the Division of Population Sciences, Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA

    ABSTRACT

    PURPOSE: Black patients with early-stage non–small-cell lung cancer (NSCLC) have worse overall survival than white patients. Decreased likelihood of resection has been implicated. To isolate the effect of decision making from access to care, we used receipt of surgical staging as a proxy for access and willingness to undergo invasive procedures, and examined treatments and outcomes by race.

    PATIENTS AND METHODS: We examined registry and claims data of Medicare-eligible patients with nonmetastatic NSCLC in areas monitored by the Surveillance, Epidemiology, and End Results program from 1991 to 2001. Patients who obtained invasive staging, defined as bronchoscopy, mediastinoscopy, or thoracoscopy, were included. Logistic regression and Cox modeling calculated the odds of having staging and surgery, and survival outcomes.

    RESULTS: A total of 14,224 patients underwent staging, and 6,972 had surgery for lung cancer. Black patients were less likely to undergo staging (odds ratio [OR] = 0.75; 95% CI, 0.67 to 0.83), and once staged, were still less likely to have surgery than whites (OR = 0.55; 95% CI, 0.47 to 0.64). Survival for blacks and whites was equivalent after resection (hazard ratio = 1.02; P = .06). Staged black patients were less likely to receive a recommendation for surgery when it was not clearly contraindicated (67.0% v 71.4%; P < .05), and were more likely to decline surgery (3.4% v 2.0%; P < .05).

    CONCLUSION: Black patients obtain surgery for lung cancer less often than whites, even after access to care has been demonstrated. They are more likely not to have surgery recommended, and more likely to refuse surgery. Additional research should focus on the physician-patient encounter as a potential source of racial disparities.

    INTRODUCTION

    Lung cancer is the leading cause of cancer mortality in the United States, with 157,000 deaths each year.1 During the last 40 years, there has been a decrease in lung cancer incidence and mortality in all races; however, a significant differential in outcomes between black and white patients remains; both incidence and mortality rates are highest in black men.2 Previous investigators have found that black race was a negative predictor of obtaining curative surgery for early-stage lung cancer and chemotherapy for metastatic disease.3,4 It has been suggested that some of the disparities in early-stage lung cancer survival are related to the gap in obtaining potentially curative surgery.3

    Invasive staging, which can include mediastinoscopy, bronchoscopy, and thoracoscopy, is usually part of the evaluation of non–small-cell lung cancer (NSCLC).5,6 It is also a marker of access to health care and a crude indication of willingness and ability to undergo invasive procedures. If access to care or willingness to accept surgery are the main reasons for observed disparities, as has been suggested,7 then disparities should not be apparent among a group of patients who have undergone invasive staging. Therefore, we have undertaken this analysis of the racial determinants of care among staged patients to evaluate whether disparities in treatment and outcomes still exist when these factors are controlled for in this way.

    PATIENTS AND METHODS

    Data Sources

    Patients from 11 tumor registries participating in the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program were studied. The registries capture 97% of the incident cases,8 covering a nearly representative sample of 14% of the population in the United States.9 The registries collect data on patient age, sex, race, ethnicity, cancer site, stage, histology, and dates of death and diagnosis. Medicare claims, both inpatient and outpatient, have been linked to SEER for patients aged 65 and older.10 Census-level demographic data have been linked to these patients as well.

    Cohort Selection

    Our study sample consisted of all Medicare-eligible patients older than age 65 who were diagnosed with NSCLC between January 1, 1991, and December 31, 1999. Because stage is determined partly at the time of definitive surgery, patients with stage I, II, and III disease were included for analysis. Patients with stage IV disease or unknown stage were excluded, except for those patients who had undergone a cancer-directed surgical procedure (1.7% of study sample), with the rationale that the extent of disease in these patients probably was not appreciated before surgery. It is important to note that results were similar when these surgical stage IV patients were excluded, and if analysis was limited only to stage I and stage II patients; thus, these subgroup analyses are not reported. Patients were excluded if they had any prior cancers. Those who were enrolled in a health maintenance organization at any time during the study period were also excluded, given that complete treatment data are not available for them. In addition, patients were excluded if the lung cancer was first identified at the time of death or autopsy.

