Background
Race/Ethnicity has been implied to play a role in both outcomes1,2 and treatment decisions2–7 in lung cancer. Despite superior survival in patients who receive therapy and a breadth of diagnostic and treatment options, a large proportion of older and elderly patients with advanced lung cancer do not undergo treatment.8–12 To our knowledge, investigations of outcomes, nontreatment, and race/ethnicity in lung cancer11,13–18 have included and focused on elderly populations (age >65 years) and have focused on either a single histology (non–small cell lung cancer [NSCLC]) or metastatic disease, all of which may play a role in the decision not to treat. Little attention has been given to understanding differences in treatment patterns for younger patients with lung cancer, whose comorbidity profiles and performance status differ from those that may impact treatment decisions in elderly patients. We have previously shown that in nonelderly patients (age ≤65 years) with lung cancer, after adjusting for clinical and sociodemographic factors (including socioeconomic status [SES] and distance to treatment facility), treatment at a facility other than an NCI-designated Comprehensive Cancer Center (CCC) was associated with a 20% to 50% higher risk of mortality compared with treatment at a CCC19; however, the impact of CCC care on nontreatment remains unknown in this population. We also examined race/ethnicity and SES as barriers to care in nonelderly patients with lung cancer19; however, these factors have not been examined in the context of nontreatment. We sought to address these knowledge gaps in nonelderly patients (age 22–65 years) with all stages of lung cancer, examining the impact of race/ethnicity and care at CCCs on nonreceipt of lung cancer treatment.
Patients and Methods
Using data from the Los Angeles County Cancer Surveillance Program (CSP), a population-based cohort was assembled. Patients were newly diagnosed with lung cancer (small cell lung cancer [SCLC] or NSCLC) between 1998 and 2008, when they were between ages 22 and 65 years. CSP collects data on all new cancer diagnoses among county residents and is a member of the California Cancer Registry and the NCI-funded SEER Program.20 Eligible patients were diagnosed with a first primary malignancy of lung cancer and treated within Los Angeles County. This project was approved by the Institutional Review Boards of the State of California, City of Hope, and the University of Alabama at Birmingham.
Nontreatment
CSP collects data regarding a patient’s first course of treatment for each tumor diagnosis. Treatment modalities included surgery, radiation, chemotherapy, hormone therapy, immunotherapy, and other therapy, with each reported in a binary fashion (yes/no). Patients were categorized as not receiving any type of treatment (“no treatment”) if all of the 6 treatment modality indicators were “no.” If any treatment modality was “yes,” the patient was categorized as receiving treatment (“treatment”).
Sociodemographic Characteristics
The following sociodemographic variables were included in multivariable analyses: (1) age (continuous variable); (2) year of diagnosis (11 categories); (3) race/ethnicity (non-Hispanic white [NHW], black, Hispanic, and Asian/Pacific Islander); (4) SES; and (5) distance to treatment facility. CSP calculated patients’ SES using 2000 census block–level education and median household income data and then ranked these into quintiles (low, middle-low, middle, middle-high, high).21 Using the SEER-based payer variable, insurance status was collapsed into 3 categories: private (private insurance, fee for service, insurance not otherwise specified [“no specifics”], managed care, HMO, preferred provider organization), public (Medicare, Medicaid, county-funded not otherwise specified, military, TRICARE, Veterans Health Administration, Indian Health Service/US Public Health Service Commissioned Corps), and uninsured.
Distance traveled by a patient to the identified treatment facility was calculated using the residential address at diagnosis, which was provided by CSP. Hospital addresses were then geocoded and straight-line distance was measured between patient residence and treatment facility using the ArcMap 10.2 geographic information system (Esri). We chose this approach because Euclidean distance has been shown to be correlated with drive time.22
Clinical Characteristics
Disease stage was defined using CSP summary staging; this method is based on the Collaborative Stage coding system and integrates TNM categories, stage groupings, and SEER Extent of Disease coding. For our analysis, localized/regional disease included localized and microinvasive disease and regional disease with either extension only, nodes only, or extension and nodes; advanced disease included remote disease.
Cases were selected using ICD-O-3–based disease site and histology coding, and all disease site codes were consistent with lung (C34). Only invasive cancers were included; both NSCLC and SCLC were included in the cohort.
