
Health Caxe System (HCS)
HCS#1 134 (38) 282 (37)
HCS #2 110 (31) 231 (30)
HCS #3 108 (31) 252 (33)
Clinical characteristics
Diagnosis index year
2000-2003 93 (26) 109 (14)
2004-2006 84 (24) 183 (24)
2007-2009 94 (27) 234 (31)
2010-2012 81 (23) 239 (31)
Stage ([dagger])
I 0 (0) 200 (26)
II 0 (0) 348 (46)
III 0 (0) 217 (28)
IV 352 (100) 0 (0)
Comorbidities
0 246 (70) 554 (72)
1 64 (18) 134 (18)
2+ 42 (12) 77 (10)
Recurrence within 12 months of - 136 (18)
definitive therapy ([double dagger])
Death within 12 months post-index 95 (27) 242 (32)
date
Inpatient event within 12 months 186 (53) 319 (42)
post-index date
Chemotherapy within 12 months 240 (68) 529 (69)
post-index date
Chemotherapy at initial diagnosis and NA 386 (50)
within 12 months post-index
Radiotherapy within 12 months 137 (39) 327 (43)
post-index
Radiotherapy at initial diagnosis and NA 194 (25)
within 12 months post-index
Hospice within 12 months post-index 22 (6) 92 (12)
Colorectal Cancer
De Novo
p-Value N(%)
Total cases 1,072
Demographics
Race/ethnicity (*)
White .6241 840 (78)
Nonwhite 115 (11)
Unknown 117 (11)
Female NA 525 (49)
Age on index date
21-54 years .8243 244 (23)
55-69 years 379 (35)
70+ years 449 (42)
<65 years .0708 486 (45)
Mean age at index (SD) .3240 66.1 (13.6)
Income (census block level based)
<$40,000 .0821 110 (10)
$40,000-$60,000 397 (37)
$60,000+ 565 (53)
Health Caxe System (HCS)
HCS#1 .7544 356 (33)
HCS #2 364 (34)
HCS #3 352 (33)
Clinical characteristics
Diagnosis index year
2000-2003 <.0001 341 (32)
2004-2006 270 (25)
2007-2009 287 (27)
2010-2012 174 (16)
Stage ([dagger])
I NA 0(0)
II 0(0)
III 0(0)
IV 1,072 (100)
Comorbidities
0 .5873 609 (57)
1 231 (21)
2+ 232 (22)
Recurrence within 12 months of -
definitive therapy ([double dagger])
Death within 12 months post-index .1161 532 (50)
date
Inpatient event within 12 months .0005 819 (76)
post-index date
Chemotherapy within 12 months .7454 704 (66)
post-index date
Chemotherapy at initial diagnosis and - NA
within 12 months post-index
Radiotherapy within 12 months .2282 165 (15)
post-index
Radiotherapy at initial diagnosis and - NA
within 12 months post-index
Hospice within 12 months post-index .0031 134 (13)
Lung Cancer
Recurrent
N(%) p-Value
Total cases 542
Demographics
Race/ethnicity (*)
White 434 (80) .1070
Nonwhite 66 (12)
Unknown 42 (8)
Female 248 (46) .2217
Age on index date
21-54 years 76 (14) .0002
55-69 years 214 (40)
70+ years 252 (47)
<65 years 201 (37) .0015
Mean age at index (SD) 68.2 (12.3) .0032
Income (census block level based)
<$40,000 47 (9) .0331
$40,000-$60,000 190 (35)
$60,000+ 305 (56)
Health Caxe System (HCS)
HCS#1 188 (35) .0180
HCS #2 148 (27)
HCS #3 206 (38)
Clinical characteristics
Diagnosis index year
2000-2003 114 (21) <.0001
2004-2006 151 (28)
2007-2009 165 (30)
2010-2012 112 (21)
Stage ([dagger])
I 51 (9) NA
II 165 (30)
III 326 (60)
IV 0 (0)
Comorbidities
0 283 (52) .1186
1 140 (26)
2+ 119 (22)
Recurrence within 12 months of 170 (31)
definitive therapy ([double dagger])
Death within 12 months post-index 211 (39) <.0001
date
Inpatient event within 12 months 307 (57) <.0001
post-index date
