- Email: [email protected]
Access to Cancer Specialist Care and Treatment in Patients With Advanced Stage Lung Cancer
Original Study
Access to Cancer Specialist Care and Treatment in Patients With Advanced Stage Lung Cancer Apar Kishor Ganti,1 Fred R. Hirsch,2 Murry W. Wynes,3 Arliene Ravelo,4 Suresh S. Ramalingam,5 Raluca Ionescu-Ittu,6 Irina Pivneva,6 Hossein Borghaei7 Abstract MarketScan and Surveillance, Epidemiology, and End ResultseMedicare databases were analyzed separately to evaluate the access to cancer specialists and treatment of patients with advanced stage lung cancer. Between 4% and 12% of the patients were never seen by a cancer specialist, and between 6% and 10% did not receive cancer-directed therapy. Patients seen by a cancer specialist were more likely to receive cancerdirected therapy. Background: Access to specialty care is critical for patients with advanced stage lung cancer. This study assessed access to cancer specialists and cancer treatment in a broad population of patients with advanced stage lung cancer. Materials and Methods: Two study samples were extracted from 2 claims databases and analyzed independently: patients aged 18 years with de novo diagnosis of metastatic lung cancer in the MarketScan database between 2008 and 2014 (commercially insured adult patients; n ¼ 22,268); and patients aged 65 years in the Surveillance, Epidemiology, and End ResultseMedicare database with a diagnosis of advanced nonesmall-cell lung cancer between 2007 and 2011 (Medicare-insured elderly patients; n ¼ 9651). The study period spanned from 6 weeks before the first lung biopsy tied to the initial lung cancer diagnosis until the end of continuous health insurance enrollment, or data availability, or death. Results: Among the commercially insured adults (MarketScan), most patients were seen by a cancer specialist within a month of first lung biopsy (80%), 12% were never seen by a cancer specialist, and 6% did not receive cancer-directed therapy. Among the Medicare-insured elderly patients (SEEReMedicare), the proportions were 79%, 4%, and 10%, respectively. Patients seen by a cancer specialist were more likely to receive cancerdirected therapy (95% vs. 92%, P < .001 and 92% vs. 38%, P < .001, respectively). Conclusion: Between 4% and 12% of patients with advanced stage lung cancer do not have appropriate access to cancer specialist, which appears to negatively affect access to optimal and timely treatment. Clinical Lung Cancer, Vol. -, No. -, 1-11 Published by Elsevier Inc. Keywords: Access to care, Cancer-directed therapy, Metastatic lung cancer, Nonesmall-cell lung cancer, Referral
Introduction Lung cancer is the most common cancer and the leading cause of cancer deaths worldwide.1 Approximately 75% of patients present 1 Veteran’s Affairs NebraskaeWestern Iowa Health Care System, University of Nebraska Medical Center, Omaha, NE 2 Division of Medical Oncology, University of Colorado Anschutz Medical Campus, Aurora, CO 3 International Association for the Study of Lung Cancer (IASLC), Aurora, CO 4 Genentech Inc, South San Francisco, CA 5 Medical Oncology, Emory University Winship Cancer Institute, Atlanta, GA 6 Analysis Group Inc, Montreal, Quebec, Canada 7 Medical Oncology, Fox Chase Cancer Center, Philadelphia, PA
Submitted: Feb 6, 2017; Accepted: Apr 18, 2017 Address for correspondence: Apar Kishor Ganti, MD, MS, FACP, Division of Oncology-Hematology, University of Nebraska Medical Center, 987680 Nebraska Medical Center, Omaha, NE 68198-7680 Fax: (402) 559-6520; e-mail contact: [email protected]
1525-7304/$ - see frontmatter Published by Elsevier Inc. http://dx.doi.org/10.1016/j.cllc.2017.04.010
with advanced disease in stages III and IV at the time of diagnosis,2,3 and approximately 85% have nonesmall-cell lung cancer (NSCLC).4 The prognosis of advanced stage lung cancer is poor, with only about 4% of patients with stage IV NSCLC surviving 5 years after diagnosis.5 Because of the aggressive and progressive nature of the disease, the management of advanced stage lung cancer typically involves different options, including targeted therapy, chemotherapy, and radiotherapy.6 As such, treatment decisions are often complex, and access to cancer specialists is particularly important to ensure that patients receive appropriate care. Indeed, specialist care has been associated with improved survival and greater use of anticancer treatments among lung cancer patients.7-9 Several studies have reported gaps in specialist care for patients with advanced NSCLC in the United States: Goulart et al8 found that 16% of patients diagnosed between 2000 and 2005 were not
Clinical Lung Cancer Month 2017
-1
Access to Cancer Specialist Care seen by a cancer specialist within 6 months of diagnosis; Small et al10 found that 25% of patients diagnosed between 2000 and 2008 did not receive radiotherapy or systemic therapy; and Owonikoko et al11 found that 66% of patients diagnosed between 2004 and 2005 did not receive systemic therapy. Suboptimal cancer care is a particular concern for elderly patients,10-14 who represent the majority of the patient population with advanced stage lung cancer (median age at diagnosis is 70 years4). Because most studies to date focusing on access to specialist care and treatment in advanced stage lung cancer were conducted among patients diagnosed more than a decade ago, their findings may not reflect the current clinical practice, which has been rapidly evolving as a result of recent therapeutic advances. A better understanding of how patients with advanced stage lung cancer are currently managed in clinical practice and which factors determine treatment strategies and referral patterns is crucial to improve patients’ access to specialist care and treatment. This study aimed to provide insights into the patterns of access to cancer specialist care and treatment in 2 diverse populations of patients with advanced stage lung cancer in the United States identified in 2 large population-based data sources: adult patients with commercial health care insurance and elderly patients with Medicare health care insurance.
Materials and Methods Data Sources The study used data from 2 large population-based claims databases: the Truven Health Analytics MarketScan (MarketScan) between 2008 and 2014, and the Surveillance, Epidemiology, and End Results (SEER)eMedicare database between 2007 and 2011 (SEER component) and 2007 and 2013 (Medicare component). This study did not compare patients between the 2 data sources; rather, it was designed to leverage each data source to analyze outcomes among 2 diverse populations of patients with advanced lung cancer. MarketScan Database. The MarketScan is a commercial insurance database that contains data on approximately 50 million individuals covered annually by over 130 health plans and self-insured employers. A mix of academic and community settings and all US census regions are represented. The database contains information on patient demographics, enrollment history, claims for inpatient and outpatient medical services, and pharmacy claims. The years 2008 to 2014 were retrospectively analyzed for this study.
2
-
SEEReMedicare Database. The SEEReMedicare is a federal insurance database that comprises 2 databases linked at the patient level: the SEER cancer registry and the Medicare claims database. The SEER cancer registry contains data on patients diagnosed with cancer across 20 geographic areas representing approximately 30% of the US population. Available information in the SEER registry include patient demographics and detailed cancer characteristics (eg, cancer site, stage, and histology) at the time of the cancer diagnosis. The Medicare claims database contains administrative claims related to hospital care (Part A), outpatient medical services (Part B), and outpatient drug prescriptions (Part D) for patients with Medicare insurance coverage (ie, primarily patients aged 65 years).
Clinical Lung Cancer Month 2017
Approximately 94% of SEER patients aged 65 years are linked to the Medicare database.15,16 Patients with a lung cancer diagnosis in the SEER registry from 2007 to 2011 were selected for the study. Medicare linkage was available for these patients from 2006 to 2013 inclusive. In both databases, data are deidentified and fully comply with the patient confidentiality requirements of the Health Insurance Portability and Accountability Act; thus, studies involving these databases are exempt from institutional review board approval.
Study Design The study used a retrospective cohort design. Patients newly diagnosed with advanced lung cancer in each of the 2 databases were followed until either (1) the end of continuous commercial insurance enrollment in the MarketScan database and the end of continuous Medicare Parts A/B/D insurance plan enrollment in the SEEReMedicare database, (2) the date of death (available in SEEReMedicare data only), or (3) the end of data availability, which was March 31, 2014, for the MarketScan database and December 31, 2013, for the SEEReMedicare database. By design, all patients were required to have 6 months of continuous health insurance plan enrollment before the lung cancer diagnosis (ie, baseline period) and 4 months of follow-up after the lung cancer diagnosis. For the commercially insured adult patients in the MarketScan database, where date of the metastatic lung cancer was not directly available, the date of the first lung biopsy that was followed by a lung cancer diagnosis within a 6-week period was used as a proxy for the date of diagnosis. For the Medicare-insured elderly patients in the SEEReMedicare database, the date of diagnosis was extracted from the SEER records. Access to cancer specialists and treatments was measured from the first lung biopsy date.
Study Samples Figure 1 details the selection of the 2 study samples that were analyzed independently: the sample of commercially insured adult patients in the MarketScan database (study sample 1) and the sample of Medicare-insured elderly patients in the Medicare database (study sample 2). The sample of commercially insured adult patients in the MarketScan database comprised individuals aged 18 years diagnosed with de novo metastatic lung cancer between January 1, 2008, and March 31, 2014. Metastatic lung cancer at the time of diagnosis was identified in the MarketScan claims data based on the presence of International Classification of Diseases, 9th Revision (ICD-9), diagnosis codes for both primary lung cancer (codes 162.xx, excluding 162.0x) and secondary malignant neoplasms (codes 196.xx-198.xx, excluding 196.1x) using previously published algorithms.17 Such algorithms are used to proxy advanced stage lung cancer because information on histology and cancer grade is not available in the MarketScan database. The sample of Medicare-insured elderly patients in the SEEReMedicare database included individuals aged 65 years with a pathologically confirmed lung cancer diagnosis from January 2007 to December 2011. NSCLC histology and advanced cancer stage (IIIB or IV) at the time of diagnosis were identified on the basis of the cancer records in the SEER database.
Apar Kishor Ganti et al Figure 1 Sample Selection Flowchart. aMarketScan Indicates Commercial Insurance Enrollment Both in 6 Months Before and 4 Months After First Eligible Biopsy; SEEReMedicare: Medicare Parts A/B Enrollment Without Enrollment in HMO Both in 6 Months Before and 4 Months After Diagnosis, and Medicare Part D in 4 Months After Diagnosis (Most Patients Were Excluded Because They Did Not Have Medicare Part D Insurance, Followed by Enrollment in HMOs). bMetastatic Lung Cancer Was Identified by Presence of ICD-9 Diagnosis Codes for Secondary Neoplasm (196.x-198.x, excluding 196.1) on ‡ 2 Distinct Dates Starting From 30 Days Before First Eligible Biopsy, of Which One Occurred in 30 Days Before and up to 42 Days After First Eligible Biopsy
Abbreviations: HMO ¼ Health Maintenance Organization; ICD-9 ¼ International Classification of Diseases, 9th Revision; LC ¼ lung cancer; NSCLC ¼ nonesmall-cell lung cancer; SEER ¼ surveillance, epidemiology, and end results.
