Patterns and predictors of opioid dispensing among older cancer patients from 2008 to 2015.

BACKGROUND: Understanding factors associated with opioid dispensing in cancer patients is important for developing tailored guidelines and ensuring equitable access to pain management. We examined patterns and predictors of opioid dispensing among older cancer patients from 2008 to 2015. METHODS: We analyzed data from the Surveillance, Epidemiology, and End Results (SEER) database linked to Medicare claims. We included the most common cancer types among patients aged 66-95 years. Opioids dispensed within 30 days before and 120 days after cancer diagnosis were assessed. We used logistic regression models to examine trends, adjusted odds ratios (aORs), and 95% confidence intervals (CIs) for opioid dispensing, considering patient demographics, geography, cancer stage, comorbidities, and treatment options. Models were stratified by sex. RESULTS: A total of 211,759 cancer patients aged 66-95 years were included in the study. For cancers combined, non-Hispanic Black men had a significantly lower odds of receiving opioids during the 120 days post-diagnosis (aOR = 0.89, 95% CI = 0.84-0.94) compared to non-Hispanic White men. Factors such as pre-diagnosis opioid dispensing, age, geography, cancer stage, comorbidities, and type of cancer treatment were associated with opioid dispensing during the 120 days post-diagnosis. Surgery had the strongest association, with men undergoing surgery being 4.4 times more likely to receive opioids within 120 days post-diagnosis (aOR = 4.41, 95% CI = 4.23-4.60), while women had an odds ratio of 2.72 (95% CI = 2.62-2.83). Chemotherapy and radiotherapy were also positively associated with opioid dispensing, with less pronounced estimates. CONCLUSIONS: We observed significant variations in opioid dispensing among cancer patients aged 66-95 years across cancer types and demographic and clinical factors.

Assessing 1 year Comorbidity Prevalence and Its Survival Implications in Medicare Beneficiaries Diagnosed with Cancer: Insights from a new SEER-Medicare Resource.

BACKGROUND: Almost half of Medicare beneficiaries diagnosed with cancer from 1992-2005 had at least one comorbid condition. Conditions impact a range of domains from clinical decision making to quality of life which are important to consider when conducting cancer research. We introduce a new SEER-Medicare resource to facilitate using claims data for cancer patients. METHODS: We use the SEER-Medicare resource to estimate prevalence of comorbidities, 5-year survival rate by cancer site, stage, age and comorbidity severity, and prevalence of surgery by comorbidity for breast, prostate, colorectal and lung cancer. RESULTS: Overall, the most prevalent comorbidities in the year prior to cancer diagnosis were diabetes (27%), COPD (22%), peripheral vascular disease (14%), and congestive heart failure (12%). Comorbidity severity had a greater impact on the probability of dying from non-cancer causes than from dying from cancer. Severity of comorbidity and age consistently increased the probability of non-cancer death. The percentage of persons receiving surgery tended to be lower among those with severe comorbidity. CONCLUSIONS: This study demonstrates the utility of new SEER*stat databases that contain Medicare beneficiaries and claims-based measures of comorbidity. Our results demonstrate that comorbidity is common among older persons diagnosed with cancer and the impact of comorbidity on the probability of dying from cancer varies by cancer site, stage at diagnosis and age. IMPACT: Comorbidity is common among persons with cancer and impacts survival. Future research on the impact of comorbidity among cancer survivors is facilitated by new databases.

Real-world lessons: combining cancer registry and retail pharmacy data for oral cancer drugs.