Treatment Facility
To assign treatment facility, the facility associated with each episode of care was systematically identified. When more than one facility reported contact with a patient, we prioritized the facility where the patient had all or part of the first treatment course (with or without initial diagnosis) or where the decision was made not to treat. Facility–patient combinations were excluded if the patient did not appear in person, presented with disease recurrence/persistence, presented for second opinion, received in-transit care, or was seen for diagnostic workup only. At least one facility was therefore determined to be a patient’s a priori facility for treatment/nontreatment. Patients were considered to have been treated at a CCC if their identified facility was 1 of the 3 CCCs in Los Angeles County (UCLA Jonsson, USC Norris, and City of Hope). All other patients were considered to have received care at non-CCCs.
Statistical Analysis
Univariable and multivariable logistic regressions were used to identify factors associated with nontreatment for lung cancer. The magnitudes of association are presented as odds ratios (ORs) with associated 95% CIs. Analyses described later were repeated in patients with only advanced disease and produced identical findings (supplemental eTables 1 and 2, available with this article at JNCCN.org). Two-sided tests with P<.05 were considered statistically significant. SAS 9.3 (SAS Institute Inc.) was used for all analyses.
Results
Patients
Clinical and sociodemographic characteristics of the cohort (N=9,877) are summarized in Table 1. Most patients were men (55.8%), aged ≥40 years (97.9%), and NHW (54.5%), with representation of patients who were black (19.9%), Hispanic (13.2%), and Asian/Pacific Islander (12.4%). Most individuals had private insurance (59.5%), with smaller numbers having public (31.5%) or no insurance (9.0%). There was an even distribution among SES strata, with 19.3% in the highest SES group and 15.8% in the lowest SES group. Most tumors were NSCLC (86.3% vs 13.7% for SCLC) and were at an advanced stage (65.3%). Most patients sought care at non-CCCs (95.5%).
Patient Characteristics
No difference was seen between the “treatment” and “no treatment” groups with respect to age (P=.22) or histology (P=.46). Larger proportions of the “no treatment” group were black (23.8% vs 19.0%) or Hispanic (15.5% vs 12.6%; P<.001), were from the lowest SES group (22.8% vs 14.2%; P<.001), had public (38.6% vs 29.8%) or no insurance (13.6% vs 7.9%; P<.001), or were cared for at non-CCCs (98.7% vs 94.8%; P<.001). In addition, a larger proportion of the “no treatment” group had advanced disease (81.5% vs 61.5%; P<.001).
Treatment
A higher proportion of black (22.6%; n=446) and Hispanic (22.4%; n=291) patients diagnosed with lung cancer did not receive treatment compared with Asian/Pacific Islander (17.3%; n=211) and NHW patients (17.2%; n=925) (P<.001) (Figure 1). The proportion of patients not receiving lung cancer treatment was highest in the low SES group and increased with SES stratum (low, 27.3%; middle-low, 20.9%; middle, 20.5%; middle-high, 16.1%; high, 11.6%; P<.001). A larger proportion of publicly insured (23.2%; n=723) and uninsured (28.6%; n=254) patients did not receive treatment compared with those who were privately insured (15.2%; n=896; P<.001). Similarly, a larger proportion of patients cared for at non-CCC facilities did not receive treatment (19.6%; n=1,849) compared with those cared for at CCCs (5.4%; n=24; P<.001). Finally, a larger proportion of patients with advanced disease (23.7%; n=1,527) than with localized or regional disease (10.1%; n=346; P<.001) did not receive treatment.
Proportion of patients not receiving treatment of lung cancer, according to (A) race/ethnicity, (B) socioeconomic status, (C) insurance status, (D) treatment facility, and (E) disease stage.
Abbreviation: CCC, NCI-designated Comprehensive Cancer Center.
Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 17, 8; 10.6004/jnccn.2019.7294
Characteristics Associated With Nontreatment
As presented in Table 2, in univariable analysis (model 1), both black (OR, 1.41; 95% CI, 1.24–1.60; P<.001) and Hispanic patients (OR, 1.39; 95% CI, 1.20–1.61; P<.001) had higher odds of not receiving treatment than NHW patients (reference group [ref]). The odds of nontreatment by race/ethnicity were not substantially altered by adjusting for age (model 2; black: OR, 1.44; 95% CI, 1.27–1.64; P<.001; Hispanic: OR, 1.46; 95% CI, 1.26–1.69; P<.001), nor were the odds of nontreatment by race/ethnicity substantially altered by further adjustment for sex, disease stage, histology, distance to nearest CCC, year of diagnosis, and treatment facility (model 3; black: OR, 1.33; 95% CI, 1.17–1.52; P<.001; Hispanic: OR, 1.28; 95% CI, 1.10–1.49; P=.002). There was minimal change in odds of nontreatment by race/ethnicity when adjusting for insurance status in addition to the previously mentioned variables (model 4; black: OR, 1.22; 95% CI, 1.06–1.39; P=.004; Hispanic: OR, 1.17; 95% CI, 1.01–1.37; P=.04). However, in model 5, when including SES in the multivariable model with all previous variables, the increased odds of nontreatment by race/ethnicity was mitigated (black: OR, 1.02; 95% CI, 0.88–1.17; P=.80; Hispanic: OR, 1.00; 95% CI, 0.85–1.18; P=1.00).
Odds of Nontreatment
In the final multivariable model (model 5), including race/ethnicity, age, sex, diagnosis year, disease stage, histology, treatment facility, insurance status, and SES, the following characteristics remained independently associated with nontreatment: treatment facility ([ref: CCC]; non-CCC: OR, 3.22; 95% CI, 2.12–4.90; P<.001), insurance status ([ref: private]; public, OR, 1.71; 95% CI, 1.44–2.03; P<.001; uninsured: OR, 1.30; 95% CI, 1.16–1.47; P<.001), and SES ([ref: high]; middle-high: OR, 1.37; 95% CI, 1.14–1.64; P=.001; middle: OR, 1.73; 95% CI, 1.44–2.08; P<.001; middle-low: OR, 1.66; 95% CI, 1.37–2.00; P<.001; low: OR, 2.15; 95% CI, 1.75–2.64; P<.001). Additional characteristics associated with nontreatment included age (OR, 1.03; 95% CI, 1.02–1.04; P<.001) and disease stage ([ref: localized/regional]; advanced: OR, 2.69; 95% CI, 2.36–3.06; P<.001).
Discussion
Our population-level findings among nonelderly patients (aged 22–65 years) with lung cancer indicate that although race/ethnicity at first appears to be associated with not receiving treatment, it is the patient's SES that is associated with nontreatment. Essentially, SES mitigates the higher odds of nontreatment in this population. Factors independently associated with not receiving lung cancer treatment include both SES and insurance status together with receiving treatment at a non-CCC, taking into account age, diagnosis year, distance to treatment facility, and clinical factors.
Few studies have examined younger adults with lung cancer who do not receive treatment.9,11,23 It is key to examine younger patients separately, because older patients inherently face more comorbidities and functional challenges that inform their treatment decisions.24,25 In younger patients, it is crucial to identify barriers to treatment that are related to healthcare delivery (insurance, SES, treatment facility).
Our results suggest that race/ethnicity is not associated with nontreatment in younger adults with lung cancer, once SES is accounted for. Among older patients with lung cancer, those who are black have inferior survival compared with NHW patients1,2 and receive surgical intervention and radiation less often,2–7 with comparable survival among patients undergoing surgical intervention.3,4 Consistent with our findings, those of single-institution studies15 and large studies of older patients4,26,27 emphasize the importance of accounting for not only clinical factors and receipt of treatment but also SES and access to treatment in black patients with lung cancer when considering reports of poor survival.4,27 According to our findings, SES effectively mitigates the association between race/ethnicity and nontreatment. These novel population-level results expand on conceptual frameworks from single-institution and small cohort studies in which SES is a mitigating factor in apparent racial disparities in staging and outcomes for adult-onset cancers.15,28,29 Because SES encompasses aspects of both income and education, this implies that treatment decisions may be influenced by both health literacy and the ability to afford treatment.
In our young cohort, nontreatment was more likely in patients without private insurance, echoing treatment patterns in studies of older patients.30,31 Lack of private insurance has been associated with inferior overall survival in both elderly and nonelderly populations.32–34 The association between treatment decisions and insurance status could be related to referral patterns, but also to out-of-pocket costs. For example, the study period was before implementation of the Patient Protection and Affordable Care Act, and therefore patients with private insurance likely had employer-sponsored coverage and may have been able to afford higher out-of-pocket costs than those with public or no insurance, thereby influencing treatment decisions. Distance to treatment facility was not associated with nontreatment, suggesting that SES and insurance status play a role in treatment decisions but that physical distance and transportation barriers do not. This is important to note in this Los Angeles County cohort, because patients in this population-dense area, which is home to >10 million residents in an area of <5,000 square miles,35 often face significant transportation challenges.