Study Cohorts Within each study sample, patients were grouped into 2 mutually exclusive study cohorts: (1) patients who were seen by a cancer specialist in the period starting from 6 weeks before the first lung biopsy until the end of follow-up and (2) patients who were not seen by a cancer specialist over the same time period. Cancer specialists were defined on the basis of physician specialty information available in each data source. In the MarketScan database, cancer specialists included oncologists, hematologists, and radiologists. However, because the MarketScan database does not distinguish between radiology and radiation oncology specialties, radiologists were considered as cancer specialists only if the visits were made > 1 week after the first lung biopsy and no biopsy or biomarker test was performed on the same date. In the SEEReMedicare database, cancer specialists included oncologists, hematologist oncologists, surgical oncologists, radiation oncologists, and other oncologists (ie, gynecologic oncologists, orthopedic oncologists, and orthopedic musculoskeletal oncologists). The physician specialty in the SEEReMedicare database was obtained through linkage to the American Medical Association physician files.18
Outcomes and Statistical Analyses All analyses were performed independently in each of the 2 study samples (ie, the sample of commercially insured adult patients in the MarketScan database and the sample of Medicare-insured elderly
patients in the SEEReMedicare database). Patient characteristics measured during the baseline period were compared between the study cohorts in each sample by the Wilcoxon rank sum test for continuous variables or chi-square tests for categorical variables. The timing of the first visit to the cancer specialist was evaluated in terms of (1) when the first visit with a cancer specialist occurred (ie, the 6 weeks before or at the time of their first lung biopsy, between their first lung biopsy and initiation of their first treatment, after initiation of their first treatment, or no visit) and (2) the time from the first lung biopsy to the first cancer specialist visit, which was assessed by reversed Kaplan-Meier analysis. Referral to cancer specialists was assessed as the proportion of patients seen by a cancer specialist in the 8 weeks after the first lung biopsy among mutually exclusive patient groups defined on the basis of the level of specialization in cancer/lung disease of the physicians seen in the 6 weeks before the first lung biopsy. Four mutually exclusive groups were defined hierarchically: (1) patients seen by cancer specialists with or without other specialists (ie, highest level of specialization in cancer/lung disease); (2) patients seen by pulmonologists with or without specialists other than cancer specialists; (3) patients seen by internists or family physicians with or without specialists other than cancer specialists or pulmonologists; and (4) patients not seen by any of the specialists listed above. The use of cancer-directed therapy after the first lung biopsy was compared between the study cohorts within each study sample by
Clinical Lung Cancer Month 2017
-3
Access to Cancer Specialist Care Table 1 Characteristics of 2 Diverse Populations of Patients With Advanced Lung Cancer
Characteristic
Commercially Insured Adult Medicare-insured Patients With Elderly Patients With Advanced Metastatic NSCLC Lung Cancer (N [ 9651) (N [ 22,268)
Demographic Data Age Age at lung cancer diagnosis, y, median (IQR) <65 years, n (%) 65-79 years, n (%) 80þ years, n (%) Male, n (%) Non-Hispanic white race/ethnicity, n (%)
63 (57-73) 12,684 (57.0) 7426 (33.3)
74 (70-79) — 7244 (75.1)
2158 (9.7)
2407 (24.9)
11,370 (51.1)
4658 (48.3)
NA
7808 (80.9)
NA
IIIB
NA
3409 (35.3)
IV
NA
6242 (64.7)
Tumor Size
NA
Median (IQR), mmc
NA
2 cm
NA
880 (9.1)
2-5 cm
NA
3930 (40.7)
5þ cm
NA
2890 (29.9)
Not reported
NA
1951 (20.2) 118 (1.2)
40 (27-60)
Adenosquamous
NA
Large-cell carcinoma
NA
251 (2.6)
NA
4675 (48.4)
Married/common law
NA
4511 (46.7)
Adenocarcinoma, NOS
Widowed/divorced/separated
NA
3903 (40.4)
Squamous-cell carcinoma
NA
2898 (30.0)
Other/unknown
NA
314 (3.3)
NSCLC, NOSd
NA
1691 (17.5)
Other types (eg, giant-cell carcinoma, spindle-cell carcinoma, pleomorphic carcinoma)
NA
18 (0.2)
NA
6446 (66.8)
2007
NA
1897 (19.7)
2008
1994 (9.0)
1898 (19.7)
2009
4154 (18.7)
1924 (19.9)
20% residents living below poverty line in zip code of patient residence
NA
2316 (24.0)
Distant metastasese at diagnosis, n (%)
Living in Area With Lower Education,a N (%) 20% residents with <12 grades in zip code of patient residence
NA
2441 (25.3)
Region of Residence, N (%)
Year of Lung Cancer Diagnosis, N (%)
Northeast
5297 (23.8)
1895 (19.6)
2010
3991 (17.9)
1952 (20.2)
North Central
6105 (27.4)
1307 (13.5)
2011
4622 (20.8)
1980 (20.5)
South
7136 (32.0)
2736 (28.3)
2012
4385 (19.7)
NA
West
3117 (14.0)
3609 (37.4)
2013
3122 (14.0)
NA
613 (2.8)
104 (1.1)
Big metropolitan (counties of metro areas of 1 million population)
NA
5066 (52.5)
Metropolitan (counties in metro areas of up to 1 million population)
NA
2673 (27.7)
Urban (urban population of 20,000 adjacent or nonadjacent to a metro area)
NA
Less urban (urban population of 2500 to <20,000 adjacent or nonadjacent to a metro area)
NA
Rural (completely rural or <2500 adjacent or nonadjacent to a metro area)/Unknown
NA
Unknown/Hawaii Urban Residency,b N (%)
Cancer Characteristics at Time of Lung Cancer Diagnosis
-
AJCC 6 Cancer Stage, N (%)
923 (9.6)
Living in Area With High Poverty,a N (%)
4
Characteristic
Commercially Insured Adult Medicare-insured Patients With Elderly Patients With Advanced Metastatic NSCLC Lung Cancer (N [ 9651) (N [ 22,268)
NSCLC Histology, N (%)
Marital Status, N (%) Never married
Table 1 Continued
Clinical Lung Cancer Month 2017
591 (6.1)
1060 (11.0)
261 (2.7)
Comorbid Conditions in 6 Months Before Lung Cancer Diagnosis Overall Comorbidity Burden (Charlson Comorbidity Indexf), Median (IQR)
1 (0-2)
2 (1-3)
0-2
19,397 (87.1)
6556 (67.9)
3-6
2778 (12.5)
2858 (29.6)
7þ
93 (0.4)
237 (2.5)
Physical Comorbidities (Frequency >5% in ‡1 of samples), n (%) Anemia
3001 (13.5)
2673 (27.7)
Bleeding
3485 (15.7)
2130 (22.1)
Chronic pulmonary disease
8013 (36.0)
5486 (56.8)
Congestive heart failure
1244 (5.6)
1578 (16.4)
Diabetes
4142 (18.6)
2938 (30.4)
Fluid and electrolyte disorders
2399 (10.8)
1505 (15.6)
Apar Kishor Ganti et al brachytherapy, and surgical resection. The use of palliative care service was assessed among untreated patients.
Table 1 Continued
Characteristic Hypertension
Commercially Insured Adult Medicare-insured Patients With Elderly Patients With Advanced Metastatic NSCLC Lung Cancer (N [ 9651) (N [ 22,268) 10,376 (46.7)
7136 (73.9)
Hypothyroidism
2005 (9.0)
1643 (17.0)
Peripheral vascular disease
2259 (10.2)
2356 (24.4)
Pulmonary circulation disorders
851 (3.8)
529 (5.5)
Renal failure
840 (3.8)
966 (10.0)
Valvular disease
1700 (7.6)
1402 (14.5)
Weight loss
1617 (7.3)
1275 (13.2)
16,207 (72.9)
8595 (89.1)
2354 (10.6)
1360 (14.1)
Any of the above Mental Comorbidities, N (%) Mood disordersg Anxiety disorders
1203 (5.4)
679 (7.0)
Depressive disorders
1355 (6.1)
850 (8.8)
Other Conditions (Including Psychologic Factors Affected or Determined by Physical Condition)h
1713 (7.7)
864 (9.0)
Substance-related disorders
3538 (15.9)
1926 (20.0)
1 mental disorderi
7424 (33.4)
3916 (40.6)
Abbreviations: AJCC 6 ¼ 6th edition of the American Joint Committee on Cancer (AJCC) Cancer Staging Manual (used by SEER to derive cancer stage); DSM ¼ Diagnostic and Statistical Manual of Mental Disorders; IQR ¼ interquartile range; mLC ¼ metastatic lung cancer; NA ¼ not available; NOS ¼ not otherwise specified; NSCLC ¼ nonesmall-cell lung cancer. a Estimated from the American Community Survey (ACS), 2008-2012. b Refers to a classification scheme that distinguishes metropolitan (metro) counties by the population size of their metro area and nonmetropolitan (nonmetro) counties by degree of urbanization and adjacency to a metro area or nonmetro areas. c Four patients with tumor sizes > 20 cm were reset to a maximum value of 20 cm. d A nonspecific histology code is used in the SEER database when the cancer registry abstractor did not find enough information (eg, test results) to confirm with certitude the specific cancer histology (patients in this group may have in truth any specific histologies listed above). e Could include malignant pleural effusion; malignant pericardial effusion; extension to contralateral lung, extension to contralateral main stem bronchus, separate tumor nodules in contralateral lung, pleural tumor foci or nodules on contralateral lung; pleural tumor foci or nodules on the ipsilateral lung separate from direct invasion; distant lymph nodes, including cervical nodes extension to skeletal muscle, sternum, skin of chest; abdominal organs, other distant metastases (alone or in combination); could also include stated as M1a, M1b, or M1 (NOS) without additional information on distant metastasis. f Overall burden of disease due to comorbidities other than the cancer/metastatic cancer (ie, no points were accounted for the cancer/metastatic cancer diagnosis). g Mood disorders included anxiety, bipolar, depressive, dissociative, factitious, and somatoform disorders. h Other conditions that may be a focus of clinical attention, including psychologic factors affecting medical condition, medication-induced movement disorders, other medication-induced disorders, additional conditions that may be a focus of clinical attention, among others. i Mental disorders were based on Diagnostic and Statistical Manual of Mental Disorders, 5th edition.
chi-square tests. Cancer-directed therapy included systemic therapy, radiotherapy, stereotactic radiosurgery, radiofrequency ablation, and lung surgery. Systemic therapy was further divided into the following categories: epidermal growth factor receptor (EGFR) inhibitors (ie, afatinib, erlotinib, gefitinib), anaplastic lymphoma kinase (ALK) inhibitors (ie, crizotinib, which was the only ALK inhibitor approved during the period covered by the data), other targeted agents (eg, bevacizumab, cetuximab), and chemotherapy (eg, taxanes, platinum-based agents, pemetrexed). Lung surgery was further divided into the following categories: cryoablation,
Results A total of 22,268 commercially insured adult patients from the MarketScan database and 9651 Medicare-insured elderly patients from the SEEReMedicare database were included in the analysis (Figure 1). Patients in the 2 samples were followed for a median of 10 and 11 months after the index date, respectively.
Commercially Insured Adult Patients (MarketScan Database) Among the commercially insured adult patients, median age the time of lung cancer diagnosis was 63 years and about half were males (Table 1). The median Charlson comorbidity index exclusive of the burden associated with the cancer and metastatic cancer diagnoses was 1 (Table 1). With respect to cancer specialist access, as many as 2603 (12%) commercially insured adult patients were not seen by a cancer specialist over the entire follow-up (Table 2, Figure 2), with the remaining 88% being distributed as follows with respect to the timing of the first visit to a cancer specialist: 28% were first seen before or at the time of their first lung biopsy, 46% after the lung biopsy but before treatment initiation, and 14% after treatment initiation (Figure 2). Overall, 80% of the commercially insured adult patients were seen by a cancer specialist before, at, or within 1 month of biopsy, and 95% were seen within 6 months of biopsy (Figure 3). Because the MarketScan database has limited data on sociodemographic and cancer characteristics, few differences could be observed between the cohorts of patients seen and not seen by a cancer specialist. However, some differences were noted: compared to patients seen by a cancer specialist, those not seen were slightly older, they were more likely to reside in the Northeast and North Central regions, and a greater proportion had diabetes with chronic complications (P < .05; Supplemental Table 1 in the online version). Furthermore, a higher proportion of patients seen by a cancer specialist in the 6 weeks before the first lung biopsy were seen by cancer specialists in the 8-week postbiopsy period compared to those seen before biopsy by other physicians (99% vs. 40%-96%; P < .05 for all comparisons; Figure 4). The rates of referral to cancer specialist increased in line with the level of specialization in cancer/ lung disease of the physicians seen in the 6 weeks before the first lung biopsy (Figure 4). With respect to treatment, a significantly greater proportion of commercially insured adult patients who were seen by a cancer specialist received cancer-directed therapy compared to patients who were not seen by a cancer specialist (95% vs. 92%; P < .01; Table 2). The most common cancer-directed therapy was chemotherapy, accounting for 79% of each study cohort (P ¼ .99). The proportion of patients receiving radiotherapy was similar between the commercially insured adult patients seen and not seen by a cancer specialist (72% for both; P ¼ .94). However, slightly more patients seen by a cancer specialist had a surgical resection compared to those not seen by a cancer specialist (18% vs. 15%, respectively, P < .01; Table 2). Palliative services were received by 33% of the patients seen by cancer specialist and 26% of those not seen (P ¼ .05; Table 2).
Clinical Lung Cancer Month 2017
-5
Access to Cancer Specialist Care Medicare-insured Elderly Patients (SEEReMedicare Database) Among the Medicare-insured elderly patients, median age at the time of lung cancer diagnosis was 74 years, and about half were men; the median Charlson comorbidity index was 2 (Table 1). With respect to cancer specialist access, 367 (4%) Medicareinsured elderly patients were not seen by a cancer specialist over the entire follow-up (Table 2; Figure 2), with the remaining 96% being distributed as follows with respect to the timing of the first visit to a cancer specialist: 24% were first seen before or at the time of their first lung biopsy, 62% after the lung biopsy but before treatment initiation, and 9% after treatment initiation (Figure 2). Overall, 79% of the Medicare-insured elderly patients were seen by a cancer specialist before, at, or within 1 month of biopsy, and 87% were seen within 6 months of biopsy (Figure 3). Compared to patients seen by a cancer specialist, the patients not seen were older (median age 78 vs. 74 years); more likely to be female (61% vs. 51%); more likely to be widowed, divorced, or separated (48% vs. 40%); more likely to reside in areas with higher poverty (35% vs. 24%) and lower education levels (31% vs. 25%); more likely to have smaller tumor size (median 35 vs. 40 mm) and stage IIIB (vs. stage IV) NSCLC (43% vs. 35%); and showed a greater comorbidity burden (P < .01 for all listed differences, but the 2 cohorts were not statistically significant different in terms of squamous vs. nonsquamous histology; Supplemental Table 1 in the online version). In addition, a higher proportion of patients seen by a cancer specialist in the 6 weeks before the first lung biopsy were seen by cancer specialists in the 8-week postbiopsy period compared to those seen before biopsy by other physicians (97% vs. 82%-86%; P < .05 for all comparisons; Figure 4). The rates of referral to cancer specialist increased in line with the level of specialization in cancer/ lung disease of the physicians seen in the 6 weeks before the first lung biopsy, with the exception of the referral rates of the patients seen before biopsy by pulmonologists versus internists or family physicians (Figure 4). With respect to treatment, a significantly greater proportion of patients who were seen by a cancer specialist received cancerdirected therapy compared to those not seen by a cancer specialist (92% vs. 38%; P < .01; Table 2). For the cohort of patients seen by a cancer specialist, the most common cancer-directed therapy was chemotherapy (75%), while for the cohort of patients not seen by a cancer specialist, surgical resection was most common (57%). Radiotherapy was used by 72% of the patients seen by a cancer specialist versus 33% of those not seen (P < .01; Table 2). Palliative services were received by most patients in both study cohorts (86% and 87%, P ¼ .70; Table 2).