Chemotherapy within 12 months 349 (64) .6099
post-index date
Chemotherapy at initial diagnosis and 241 (44) _
within 12 months post-index
Radiotherapy within 12 months 88 (16) .6595
post-index
Radiotherapy at initial diagnosis and 19 (3.5) -
within 12 months post-index
Hospice within 12 months post-index 54 (10) .1335
Lung Cancer
De Novo Recurrent
N(%) N(%)
Total cases 4,041 340
Demographics
Race/ethnicity (*)
White 3,424 (85) 287 (84)
Nonwhite 353 (9) 33 (10)
Unknown 264 (7) 20 (6)
Female 1,949 (48) 183 (54)
Age on index date
21-54 years 432 (11) 21 (6)
55-69 years 1,672 (41) 144 (42)
70+ years 1,937 (48) 175 (52)
<65 years 1,450 (36) 109 (32)
Mean age at index (SD) 68.3 (10.5) 69.0 (9.4)
Income (census block level based)
<$40,000 490 (12) 34 (10)
$40,000-$60,000 1,610 (40) 140 (41)
$60,000+ 1,941 (48) 166 (49)
Health Caxe System (HCS)
HCS#1 1,378 (34) 124 (36)
HCS #2 1,201 (30) 80 (24)
HCS #3 1,462 (36) 136 (40)
Clinical characteristics
Diagnosis index year
2000-2003 1,268 (31) 68 (20)
2004-2006 1,074 (27) 93 (27)
2007-2009 1,018 (25) 94 (28)
2010-2012 681 (17) 85 (25)
Stage ([dagger])
I 0 (0) 177 (52)
II 0 (0) 86 (2.5)
III 0 (0) 76 (22)
IV 4,042 (100) 0 (0)
Comorbidities
0 2,047 (51) 96 (28)
1 953 (24) 130 (38)
2+ 1,041 (26) 114 (34)
Recurrence within 12 months of - 152 (45)
definitive therapy ([double dagger])
Death within 12 months post-index 2,879 (71) 177 (52)
date
Inpatient event within 12 months 2,414 (60) 170 (50)
post-index date
Chemotherapy within 12 months 2,601 (64) 223 (66)
post-index date
Chemotherapy at initial diagnosis and NA 76 (22)
within 12 months post-index
Radiotherapy within 12 months 1,997 (49) 157 (46)
post-index
Radiotherapy at initial diagnosis and NA 39 (11)
within 12 months post-index
Hospice within 12 months post-index 732 (18) 58 (17)
p-Value
Total cases
Demographics
Race/ethnicity (*)
White .7619
Nonwhite
Unknown
Female .0475
Age on index date
21-54 years .0292
55-69 years
70+ years
<65 years .1573
Mean age at index (SD) .2067
Income (census block level based)
<$40,000 .0541
$40,000-$60,000
$60,000+
Health Caxe System (HCS)
HCS#1 .0529
HCS #2
HCS #3
Clinical characteristics
Diagnosis index year
2000-2003 <.0001
2004-2006
2007-2009
2010-2012
Stage ([dagger])
I NA
II
III
IV
Comorbidities
0 <.0001
1
2+
Recurrence within 12 months of
definitive therapy ([double dagger])
Death within 12 months post-index <.0001
date
Inpatient event within 12 months .0007
post-index date
Chemotherapy within 12 months .6509
post-index date
Chemotherapy at initial diagnosis and -
within 12 months post-index
Radiotherapy within 12 months .2738
post-index
Radiotherapy at initial diagnosis and -
within 12 months post-index
Hospice within 12 months post-index .6283
(*) Nonwhite includes Hispanic, black, Asian, and other races and/or
ethnicities.
([dagger]) For lung cancer cohorts, de novo includes stages IIIB and
IV, while recurrent includes stages I, II, or IIIA (at initial
diagnosis).
([double dagger]) Definitive therapy is defined mastectomy/lumpectomy
for breast cancer, colectomy for CRC, and lobectomy for lung cancer,
with or without radiation.