BACKGROUND: Recent cancer care advances have introduced new oral therapies, and yet population registries lack detailed treatment data, hampering investigations into therapy uptake, adherence, and outcomes. OBJECTIVE: This study aimed to assess the representativeness and completeness of linking Surveillance, Epidemiology, and End Results (SEER) cancer registry data with data from two major retail pharmacy chains, collectively covering a large segment of the US market. METHODS: A deterministic data linkage between 11 SEER cancer registries and retail pharmacy data (excluding mail order fills) was conducted for individuals diagnosed with selected cancers from 2013 to 2017, with follow-up through 2019. Descriptive characteristics of the linked and unlinked populations were examined. In a selected subcohort of older women (aged ≥65) with first and only primary breast cancer who had Medicare Part D claims for tamoxifen, we further validated the linkage using Medicare Part D event data as the reference standard. RESULTS: Among 758 068 eligible individuals, only 6.4% were linked to CVS/Walgreens data; the linkage percentage varied by age, sex, race, ethnicity, registry, and cancer type. Within the subcohort of 5963 older women with breast cancer and a claim for tamoxifen in Part D data, 25% were identified as tamoxifen users in retail pharmacy data. Out of these 1490 women, 749 (50.3%) had complete longitudinal tamoxifen dispensing information from retail pharmacy data. CONCLUSION: Retail pharmacy data show promise in identifying oral anticancer treatments, enhancing SEER registry efforts, but they require further validation. We propose an evaluation framework, sharing insights and potential use cases for this resource.

Evolving patterns in systemic treatment utilization and survival among older patients with advanced cutaneous melanoma.

INTRODUCTION: In the last decade, melanoma treatment has improved significantly. However, data on population-level treatment utilization and survival trends among older patients is limited. This study aimed to analyze trends in systemic anticancer therapy (Rx), including the uptake of immune checkpoint inhibitors (ICIs), in conjunction with trends in cause-specific survival among older patients (66+) diagnosed with advanced melanoma (2008-2019). METHODS: We used the Surveillance, Epidemiology, and End Results (SEER)-Medicare Condensed Resource to assess any Rx utilization among patients first diagnosed with advanced melanoma in 2008-2010, 2011-2014, and 2015-2019, stratified by stage, and type of first-line Rx among patients receiving Rx. The SEER dataset was used to evaluate trends in cause-specific survival by year of diagnosis. RESULTS: Rx utilization (any type) almost doubled, from 28.6% (2008-2010) to 55.4% (2015-2019) for stage 3 melanoma, and from 35.5% to 68.0% for stage 4 melanoma. In 2008-2010, the standard first-line treatment was cytokines/cytotoxic chemotherapy/other. By 2015-2019, only 5.1% (stage 3) and <3.6% (stage 4) of patients receiving Rx received these agents, as ICIs emerged as the dominant treatment. Both 1-year and 5-year cause-specific survival significantly improved since 2010 for stage 4 and since 2013 for stage 3. CONCLUSIONS: This study shows a significant rise in Rx utilization and a rapid transition from cytokines/cytotoxic chemotherapy to ICIs, reflecting a rapid uptake of highly effective treatment in a previously challenging disease with limited options before 2011. The documented survival improvement aligns with the adoption of these novel treatments, underscoring their significant impact on real-world patient outcomes.

Housing assistance among patients with cancer: SEER-Medicare US Department of Housing and Urban Development data linkage.

BACKGROUND: Lack of stable, affordable housing is an important social determinant of health. Federal housing assistance may buffer against housing vulnerabilities among low-income households, but research examining the association of housing assistance and cancer care has been limited. We introduce a new linkage of Surveillance, Epidemiology, and End Results (SEER) program-Medicare and US Department of Housing and Urban Development (HUD) administrative data. METHODS: Individuals enrolled in HUD public and assisted housing programs between 2006 and 2021 were linked with cancer diagnoses between 2006 and 2019 identified in the SEER-Medicare data from 16 states using Match*Pro (National Institutes of Health, Bethesda, MD) probabilistic linkage software. HUD administrative data include timing and type of housing assistance as well as verified household income. Medicare administrative data are available through 2020. RESULTS: A total of 335 490 unique individuals who received housing assistance at any time point, including 156 794 who received housing assistance around the time of their diagnosis (at least 6 months before diagnosis until 6 months after diagnosis or death), were matched to SEER-Medicare data. A total of 63 251 individuals receiving housing assistance at the time of their diagnosis were aged 66 years and older and continuously enrolled in Medicare parts A and B fee for service; 12 035 had a diagnosis of lung cancer, 8866 of breast cancer, 7261 of colorectal cancer, and 4703 of prostate cancer. CONCLUSIONS: This novel data linkage will be available through the National Cancer Institute and can be used to explore the ways in which housing assistance is associated with cancer diagnosis, care, and outcomes, including the role of housing assistance status in potentially reducing or contributing to inequities across racialized and ethnic groups.