Discussion
6
-
Using 2 large patient samples from the MarketScan database and the SEEReMedicare database, this study assessed patterns of access to specialist care and treatment among patients with advanced stage lung cancer in the United States. This analysis fills a gap in the literature by providing an overview of the recent trends in the management of advanced stage lung cancer in US clinical practice (years 2007-2014). The study found that a substantial number of patients with advanced stage lung cancer did not have appropriate access to cancer specialists: approximately 20% of patients in both
Clinical Lung Cancer Month 2017
samples were not seen by a cancer specialist within 1 month of the first lung biopsy, while 5% and 13%, respectively, were not seen by a cancer specialist within 6 months of the first lung biopsy. Access to cancer-directed therapy was also found to be limited, with 6% of the commercially insured adult patients from the MarketScan database and 10% of the Medicare-insured elderly patients from the SEEReMedicare database not receiving any cancer-directed therapy. To our knowledge, 2 previous studies investigated access to cancer specialists, but both were conducted before the availability of many of the currently used therapies: Goulart et al8 found that 16% of Medicare patients diagnosed with advanced NSCLC in the United States in 2000-2005 were not seen by a cancer specialist within 6 months from diagnosis, and Vinod et al9 found that 11% of patients of all ages diagnosed with lung cancer in Australia in 2001-2002 were not seen by a lung cancer specialist after diagnosis. With respect to access to cancer-directed therapy, the most up-todate estimates are also from 2 older studies: Small et al10 found 25% untreated among patients of all ages with a diagnosis of NSCLC in the National Cancer Database in 2000-2008, and Owonikoko et al11 found 66% untreated in the Medicare population with advanced stage NSCLC in 1988-2005. The findings of the current study for the Medicare-insured sample in 2007-2011 suggest an improvement over time in access to cancer-directed therapy. Given that none of the studies to date could ascertain whether the therapy received was appropriate for the patients’ specific situation, further studies are needed to assess whether or not patients receive optimal treatment. While it is possible that best supportive care could in some circumstances be the optimal treatment, the fact that lack of treatment was more common among patients not seen by a cancer specialist suggests that some patients who opt out of treatment may not make an informed decision that is based on a good understanding of all treatment options available to them. On the basis of the analysis of the sample of Medicare-insured elderly patients from the SEEReMedicare database, an important finding of the current study is that not seeing a cancer specialist reduces patient access to optimal and timely treatment. This finding is in line with a prior study that found that seeing an oncologist was one of the strongest predictors of receiving chemotherapy for stage IV NSCLC.19 Other factors found in the current study to be associated with not seeing a cancer specialist among Medicareinsured elderly and/or commercially insured adult patients included older age; female sex; being widowed, divorced, or separated; residing in areas with high poverty and lower education; being diagnosed with earlier stage disease and having a smaller tumor size; and having a greater comorbidity burden. However, squamous versus nonsquamous histology did not seem to affect access to cancer specialists. Previous studies also found that patients with advanced stage lung cancer were less likely to be seen by or referred to a cancer specialist if they were older, had a lower socioeconomic status, and exhibited a greater comorbidity burden.8,9,19-22 By contrast, an older study by Earle et al19 found no correlation between being older and not seeing an oncologist among patients with stage IV NSCLC diagnosed in 1991-1993, although patients with low socioeconomic status and those not receiving care in a teaching hospital were less likely to be seen by an oncologist. While the study
Apar Kishor Ganti et al Table 2 Use of Cancer-directed Therapies and Palliative Care Services Commercially Insured Adult Patients With Metastatic Lung Cancer
a
Cancer-directed Therapy
Patients Seen by Cancer Specialist
Patients Not Seen by Cancer Specialist
N [ 19,665 (88.3%)
N [ 2603 (11.7%)
P
Medicare-insured Elderly Patients With Advanced NSCLC Patients Seen by Cancer Specialist
Patients Not Seen by Cancer Specialist
N [ 9284 (96.2%)
N [ 367 (3.8%)
P
[A]
[B]
[A] vs. [B]
[C]
[D]
[C] vs. [D]
18,627 (94.7)
2384 (91.6)
<.01*
8521 (91.8)
139 (37.9)
<.01*
Systemic therapy
15,391 (82.6)
1962 (82.3)
.69
6910 (81.1)
63 (45.3)
<.01*
EGFR inhibitors
1896 (10.2)
231 (9.7)
.46
1876 (22.0)
21 (15.1)
.05
ALK inhibitors
104 (0.6)
7 (0.3)
.09
<11 (<0.1)c
0 (0.0)
1.00
4125 (22.1)
468 (19.6)
<.01*
<11 (<7.9)c
<.01*
Patients Treated With Cancer-Directed Therapy (Any), N (%) Therapy Type,b N (%)
Other Targeted Therapies
1466 (17.2)
14,624 (78.5)
1872 (78.5)
.99
6415 (75.3)
46 (33.1)
<.01*
13,413 (72.0)
1715 (71.9)
.94
6094 (71.5)
12 (8.6)
<.01*
Stereotactic radiosurgery
2917 (15.7)
294 (12.3)
<.01*
901 (10.6)
<11 (<7.9)c
<.01*
Radiofrequency ablation
281 (1.5)
30 (1.3)
.34
134 (1.6)
<11 (<7.9)c
.48
Cryoablation
320 (1.7)
49 (2.1)
.24
116 (1.4)
<11 (<7.9)c
.71
Brachytherapy
145 (0.8)
22 (0.9)
.45
68 (0.8)
<11 (<7.9)c
1.00
3388 (18.2)
360 (15.1)
<.01*
1340 (15.7)
79 (56.8)
<.01*
1038 (5.3)
219 (8.4)
<.01*
763 (8.2)
228 (62.1)
<.01*
57 (26.0)
.05
655 (85.8)
198 (86.8)
.70
Chemotherapy Radiotherapy
Surgery
Surgical resection Patients Not Treated, N (%) Use of palliative care services, n (% of nonrelated patients) Hospice or skilled nursing facility
339 (32.7)
Hospice only
163 (15.7)
16 (7.3)
<.01*
414 (54.3)
104 (45.6)
.02*
Skilled nursing facility only
138 (13.3)
39 (17.8)
.08
87 (11.4)
45 (19.7)
<.01*
Hospice and skilled nursing facility
38 (3.7)
2 (0.9)
.03*
154 (20.2)
49 (21.5)
.67
699 (67.3)
162 (74.0)
.05
108 (14.2)
30 (13.2)
.70
No hospice or skilled nursing facility
*P .05 by chi-square test. Abbreviations: ALK ¼ anaplastic lymphoma kinase; EGFR ¼ epidermal growth factor receptor; NSCLC ¼ non-small-cell lung cancer. a For patients who were seen or were not seen by a cancer specialist from 6 weeks before index date until the end of follow-up. b Used alone, in combination, or sequentially. c Per data user agreement with the National Cancer Institute, exact count cannot be reported for < 11 patients.
by Earle et al was conducted among patients with stage IV NSCLC more than 25 years ago, their findings nevertheless suggest that the type of facility where patients receive care might impact timely and appropriate access to cancer specialist care, an association that was not studied in the current study because this information was unavailable. While further studies are needed to clarify the interplay between these factors, physicians need to be aware that nonmedical factors such as patient age or sociodemographic characteristics may create barriers to appropriate care and therapy. Some noteworthy differences were observed in the current study between the treatment patterns of the commercially insured adult patients identified in the MarketScan database and the Medicareinsured elderly patients identified in the SEEReMedicare database: among the commercially insured adult patients who were not
seen by a cancer specialist, 92% received cancer-directed therapy, 15% received treatment that may not be consistent with clinical guidelines (surgical resection), and 26% received palliative care; among the Medicare-insured elderly patients not seen by a cancer specialist, the respective proportions were 38%, 57%, and 87%. Given the age difference between the 2 samples (adult patients of all ages vs. elderly patients), these observed differences in the use of cancer-directed therapies and palliative care services suggest a possible treatment bias against the elderly among physicians other than cancer specialists. However, other factors may have also contributed to these differences, such as the data structure differences between the MarketScan and SEEReMedicare databases. For example, because physician specialty measurement is less refined in the MarketScan database than in SEEReMedicare database
Clinical Lung Cancer Month 2017
-7
Access to Cancer Specialist Care Figure 2 Time of First Visit With Cancer Specialist. Time of First Visit With Cancer Specialist in 2 Diverse Populations of Patients With Advanced Lung Cancer: Commercially Insured Adult Patients and Medicare-insured Elderly Patients
Abbreviation: TX ¼ treatment.
(because of the linkage with American Medical Association physician files in the latter database), it is possible that more commercially insured adult patients were seen by a cancer specialist than what was reported in the current study. It is also possible that some of the patients not seen by a cancer specialist were in fact seen by a midlevel provider who could have worked in collaboration with a cancer specialist to provide treatment, a situation that is likely to
occur more frequently for commercially insured patients compared to Medicare-insured patients. Finally, the 62% of the commercially insured adult patients in the MarketScan database who did not receive any cancer-directed therapy aged 65 years (data not shown) may have had dual commercial and Medicare coverage and may have had their palliative services fully covered through Medicare.
Figure 3 Time From First Lung Biopsy to First Cancer Specialist Visit. Time From First Lung Biopsy to First Cancer Specialist Visit in 2 Diverse Populations of Patients With Advanced Lung Cancer: Commercially Insured Adult Patients and Medicare-insured Elderly Patients. The x-axis Was Censored at 1 Year Because Most Events Occur Within 1 Year. Patients Who Had Visit With Cancer Specialist in 6 Weeks Before First Lung Biopsy Were Considered to Have Seen Cancer Specialist at Time 0 (Date of First Lung Biopsy)
8
-
Clinical Lung Cancer Month 2017
Apar Kishor Ganti et al Figure 4 Patterns of Referral to Cancer Specialist. Patterns of Referral to Cancer Specialist in 2 Diverse Populations of Patients With Advanced Lung Cancer: Commercially Insured Adult Patients and Medicare-insured Elderly Patients. Specialist Groups in 6 Weeks Before First Lung Biopsy Were Mutually Exclusive and Are Defined Hierarchically on Basis of Level of Specialization of Physicians. a“± Other Specialists” Indicates “Including or Excluding Other Specialists.” bPatients Were Stratified by Type of Specialist Seen in 6 Weeks Before First Lung Biopsy. Each Color Indicates Type of Specialist Seen in 6 Weeks Before First Lung Biopsy, as Reported at Left. cAll Differences Between Groups Were Statistically Significant at P < .05. All Differences Between Groups Were Statistically Significant at P < .05, With Exception of Difference Between Groups Seen by Pulmonologist and Internist/Family Doctor Before First Lung Biopsy (P [ .59)
Abbreviation: CS ¼ cancer specialist.
The findings of this study should be interpreted in light of some limitations. First, MarketScan commercial claims databases do not include information on lung cancer histology or disease staging. While the sample of commercially insured adults could not be restricted to patients with advanced (stage IIIB and IV) NSCLC, stage IV is likely captured by the metastatic disease proxy used in this study. Also, based on the 85% prevalence of NSCLC in lung cancer,4 it is likely that patients with NSCLC histology represent the majority of the patients included in the study. Second, although claims data were available until the end of 2013 in the SEEReMedicare database, patients diagnosed with advanced
NSCLC after the year 2011 were not available for inclusion in the study. Thus, potential changes in the use of targeted therapies that occurred in patients diagnosed after 2011 were not captured in the results for the Medicare-insured elderly patients. Third, claims data did not include sufficient information to determine which of the specialists seen before therapy initiation actually made treatment decisions. However, in the case of patients seen by cancer specialists, it is likely that treatment decisions were made by the cancer specialists. Last, the study may have been subject to general limitations of claims data, including occasional inaccuracies in coding diagnoses, procedures, or pharmacy claims.
Clinical Lung Cancer Month 2017
-9
Access to Cancer Specialist Care Conclusion The current study found that a substantial proportion of patients with advanced stage lung cancer and commercial or Medicare insurance in the period 2007-2014 did not see a cancer specialist within 1 month of the first lung biopsy. Furthermore, access to cancer-directed therapy among Medicare-insured elderly patients not seen by a cancer specialist was significantly lower than that of those seen by a cancer specialist. While older age and lower socioeconomic status appear to be correlated with reduced access to a cancer specialist, further research should be directed toward understanding and addressing disparities in access to appropriate cancer care.
Clinical Practice Points This analysis fills a gap in the literature by providing an overview
of the recent trends in the access to cancer specialist care and treatment of patients with advanced stage lung cancer in US clinical practice (years 2007-2014). Using 2 data sources, the study identified 2 diverse populations of patients: commercially insured adult patients with metastatic lung cancer (MarketScan database) and Medicare-insured elderly patients with advanced NSCLC (SEEReMedicare database). The study found that a substantial number of patients with advanced stage lung cancer did not have access to cancer specialists: approximately 20% of patients in both samples were not seen by a cancer specialist within 1 month of the first lung biopsy, while 5% to 13%, depending on the sample, were not seen by a cancer specialist within 6 months of the first lung biopsy. Several demographic (eg, older age; being widowed, divorced, or separated; residing in areas with high poverty) and clinical (eg, having a smaller tumor size; and having a greater comorbidity burden) factors were found to be associated with limited access to cancer specialists. Physicians need to be aware that both medical and nonmedical factors can create barriers to appropriate care and therapy. Access to cancer-directed therapy was also found to be limited, with 6% to 10% of the patients not receiving any cancer-directed therapy, especially among patients not seen by a cancer specialist. Patients not seen by a cancer specialist may not be aware of all treatment options available to them.
Acknowledgments Medical writing assistance was provided by Cinzia Metallo, PhD, an employee of Analysis Group Inc. supported by Genentech Inc, South San Francisco, CA.
Disclosure A.R. is an employee of Genentech Inc and owns stock/stock options. R.I.-I. and I.P. are employees of Analysis Group Inc, which has received consultancy fees from Genentech Inc. A.K.G. has received institutional clinical research grants from Pfizer, New Link Genetics, Amgen, AstraZeneca, Merck, Janssen, and Bristol-Myers Squibb and has acted as a paid consultant for Pfizer, Ariad, and
10
-
Clinical Lung Cancer Month 2017
Biodesix for work performed outside of the current study. H.B. has acted as a paid consultant for BMS, Genentech, Lilly, Pfizer, Boerhringer-Ingelheim, Trovegen, and Celgene for work performed outside of the current study and as an uncompensated consultant for Merck for work performed outside of the current study, and has received honoraria from Celgene. F.R.H. has received institutional clinical research grants from Lilly, InClone Systems, Genentech, Bristol-Myers Squibb, Amgen, Celgene, and Bayer; holds an institutional (University of Colorado) patent for EGFR fluorescence insitu hybridization and immunohistochemistry for prediction of outcome in patients treated with EGFR inhibitors; and has acted as a paid consultant for Bristol-Myers Squibb, AstraZeneca, Genentech, Roche, HTG Inc, Pfizer, Lilly, Merck, Clovis Oncology, and Ventana Medical Systems for work performed outside of the current study. S.S.R. has acted as a paid consultant for Abbvie, Amgen, AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Genentech/Roche, and Lilly/InClone for work performed outside of the current study. M.W.W. has no conflict of interest to disclose.