Table 2: Cost Estimates by Cancer Site and Age Group
Adjusted Average Total Monthly Costs
[greater than or equal to]65 Years of
Age at Index Date
Average Monthly Cost [$] (95% CI)
De Novo Recurrent
Breast
12 months pre 305 (236-394.-) 2,737 (2,337-3,205)
Index month 7,926 (6,564-9,572.-) 7,099 (5,985-8,421)
12 months post 8,559 (7,398-9,904.-) 7,817 (6,896-8,859)
Colorectal cancer
12 months pre 448 (336-595.-) 3,667 (3,067-4,383)
Index month 13,588 (11,838-15,596.-) 5,601 (4,450-7,050)
12 months post 8,694 (7,724-9,785.-) 6,217 (5,334-7,247)
Lung cancer-Outliers
Removed ([dagger])
12 months pre 582 (507-669.-) 3,669 (3,140-4,287)
Index month 10,980 (10,152-11,876.-) 8,724 (6,757-11,263)
12 months post 7,148 (6,722-7,602.-) 5,592 (4,673-6,690)
Breast
12 months 3,310 (2,563-4,274.-) 30,531 (26,149-35,649)
pre
Index 90,391 (79,197-103,166.-) 72,243 (64,492-80,926)
month +
11 months
post
Colorectal cancer
12 months 4,721 (3,331-6,690.-) 40,424 (33,359-48,991)
pre
Index 79,809 (70,735-90,048.-) 54,956 (46,859-64,453)
month +
11 months
post
Lung cancer-Outliers
removed ([dagger])
12 months 6,508 (5,628-7,525.-) 41,973 (35,789-49,232)
pre
Index 64,582 (60,240-69,230.-) 52,759 (45,437-61,261)
month +
11 months
post
[greater than or equal to]65 Years of Age at
Index Date
Average Monthly Cost [$] (95% CI)
p-Value
Cost on Cost
Difference Difference (*)
Breast
12 months pre 2,431 <.0001
Index month -827 .1087
12 months post -742 .0690
Colorectal cancer
12 months pre 3,219 <.0001
Index month -7,987 <.0001
12 months post -2,476 <.0001
Lung cancer-Outliers
Removed ([dagger])
12 months pre 3,086 <.0001
Index month -2,256 .0821
12 months post -1,556 .0071
Breast
12 months 27,221 <.0001
pre
Index -18,148 .0005
month +
11 months
post
Colorectal cancer
12 months 35,703 <.0001
pre
Index -24,853 <.0001
month +
11 months
post
Lung cancer-Outliers
removed ([dagger])
12 months 35,465 <.0001
pre
Index -11,823 .0057
month +
11 months
post
[greater than or equal to]65 Years of Age at
Index Date
Average Monthly Cost [$] (95% CI)
De Novo Recurrent
Breast
12 months pre 818 (608-1,099) 2,178 (1,843-2,776)
Index month 9,210 (7,309-11,607) 6,530 (5,396-7,903)
12 months post 5,816 (4,852-6,972) 4,772 (4,213-5,406)
Colorectal cancer
12 months pre 878 (749-1,029) 3,125 (2,701-3,616)
Index month 15,500 (13,468-17,836) 7,093 (5,778-8,708)
12 months post 7,024 (6,221-7,932) 5,274 (4,587-$6,063)
Lung cancer-Outliers
Removed ([dagger])
12 months pre 1,025 (938-1,121) 3,286 (2,914-3,705)
Index month 10,314 (9,946-11,029) 6,138 (5,018-7,509)
12 months post 5,498 (5,203-5,811) 4,505 (3,989-5,089)
Breast
12 months 9,596 (7,103-12,964) 24,797 (21,118-29,117)
pre
Index 55,436 (47,112-65,238) 44,378 (39,703-49,602)
month +
11 months
post
Colorectal cancer
12 months 11,558 (9,730-13,729) 36,702 (31,650-42,565)
pre
Index 59,623 (53,690-66,210) 43,726 (37,702-50,322)
month +
11 months
post
Lung cancer-Outliers
removed ([dagger])
12 months 11,286 (10,288-12,383) 36,195 (31,758-41,249)
pre
Index 41,919 (39,410-44,587) 39,854 (35,710-44,480)
month +
11 months
post
[greater than or equal to]65 Years of Age at
Index Date
Average Monthly Cost [$] (95% CI)
p-Value on
Cost Cost
Difference Difference (*)
Breast
12 months pre 1,360 <.0001
Index month -2,679 .1391
12 months post -1,043 .0967
Colorectal cancer
12 months pre 2,247 <.0001
Index month -8,406 <.0001
12 months post -1,750 .0004
Lung cancer-Outliers
Removed ([dagger])
12 months pre 2,260 <.0001
Index month -4,175 <.0001
12 months post -993 .0013
Breast
12 months 15,201 <.0001
pre
Index -11,058 .0086
month +
11 months
post
Colorectal cancer
12 months 25,144 <.0001
pre
Index -15,897 .0001
month +
11 months
post
Lung cancer-Outliers
removed ([dagger])
12 months 24,909 <.0001
pre
Index -2,065 .3236
month +
11 months
post
(*) Differences in cost based on log values.
([dagger]) Outliers removed by adjusting all costs outside of the 99th
percentile to overall mean cost.
 Printer friendly
 Cite/link
 Email
 Feedback
| |
| Title Annotation: | RESEARCH ARTICLE |
|---|---|
| Author: | Ritzwoller, Debra P.; Fishman, Paul A.; Banegas, Matthew P.; Carroll, Nikki M.; O'Keeffe-Rosetti, Ma |
| Publication: | Health Services Research |
| Date: | Dec 1, 2018 |
| Words: | 8280 |
| Previous Article: | Drug Shortage Impacts Patient Receipt of Induction Treatment. |
| Next Article: | HSR Commentary: Linking VA and Non-VA Data to Address Important US Veteran Health Services Research Issues. |
| Topics: | |