Incidence of papillary thyroid cancer: Comparison of the military and the general population by race and tumor stage/size.

BACKGROUND: A previous study found higher papillary thyroid cancer incidence in the US military than the general population with larger differences among Black than White individuals. This study compared the two populations in the incidence by sex, race, tumor stage, and size to assess possible factors related to identified differences. METHODS: Subjects were aged 18-59 in the military and general populations. Papillary thyroid cancer patients diagnosed during 1990-2013 were identified from the Department of Defense's Automated Central Tumor Registry (ACTUR) and the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) program. Age-adjusted rates and incidence rate ratios (IRR) comparing ACTUR to SEER were calculated. RESULTS: Higher incidence rates in ACTUR than SEER were more obvious for Black (IRR=2.07, 95%CI=1.56-2.70) than White men (IRR=1.17, 95%CI=1.07-1.26) and for Black (IRR=2.30, 95%CI=1.91-2.71) than White women (IRR=1.50, 95%CI=1.38-1.64). Population differences by race were observed for localized tumors among both men and women and were larger for Black individuals. Differences were observed regardless of tumor size among Black men and White women, and in smaller tumors among Black women. CONCLUSION: Higher incidence in the military than general population primarily in localized tumors suggests universal healthcare in the military may lead to earlier detection. The differences were larger among Blacks than Whites, suggesting universal access in the military may be more impactful among Black persons, who are less likely to have timely care than White persons in the general population. Nevertheless, observed differences for tumors > 2 cm suggest other factors may also play a role.

Accuracy of Cancer Registry Primary Payer Information and Implications for Policy Research.

BACKGROUND: Cancer registry-based "primary payer at diagnosis" (PPDx) data are commonly used to evaluate the effect of insurance on cancer care outcomes, yet little is known about how well they capture Medicaid or Medicare enrollment. METHODS: We linked the National Cancer Institute's Surveillance, Epidemiology, and End Results registry data to monthly Centers for Medicare and Medicaid Services (CMS) Medicaid and Medicare enrollment records, state-year Medicaid policy, and managed care enrollment. We selected adults aged 19-64 years diagnosed between 2007 and 2011. We used bivariate analyses to compare PPDx to CMS enrollment at diagnosis month and assessed underreporting rates by patient characteristics and state-year policy. RESULTS: PPDx reported 7.8% Medicare and 10.1% Medicaid, whereas CMS enrollment indicated 5.5% Medicare, 10.4% Medicaid, and 3.4% dual Medicare-Medicaid (N = 896,031). Positive predictive values for PPDx assignment to Medicaid and Medicare were 65.3% and 75.4%, with false negative rates of 52.0% and 33.8%, respectively. Medicaid underreporting was higher in low (56.5%) versus high (50.8%) poverty areas, for males (56.1%) versus females (48.9%), for Medicaid poverty expansion or waiver enrolled (63.8%) versus cash assistance-related eligibility (47.3%), and in states with large managed care enrollment (all P < 0.001). If Medicaid and Medicare enrollment data were used to edit PPDx, 12.0% of persons would switch primary payer assignment. CONCLUSIONS: Registry-reported PPDx fails to fully capture Medicaid and Medicare enrollment, which may result in biased estimates of insurance-related policy impacts. Enhancement with objective enrollment data could reduce measurement error and bias in estimates necessary to support policy assessment.

Use of pembrolizumab among older adults with cancer in the United States, before and after FDA approval of its tumor-agnostic indication.