Supplemental Data Supplemental table accompanying this article can be found in the online version at http://dx.doi.org/10.1016/j.cllc.2017.04.010.
References 1. World Health Organization; International Agency for Research on Cancer. GLOBOCAN 2012: estimated cancer incidence, mortality and prevalence worldwide in 2012, Available at: http://globocan.iarc.fr/Pages/fact_sheets_cancer. aspx?cancer¼lung. Accessed: July 31, 2016. 2. Hurria A, Kris MG. Management of lung cancer in older adults. CA Cancer J Clin 2003; 53:325-41. 3. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2007. CA Cancer J Clin 2007; 57:43-66. 4. Key statistics for lung cancer, Available at: http://www.cancer.org/cancer/ lungcancer-non-smallcell/detailedguide/non-small-cell-lung-cancer-what-is-nonsmall-cell-lung-cancer. Accessed: July 31, 2016. 5. National Cancer Institute. SEER Stat Fact Sheets: lung and bronchus cancer, Available at: http://seer.cancer.gov/statfacts/html/lungb.html. Accessed: August 1, 2016. 6. National Cancer Institute. Nonesmall cell lung cancer treatment, Available at: http://www.cancer.gov/types/lung/hp/non-small-cell-lung-treatment-pdq#section/_ 514. Accessed: July 31, 2016. 7. Gregor A, Thomson CS, Brewster DH, et al. Management and survival of patients with lung cancer in Scotland diagnosed in 1995: results of a national population based study. Thorax 2001; 56:212-7. 8. Goulart BH, Reyes CM, Fedorenko CR, et al. Referral and treatment patterns among patients with stages III and IV nonesmall-cell lung cancer. J Oncol Pract 2013; 9:42-50. 9. Vinod SK, O’Connell DL, Simonella L, et al. Gaps in optimal care for lung cancer. J Thorac Oncol 2008; 3:871-9. 10. Small AC, Tsao CK, Moshier EL, et al. Prevalence and characteristics of patients with metastatic cancer who receive no anticancer therapy. Cancer 2012; 118: 5947-54. 11. Owonikoko TK, Ragin C, Chen Z, et al. Real-world effectiveness of systemic agents approved for advanced nonesmall cell lung cancer: a SEEReMedicare analysis. Oncologist 2013; 18:600-10. 12. Sacher AG, Le LW, Lau A, Earle CC, Leighl NB. Real-world chemotherapy treatment patterns in metastatic nonesmall cell lung cancer: are patients undertreated? Cancer 2015; 121:2562-9. 13. Coate LE, Massey C, Hope A, et al. Treatment of the elderly when cure is the goal: the influence of age on treatment selection and efficacy for stage III nonesmall cell lung cancer. J Thorac Oncol 2011; 6:537-44. 14. Ramesh HSJ, Boase T, Audisio RA. Risk assessment for cancer surgery in elderly patients. Clin Interv Aging 2006; 1:221-7. 15. Potosky AL, Riley GF, Lubitz JD, Mentnech RM, Kessler LG. Potential for cancer related health services research using a linked Medicareetumor registry database. Med Care 1993; 31:732-48. 16. National Cancer Institute. SEEReMedicare: how the SEER and Medicare data are linked, Available at: http://healthcaredelivery.cancer.gov/seermedicare/overview/ linked.html. Accessed: July 31, 2016.
Apar Kishor Ganti et al 17. Whyte JL, Engel-Nitz NM, Teitelbaum A, Gomez Rey G, Kallich JD. An evaluation of algorithms for identifying metastatic breast, lung, or colorectal cancer in administrative claims data. Med Care 2015; 53:e49-57. 18. National Cancer Institute. SEEReMedicare: encrypted variables, Available at: http://healthcaredelivery.cancer.gov/seermedicare/privacy/variables.html?&url¼/ seermedicare/privacy/variables.html#release. Accessed: July 31, 2016. 19. Earle CC, Neumann PJ, Gelber RD, Weinstein MC, Weeks JC. Impact of referral patterns on the use of chemotherapy for lung cancer. J Clin Oncol 2002; 20: 1786-92.
20. Li X, Butts C, Fenton D, King K, Scarfe A, Winget M. Utilization of oncology services and receipt of treatment: a comparison between patients with breast, colon, rectal, or lung cancer. Ann Oncol 2011; 22:1902-9. 21. Dalton SO, Frederiksen BL, Jacobsen E, et al. Socioeconomic position, stage of lung cancer and time between referral and diagnosis in Denmark, 2001-2008. Br J Cancer 2011; 105:1042-8. 22. Wang J, Kuo YF, Freeman J, Goodwin JS. Increasing access to medical oncology consultation in older patients with stage II-IIIA nonesmall-cell lung cancer. Med Oncol 2008; 25:125-32.
Clinical Lung Cancer Month 2017
- 11
Access to Cancer Specialist Care Supplemental Table 1 Characteristics of 2 Populations of Patients With Advanced Lung Cancer Commercially Insured Adult Patients
Seen by Cancer Not Seen by Specialist Cancer Specialist
Seen by Cancer Not Seen by Specialist Cancer Specialist Characteristica
N [ 19,665
N [ 2603
Medicare-insured Elderly Patients
P
N [ 9284
N [ 367
P
Demographic Data Age Age at Diagnosis, y, Median (IQR) <65 years, n (%)
63 (57-72)
65 (58-74)
<.01*
11,387 (57.9)
1297 (49.8)
<.01*
65-79 years, n (%)
6421 (32.7)
1005 (38.6)
<.01*
80þ years, n (%)
1857 (9.4)
301 (11.6)
10,003 (50.9)
1367 (52.5)
74 (70-79) —
78 (72-84)
<.01*
—
—
7044 (75.9)
200 (54.5)
<.01*
<.01*
2240 (24.1)
167 (45.5)
<.01*
.11
4514 (48.6)
144 (39.2)
<.01*
NA
NA
—
7517 (81.0)
291 (79.3)
.42
Never married
NA
NA
—
879 (9.5)
44 (12.0)
.11
Married/common law
NA
NA
—
4382 (47.2)
129 (35.1)
<.01*
Widowed/divorced/separated
NA
NA
—
3726 (40.1)
177 (48.2)
<.01*
Other/unknown
NA
NA
—
297 (3.2)
17 (4.6)
.13
NA
NA
—
2189 (23.6)
127 (34.6)
<.01*
NA
NA
—
2326 (25.1)
115 (31.3)
<.01*
Male sex, n (%) Non-Hispanic white race/ethnicity, n (%) Marital Status, N (%)
Living in Area With High Poverty,b N (%) 20% residents living below poverty line in zip code of patient residence Living in Area With Lower Education,b N (%) 20% residents with <12 grades in zip code of patient residence Region of Residence, N (%) Northeast
4599 (23.4)
698 (26.8)
<.01*
1836 (19.8)
59 (16.1)
.08
North Central
5341 (27.2)
764 (29.4)
.02*
1268 (13.7)
39 (10.6)
.10
South
6468 (32.9)
668 (25.7)
<.01*
2623 (28.3)
113 (30.8)
.29
West
2734 (13.9)
383 (14.7)
.26
3454 (37.2)
155 (42.2)
.05
523 (2.7)
90 (3.5)
.02*
<11 (<0.1)
<11 (<3.0)
.13
Big metropolitan (counties of metro areas of 1 million population)
NA
NA
—
4891 (52.7)
175 (47.7)
.06
Metropolitan (counties in metro areas of up to 1 million population)
NA
NA
—
2568 (27.7)
105 (28.6)
.69
Urban (urban population of 20,000 adjacent or nonadjacent to a metro area)
NA
NA
—
566 (6.1)
25 (6.8)
.58
Less urban (urban population of 2500 to <20,000 adjacent or nonadjacent to a metro area)
NA
NA
—
1011 (10.9)
49 (13.4)
.14
Rural (completely rural or <2500 adjacent or nonadjacent to a metro area)/unknown
NA
NA
—
248 (2.7)
13 (3.5)
.31
Unknown/otherc Urban Residency,d N (%)
Cancer Characteristics at Time of Diagnosis AJCC 6 Cancer Stage, N (%) IIIB
NA
NA
—
3250 (35.0)
159 (43.3)
<.01*
IV
NA
NA
—
6034 (65.0)
208 (56.7)
<.01*
2 cm
NA
NA
—
838 (9.0)
42 (11.4)
.12
2-5 cm
NA
NA
—
3780 (40.7)
150 (40.9)
.95
5þ cm
NA
NA
—
2807 (30.2)
83 (22.6)
<.01*
Not reported
NA
NA
—
1859 (20.0)
92 (25.1)
.02*
Tumor Size Median (IQR), mm,e
11.e1
-
Clinical Lung Cancer Month 2017
40 (27-60)
35 (25-55)
<.01*
Apar Kishor Ganti et al Supplemental Table 1 Continued Commercially Insured Adult Patients
Seen by Cancer Not Seen by Specialist Cancer Specialist
Seen by Cancer Not Seen by Specialist Cancer Specialist Characteristica
N [ 19,665
Medicare-insured Elderly Patients
N [ 2603
P
N [ 9284
N [ 367
P
NSCLC Histology, N (%) Adenosquamousc
NA
NA
—
<11 (<0.1)
<11 (<3.0)
.09
Large-cell carcinoma
NA
NA
—
240 (2.6)
11 (3.0)
.63
Adenocarcinoma, NOS
NA
NA
—
4496 (48.4)
179 (48.8)
.90
Squamous-cell carcinoma
NA
NA
—
2787 (30.0)
111 (30.2)
.93
f
NSCLC, NOS
NA
NA
—
1626 (17.5)
65 (17.7)
.92
Otherc
NA
NA
—
<11 (<0.1)
<11 (<3.0)
.40
6227 (67.1)
219 (59.7)
<.01*
1824 (19.6)
73 (19.9)
.91
Distant metastasesg at diagnosis, n (%) Year of Diagnosis, N (%) 2007
NA
NA
—
2008
1707 (8.7)
287 (11.0)
<.01*
1821 (19.6)
77 (21.0)
.52
2009
3638 (18.5)
516 (19.8)
.10
1841 (19.8)
83 (22.6)
.19
2010
3532 (18.0)
459 (17.6)
.68
1890 (20.4)
62 (16.9)
.11
2011
4133 (21.0)
489 (18.8)
<.01*
1908 (20.6)
72 (19.6)
.66
2012
3878 (19.7)
507 (19.5)
.77
NA
NA
—
2013
2777 (14.1)
345 (13.3)
.23
NA
NA
—
1 (0-2)
.56
Comorbid Conditions in 6 Months Before NSCLC Diagnosis Overall Comorbidity Burden (Charlson Comorbidity Indexh) 2 (1-3)
<.01*
0-2
17,145 (87.2)
2252 (86.5)
.35
6334 (68.2)
222 (60.5)
<.01*
3-6
2440 (12.4)
338 (13.0)
.42
2730 (29.4)
128 (34.9)
7þ
80 (0.4)
13 (0.5)
.60
220 (2.4)
17 (4.6)
Median (IQR)
1 (0-2)
2 (1-3)
.02* <.01*
Physical Comorbidities (Statistically Significant Differences in ‡1 Sample), N (%) Anemia
2656 (13.5)
345 (13.3)
.73
2544 (27.4)
129 (35.1)
<.01*
Congestive heart failure
1085 (5.5)
159 (6.1)
.22
1473 (15.9)
105 (28.6)
<.01*
Complicated diabetes
815 (4.2)
130 (5.0)
.04*
835 (9.0)
46 (12.5)
.02*
Uncomplicated diabetes
2827 (14.4)
370 (14.2)
.83
2001 (21.6)
56 (15.3)
<.01*
Peripheral vascular disease
1982 (10.1)
277 (10.7)
.37
2241 (24.1)
115 (31.3)
<.01*
742 (3.8)
109 (4.2)
.30
500 (5.4)
29 (7.9)
1414 (7.2)
203 (7.8)
.26
1208 (13.0)
67 (18.3)
<.01*
Mood disordersi
2079 (10.6)
275 (10.6)
.99
1291 (13.9)
69 (18.8)
<.01*
Anxiety disorders
1068 (5.4)
135 (5.2)
.61
642 (6.9)
37 (10.1)
.02*
Pulmonary circulation disorders Weight loss
.04*
Mental Comorbidities, N (%)
*P .05. Abbreviations: AJCC 6 ¼ 6th edition of the American Joint Committee on Cancer (AJCC) Cancer Staging Manual (used by SEER to derive cancer stage); CS ¼ cancer specialist; DSM ¼ Diagnostic and Statistical Manual of Mental Disorders; IQR ¼ interquartile range; NA ¼ not available. a Patients who have seen or not seen a cancer specialist from 6 weeks before index date until end of follow-up. b Estimated from the American Community Survey (ACS), 2008-2012. c Data user agreement with the National Cancer Institute forbids the presentation of frequencies/percentages for subgroups of < 11 patients. d Refers to a classification scheme that distinguishes metropolitan (metro) counties by the population size of their metro area and nonmetropolitan (nonmetro) counties by degree of urbanization and adjacency to a metro area or nonmetro areas. e Four patients with tumor sizes > 20 cm were reset to a maximum value of 20 cm. f A nonspecific histology code is used in the SEER database when the cancer registry abstractor did not find enough information (eg, test results) to confirm with certitude the specific cancer histology (patients in this group may have in truth any specific histologies listed above). g Could include malignant pleural effusion; malignant pericardial effusion; extension to contralateral lung, extension to contralateral main stem bronchus, separate tumor nodules in contralateral lung, pleural tumor foci or nodules on contralateral lung; pleural tumor foci or nodules on the ipsilateral lung separate from direct invasion; distant lymph nodes, including cervical nodes extension to skeletal muscle, sternum, skin of chest; abdominal organs, other distant metastases (alone or in combination); could also include “stated as M1a, M1b or M1 (NOS) without additional information on distant metastasis.” h Overall burden of disease due to comorbidities other than the cancer/metastatic cancer (ie, no points were accounted for the cancer/metastatic cancer diagnosis). i Mood disorders included anxiety, bipolar, depressive, dissociative, factitious, and somatoform disorders.