Nontreatment was 3 times more likely in patients cared for at non-CCCs versus CCCs. This population-level finding suggests that, adjusting for clinical and sociodemographic characteristics together with age and year of diagnosis, nonelderly patients with lung cancer who seek care at a CCC are more likely to receive treatment than comparable patients who seek care at a non-CCC; this expands on findings in older populations with a number of adult-onset cancers.9 Our team has previously shown that patients with lung cancer who receive care at CCCs have superior overall survival19; our present study indicates that one of the differences contributing to these outcomes includes nontreatment. Variations in therapy between CCCs and non-CCCs can be nuanced between facilities and not easily identified at the population level (eg, treatment protocols, multidisciplinary team approaches to care, availability of clinical trials, supportive care practices, availability of targeted therapy or immunotherapy). However, our findings show that such variations between CCCs and non-CCCs can also be basic differences, such as treatment/no treatment.
Patients with advanced-stage lung cancer, as expected,8,9,18,36,37 were less likely to receive treatment. Studies limited to advanced or metastatic disease report rates of nontreatment comparable to those in our study (21.0% vs 19.0%).11 It is conceivable that this trend is caused by either a real or perceived inability to treat the cancer, especially in older patients with shorter life expectancy. Although advanced stage is associated with nontreatment, a substantial proportion of our cohort (34.7%) did not have advanced disease; these patients represented nearly 20% of the “no treatment” cohort. As expected, age is also associated with nontreatment. In this young cohort, the trend seen when age was considered categorically clarifies that the magnitude of effect increases as age increases; specifically, when compared with patients diagnosed between ages 22 and 49 years, the highest odds of nontreatment are seen in the oldest patients (60–65 years: OR, 1.46; 95% CI, 1.24–1.71; 50–59 years: OR, 1.21; 95% CI, 1.04–1.43; P<.001) (supplemental eTable 3). This finding underscores the importance of recognizing not only the clinical factors but also the factors related to healthcare delivery (eg, SES, insurance) when considering these treatment decisions and potential barriers.
Due to the nature of registry data, our study was unable to account for comorbidities or performance status. However, this cohort was limited to younger patients (aged ≤65 years), thus minimizing the effect of functional status, comorbidities, and patient preference on treatment decision-making. Although registry data rely on case abstraction for data accuracy, recent literature supports minimal underascertainment of chemotherapy (10%), radiation (12%), and nontreatment (4%) in SEER38,39; the magnitude of effect in our findings suggests that underascertainment would not bias these results. Furthermore, the data were unable to differentiate between treatments that were recommended but declined and those that were not recommended. Finally, there is minimal granularity to the type of treatment received (ie, type of surgery or chemotherapy [eg, targeted agent vs traditional therapeutic approach]); thus, treatments were categorized broadly. With that in mind, new advancements in relevant lung cancer therapy occurred during the study period, including targeted therapy (tyrosine kinase inhibitors targeting EGFR, ALK) and immunotherapy. We categorized the treatment variable broadly to include all of these modalities; in addition, all analyses were adjusted for diagnosis year to account for the advent of new therapy over time. However, future studies are warranted to further understand the impact of the cost of and access to targeted therapy and their role in nontreatment. Finally, although the data do not include referral source to identify patients who self-referred to a CCC, it is possible that patients with a higher intent to receive treatment would self-refer to a CCC. Nevertheless, the fact that SES, insurance, and CCC remain independently associated with nontreatment in multivariable analysis indicates that they each maintain an effect when considering the other variables. This finding is consistent with the notion that education, income, and insurance are each a part of the construct of nontreatment in this population.
Conclusions
Disparities in race/ethnicity and nonreceipt of lung cancer treatment were explained by SES in nonelderly patients with varying stages of NSCLC and SCLC. Factors found to be independently associated with nonreceipt of lung cancer treatment in this population included SES, insurance status, and treatment at a non-CCC. These findings highlight the importance of future work to understand the underlying reasons why SES, insurance status, and treatment facility serve as barriers to receiving treatment among young patients with lung cancer. Specifically, understanding whether these factors serve as barriers to being offered or receiving treatment will be crucial in the development of targeted interventions, which could range from health literacy/educational programs to insurance-related policy and advocacy or even transportation assistance, depending on the ultimate conceptual model. Such work could potentially ameliorate disparities in treatment, leading to the optimization of healthcare delivery and improved outcomes among all patients.