INTRODUCTION: Pembrolizumab, an anticancer immunotherapy agent, has received multiple approvals since its first approval by the U.S. Food and Drug Administration (FDA) in 2014. Limited data exist on its real-world use and shifts post tumor-agnostic approval in 2017 for the treatment of patients with any microsatellite instability-high/mismatch repair deficient (MSI-H/dMMR) solid tumors. This study analyzes pembrolizumab's pre and post-tumor-agnostic approval use among older U.S. adults, revealing its evolving role in oncology practice. METHODS: Using the Surveillance, Epidemiology and End Results (SEER)-Medicare data (2014-2019), we examined the cancer sites of pembrolizumab recipients before and after tumor-agnostic approval. Cancer sites were classified based on the timing of site-specific approvals (before/after tumor-agnostic approval) or no site-specific approval, and inclusion in MSI-H/dMMR clinical trials. RESULTS: The total number of pembrolizumab recipients increased from 4221 in the pre-agnostic period to 20 479 in the post-agnostic period. Pembrolizumab was used for a broad range of cancer types, including cancers that had no FDA-approved site-specific indications at the time of use (25.8% in pre- and 24.6% in post-agnostic periods). The proportion of pembrolizumab recipients receiving pembrolizumab for cancers with site-specific approvals before tumor-agnostic approval decreased from 77.3% to 70.8%. The proportion of pembrolizumab recipients receiving pembrolizumab for cancers that gained site-specific approvals following tumor-agnostic approval almost doubled (6.8% to 13.0%). The proportion of pembrolizumab recipients with cancers included in MSI-H/dMMR trials also doubled (12.3% to 25.5%) following tumor-agnostic approval. CONCLUSIONS: Pembrolizumab use has expanded over time among older adults with cancer, extending beyond those with FDA-approved site-specific indications.

Evaluation of the Completeness of Managed Care Data to Identify Cancer Diagnoses and Treatments for Patients in the SEER-Medicare Data.

BACKGROUND: The utility of codes on Medicare Advantage (MA) data to capture cancer diagnoses and treatment for cancer patients is unknown. OBJECTIVE: This study compared cancer diagnoses and treatments on MA encounter data (MA data) with the Surveillance, Epidemiology, and End-Results (SEER) data. SUBJECTS: Subjects were patients enrolled in either MA or Medicare fee-for-service (MFFS) when diagnosed with incident breast, colorectal, prostate, or lung cancer, 2015-2017, in a SEER cancer registry. MEASURES: MA data, from 2 months before to 12 months following SEER diagnosis, were reviewed to identify cancer diagnoses, surgery, chemotherapy, and radiotherapy (RT). MA data were compared with SEER to determine their sensitivity to capture cancer diagnoses and sensitivity/specificity to identify surgeries. The agreement between SEER and Medicare data regarding receipt of chemotherapy and RT was measured by Kappa statistics. A similar comparison to SEER diagnoses/treatments was made using MFFS claims to provide context for the SEER-MA comparison. RESULTS: The study included 186,449 patients, 38% in MA. MA data had 92%+ sensitivity to identify SEER cancer diagnosis and 90%+ sensitivity for cancer surgery. Specificity for surgery was >84%, except for breast cancer (52%). Kappa statistics for agreement between SEER and MA data regarding chemotherapy varied by cancer, 0.61-0.82, and for receipt of RT exceeded 0.75 for all cancers. Results observed for MFFS claims were similar to those in MA data. CONCLUSION: For 4 common cancers, MA data included most cancer diagnoses and general types of cancer treatment reported in the SEER data. More research is needed to assess additional cancers and detailed treatments.

Role of Medicaid in Early Detection of Screening-Amenable Cancers.