Clinical Lung Cancer Month 2017
- 11.e2
Access to Cancer Specialist Care and Treatment in Patients With Advanced Stage Lung Cancer Apar Kishor Ganti,1 Fred R. Hirsch,2 Murry W. Wynes,3 Arliene Ravelo,4 Suresh S. Ramalingam,5 Raluca Ionescu-Ittu,6 Irina Pivneva,6 Hossein Borghaei7 Abstract MarketScan and Surveillance, Epidemiology, and End ResultseMedicare databases were analyzed separately to evaluate the access to cancer specialists and treatment of patients with advanced stage lung cancer. Between 4% and 12% of the patients were never seen by a cancer specialist, and between 6% and 10% did not receive cancer-directed therapy. Patients seen by a cancer specialist were more likely to receive cancerdirected therapy. Background: Access to specialty care is critical for patients with advanced stage lung cancer. This study assessed access to cancer specialists and cancer treatment in a broad population of patients with advanced stage lung cancer. Materials and Methods: Two study samples were extracted from 2 claims databases and analyzed independently: patients aged 18 years with de novo diagnosis of metastatic lung cancer in the MarketScan database between 2008 and 2014 (commercially insured adult patients; n ¼ 22,268); and patients aged 65 years in the Surveillance, Epidemiology, and End ResultseMedicare database with a diagnosis of advanced nonesmall-cell lung cancer between 2007 and 2011 (Medicare-insured elderly patients; n ¼ 9651). The study period spanned from 6 weeks before the first lung biopsy tied to the initial lung cancer diagnosis until the end of continuous health insurance enrollment, or data availability, or death. Results: Among the commercially insured adults (MarketScan), most patients were seen by a cancer specialist within a month of first lung biopsy (80%), 12% were never seen by a cancer specialist, and 6% did not receive cancer-directed therapy. Among the Medicare-insured elderly patients (SEEReMedicare), the proportions were 79%, 4%, and 10%, respectively. Patients seen by a cancer specialist were more likely to receive cancerdirected therapy (95% vs. 92%, P < .001 and 92% vs. 38%, P < .001, respectively). Conclusion: Between 4% and 12% of patients with advanced stage lung cancer do not have appropriate access to cancer specialist, which appears to negatively affect access to optimal and timely treatment. Clinical Lung Cancer, Vol. -, No. -, 1-11 Published by Elsevier Inc. Keywords: Access to care, Cancer-directed therapy, Metastatic lung cancer, Nonesmall-cell lung cancer, Referral
Introduction Lung cancer is the most common cancer and the leading cause of cancer deaths worldwide.1 Approximately 75% of patients present 1 Veteran’s Affairs NebraskaeWestern Iowa Health Care System, University of Nebraska Medical Center, Omaha, NE 2 Division of Medical Oncology, University of Colorado Anschutz Medical Campus, Aurora, CO 3 International Association for the Study of Lung Cancer (IASLC), Aurora, CO 4 Genentech Inc, South San Francisco, CA 5 Medical Oncology, Emory University Winship Cancer Institute, Atlanta, GA 6 Analysis Group Inc, Montreal, Quebec, Canada 7 Medical Oncology, Fox Chase Cancer Center, Philadelphia, PA
Submitted: Feb 6, 2017; Accepted: Apr 18, 2017 Address for correspondence: Apar Kishor Ganti, MD, MS, FACP, Division of Oncology-Hematology, University of Nebraska Medical Center, 987680 Nebraska Medical Center, Omaha, NE 68198-7680 Fax: (402) 559-6520; e-mail contact: [email protected]
1525-7304/$ - see frontmatter Published by Elsevier Inc. http://dx.doi.org/10.1016/j.cllc.2017.04.010
with advanced disease in stages III and IV at the time of diagnosis,2,3 and approximately 85% have nonesmall-cell lung cancer (NSCLC).4 The prognosis of advanced stage lung cancer is poor, with only about 4% of patients with stage IV NSCLC surviving 5 years after diagnosis.5 Because of the aggressive and progressive nature of the disease, the management of advanced stage lung cancer typically involves different options, including targeted therapy, chemotherapy, and radiotherapy.6 As such, treatment decisions are often complex, and access to cancer specialists is particularly important to ensure that patients receive appropriate care. Indeed, specialist care has been associated with improved survival and greater use of anticancer treatments among lung cancer patients.7-9 Several studies have reported gaps in specialist care for patients with advanced NSCLC in the United States: Goulart et al8 found that 16% of patients diagnosed between 2000 and 2005 were not
Clinical Lung Cancer Month 2017
-1
Access to Cancer Specialist Care seen by a cancer specialist within 6 months of diagnosis; Small et al10 found that 25% of patients diagnosed between 2000 and 2008 did not receive radiotherapy or systemic therapy; and Owonikoko et al11 found that 66% of patients diagnosed between 2004 and 2005 did not receive systemic therapy. Suboptimal cancer care is a particular concern for elderly patients,10-14 who represent the majority of the patient population with advanced stage lung cancer (median age at diagnosis is 70 years4). Because most studies to date focusing on access to specialist care and treatment in advanced stage lung cancer were conducted among patients diagnosed more than a decade ago, their findings may not reflect the current clinical practice, which has been rapidly evolving as a result of recent therapeutic advances. A better understanding of how patients with advanced stage lung cancer are currently managed in clinical practice and which factors determine treatment strategies and referral patterns is crucial to improve patients’ access to specialist care and treatment. This study aimed to provide insights into the patterns of access to cancer specialist care and treatment in 2 diverse populations of patients with advanced stage lung cancer in the United States identified in 2 large population-based data sources: adult patients with commercial health care insurance and elderly patients with Medicare health care insurance.
Materials and Methods Data Sources The study used data from 2 large population-based claims databases: the Truven Health Analytics MarketScan (MarketScan) between 2008 and 2014, and the Surveillance, Epidemiology, and End Results (SEER)eMedicare database between 2007 and 2011 (SEER component) and 2007 and 2013 (Medicare component). This study did not compare patients between the 2 data sources; rather, it was designed to leverage each data source to analyze outcomes among 2 diverse populations of patients with advanced lung cancer. MarketScan Database. The MarketScan is a commercial insurance database that contains data on approximately 50 million individuals covered annually by over 130 health plans and self-insured employers. A mix of academic and community settings and all US census regions are represented. The database contains information on patient demographics, enrollment history, claims for inpatient and outpatient medical services, and pharmacy claims. The years 2008 to 2014 were retrospectively analyzed for this study.
2
-
SEEReMedicare Database. The SEEReMedicare is a federal insurance database that comprises 2 databases linked at the patient level: the SEER cancer registry and the Medicare claims database. The SEER cancer registry contains data on patients diagnosed with cancer across 20 geographic areas representing approximately 30% of the US population. Available information in the SEER registry include patient demographics and detailed cancer characteristics (eg, cancer site, stage, and histology) at the time of the cancer diagnosis. The Medicare claims database contains administrative claims related to hospital care (Part A), outpatient medical services (Part B), and outpatient drug prescriptions (Part D) for patients with Medicare insurance coverage (ie, primarily patients aged 65 years).
Clinical Lung Cancer Month 2017
Approximately 94% of SEER patients aged 65 years are linked to the Medicare database.15,16 Patients with a lung cancer diagnosis in the SEER registry from 2007 to 2011 were selected for the study. Medicare linkage was available for these patients from 2006 to 2013 inclusive. In both databases, data are deidentified and fully comply with the patient confidentiality requirements of the Health Insurance Portability and Accountability Act; thus, studies involving these databases are exempt from institutional review board approval.
Study Design The study used a retrospective cohort design. Patients newly diagnosed with advanced lung cancer in each of the 2 databases were followed until either (1) the end of continuous commercial insurance enrollment in the MarketScan database and the end of continuous Medicare Parts A/B/D insurance plan enrollment in the SEEReMedicare database, (2) the date of death (available in SEEReMedicare data only), or (3) the end of data availability, which was March 31, 2014, for the MarketScan database and December 31, 2013, for the SEEReMedicare database. By design, all patients were required to have 6 months of continuous health insurance plan enrollment before the lung cancer diagnosis (ie, baseline period) and 4 months of follow-up after the lung cancer diagnosis. For the commercially insured adult patients in the MarketScan database, where date of the metastatic lung cancer was not directly available, the date of the first lung biopsy that was followed by a lung cancer diagnosis within a 6-week period was used as a proxy for the date of diagnosis. For the Medicare-insured elderly patients in the SEEReMedicare database, the date of diagnosis was extracted from the SEER records. Access to cancer specialists and treatments was measured from the first lung biopsy date.
Study Samples Figure 1 details the selection of the 2 study samples that were analyzed independently: the sample of commercially insured adult patients in the MarketScan database (study sample 1) and the sample of Medicare-insured elderly patients in the Medicare database (study sample 2). The sample of commercially insured adult patients in the MarketScan database comprised individuals aged 18 years diagnosed with de novo metastatic lung cancer between January 1, 2008, and March 31, 2014. Metastatic lung cancer at the time of diagnosis was identified in the MarketScan claims data based on the presence of International Classification of Diseases, 9th Revision (ICD-9), diagnosis codes for both primary lung cancer (codes 162.xx, excluding 162.0x) and secondary malignant neoplasms (codes 196.xx-198.xx, excluding 196.1x) using previously published algorithms.17 Such algorithms are used to proxy advanced stage lung cancer because information on histology and cancer grade is not available in the MarketScan database. The sample of Medicare-insured elderly patients in the SEEReMedicare database included individuals aged 65 years with a pathologically confirmed lung cancer diagnosis from January 2007 to December 2011. NSCLC histology and advanced cancer stage (IIIB or IV) at the time of diagnosis were identified on the basis of the cancer records in the SEER database.
Apar Kishor Ganti et al Figure 1 Sample Selection Flowchart. aMarketScan Indicates Commercial Insurance Enrollment Both in 6 Months Before and 4 Months After First Eligible Biopsy; SEEReMedicare: Medicare Parts A/B Enrollment Without Enrollment in HMO Both in 6 Months Before and 4 Months After Diagnosis, and Medicare Part D in 4 Months After Diagnosis (Most Patients Were Excluded Because They Did Not Have Medicare Part D Insurance, Followed by Enrollment in HMOs). bMetastatic Lung Cancer Was Identified by Presence of ICD-9 Diagnosis Codes for Secondary Neoplasm (196.x-198.x, excluding 196.1) on ‡ 2 Distinct Dates Starting From 30 Days Before First Eligible Biopsy, of Which One Occurred in 30 Days Before and up to 42 Days After First Eligible Biopsy
Abbreviations: HMO ¼ Health Maintenance Organization; ICD-9 ¼ International Classification of Diseases, 9th Revision; LC ¼ lung cancer; NSCLC ¼ nonesmall-cell lung cancer; SEER ¼ surveillance, epidemiology, and end results.