    Identification of Staging and Surgical Procedures

    We defined invasive staging as mediastinoscopy, bronchoscopy, or thoracoscopy (video-assisted thoracic surgery was included in thoracoscopy) that occurred within ± 6 months of a diagnosis of NSCLC. International Classification of Diseases (9th revision, clinical modification [ICD9-CM]) and Current Procedural Terminology (CPT) codes were taken from outpatient and inpatient billing claims and used to define the following procedures: bronchoscopy used ICD9-CM 33.22, 33.24, 33.23, 33.27, and CPT 31622, 31625, 31628, 31629, 31640, 31641, 31646; mediastinoscopy used ICD9-CM 34.22, 34.29, and CPT 39400; thoracoscopy used ICD9-CM 34.21, 34.23, 34.24, and CPT 32601, 32602, 32604, 32605, 32606, 32657, 32661, 32662, 32663; pneumonectomy used ICD9-CM 32.50, 32.60, and CPT 32440, 32442, 32445; lobectomy used ICD9-CM 32.40, and CPT 32480, 32482, 32484, 32486; wedge resection used ICD9-CM 32.29, 32.30, and CPT 32500; local surgery used ICD9-CM 32.09, 32.10, and CPT 32520; and miscellaneous surgery used ICD9-CM 32.90, and CPT 32999. SEER identification of cancer-directed surgery was also used if these codes were not found in the Medicare claims.

    Definition of Explanatory Variables

    We categorized race and ethnicity into black (non-Hispanic African Americans) and white (non-Hispanic white patients). If patients did not fall into either of these two groups, they were classified as other. A total of 1,105 patients were classified as other. For the crude analysis we used black versus white only but included other in our multivariable analysis. We calculated the Charlson comorbidity index by identifying billing codes11 for various conditions during the year before diagnosis of cancer using the Deyo12 implementation of the Charlson score applied to both inpatient and outpatient claims, as suggested by Klabunde.13 Socioeconomic quintiles were developed based on the availability of information according to the following hierarchy: race- and age-specific median household income by census tract, unadjusted median household income by census tract, and median per capita income.14 Patients were classified as being treated in a teaching hospital if their records contained a charge for indirect medical education. To assess geographic variation, we divided the 11 SEER registries into five regions, Northeast, South, Midwest, Mountain, and Pacific. The Pacific region was used as a reference in logistic regression analysis given that this region contained the largest sample. Patients were considered to live in an urban area if they lived in or adjacent to a metropolitan county, or if they lived in an urban area as defined by SEER.

    Statistical Methods

    Univariable comparisons were performed using t tests for continuous variables and 2 tests for categoric variables. Given significant (P < .05) explanatory variables, we then constructed multivariable regression models to predict the likelihood of obtaining invasive staging, controlling for known confounders. We used multivariable regression models to predict the likelihood of surgery, stratifying by patients who had already received invasive staging versus those who did not. Kaplan-Meier curves were used to estimate survival statistics, and Cox regression analysis was performed to generate hazard ratios and adjust for known confounders.

    RESULTS

    Patient Selection

    an initial cohort of 21,219 patients, 14,224 patients underwent invasive staging: 13,030 (91.6%) white, and 1194 (8.4%) black. Table 1 shows the characteristics of the patients who underwent invasive staging by race. The most common histology for both races was squamous cell, followed by adenocarcinoma. The most common staging procedure in this cohort was bronchoscopy followed by mediastinoscopy. More black patients had bronchoscopy than white patients, and more white patients had mediastinoscopy and thoracoscopy. Black patients had a younger median age compared with white patients, and a much lower median income. Some patients had more than one surgical and/or staging procedure. In these patients, the most aggressive procedure is reported.