BACKGROUND: This study examines the association between Medicaid enrollment, including through the National Breast and Cervical Cancer Early Detection Program (NBCCEDP), and distant stage for three screening-amenable cancers: breast, cervical, and colorectal. METHODS: We use the Surveillance, Epidemiology, and End Results Cancer Registry linked with Medicaid enrollment data to compare patients who were Medicaid insured with patients who were not Medicaid insured. We estimate the likelihood of distant stage at diagnosis using logistic regression. RESULTS: Medicaid enrollment following diagnosis was associated with the highest likelihood of distant stage. Medicaid enrollment through NBCCEDP did not mitigate the likelihood of distant stage disease relative to Medicaid enrollment prior to diagnosis. Non-Hispanic Black patients had a greater likelihood of distant stage breast and colorectal cancer. Residing in higher socioeconomic areas was associated with a lower likelihood of distant stage breast cancer. CONCLUSIONS: Medicaid enrollment prior to diagnosis is associated with a lower likelihood of distant stage in screen amenable cancers but does not fully ameliorate disparities. IMPACT: Our study highlights the importance of health insurance coverage prior to diagnosis and demonstrates that while targeted programs such as the NBCCEDP provide critical access to screening, they are not a substitute for comprehensive insurance coverage.

Workforce Caring for Cancer Survivors in the United States: Estimates and Projections of Use.

BACKGROUND: This study aims to quantify the extent and diversity of the cancer care workforce, beyond medical oncologists, to inform future demand because the number of cancer survivors is expected to grow in the United States. METHODS: Surveillance, Epidemiology, and End Results-Medicare data were used to evaluate health-care use of cancer survivors diagnosed between 2000 and 2014, enrolled in fee-for-service Medicare Parts A and B, and 65 years or older in 2008-2015. We calculated percentage of cancer survivors who saw each clinician specialty and their average annual number of visits in each phase of care. We projected the national number of individuals receiving care and number of annual visits by clinician specialty and phase of care through 2040. RESULTS: Cancer survivors had higher care use in the first year after diagnosis and last year of life phases. During the initial year after cancer diagnosis, most survivors were seen for cancer-related care by a medical oncologist (59.1%), primary care provider (55.9%), and/or other cancer-treating physicians (42.2%). The percentage of survivors with cancer-related visits to each specialty declined after the first year after diagnosis, plateauing after year 6-7. However, at 10 or more years after diagnosis, approximately 20% of cancer survivors had visits to medical oncologists and an average of 4 visits a year. CONCLUSIONS: Cancer survivors had higher care use in the first year after diagnosis and last year of life. High levels of care use across specialties in all phases of care have important implications for models of survivorship care coordination and workforce planning.

The Friends of Cancer Research Real-World Data Collaboration Pilot 2.0: Methodological Recommendations from Oncology Case Studies.

The purpose of this study was to evaluate the potential collective opportunities and challenges of transforming real-world data (RWD) to real-world evidence for clinical effectiveness by focusing on aligning analytic definitions of oncology end points. Patients treated with a qualifying therapy for advanced non-small cell lung cancer in the frontline setting meeting broad eligibility criteria were included to reflect the real-world population. Although a trend toward improved outcomes in patients receiving PD-(L)1 therapy over standard chemotherapy was observed in RWD analyses, the magnitude and consistency of treatment effect was more heterogeneous than previously observed in controlled clinical trials. The study design and analysis process highlighted the identification of pertinent methodological issues and potential innovative approaches that could inform the development of high-quality RWD studies.

Racial disparities in survival among women with endometrial cancer in an equal access system.

OBJECTIVE: The mortality rate for Black women with endometrial cancer (EC) is double that of White women, although the incidence rate is lower among Black women. Unequal access to care may contribute to this racial disparity. This study aimed to assess whether survival varied between non-Hispanic Black (NHB) and non-Hispanic White (NHW) women with EC in the Military Health System (MHS) which provides equal access care to its beneficiaries despite racial/ethnic background. METHODS: The study was conducted using data from the U.S. Department of Defense's (DoD) Automated Central Tumor Registry (ACTUR). Study subjects included NHB and NHW women with histologically confirmed and surgically managed EC diagnosed between 1988 and 2013. The study outcome was all-cause death. Overall survival between NHB and NHW women was compared using multivariable Cox modeling. RESULTS: The study included 144 NHB and 1439 NHW women with EC. Kaplan-Meier curves showed NHB women had worse survival than NHW women (log-rank P < 0.0001). The disparity in survival between NHB and NHW women persisted after adjusting for age, diagnosis period, tumor stage, tumor histology/grade, and adjuvant treatment (HR = 1.64, 95% CI = 1.19 to 2.27). Multivariable analyses stratified by tumor features or treatment showed that the racial disparity was confined to women with low-risk features (stage I/II disease or low-grade EC) or no adjuvant treatment. CONCLUSION: There were racial differences in overall survival between NHB and NHW women with EC in the MHS equal access healthcare system, suggesting that factors other than access to care may be related to this racial disparity.