Study Cohorts Within each study sample, patients were grouped into 2 mutually exclusive study cohorts: (1) patients who were seen by a cancer specialist in the period starting from 6 weeks before the first lung biopsy until the end of follow-up and (2) patients who were not seen by a cancer specialist over the same time period. Cancer specialists were defined on the basis of physician specialty information available in each data source. In the MarketScan database, cancer specialists included oncologists, hematologists, and radiologists. However, because the MarketScan database does not distinguish between radiology and radiation oncology specialties, radiologists were considered as cancer specialists only if the visits were made > 1 week after the first lung biopsy and no biopsy or biomarker test was performed on the same date. In the SEEReMedicare database, cancer specialists included oncologists, hematologist oncologists, surgical oncologists, radiation oncologists, and other oncologists (ie, gynecologic oncologists, orthopedic oncologists, and orthopedic musculoskeletal oncologists). The physician specialty in the SEEReMedicare database was obtained through linkage to the American Medical Association physician files.18
Outcomes and Statistical Analyses All analyses were performed independently in each of the 2 study samples (ie, the sample of commercially insured adult patients in the MarketScan database and the sample of Medicare-insured elderly
patients in the SEEReMedicare database). Patient characteristics measured during the baseline period were compared between the study cohorts in each sample by the Wilcoxon rank sum test for continuous variables or chi-square tests for categorical variables. The timing of the first visit to the cancer specialist was evaluated in terms of (1) when the first visit with a cancer specialist occurred (ie, the 6 weeks before or at the time of their first lung biopsy, between their first lung biopsy and initiation of their first treatment, after initiation of their first treatment, or no visit) and (2) the time from the first lung biopsy to the first cancer specialist visit, which was assessed by reversed Kaplan-Meier analysis. Referral to cancer specialists was assessed as the proportion of patients seen by a cancer specialist in the 8 weeks after the first lung biopsy among mutually exclusive patient groups defined on the basis of the level of specialization in cancer/lung disease of the physicians seen in the 6 weeks before the first lung biopsy. Four mutually exclusive groups were defined hierarchically: (1) patients seen by cancer specialists with or without other specialists (ie, highest level of specialization in cancer/lung disease); (2) patients seen by pulmonologists with or without specialists other than cancer specialists; (3) patients seen by internists or family physicians with or without specialists other than cancer specialists or pulmonologists; and (4) patients not seen by any of the specialists listed above. The use of cancer-directed therapy after the first lung biopsy was compared between the study cohorts within each study sample by
Clinical Lung Cancer Month 2017
-3
Access to Cancer Specialist Care Table 1 Characteristics of 2 Diverse Populations of Patients With Advanced Lung Cancer
Characteristic
Commercially Insured Adult Medicare-insured Patients With Elderly Patients With Advanced Metastatic NSCLC Lung Cancer (N [ 9651) (N [ 22,268)
Demographic Data Age Age at lung cancer diagnosis, y, median (IQR) <65 years, n (%) 65-79 years, n (%) 80þ years, n (%) Male, n (%) Non-Hispanic white race/ethnicity, n (%)
63 (57-73) 12,684 (57.0) 7426 (33.3)
74 (70-79) — 7244 (75.1)
2158 (9.7)
2407 (24.9)
11,370 (51.1)
4658 (48.3)
NA
7808 (80.9)
NA
IIIB
NA
3409 (35.3)
IV
NA
6242 (64.7)
Tumor Size
NA
Median (IQR), mmc
NA
2 cm
NA
880 (9.1)
2-5 cm
NA
3930 (40.7)
5þ cm
NA
2890 (29.9)
Not reported
NA
1951 (20.2) 118 (1.2)
40 (27-60)
Adenosquamous
NA
Large-cell carcinoma
NA
251 (2.6)
NA
4675 (48.4)
Married/common law
NA
4511 (46.7)
Adenocarcinoma, NOS
Widowed/divorced/separated
NA
3903 (40.4)
Squamous-cell carcinoma
NA
2898 (30.0)
Other/unknown
NA
314 (3.3)
NSCLC, NOSd
NA
1691 (17.5)
Other types (eg, giant-cell carcinoma, spindle-cell carcinoma, pleomorphic carcinoma)
NA
18 (0.2)
NA
6446 (66.8)
2007
NA
1897 (19.7)
2008
1994 (9.0)
1898 (19.7)
2009
4154 (18.7)
1924 (19.9)
20% residents living below poverty line in zip code of patient residence
NA
2316 (24.0)
Distant metastasese at diagnosis, n (%)
Living in Area With Lower Education,a N (%) 20% residents with <12 grades in zip code of patient residence
NA
2441 (25.3)
Region of Residence, N (%)
Year of Lung Cancer Diagnosis, N (%)
Northeast
5297 (23.8)
1895 (19.6)
2010
3991 (17.9)
1952 (20.2)
North Central
6105 (27.4)
1307 (13.5)
2011
4622 (20.8)
1980 (20.5)
South
7136 (32.0)
2736 (28.3)
2012
4385 (19.7)
NA
West
3117 (14.0)
3609 (37.4)
2013
3122 (14.0)
NA
613 (2.8)
104 (1.1)
Big metropolitan (counties of metro areas of 1 million population)
NA
5066 (52.5)
Metropolitan (counties in metro areas of up to 1 million population)
NA
2673 (27.7)
Urban (urban population of 20,000 adjacent or nonadjacent to a metro area)
NA
Less urban (urban population of 2500 to <20,000 adjacent or nonadjacent to a metro area)
NA
Rural (completely rural or <2500 adjacent or nonadjacent to a metro area)/Unknown
NA
Unknown/Hawaii Urban Residency,b N (%)
Cancer Characteristics at Time of Lung Cancer Diagnosis
-
AJCC 6 Cancer Stage, N (%)
923 (9.6)
Living in Area With High Poverty,a N (%)
4
Characteristic
Commercially Insured Adult Medicare-insured Patients With Elderly Patients With Advanced Metastatic NSCLC Lung Cancer (N [ 9651) (N [ 22,268)
NSCLC Histology, N (%)
Marital Status, N (%) Never married
Table 1 Continued
Clinical Lung Cancer Month 2017
591 (6.1)
1060 (11.0)
261 (2.7)
Comorbid Conditions in 6 Months Before Lung Cancer Diagnosis Overall Comorbidity Burden (Charlson Comorbidity Indexf), Median (IQR)
1 (0-2)
2 (1-3)
0-2
19,397 (87.1)
6556 (67.9)
3-6
2778 (12.5)
2858 (29.6)
7þ
93 (0.4)
237 (2.5)
Physical Comorbidities (Frequency >5% in ‡1 of samples), n (%) Anemia
3001 (13.5)
2673 (27.7)
Bleeding
3485 (15.7)
2130 (22.1)
Chronic pulmonary disease
8013 (36.0)
5486 (56.8)
Congestive heart failure
1244 (5.6)
1578 (16.4)
Diabetes
4142 (18.6)
2938 (30.4)
Fluid and electrolyte disorders
2399 (10.8)
1505 (15.6)
Apar Kishor Ganti et al brachytherapy, and surgical resection. The use of palliative care service was assessed among untreated patients.
Table 1 Continued
Characteristic Hypertension
Commercially Insured Adult Medicare-insured Patients With Elderly Patients With Advanced Metastatic NSCLC Lung Cancer (N [ 9651) (N [ 22,268) 10,376 (46.7)
7136 (73.9)
Hypothyroidism
2005 (9.0)
1643 (17.0)
Peripheral vascular disease
2259 (10.2)
2356 (24.4)
Pulmonary circulation disorders
851 (3.8)
529 (5.5)
Renal failure
840 (3.8)
966 (10.0)
Valvular disease
1700 (7.6)
1402 (14.5)
Weight loss
1617 (7.3)
1275 (13.2)
16,207 (72.9)
8595 (89.1)
2354 (10.6)
1360 (14.1)
Any of the above Mental Comorbidities, N (%) Mood disordersg Anxiety disorders
1203 (5.4)
679 (7.0)
Depressive disorders
1355 (6.1)
850 (8.8)
Other Conditions (Including Psychologic Factors Affected or Determined by Physical Condition)h
1713 (7.7)
864 (9.0)
Substance-related disorders
3538 (15.9)
1926 (20.0)
1 mental disorderi
7424 (33.4)
3916 (40.6)
Abbreviations: AJCC 6 ¼ 6th edition of the American Joint Committee on Cancer (AJCC) Cancer Staging Manual (used by SEER to derive cancer stage); DSM ¼ Diagnostic and Statistical Manual of Mental Disorders; IQR ¼ interquartile range; mLC ¼ metastatic lung cancer; NA ¼ not available; NOS ¼ not otherwise specified; NSCLC ¼ nonesmall-cell lung cancer. a Estimated from the American Community Survey (ACS), 2008-2012. b Refers to a classification scheme that distinguishes metropolitan (metro) counties by the population size of their metro area and nonmetropolitan (nonmetro) counties by degree of urbanization and adjacency to a metro area or nonmetro areas. c Four patients with tumor sizes > 20 cm were reset to a maximum value of 20 cm. d A nonspecific histology code is used in the SEER database when the cancer registry abstractor did not find enough information (eg, test results) to confirm with certitude the specific cancer histology (patients in this group may have in truth any specific histologies listed above). e Could include malignant pleural effusion; malignant pericardial effusion; extension to contralateral lung, extension to contralateral main stem bronchus, separate tumor nodules in contralateral lung, pleural tumor foci or nodules on contralateral lung; pleural tumor foci or nodules on the ipsilateral lung separate from direct invasion; distant lymph nodes, including cervical nodes extension to skeletal muscle, sternum, skin of chest; abdominal organs, other distant metastases (alone or in combination); could also include stated as M1a, M1b, or M1 (NOS) without additional information on distant metastasis. f Overall burden of disease due to comorbidities other than the cancer/metastatic cancer (ie, no points were accounted for the cancer/metastatic cancer diagnosis). g Mood disorders included anxiety, bipolar, depressive, dissociative, factitious, and somatoform disorders. h Other conditions that may be a focus of clinical attention, including psychologic factors affecting medical condition, medication-induced movement disorders, other medication-induced disorders, additional conditions that may be a focus of clinical attention, among others. i Mental disorders were based on Diagnostic and Statistical Manual of Mental Disorders, 5th edition.
chi-square tests. Cancer-directed therapy included systemic therapy, radiotherapy, stereotactic radiosurgery, radiofrequency ablation, and lung surgery. Systemic therapy was further divided into the following categories: epidermal growth factor receptor (EGFR) inhibitors (ie, afatinib, erlotinib, gefitinib), anaplastic lymphoma kinase (ALK) inhibitors (ie, crizotinib, which was the only ALK inhibitor approved during the period covered by the data), other targeted agents (eg, bevacizumab, cetuximab), and chemotherapy (eg, taxanes, platinum-based agents, pemetrexed). Lung surgery was further divided into the following categories: cryoablation,
Results A total of 22,268 commercially insured adult patients from the MarketScan database and 9651 Medicare-insured elderly patients from the SEEReMedicare database were included in the analysis (Figure 1). Patients in the 2 samples were followed for a median of 10 and 11 months after the index date, respectively.
Commercially Insured Adult Patients (MarketScan Database) Among the commercially insured adult patients, median age the time of lung cancer diagnosis was 63 years and about half were males (Table 1). The median Charlson comorbidity index exclusive of the burden associated with the cancer and metastatic cancer diagnoses was 1 (Table 1). With respect to cancer specialist access, as many as 2603 (12%) commercially insured adult patients were not seen by a cancer specialist over the entire follow-up (Table 2, Figure 2), with the remaining 88% being distributed as follows with respect to the timing of the first visit to a cancer specialist: 28% were first seen before or at the time of their first lung biopsy, 46% after the lung biopsy but before treatment initiation, and 14% after treatment initiation (Figure 2). Overall, 80% of the commercially insured adult patients were seen by a cancer specialist before, at, or within 1 month of biopsy, and 95% were seen within 6 months of biopsy (Figure 3). Because the MarketScan database has limited data on sociodemographic and cancer characteristics, few differences could be observed between the cohorts of patients seen and not seen by a cancer specialist. However, some differences were noted: compared to patients seen by a cancer specialist, those not seen were slightly older, they were more likely to reside in the Northeast and North Central regions, and a greater proportion had diabetes with chronic complications (P < .05; Supplemental Table 1 in the online version). Furthermore, a higher proportion of patients seen by a cancer specialist in the 6 weeks before the first lung biopsy were seen by cancer specialists in the 8-week postbiopsy period compared to those seen before biopsy by other physicians (99% vs. 40%-96%; P < .05 for all comparisons; Figure 4). The rates of referral to cancer specialist increased in line with the level of specialization in cancer/ lung disease of the physicians seen in the 6 weeks before the first lung biopsy (Figure 4). With respect to treatment, a significantly greater proportion of commercially insured adult patients who were seen by a cancer specialist received cancer-directed therapy compared to patients who were not seen by a cancer specialist (95% vs. 92%; P < .01; Table 2). The most common cancer-directed therapy was chemotherapy, accounting for 79% of each study cohort (P ¼ .99). The proportion of patients receiving radiotherapy was similar between the commercially insured adult patients seen and not seen by a cancer specialist (72% for both; P ¼ .94). However, slightly more patients seen by a cancer specialist had a surgical resection compared to those not seen by a cancer specialist (18% vs. 15%, respectively, P < .01; Table 2). Palliative services were received by 33% of the patients seen by cancer specialist and 26% of those not seen (P ¼ .05; Table 2).
Clinical Lung Cancer Month 2017
-5
Access to Cancer Specialist Care Medicare-insured Elderly Patients (SEEReMedicare Database) Among the Medicare-insured elderly patients, median age at the time of lung cancer diagnosis was 74 years, and about half were men; the median Charlson comorbidity index was 2 (Table 1). With respect to cancer specialist access, 367 (4%) Medicareinsured elderly patients were not seen by a cancer specialist over the entire follow-up (Table 2; Figure 2), with the remaining 96% being distributed as follows with respect to the timing of the first visit to a cancer specialist: 24% were first seen before or at the time of their first lung biopsy, 62% after the lung biopsy but before treatment initiation, and 9% after treatment initiation (Figure 2). Overall, 79% of the Medicare-insured elderly patients were seen by a cancer specialist before, at, or within 1 month of biopsy, and 87% were seen within 6 months of biopsy (Figure 3). Compared to patients seen by a cancer specialist, the patients not seen were older (median age 78 vs. 74 years); more likely to be female (61% vs. 51%); more likely to be widowed, divorced, or separated (48% vs. 40%); more likely to reside in areas with higher poverty (35% vs. 24%) and lower education levels (31% vs. 25%); more likely to have smaller tumor size (median 35 vs. 40 mm) and stage IIIB (vs. stage IV) NSCLC (43% vs. 35%); and showed a greater comorbidity burden (P < .01 for all listed differences, but the 2 cohorts were not statistically significant different in terms of squamous vs. nonsquamous histology; Supplemental Table 1 in the online version). In addition, a higher proportion of patients seen by a cancer specialist in the 6 weeks before the first lung biopsy were seen by cancer specialists in the 8-week postbiopsy period compared to those seen before biopsy by other physicians (97% vs. 82%-86%; P < .05 for all comparisons; Figure 4). The rates of referral to cancer specialist increased in line with the level of specialization in cancer/ lung disease of the physicians seen in the 6 weeks before the first lung biopsy, with the exception of the referral rates of the patients seen before biopsy by pulmonologists versus internists or family physicians (Figure 4). With respect to treatment, a significantly greater proportion of patients who were seen by a cancer specialist received cancerdirected therapy compared to those not seen by a cancer specialist (92% vs. 38%; P < .01; Table 2). For the cohort of patients seen by a cancer specialist, the most common cancer-directed therapy was chemotherapy (75%), while for the cohort of patients not seen by a cancer specialist, surgical resection was most common (57%). Radiotherapy was used by 72% of the patients seen by a cancer specialist versus 33% of those not seen (P < .01; Table 2). Palliative services were received by most patients in both study cohorts (86% and 87%, P ¼ .70; Table 2).