    Likelihood of Obtaining Staging

    We evaluated the odds of patients obtaining any of the aforementioned staging procedures, while controlling for known confounders by logistic regression modeling. As listed in Table 2, black race (odds ratio [OR] = 0.75; 95% CI, 0.67 to 0.83) and older age significantly decreased the odds of obtaining an invasive staging procedure. Being treated in a teaching hospital, having stage II disease, and living in the South or the Mountain region all significantly increased the likelihood that the patients would obtain invasive staging. Notably, comorbidity, socioeconomic status, sex, and having stage III disease did not affect the likelihood of undergoing invasive staging.

    Evaluation of Staged Patients

    We next examined the likelihood of surgery among staged patients (Table 3). Black race remained a powerful negative predictor of having surgery (OR = 0.55; 95% CI, 0.47 to 0.64). Increasing age, Charlson score, stage III disease, other race, female sex, and living in the Midwest and Mountain regions were also significant negative predictors. Being treated in a teaching hospital and having stage II disease were significantly positively associated with having surgery once staged. For sensitivity analysis we performed an additional analysis using the same logistic regression model on a cohort restricted to patients who obtained mediastinoscopy (data not shown). Black race remained a negative predictor of having surgery (OR = 0.58; 95% CI, 0.37 to 0.90).

    Kaplan-Meier and Cox regression analyses demonstrated no statistically significant difference in survival between black and white patients who underwent surgery. The median survival for black patients who underwent a surgical procedure was 1,045 (95% CI, 888 to 1,248 days) v 1,109 days (95% CI, 1,077 to 1,157 days) for white patients. The adjusted hazard ratio for black patients was 1.02 (P = .06). Similarly, there was no difference in survival among those who did not have a surgical procedure; median survival for black patients was 246 (95% CI, 221 to 271 days) v 248 days for whites (95% CI, 226 to 271; log-rank P = .46). The hazard ratio for black race was nonsignificant (1.07; P = .09).

    Reasons Cancer-Directed Surgery Was Not Performed

    DISCUSSION

    Our study, the first to examine the role of invasive staging in obtaining surgery for early-stage lung cancer, had many interesting findings. Black patients had invasive staging in the work-up of their lung cancer significantly less than white patients, when controlling for known confounders. Even when black patients had invasive staging, they remained far less likely than their white counterparts to have potentially curative surgery. Similar treatment resulted in similar outcomes, however. Review of the reasons that surgery was not performed indicated that black patients had surgery recommended less often than whites and also refused surgery more than white patients.

    The importance of invasive staging in lung cancer as a necessary part of the clinical assessment is well documented.5,15 Current recommendations advise the use of mediastinoscopy in addition to computed tomography scanning for patients with possibly operable NSCLC.6 Mediastinoscopy and bronchoscopy are the most common methods of obtaining staging and diagnosis information; however, thoracoscopy is used in certain situations as well.5,16 [18F]fluorodeoxyglucose positron emission tomography recently has been shown to be a valuable tool in staging for NSCLC.17,18 The SEER-Medicare database as of 2001 did not have a billing code for positron emission tomography imaging, and thus this modality could not be used in our analysis. Our aim in this study was to use staging as a proxy for access to health care, but also as a measure of the willingness and ability of the patient to undergo an invasive medical procedure. To obtain invasive staging, the patient needed to be seen by a specialist and then must have consented to an invasive staging procedure. The low frequency of mediastinoscopy is a potential concern—it suggests that many patients are not finding their way to the thoracic surgeons. This trend is exacerbated for black patients. Given this concern, subset analysis was performed on the patients obtaining mediastinoscopy. Our findings were unchanged. Therefore, our findings are not dependent on which specialist performs the staging procedures (pulmonologist or surgeon). Our finding that black patients obtained invasive staging less than white patients with comparable disease and comorbidities is consistent with other studies in this field.3,4,19,20 Studies in other medical fields have depicted disparities in the use of health care as well.21-23 Thus, the finding that black patients with nonmetastatic lung cancer are not staged at the same rates as their white counterparts is yet another example of racial disparities in health care.