Brain cancer incidence: a comparison of active-duty military and general populations.

BACKGROUND: In the USA, brain cancer disproportionately affects young adults. The US military has a younger age structure than the general population and may have differential exposures related to brain cancer. This study aimed to compare the incidence rates of brain cancer in the active-duty military and general populations to provide clues for future etiologic research. The rates between military service branches were also compared. METHODS: The data for this study were from the Department of Defense's Automated Central Tumor Registry (ACTUR) and the National Cancer Institute's Surveillance, Epidemiology, and End Results 9 (SEER-9) registries. Age- and sex-adjusted incidence rates of malignant neuroepithelial brain cancer among adults 20-54 years of age from 1990-2013 were calculated and compared between the two populations, given as incidence rate ratios (IRRs) with 95% confidence intervals (CIs). RESULTS: The age and sex-adjusted incidence rate for malignant neuroepithelial brain cancer was significantly lower in the active-duty population than in the US general population (IRR = 0.62, 95% CI, 0.56-0.68). The reduced incidence rate in the active-duty population was observed in men, all races, individuals 20-44 of age, and for all histological subtypes and time periods assessed. There were no significant differences in rates between the military service branches. CONCLUSION: The incidence rates of neuroepithelial brain cancer were lower in the active-duty military population than the US general population. This study highlights the need for more research to enhance our understanding of variations in brain cancer incidence between these two populations.

Stage and mortality of low-income patients with cancer: Evidence from SEER-Medicaid.

BACKGROUND: A national data source for identifying patients with cancer enrolled in Medicaid is needed to evaluate cancer care for low-income, publicly insured patients. In this study, a population-based data set of patients diagnosed with cancer and enrolled in Medicaid was created and evaluated. The objective was to compare the characteristics of patients with cancer identified in Surveillance, Epidemiology, and End Results (SEER) data and linked to the Medicaid Analytic eXtract (MAX) Personal Summary files with the characteristics of patients who were not linked to the MAX file. METHODS: All persons in 14 SEER registries diagnosed with cancer from 2006 to 2013 who were or were not linked to the 2006-2013 nationwide MAX files were selected, and patient demographic characteristics were compared for 3 age groups. Common cancer sites and the timing of Medicaid enrollment with respect to patients' cancer diagnoses were reported, and the stage at diagnosis and 4-year mortality were compared by 3 categories of Medicaid enrollment. RESULTS: Approximately 18% of the sample was enrolled in Medicaid within 25 months of diagnosis. Enrollees had a greater proportion of racial/ethnic minorities in comparison with patients who were not enrolled. A late-stage diagnosis was more common among Medicaid patients and particularly among those who enrolled after their diagnosis. For every common cancer site, mortality was highest in the sample of Medicaid patients who enrolled after their diagnosis. CONCLUSIONS: The Medicaid enrollment data newly added to SEER records enhance researchers' ability to investigate research questions related to Medicaid policies and care delivery. For patients enrolled before their diagnosis, Medicaid appears to offer protection against late-stage disease and mortality.

Population-based utilization and costs associated with tyrosine kinase inhibitors for first-line treatment of chronic myelogenous leukemia among elderly patients.