Discussion
6
-
Using 2 large patient samples from the MarketScan database and the SEEReMedicare database, this study assessed patterns of access to specialist care and treatment among patients with advanced stage lung cancer in the United States. This analysis fills a gap in the literature by providing an overview of the recent trends in the management of advanced stage lung cancer in US clinical practice (years 2007-2014). The study found that a substantial number of patients with advanced stage lung cancer did not have appropriate access to cancer specialists: approximately 20% of patients in both
Clinical Lung Cancer Month 2017
samples were not seen by a cancer specialist within 1 month of the first lung biopsy, while 5% and 13%, respectively, were not seen by a cancer specialist within 6 months of the first lung biopsy. Access to cancer-directed therapy was also found to be limited, with 6% of the commercially insured adult patients from the MarketScan database and 10% of the Medicare-insured elderly patients from the SEEReMedicare database not receiving any cancer-directed therapy. To our knowledge, 2 previous studies investigated access to cancer specialists, but both were conducted before the availability of many of the currently used therapies: Goulart et al8 found that 16% of Medicare patients diagnosed with advanced NSCLC in the United States in 2000-2005 were not seen by a cancer specialist within 6 months from diagnosis, and Vinod et al9 found that 11% of patients of all ages diagnosed with lung cancer in Australia in 2001-2002 were not seen by a lung cancer specialist after diagnosis. With respect to access to cancer-directed therapy, the most up-todate estimates are also from 2 older studies: Small et al10 found 25% untreated among patients of all ages with a diagnosis of NSCLC in the National Cancer Database in 2000-2008, and Owonikoko et al11 found 66% untreated in the Medicare population with advanced stage NSCLC in 1988-2005. The findings of the current study for the Medicare-insured sample in 2007-2011 suggest an improvement over time in access to cancer-directed therapy. Given that none of the studies to date could ascertain whether the therapy received was appropriate for the patients’ specific situation, further studies are needed to assess whether or not patients receive optimal treatment. While it is possible that best supportive care could in some circumstances be the optimal treatment, the fact that lack of treatment was more common among patients not seen by a cancer specialist suggests that some patients who opt out of treatment may not make an informed decision that is based on a good understanding of all treatment options available to them. On the basis of the analysis of the sample of Medicare-insured elderly patients from the SEEReMedicare database, an important finding of the current study is that not seeing a cancer specialist reduces patient access to optimal and timely treatment. This finding is in line with a prior study that found that seeing an oncologist was one of the strongest predictors of receiving chemotherapy for stage IV NSCLC.19 Other factors found in the current study to be associated with not seeing a cancer specialist among Medicareinsured elderly and/or commercially insured adult patients included older age; female sex; being widowed, divorced, or separated; residing in areas with high poverty and lower education; being diagnosed with earlier stage disease and having a smaller tumor size; and having a greater comorbidity burden. However, squamous versus nonsquamous histology did not seem to affect access to cancer specialists. Previous studies also found that patients with advanced stage lung cancer were less likely to be seen by or referred to a cancer specialist if they were older, had a lower socioeconomic status, and exhibited a greater comorbidity burden.8,9,19-22 By contrast, an older study by Earle et al19 found no correlation between being older and not seeing an oncologist among patients with stage IV NSCLC diagnosed in 1991-1993, although patients with low socioeconomic status and those not receiving care in a teaching hospital were less likely to be seen by an oncologist. While the study
Apar Kishor Ganti et al Table 2 Use of Cancer-directed Therapies and Palliative Care Services Commercially Insured Adult Patients With Metastatic Lung Cancer
a
Cancer-directed Therapy
Patients Seen by Cancer Specialist
Patients Not Seen by Cancer Specialist
N [ 19,665 (88.3%)
N [ 2603 (11.7%)
P
Medicare-insured Elderly Patients With Advanced NSCLC Patients Seen by Cancer Specialist
Patients Not Seen by Cancer Specialist
N [ 9284 (96.2%)
N [ 367 (3.8%)
P
[A]
[B]
[A] vs. [B]
[C]
[D]
[C] vs. [D]
18,627 (94.7)
2384 (91.6)
<.01*
8521 (91.8)
139 (37.9)
<.01*
Systemic therapy
15,391 (82.6)
1962 (82.3)
.69
6910 (81.1)
63 (45.3)
<.01*
EGFR inhibitors
1896 (10.2)
231 (9.7)
.46
1876 (22.0)
21 (15.1)
.05
ALK inhibitors
104 (0.6)
7 (0.3)
.09
<11 (<0.1)c
0 (0.0)
1.00
4125 (22.1)
468 (19.6)
<.01*
<11 (<7.9)c
<.01*
Patients Treated With Cancer-Directed Therapy (Any), N (%) Therapy Type,b N (%)
Other Targeted Therapies
1466 (17.2)
14,624 (78.5)
1872 (78.5)
.99
6415 (75.3)
46 (33.1)
<.01*
13,413 (72.0)
1715 (71.9)
.94
6094 (71.5)
12 (8.6)
<.01*
Stereotactic radiosurgery
2917 (15.7)
294 (12.3)
<.01*
901 (10.6)
<11 (<7.9)c
<.01*
Radiofrequency ablation
281 (1.5)
30 (1.3)
.34
134 (1.6)
<11 (<7.9)c
.48
Cryoablation
320 (1.7)
49 (2.1)
.24
116 (1.4)
<11 (<7.9)c
.71
Brachytherapy
145 (0.8)
22 (0.9)
.45
68 (0.8)
<11 (<7.9)c
1.00
3388 (18.2)
360 (15.1)
<.01*
1340 (15.7)
79 (56.8)
<.01*
1038 (5.3)
219 (8.4)
<.01*
763 (8.2)
228 (62.1)
<.01*
57 (26.0)
.05
655 (85.8)
198 (86.8)
.70
Chemotherapy Radiotherapy
Surgery
Surgical resection Patients Not Treated, N (%) Use of palliative care services, n (% of nonrelated patients) Hospice or skilled nursing facility
339 (32.7)
Hospice only
163 (15.7)
16 (7.3)
<.01*
414 (54.3)
104 (45.6)
.02*
Skilled nursing facility only
138 (13.3)
39 (17.8)
.08
87 (11.4)
45 (19.7)
<.01*
Hospice and skilled nursing facility
38 (3.7)
2 (0.9)
.03*
154 (20.2)
49 (21.5)
.67
699 (67.3)
162 (74.0)
.05
108 (14.2)
30 (13.2)
.70
No hospice or skilled nursing facility
*P .05 by chi-square test. Abbreviations: ALK ¼ anaplastic lymphoma kinase; EGFR ¼ epidermal growth factor receptor; NSCLC ¼ non-small-cell lung cancer. a For patients who were seen or were not seen by a cancer specialist from 6 weeks before index date until the end of follow-up. b Used alone, in combination, or sequentially. c Per data user agreement with the National Cancer Institute, exact count cannot be reported for < 11 patients.
by Earle et al was conducted among patients with stage IV NSCLC more than 25 years ago, their findings nevertheless suggest that the type of facility where patients receive care might impact timely and appropriate access to cancer specialist care, an association that was not studied in the current study because this information was unavailable. While further studies are needed to clarify the interplay between these factors, physicians need to be aware that nonmedical factors such as patient age or sociodemographic characteristics may create barriers to appropriate care and therapy. Some noteworthy differences were observed in the current study between the treatment patterns of the commercially insured adult patients identified in the MarketScan database and the Medicareinsured elderly patients identified in the SEEReMedicare database: among the commercially insured adult patients who were not
seen by a cancer specialist, 92% received cancer-directed therapy, 15% received treatment that may not be consistent with clinical guidelines (surgical resection), and 26% received palliative care; among the Medicare-insured elderly patients not seen by a cancer specialist, the respective proportions were 38%, 57%, and 87%. Given the age difference between the 2 samples (adult patients of all ages vs. elderly patients), these observed differences in the use of cancer-directed therapies and palliative care services suggest a possible treatment bias against the elderly among physicians other than cancer specialists. However, other factors may have also contributed to these differences, such as the data structure differences between the MarketScan and SEEReMedicare databases. For example, because physician specialty measurement is less refined in the MarketScan database than in SEEReMedicare database
Clinical Lung Cancer Month 2017
-7
Access to Cancer Specialist Care Figure 2 Time of First Visit With Cancer Specialist. Time of First Visit With Cancer Specialist in 2 Diverse Populations of Patients With Advanced Lung Cancer: Commercially Insured Adult Patients and Medicare-insured Elderly Patients
Abbreviation: TX ¼ treatment.
(because of the linkage with American Medical Association physician files in the latter database), it is possible that more commercially insured adult patients were seen by a cancer specialist than what was reported in the current study. It is also possible that some of the patients not seen by a cancer specialist were in fact seen by a midlevel provider who could have worked in collaboration with a cancer specialist to provide treatment, a situation that is likely to
occur more frequently for commercially insured patients compared to Medicare-insured patients. Finally, the 62% of the commercially insured adult patients in the MarketScan database who did not receive any cancer-directed therapy aged 65 years (data not shown) may have had dual commercial and Medicare coverage and may have had their palliative services fully covered through Medicare.
Figure 3 Time From First Lung Biopsy to First Cancer Specialist Visit. Time From First Lung Biopsy to First Cancer Specialist Visit in 2 Diverse Populations of Patients With Advanced Lung Cancer: Commercially Insured Adult Patients and Medicare-insured Elderly Patients. The x-axis Was Censored at 1 Year Because Most Events Occur Within 1 Year. Patients Who Had Visit With Cancer Specialist in 6 Weeks Before First Lung Biopsy Were Considered to Have Seen Cancer Specialist at Time 0 (Date of First Lung Biopsy)
8
-
Clinical Lung Cancer Month 2017
Apar Kishor Ganti et al Figure 4 Patterns of Referral to Cancer Specialist. Patterns of Referral to Cancer Specialist in 2 Diverse Populations of Patients With Advanced Lung Cancer: Commercially Insured Adult Patients and Medicare-insured Elderly Patients. Specialist Groups in 6 Weeks Before First Lung Biopsy Were Mutually Exclusive and Are Defined Hierarchically on Basis of Level of Specialization of Physicians. a“± Other Specialists” Indicates “Including or Excluding Other Specialists.” bPatients Were Stratified by Type of Specialist Seen in 6 Weeks Before First Lung Biopsy. Each Color Indicates Type of Specialist Seen in 6 Weeks Before First Lung Biopsy, as Reported at Left. cAll Differences Between Groups Were Statistically Significant at P < .05. All Differences Between Groups Were Statistically Significant at P < .05, With Exception of Difference Between Groups Seen by Pulmonologist and Internist/Family Doctor Before First Lung Biopsy (P [ .59)
Abbreviation: CS ¼ cancer specialist.
The findings of this study should be interpreted in light of some limitations. First, MarketScan commercial claims databases do not include information on lung cancer histology or disease staging. While the sample of commercially insured adults could not be restricted to patients with advanced (stage IIIB and IV) NSCLC, stage IV is likely captured by the metastatic disease proxy used in this study. Also, based on the 85% prevalence of NSCLC in lung cancer,4 it is likely that patients with NSCLC histology represent the majority of the patients included in the study. Second, although claims data were available until the end of 2013 in the SEEReMedicare database, patients diagnosed with advanced
NSCLC after the year 2011 were not available for inclusion in the study. Thus, potential changes in the use of targeted therapies that occurred in patients diagnosed after 2011 were not captured in the results for the Medicare-insured elderly patients. Third, claims data did not include sufficient information to determine which of the specialists seen before therapy initiation actually made treatment decisions. However, in the case of patients seen by cancer specialists, it is likely that treatment decisions were made by the cancer specialists. Last, the study may have been subject to general limitations of claims data, including occasional inaccuracies in coding diagnoses, procedures, or pharmacy claims.
Clinical Lung Cancer Month 2017
-9
Access to Cancer Specialist Care Conclusion The current study found that a substantial proportion of patients with advanced stage lung cancer and commercial or Medicare insurance in the period 2007-2014 did not see a cancer specialist within 1 month of the first lung biopsy. Furthermore, access to cancer-directed therapy among Medicare-insured elderly patients not seen by a cancer specialist was significantly lower than that of those seen by a cancer specialist. While older age and lower socioeconomic status appear to be correlated with reduced access to a cancer specialist, further research should be directed toward understanding and addressing disparities in access to appropriate cancer care.
Clinical Practice Points This analysis fills a gap in the literature by providing an overview
of the recent trends in the access to cancer specialist care and treatment of patients with advanced stage lung cancer in US clinical practice (years 2007-2014). Using 2 data sources, the study identified 2 diverse populations of patients: commercially insured adult patients with metastatic lung cancer (MarketScan database) and Medicare-insured elderly patients with advanced NSCLC (SEEReMedicare database). The study found that a substantial number of patients with advanced stage lung cancer did not have access to cancer specialists: approximately 20% of patients in both samples were not seen by a cancer specialist within 1 month of the first lung biopsy, while 5% to 13%, depending on the sample, were not seen by a cancer specialist within 6 months of the first lung biopsy. Several demographic (eg, older age; being widowed, divorced, or separated; residing in areas with high poverty) and clinical (eg, having a smaller tumor size; and having a greater comorbidity burden) factors were found to be associated with limited access to cancer specialists. Physicians need to be aware that both medical and nonmedical factors can create barriers to appropriate care and therapy. Access to cancer-directed therapy was also found to be limited, with 6% to 10% of the patients not receiving any cancer-directed therapy, especially among patients not seen by a cancer specialist. Patients not seen by a cancer specialist may not be aware of all treatment options available to them.
Acknowledgments Medical writing assistance was provided by Cinzia Metallo, PhD, an employee of Analysis Group Inc. supported by Genentech Inc, South San Francisco, CA.
Disclosure A.R. is an employee of Genentech Inc and owns stock/stock options. R.I.-I. and I.P. are employees of Analysis Group Inc, which has received consultancy fees from Genentech Inc. A.K.G. has received institutional clinical research grants from Pfizer, New Link Genetics, Amgen, AstraZeneca, Merck, Janssen, and Bristol-Myers Squibb and has acted as a paid consultant for Pfizer, Ariad, and
10
-
Clinical Lung Cancer Month 2017
Biodesix for work performed outside of the current study. H.B. has acted as a paid consultant for BMS, Genentech, Lilly, Pfizer, Boerhringer-Ingelheim, Trovegen, and Celgene for work performed outside of the current study and as an uncompensated consultant for Merck for work performed outside of the current study, and has received honoraria from Celgene. F.R.H. has received institutional clinical research grants from Lilly, InClone Systems, Genentech, Bristol-Myers Squibb, Amgen, Celgene, and Bayer; holds an institutional (University of Colorado) patent for EGFR fluorescence insitu hybridization and immunohistochemistry for prediction of outcome in patients treated with EGFR inhibitors; and has acted as a paid consultant for Bristol-Myers Squibb, AstraZeneca, Genentech, Roche, HTG Inc, Pfizer, Lilly, Merck, Clovis Oncology, and Ventana Medical Systems for work performed outside of the current study. S.S.R. has acted as a paid consultant for Abbvie, Amgen, AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Genentech/Roche, and Lilly/InClone for work performed outside of the current study. M.W.W. has no conflict of interest to disclose.
Supplemental Data Supplemental table accompanying this article can be found in the online version at http://dx.doi.org/10.1016/j.cllc.2017.04.010.