Acknowledgments
The authors wish to acknowledge the contributions of Kelly Kenzik, PhD, and Smita Bhatia, MD, MPH, to this project.
References
- 1.↑
Gadgeel SM, Severson RK, Kau Y, et al.. Impact of race in lung cancer: analysis of temporal trends from a Surveillance, Epidemiology, and End Results database. Chest 2001;120:55–63.
- 2.↑
Greenwald HP, Polissar NL, Borgatta EF, et al.. Social factors, treatment, and survival in early-stage non-small cell lung cancer. Am J Public Health 1998;88:1681–1684.
- 3.↑
Lathan CS, Neville BA, Earle CC. The effect of race on invasive staging and surgery in non-small-cell lung cancer. J Clin Oncol 2006;24:413–418.
- 4.↑
Bach PB, Cramer LD, Warren JL, et al.. Racial differences in the treatment of early-stage lung cancer. N Engl J Med 1999;341:1198–1205.
- 5.↑
Polednak AP. Disparities in surgical treatment of early-stage non-small-cell lung cancer. Yale J Biol Med 2001;74:309–314.
- 6.↑
Esnaola NF, Gebregziabher M, Knott K, et al.. Underuse of surgical resection for localized, non-small cell lung cancer among whites and African Americans in South Carolina. Ann Thorac Surg 2008;86:220–226, discussion 227.
- 7.↑
Farjah F, Wood DE, Yanez ND, et al.. Racial disparities among lung cancer patients recommended operative therapy. Arch Surg 2009;144:14–18.
- 8.↑
Small AC, Tsao CK, Moshier EL, et al.. Prevalence and characteristics of patients with metastatic cancer who receive no anticancer therapy. Cancer 2012;118:5947–5954.
- 9.↑
Ward MM, Ullrich F, Matthews K, et al.. Who does not receive treatment for cancer? J Oncol Pract 2013;9:20–26.
- 10.↑
Wao H, Mhaskar R, Kumar A, et al.. Survival of patients with non-small cell lung cancer without treatment: a systematic review and meta-analysis. Syst Rev 2013;2:10.
- 11.↑
David EA, Daly ME, Li CS, et al.. Increasing rates of no treatment in advanced-stage non-small cell lung cancer patients: a propensity-matched analysis. J Thorac Oncol 2017;12:437–445.
- 12.↑
Brule SY, Al-Baimani K, Jonker H, et al.. Palliative systemic therapy for advanced non-small cell lung cancer: investigating disparities between patients who are treated versus those who are not. Lung Cancer 2016;97:15–21.
- 13.↑
Earle CC, Venditti LN, Neumann PJ, et al.. Who gets chemotherapy for metastatic lung cancer? Chest 2000;117:1239–1246.
- 14.↑
Patel MI, Schupp CW, Gomez SL, et al.. How do social factors explain outcomes in non-small-cell lung cancer among Hispanics in California? Explaining the Hispanic paradox. J Clin Oncol 2013;31:3572–3578.
- 15.↑
Bryant AS, Cerfolio RJ. Impact of race on outcomes of patients with non-small cell lung cancer. J Thorac Oncol 2008;3:711–715.
- 16.↑
Hardy D, Liu CC, Xia R, et al.. Racial disparities and treatment trends in a large cohort of elderly black and white patients with nonsmall cell lung cancer. Cancer 2009;115:2199–2211.
- 17.↑
Shugarman LR, Mack K, Sorbero ME, et al.. Race and sex differences in the receipt of timely and appropriate lung cancer treatment. Med Care 2009;47:774–781.
- 18.↑
Ramsey SD, Howlader N, Etzioni RD, et al.. Chemotherapy use, outcomes, and costs for older persons with advanced non–small-cell lung cancer: evidence from Surveillance, Epidemiology and End Results–Medicare. J Clin Oncol 2004;22:4971–4978.
- 19.↑
Wolfson JA, Sun CL, Wyatt LP, et al.. Impact of care at comprehensive cancer centers on outcome: results from a population-based study. Cancer 2015;121:3885–3893.
- 20.↑
CSPedia: Los Angeles Cancer Surveillance Program: user’s guide to data elements. Available at: http://cspedia.usc.edu/whnjs.htm. Accessed July 1, 2014.