BACKGROUND: Following approval of imatinib, a breakthrough tyrosine kinase inhibitor (TKI), survival significantly improved by more than 20% since 2001 among treated chronic myelogenous leukemia (CML) patients. Subsequently, more expensive second-generation TKIs with varying selectivity profiles have been approved. Population-based studies are needed to evaluate the real-world utilization of TKI therapies, particularly given their escalating costs and recommendations for maintenance therapy. OBJECTIVE: To assess the utilization patterns of first-line TKIs, overall and by specific agent, among elderly CML patients in the United States, and the cost implications. METHODS: CML patients aged 65 years and older at diagnosis between 2007 and 2015 were identified from population-based cancer registries in the linked Surveillance, Epidemiology, and End Results (SEER)-Medicare database. The percentage of CML patients receiving imatinib, dasatinib, or nilotinib within the first year of diagnosis was calculated along with time to first-line treatment initiation. Bivariate comparisons and Cox proportional hazards models were used to identify factors associated with TKI initiation. Average monthly patient responsibility, including patient out-of-pocket (OOP) costs, stratified by Part D low-income subsidy (LIS) status were also calculated. RESULTS: Among the 1,589 CML patients included, receipt of any TKI within 1 year of diagnosis increased from 66.2% to 78.9%. In 2015, the distribution of first-line TKI therapies was 41.3% imatinib, 28.3% dasatinib, and 9.3% nilotinib. Almost 60% of patients initiated TKI treatment within 3 months of diagnosis. Multivariable analysis indicated that TKI use in the first year was lower among the very elderly (aged > 75 years vs. 65-69 years: HR = 0.72; 95% CI = 0.63-0.83), patients with more comorbidities (Hierarchical Condition Category risk score > 2 vs. HR = 0.74, 95% CI = 0.62-0.88), and patients ineligible for LIS (HR = 0.75; 95% CI = 0.65-0.87). Average monthly patient OOP cost was significantly lower for LIS-eligible versus LIS-ineligible patients: imatinib (2016: $12 vs. $487), dasatinib (2016: $34 vs. $557), and nilotinib (2016: $1 vs. $526). CONCLUSIONS: TKI use has increased significantly since 2007. While imatinib remained the most frequently prescribed first-line agent, by 2015 newer TKIs represented one third of the market share. Utilization patterns indicated persistent age, comorbidity, and financial barriers. TKI use is indicated for long-term therapy, and increased adoption of newer, more expensive agents raises concerns about the sustained affordability of CML treatment, particularly among unsubsidized patients. DISCLOSURES: No outside funding supported this study. There are no reported conflicts of interest.

Medical Care Costs Associated with Cancer Survivorship in the United States.

BACKGROUND: The prevalence of cancer survivorship is increasing. In this study, we provide contemporary population-based estimates and projections of the overall and site-specific cancer-attributable medical care costs in the United States. METHODS: We identified survivors aged ≥65 years diagnosed with cancer between 2000 and 2012 from the Surveillance, Epidemiology, and End Results (SEER)-Medicare database and used 2007 to 2013 claims to estimate costs by cancer site, phases of care, and stage at diagnosis. Annualized average cancer-attributable costs for medical care (Medicare Parts A and B) and oral prescription drugs (Medicare Part D) were estimated by subtracting costs between patients with cancer and matched controls. Costs are reported in 2019 U.S. dollars. We combined phase-specific attributable costs with prevalence projections to estimate national costs from 2015 through 2030. RESULTS: Overall annualized average costs were highest in the end-of-life-cancer death phase, followed by the initial and continuing phases (medical care: $105,500, $41,800, and $5,300 and oral prescription drugs: $4,200, $1,800, $1,100, respectively). There was considerable variation in costs by cancer site and stage. Overall national costs in 2015 were $183 billion and projected to increase 34% to $246 billion by 2030, based only on population growth. CONCLUSIONS: Phase of care cancer-attributable cost estimates by cancer site and stage are key inputs for simulation models and cost-effectiveness analyses. IMPACT: The national cancer-attributed medical care costs in the United States are substantial and projected to increase dramatically by 2030, due to population changes alone, reflecting the rising burden of cancer care among cancer survivors.

Estimating Chemotherapy Use Among Patients With a Prior Primary Cancer Diagnosis Using SEER-Medicare Data.