    Although the observation that black patients obtained staging less than white patients was consistent with our initial hypothesis; the recognition that even when black patients did have staging, they did not have surgery at the same rate as white patients with comparable disease was a new observation. Given the marked difference in surgery between the races even when controlling for known confounders such as age, region, stage, and comorbid disease, the fact that staging did not ameliorate this result indicates that there is some selection factor occurring after the patient has demonstrated access to care, obtained a referral to a specialist, and shown willingness and ability to undergo an invasive medical procedure. Although our race category designated other is heterogeneous, those defined as other also obtained surgery less often than white patients. The most powerful negative predictor of surgery, as expected, was having stage III disease. The presence of comorbid disease did not significantly affect obtaining a staging procedure. For surgery however, comorbid disease was predictive, as we had expected. Region also seemed to have an effect on surgery in our model, but even in the face of these qualifiers, race remained a powerful negative predictor of obtaining surgery in our cohort.

    Starting with a population-based cohort with uniform insurance and examining only patients who had staging indicates that the racial effect possibly is happening at the level of the patient-physician encounter. The findings in Figure 1, that black patients were indeed having surgery recommended less often than whites and were refusing surgery at a higher rate, further supports this hypothesis. The exact reason that surgery was not recommended or was refused is not known, however. It has been proposed by some that black patients either refuse surgery because of an unclear understanding of the disease process24 or they present with symptoms that preclude surgery.7 Some have also suggested that black patients are simply not being referred to the appropriate specialist who would give adequate care.19,22 A recent study used SEER data during a 1-year period to show a wholesale decrease in therapy when blacks are compared to whites for surgery and chemotherapy.20 Clearly, the problem of racial disparities is multifactorial, including access to care, poor primary care, and perhaps some elemental mistrust of the medical system.23-25

    There are some limitations to our study. Procedure codes can suffer from coding error, leading to over- or underestimation of procedure rates. In addition, comorbidity adjustment using administrative data might miss some relevant factors that physicians use in choosing who is appropriate for surgery or who would tolerate surgery,13,26 and is not a proxy for performance status. There may be other patient-specific factors or findings on invasive staging that may affect subsequent decisions. In addition, the greatest racial difference in morbidity and mortality in lung cancer occurs in younger age groups.2 Medicare generally only covers patients older than age 65, however; therefore, we are unable to comment on the care of younger patients. In addition, we had to exclude any patients who were enrolled in a health maintenance organization, and these patients may have different patterns of care. There is no clear reason to think that these limitations would differentially affect decisions for black and white patients, however.

    Black patients do not have surgery for early-stage lung cancer at the same rate as white patients with similar stage disease and comorbidity. They are less likely to be staged, and even when staged, they are still less likely than white patients to have surgery. When they do receive surgery, their survival experience is similar to that of whites. Black patients have surgery recommended significantly less often than white patients, and are also refusing surgery more often. More attention should be paid to the physician-patient encounter that leads to the decision to have surgery, to examine whether patient and provider expectations and biases affect medical decision making in lung cancer. Efforts must continue to ensure the appropriate quality care is available for all patients, regardless of race and ethnicity.

    Authors’ Disclosures of Potential Conflicts of Interest

    The authors indicated no potential conflicts of interest.

    Author Contributions

    Conception and design: Christopher S. Lathan, Craig C. Earle

    Financial support: Craig C. Earle

    Collection and assembly of data: Bridget A. Neville

    Data analysis and interpretation: Christopher S. Lathan, Bridget A. Neville, Craig C. Earle

    Manuscript writing: Christopher S. Lathan, Bridget A. Neville, Craig C. Earle

    Final approval of manuscript: Christopher S. Lathan, Craig C. Earle

    NOTES

    Presented in part at the 39th Annual Meeting of the American Society of Clinical Oncology, Chicago, IL, May 31-June 3, 2003.

    Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.