- 21.↑
CSPedia: Los Angeles Cancer Surveillance Program: user’s guide to data elements. Patient/Demographics: introduction. Available at: http://cspedia.usc.edu/whnjs.htm. Accessed September 24, 2015.
- 22.↑
Bliss RL, Katz JN, Wright EA, et al.. Estimating proximity to care: are straight line and zipcode centroid distances acceptable proxy measures? Med Care 2012;50:99–106.
- 23.↑
Lazarus MA, Schachter L, Xavier M. Social factors, treatment, and survival in patients with advanced-stage non-small cell lung cancer. Transl Cancer Res 2014;3:146–151.
- 24.↑
de Rijke JM, Schouten LJ, ten Velde GP, et al.. Influence of age, comorbidity and performance status on the choice of treatment for patients with non-small cell lung cancer; results of a population-based study. Lung Cancer 2004;46:233–245.
- 25.↑
Li A, Zhou C, Xu J, et al.. Influence of comorbidity on the choice of treatment and survival of elderly patients with advanced non-small cell lung cancer. Chinese-German J Clin Oncol 2008;7:259–262.
- 26.↑
Mulligan CR, Meram AD, Proctor CD, et al.. Unlimited access to care: effect on racial disparity and prognostic factors in lung cancer. Cancer Epidemiol Biomarkers Prev 2006;15:25–31.
- 27.↑
Blackstock AW, Herndon JE II, Paskett ED, et al.. Similar outcomes between African American and non-African American patients with extensive-stage small-cell lung carcinoma: report from the Cancer and Leukemia Group B. J Clin Oncol 2006;24:407–412.
- 28.↑
Le H, Ziogas A, Lipkin SM, et al.. Effects of socioeconomic status and treatment disparities in colorectal cancer survival. Cancer Epidemiol Biomarkers Prev 2008;17:1950–1962.
- 29.↑
Schwartz KL, Crossley-May H, Vigneau FD, et al.. Race, socioeconomic status and stage at diagnosis for five common malignancies. Cancer Causes Control 2003;14:761–766.
- 30.↑
Groth SS, Al-Refaie WB, Zhong W, et al.. Effect of insurance status on the surgical treatment of early-stage non-small cell lung cancer. Ann Thorac Surg 2013;95:1221–1226.
- 31.↑
El Ezz KA, Olson DC, Silberstein PT, et al.. Treatment trends for stage III non-small cell lung cancer for noninsured patients compared with other types of insurances: a National Cancer Database analysis [abstract]. J Clin Oncol 2012;30(Suppl):Abstract 262.
- 32.↑
Walker GV, Grant SR, Guadagnolo BA, et al.. Disparities in stage at diagnosis, treatment, and survival in nonelderly adult patients with cancer according to insurance status. J Clin Oncol 2014;32:3118–3125.
- 33.↑
Slatore CG, Au DH, Gould MK; American Thoracic Society Disparities in Healthcare Group. An official American Thoracic Society systematic review: insurance status and disparities in lung cancer practices and outcomes. Am J Respir Crit Care Med 2010;182:1195–1205.
- 34.↑
Rajyaguru D, Park M, Awais O, et al.. The impact of insurance and race on non-small cell lung cancer outcomes [abstract]. J Clin Oncol 2015;33(Suppl):Abstract e17606.
- 35.↑
United States Census Bureau. Quick facts: Los Angeles County, California. Available at: https://www.census.gov/quickfacts/fact/table/losangelescountycalifornia/PST045217. Accessed July 18, 2018.
- 36.↑
Sacher AG, Le LW, Lau A, et al.. Real-world chemotherapy treatment patterns in metastatic non-small cell lung cancer: are patients undertreated? Cancer 2015;121:2562–2569.
- 37.↑
Lang K, Marciniak MD, Faries D, et al.. Trends and predictors of first-line chemotherapy use among elderly patients with advanced non-small cell lung cancer in the United States. Lung Cancer 2009;63:264–270.
- 38.↑
Kraus RD, Hamilton AS, Carlos M, et al.. Using hospital medical record data to assess the accuracy of the SEER Los Angeles Cancer Surveillance Program for initial treatment of prostate cancer: a small pilot study. Cancer Causes Control 2018;29:815–821.
- 39.↑
Healy MA, Morris AM, Abrahamse P, et al.. The accuracy of chemotherapy ascertainment among colorectal cancer patients in the Surveillance, Epidemiology, and End Results Registry Program. BMC Cancer 2018;18:481.