References 1. World Health Organization; International Agency for Research on Cancer. GLOBOCAN 2012: estimated cancer incidence, mortality and prevalence worldwide in 2012, Available at: http://globocan.iarc.fr/Pages/fact_sheets_cancer. aspx?cancer¼lung. Accessed: July 31, 2016. 2. Hurria A, Kris MG. Management of lung cancer in older adults. CA Cancer J Clin 2003; 53:325-41. 3. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2007. CA Cancer J Clin 2007; 57:43-66. 4. Key statistics for lung cancer, Available at: http://www.cancer.org/cancer/ lungcancer-non-smallcell/detailedguide/non-small-cell-lung-cancer-what-is-nonsmall-cell-lung-cancer. Accessed: July 31, 2016. 5. National Cancer Institute. SEER Stat Fact Sheets: lung and bronchus cancer, Available at: http://seer.cancer.gov/statfacts/html/lungb.html. Accessed: August 1, 2016. 6. National Cancer Institute. Nonesmall cell lung cancer treatment, Available at: http://www.cancer.gov/types/lung/hp/non-small-cell-lung-treatment-pdq#section/_ 514. Accessed: July 31, 2016. 7. Gregor A, Thomson CS, Brewster DH, et al. Management and survival of patients with lung cancer in Scotland diagnosed in 1995: results of a national population based study. Thorax 2001; 56:212-7. 8. Goulart BH, Reyes CM, Fedorenko CR, et al. Referral and treatment patterns among patients with stages III and IV nonesmall-cell lung cancer. J Oncol Pract 2013; 9:42-50. 9. Vinod SK, O’Connell DL, Simonella L, et al. Gaps in optimal care for lung cancer. J Thorac Oncol 2008; 3:871-9. 10. Small AC, Tsao CK, Moshier EL, et al. Prevalence and characteristics of patients with metastatic cancer who receive no anticancer therapy. Cancer 2012; 118: 5947-54. 11. Owonikoko TK, Ragin C, Chen Z, et al. Real-world effectiveness of systemic agents approved for advanced nonesmall cell lung cancer: a SEEReMedicare analysis. Oncologist 2013; 18:600-10. 12. Sacher AG, Le LW, Lau A, Earle CC, Leighl NB. Real-world chemotherapy treatment patterns in metastatic nonesmall cell lung cancer: are patients undertreated? Cancer 2015; 121:2562-9. 13. Coate LE, Massey C, Hope A, et al. Treatment of the elderly when cure is the goal: the influence of age on treatment selection and efficacy for stage III nonesmall cell lung cancer. J Thorac Oncol 2011; 6:537-44. 14. Ramesh HSJ, Boase T, Audisio RA. Risk assessment for cancer surgery in elderly patients. Clin Interv Aging 2006; 1:221-7. 15. Potosky AL, Riley GF, Lubitz JD, Mentnech RM, Kessler LG. Potential for cancer related health services research using a linked Medicareetumor registry database. Med Care 1993; 31:732-48. 16. National Cancer Institute. SEEReMedicare: how the SEER and Medicare data are linked, Available at: http://healthcaredelivery.cancer.gov/seermedicare/overview/ linked.html. Accessed: July 31, 2016.
Apar Kishor Ganti et al 17. Whyte JL, Engel-Nitz NM, Teitelbaum A, Gomez Rey G, Kallich JD. An evaluation of algorithms for identifying metastatic breast, lung, or colorectal cancer in administrative claims data. Med Care 2015; 53:e49-57. 18. National Cancer Institute. SEEReMedicare: encrypted variables, Available at: http://healthcaredelivery.cancer.gov/seermedicare/privacy/variables.html?&url¼/ seermedicare/privacy/variables.html#release. Accessed: July 31, 2016. 19. Earle CC, Neumann PJ, Gelber RD, Weinstein MC, Weeks JC. Impact of referral patterns on the use of chemotherapy for lung cancer. J Clin Oncol 2002; 20: 1786-92.
20. Li X, Butts C, Fenton D, King K, Scarfe A, Winget M. Utilization of oncology services and receipt of treatment: a comparison between patients with breast, colon, rectal, or lung cancer. Ann Oncol 2011; 22:1902-9. 21. Dalton SO, Frederiksen BL, Jacobsen E, et al. Socioeconomic position, stage of lung cancer and time between referral and diagnosis in Denmark, 2001-2008. Br J Cancer 2011; 105:1042-8. 22. Wang J, Kuo YF, Freeman J, Goodwin JS. Increasing access to medical oncology consultation in older patients with stage II-IIIA nonesmall-cell lung cancer. Med Oncol 2008; 25:125-32.
Clinical Lung Cancer Month 2017
- 11
Access to Cancer Specialist Care Supplemental Table 1 Characteristics of 2 Populations of Patients With Advanced Lung Cancer Commercially Insured Adult Patients
Seen by Cancer Not Seen by Specialist Cancer Specialist
Seen by Cancer Not Seen by Specialist Cancer Specialist Characteristica
N [ 19,665
N [ 2603
Medicare-insured Elderly Patients
P
N [ 9284
N [ 367
P
Demographic Data Age Age at Diagnosis, y, Median (IQR) <65 years, n (%)
63 (57-72)
65 (58-74)
<.01*
11,387 (57.9)
1297 (49.8)
<.01*
65-79 years, n (%)
6421 (32.7)
1005 (38.6)
<.01*
80þ years, n (%)
1857 (9.4)
301 (11.6)
10,003 (50.9)
1367 (52.5)
74 (70-79) —
78 (72-84)
<.01*
—
—
7044 (75.9)
200 (54.5)
<.01*
<.01*
2240 (24.1)
167 (45.5)
<.01*
.11
4514 (48.6)
144 (39.2)
<.01*
NA
NA
—
7517 (81.0)
291 (79.3)
.42
Never married
NA
NA
—
879 (9.5)
44 (12.0)
.11
Married/common law
NA
NA
—
4382 (47.2)
129 (35.1)
<.01*
Widowed/divorced/separated
NA
NA
—
3726 (40.1)
177 (48.2)
<.01*
Other/unknown
NA
NA
—
297 (3.2)
17 (4.6)
.13
NA
NA
—
2189 (23.6)
127 (34.6)
<.01*
NA
NA
—
2326 (25.1)
115 (31.3)
<.01*
Male sex, n (%) Non-Hispanic white race/ethnicity, n (%) Marital Status, N (%)
Living in Area With High Poverty,b N (%) 20% residents living below poverty line in zip code of patient residence Living in Area With Lower Education,b N (%) 20% residents with <12 grades in zip code of patient residence Region of Residence, N (%) Northeast
4599 (23.4)
698 (26.8)
<.01*
1836 (19.8)
59 (16.1)
.08
North Central
5341 (27.2)
764 (29.4)
.02*
1268 (13.7)
39 (10.6)
.10
South
6468 (32.9)
668 (25.7)
<.01*
2623 (28.3)
113 (30.8)
.29
West
2734 (13.9)
383 (14.7)
.26
3454 (37.2)
155 (42.2)
.05
523 (2.7)
90 (3.5)
.02*
<11 (<0.1)
<11 (<3.0)
.13
Big metropolitan (counties of metro areas of 1 million population)
NA
NA
—
4891 (52.7)
175 (47.7)
.06
Metropolitan (counties in metro areas of up to 1 million population)
NA
NA
—
2568 (27.7)
105 (28.6)
.69
Urban (urban population of 20,000 adjacent or nonadjacent to a metro area)
NA
NA
—
566 (6.1)
25 (6.8)
.58
Less urban (urban population of 2500 to <20,000 adjacent or nonadjacent to a metro area)
NA
NA
—
1011 (10.9)
49 (13.4)
.14
Rural (completely rural or <2500 adjacent or nonadjacent to a metro area)/unknown
NA
NA
—
248 (2.7)
13 (3.5)
.31
Unknown/otherc Urban Residency,d N (%)
Cancer Characteristics at Time of Diagnosis AJCC 6 Cancer Stage, N (%) IIIB
NA
NA
—
3250 (35.0)
159 (43.3)
<.01*
IV
NA
NA
—
6034 (65.0)
208 (56.7)
<.01*
2 cm
NA
NA
—
838 (9.0)
42 (11.4)
.12
2-5 cm
NA
NA
—
3780 (40.7)
150 (40.9)
.95
5þ cm
NA
NA
—
2807 (30.2)
83 (22.6)
<.01*
Not reported
NA
NA
—
1859 (20.0)
92 (25.1)
.02*
Tumor Size Median (IQR), mm,e
11.e1
-
Clinical Lung Cancer Month 2017
40 (27-60)
35 (25-55)
<.01*
Apar Kishor Ganti et al Supplemental Table 1 Continued Commercially Insured Adult Patients
Seen by Cancer Not Seen by Specialist Cancer Specialist
Seen by Cancer Not Seen by Specialist Cancer Specialist Characteristica
N [ 19,665
Medicare-insured Elderly Patients
N [ 2603
P
N [ 9284
N [ 367
P
NSCLC Histology, N (%) Adenosquamousc
NA
NA
—
<11 (<0.1)
<11 (<3.0)
.09
Large-cell carcinoma
NA
NA
—
240 (2.6)
11 (3.0)
.63
Adenocarcinoma, NOS
NA
NA
—
4496 (48.4)
179 (48.8)
.90
Squamous-cell carcinoma
NA
NA
—
2787 (30.0)
111 (30.2)
.93
f
NSCLC, NOS
NA
NA
—
1626 (17.5)
65 (17.7)
.92
Otherc
NA
NA
—
<11 (<0.1)
<11 (<3.0)
.40
6227 (67.1)
219 (59.7)
<.01*
1824 (19.6)
73 (19.9)
.91
Distant metastasesg at diagnosis, n (%) Year of Diagnosis, N (%) 2007
NA
NA
—
2008
1707 (8.7)
287 (11.0)
<.01*
1821 (19.6)
77 (21.0)
.52
2009
3638 (18.5)
516 (19.8)
.10
1841 (19.8)
83 (22.6)
.19
2010
3532 (18.0)
459 (17.6)
.68
1890 (20.4)
62 (16.9)
.11
2011
4133 (21.0)
489 (18.8)
<.01*
1908 (20.6)
72 (19.6)
.66
2012
3878 (19.7)
507 (19.5)
.77
NA
NA
—
2013
2777 (14.1)
345 (13.3)
.23
NA
NA
—
1 (0-2)
.56
Comorbid Conditions in 6 Months Before NSCLC Diagnosis Overall Comorbidity Burden (Charlson Comorbidity Indexh) 2 (1-3)
<.01*
0-2
17,145 (87.2)
2252 (86.5)
.35
6334 (68.2)
222 (60.5)
<.01*
3-6
2440 (12.4)
338 (13.0)
.42
2730 (29.4)
128 (34.9)
7þ
80 (0.4)
13 (0.5)
.60
220 (2.4)
17 (4.6)
Median (IQR)
1 (0-2)
2 (1-3)
.02* <.01*
Physical Comorbidities (Statistically Significant Differences in ‡1 Sample), N (%) Anemia
2656 (13.5)
345 (13.3)
.73
2544 (27.4)
129 (35.1)
<.01*
Congestive heart failure
1085 (5.5)
159 (6.1)
.22
1473 (15.9)
105 (28.6)
<.01*
Complicated diabetes
815 (4.2)
130 (5.0)
.04*
835 (9.0)
46 (12.5)
.02*
Uncomplicated diabetes
2827 (14.4)
370 (14.2)
.83
2001 (21.6)
56 (15.3)
<.01*
Peripheral vascular disease
1982 (10.1)
277 (10.7)
.37
2241 (24.1)
115 (31.3)
<.01*
742 (3.8)
109 (4.2)
.30
500 (5.4)
29 (7.9)
1414 (7.2)
203 (7.8)
.26
1208 (13.0)
67 (18.3)
<.01*
Mood disordersi
2079 (10.6)
275 (10.6)
.99
1291 (13.9)
69 (18.8)
<.01*
Anxiety disorders
1068 (5.4)
135 (5.2)
.61
642 (6.9)
37 (10.1)
.02*
Pulmonary circulation disorders Weight loss
.04*
Mental Comorbidities, N (%)
*P .05. Abbreviations: AJCC 6 ¼ 6th edition of the American Joint Committee on Cancer (AJCC) Cancer Staging Manual (used by SEER to derive cancer stage); CS ¼ cancer specialist; DSM ¼ Diagnostic and Statistical Manual of Mental Disorders; IQR ¼ interquartile range; NA ¼ not available. a Patients who have seen or not seen a cancer specialist from 6 weeks before index date until end of follow-up. b Estimated from the American Community Survey (ACS), 2008-2012. c Data user agreement with the National Cancer Institute forbids the presentation of frequencies/percentages for subgroups of < 11 patients. d Refers to a classification scheme that distinguishes metropolitan (metro) counties by the population size of their metro area and nonmetropolitan (nonmetro) counties by degree of urbanization and adjacency to a metro area or nonmetro areas. e Four patients with tumor sizes > 20 cm were reset to a maximum value of 20 cm. f A nonspecific histology code is used in the SEER database when the cancer registry abstractor did not find enough information (eg, test results) to confirm with certitude the specific cancer histology (patients in this group may have in truth any specific histologies listed above). g Could include malignant pleural effusion; malignant pericardial effusion; extension to contralateral lung, extension to contralateral main stem bronchus, separate tumor nodules in contralateral lung, pleural tumor foci or nodules on contralateral lung; pleural tumor foci or nodules on the ipsilateral lung separate from direct invasion; distant lymph nodes, including cervical nodes extension to skeletal muscle, sternum, skin of chest; abdominal organs, other distant metastases (alone or in combination); could also include “stated as M1a, M1b or M1 (NOS) without additional information on distant metastasis.” h Overall burden of disease due to comorbidities other than the cancer/metastatic cancer (ie, no points were accounted for the cancer/metastatic cancer diagnosis). i Mood disorders included anxiety, bipolar, depressive, dissociative, factitious, and somatoform disorders.
Clinical Lung Cancer Month 2017
- 11.e2