    REFERENCES

    American Cancer Society: Cancer Facts and Figures 2003. American Cancer Society, 2003

    Stewart J: Lung cancer in African Americans. Cancer 91:2476-2481, 2001

    Bach PB, Cramer CL, Warren JL, et al: Racial differences in the treatment of early-stage lung cancer. N Engl J Med 341:1198-1205, 1999

    Earle C, Venditti LN, Nuemann PJ, et al: Who gets chemotherapy for metastatic lung cancer Chest 117:1239-1246, 2000

    Mentzer S, Swanson SJ, DeCamp MM, et al: Mediastinoscopy, thoracoscopy and VATS in the diagnosis and staging of lung cancer. Chest 112:239S-241S, 1997 (suppl 4)

    National Comprehensive Cancer Network: Clinical Practice Guidelines in Oncology v. 1.2004 Non Small Cell Lung Cancer, 2004 http//:www.nccn.org/professionals/physician_gls/PDF/nscl.pdf

    Blackstock AW, Herndon JE III, Paskett ED, et al: Outcomes among African-American/Non-African American patients with advanced non small cell lung carcinoma: Report from the Cancer and Leukemia Group B. J Natl Cancer Inst 94:284-290, 2002

    Zippin C, Lum D, Hankey BF: Completeness of hospital cancer case reporting from the SEER program of the National Cancer Institute. Cancer 76:2343-2350, 1995

    Nattinger AB, McAuliffe T, Schapira MM: Generalizability of the Surveillance, Epidemiology and End Results Registry population: Factors relevant to epidemiologic and health care research. J Clin Epidemiol 50:939-945, 1997

    Potosky AL, Riley GF, Lubitz JD, et al: Potential for cancer related health services research using a linked Medicare-tumor registry database. Med Care 31:732-748 1993

    Charlson ME, Pompei P, Ales KL, et al: A new method for classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis 40:373-383, 1986

    Deyo RA, Cherkin DC, Ciol MA, Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol 45:613-619, 1992

    Kalbunde C: Development of a comorbidity index using physician claims data. J Clin Epidemiol 53:1258-1267, 2000

    Krieger N: Overcoming the absence of socioeconomic data in medial records: Validation and application of a census-based methodology. Am J Public Health 82:703-710, 1992

    Sihoe AD, Yim AP: Lung cancer staging. J Surg Res 117:92-106, 2004

    Toloza E, Harpole L, Detterbeck F, et al: Invasive staging of non-small cell lung cancer. Chest 123:157-166, 2003 (suppl)

    Detterbeck FC, Falen S, Revera MP, et al: Seeking a home for PET: Part 2. Defining the appropriate place for positron emission tomography imaging in the staging of patients with suspected lung cancer. Chest 125:2300-2308, 2004

    Lardinosis D, Weder W, Hany TF, et al: Staging of non-small-cell lung cancer with integrated positron-emission tomography and computed tomography. N Engl J Med 348:2500-2508, 2003

    Earle CC, Neumann PJ, Gelber RD, et al: Impact of referral patterns on the use of chemotherapy for lung cancer. J Clin Oncol 20:1786-1792, 2002

    Potosky AL, Saxman S, Wallace RB, et al: Population variations in the initial treatment of non-small cell lung cancer. J Clin Oncol 22:3261-3268, 2004

    Ayanian JZ, Udvarhelyi IS, Gatsonis CA, et al: Racial differences in the use of revascularization procedures after coronary angiography. JAMA 269:2642-2646, 1993

    Bach PB, Pham H, Schrag D, et al: Primary care physicians who treat blacks and whites. N Engl J Med 351:575-584, 2004

    Gadgeel SM, Kalemkerian GP: Racial differences in lung cancer. Cancer Met Rev 22:39-46, 2003

    Margolis M, Christie JD, Silvestri GA, et al: Racial differences pertaining to a belief about lung cancer surgery. Ann Intern Med 139:558-563, 2003

    McDavid K, Tucker CT, Slogett A, et al: Cancer survival in Kentucky and health insurance coverage. Arch Intern Med 163:2135-2144, 2003

    Jazieh A, Kyasa MJ, Sethuraman S, et al: Disparities in surgical resection of early stage non small cell lung cancer. J Thorac Cardiovasc Surg 123:1173-1176, 2002

查询更多Cancer相关信息在本站>>

《临床肿瘤学医学期刊》2006年1月第24卷第1期