Cancer treatment studies commonly exclude patients with prior primary cancers due to difficulties in ascertaining for which site treatment is intended. Surveillance, Epidemiology, and End Results-Medicare patients 65 years and older diagnosed with an index colon or rectal cancer (CRC) or female breast cancer (BC) between 2004 and 2013 were included. Chemotherapy, defined as "any chemotherapy" and more restrictively as "chemotherapy with confirmatory diagnoses," was ascertained based on claims data within 6 months of index cancer diagnosis by prior cancer history. Any chemotherapy use was slightly lower among patients with a prior cancer (CRC: no prior = 17.4%, prior = 16.1%; BC: no prior = 12.9%, prior = 12.0%). With confirmatory diagnoses required, estimates were lower, especially among patients with a prior cancer (CRC: no prior = 16.8%, prior = 13.6%; BC: no prior = 12.6%, prior = 11.0%). These findings suggest that patients with prior cancers can be included in studies of chemotherapy use; requiring confirmatory diagnoses can increase treatment assignment confidence.

Comparison of Physician Data in Two Data Files Available for Cancer Health Services Research.

INTRODUCTION: Physicians are vital to health-care delivery, but assessing their impact on care can be challenging given limited data. Historically, health services researchers have obtained physician characteristics data from the American Medical Association (AMA) Physician Masterfile. The Center for Medicare and Medicaid Services' Medicare Data on Provider Practice and Specialty (MD-PPAS) file was assessed, as an alternative source of physician data, particularly in the context of cancer health services research. METHODS: We used physician National Provider Identifiers in the MD-PPAS data (2008-2014) to identify physicians in the AMA data current as of July 18, 2016. Within each source, we grouped physicians into six broad specialty groups. Percent agreement and Cohen's kappa coefficient (k) were calculated for age, sex, specialty, and practice state. RESULTS: Among the 698 202 included physicians, there was excellent agreement for age (percent agreement = 97.7%, k = 0.97) and sex (99.4%, k = 0.99) and good agreement for specialty (86.1%, k = 0.80). Within specialty, using AMA as the reference, agreement was lowest for oncologists (77%). Approximately 85.9% of physicians reported the same practice state in both data sets. CONCLUSION: Although AMA data have been commonly used to account for physician-level factors in health services research, MD-PPAS data provide researchers with an alternative option depending on study needs. MD-PPAS data may be optimal if nonphysicians, provider utilization, practice characteristics, and/or temporal changes are of interest. In contrast, the AMA data may be optimal if more granular specialty, physician training, and/or a broader inclusion of physicians is of interest.

Sensitivity of Medicare Data to Identify Oncologists.

BACKGROUND: Health services researchers have studied how care from oncologists impacts treatment and outcomes for cancer patients. These studies frequently identify physician specialty using files from the Center for Medicare and Medicaid Services (CMS) or the American Medical Association (AMA). The completeness of the CMS data resources, individually or combined, to identify oncologists is unknown. This study assessed the sensitivity of CMS data to capture oncologists included in the AMA Physician Masterfile. METHODS: Oncologists were identified from three CMS data resources: physician claims, the National Plan and Provider Enumeration System Registry, and the Medicare Data on Provider Practice and Specialty file. CMS files and AMA data were linked using a unique physician identifier. Sensitivity to identify any oncologists, radiation oncologists (ROs), surgical oncologists (SOs), and medical oncologists (MOs) was calculated for individual and combined CMS files. For oncologists in the AMA data not identified as oncologists in the CMS data, their CMS specialty was assessed. RESULTS: Individual CMS files each captured approximately 83% of the 17 934 oncologists in the AMA Masterfile; combined CMS files captured 90.4%. By specialty, combined CMS data captured 98.2% of ROs, 89.3% of MOs, and 70.1% of SOs. For ROs and SOs in the AMA data not identified as oncologists in the CMS data, their CMS specialty was usually similar to the AMA subspecialty; ROs were radiologists and SOs were surgeons. CONCLUSION: Using combined files from CMS identified most ROs and MOs found in the AMA, but not most SOs. Determining whether to use the AMA data or CMS files for a particular research project will depend on the specific research question and the type of oncologist included in the study.