Effect of Insurance Status on the Surgical Treatment of Early-Stage Non-Small Cell Lung Cancer

Transcription

1 Effect of Insurance Status on the Surgical Treatment of Early-Stage Non-Small Cell Lung Cancer Shawn S. Groth, MD, MS, Waddah B. Al-Refaie, MD, Wei Zhong, MS, Selwyn M. Vickers, MD, Michael A. Maddaus, MD, Jonathan D Cunha, MD, PhD, and Elizabeth B. Habermann, MPH, PhD Department of Surgery, Division of Thoracic Surgery, Brigham and Women s Hospital, Boston Massachusetts; Department of Surgery, Division of Surgical Oncology, Georgetown University, Washington, DC; Division of Biostatistics, School of Public Health, Department of Surgery, Division of Surgical Oncology, and Department of Surgery, Division of Thoracic and Foregut Surgery, University of Minnesota, Minneapolis, Minnesota; Department of Surgery, Division of Thoracic and Foregut Surgery, University of Pittsburgh, Pittsburgh, Pennsylvania; and Mayo Clinic Department of Health Sciences Research, Division of Health Care Policy and Research, Rochester, Minnesota GENERAL THORACIC Background. Social disparities permeate non-small cell lung cancer (NSCLC) treatment, yet little is known about the effect of insurance status on the delivery of guideline surgical treatment for early-stage (I or II) NSCLC. Methods. We used the California Cancer Registry (1996 through 2008) to identify patients 50 to 94 years old with early-stage NSCLC. We used logistic regression models to assess whether or not insurance status (private insurance, Medicare, Medicaid, no insurance, and unknown) had an effect on whether or not a lobectomy (or bilobectomy) is performed. Results. A total of 10,854 patients met our inclusion criteria. Compared with patients with private insurance, we found that patients with Medicare (adjusted odds ratio [aor] 0.87; 95% confidence interval [CI]: 0.79 to 0.95), Medicaid (aor 0.45; 95% CI: 0.36 to 0.57), or no insurance (aor 0.45; 95% CI: 0.29 to 0.70) were significantly less likely to undergo lobectomy, even after adjusting for patient factors (age, race, and gender) and tumor characteristics (histology and tumor size). Increasing age, African American race, squamous cell carcinoma, and increasing tumor size were significant independent negative predictors of whether or not a lobectomy was performed. Conclusions. Patients without private insurance were significantly less likely than patients with private insurance to undergo a lobectomy for early-stage NSCLC. The variables(s) contributing to this disparity have yet to be elucidated. (Ann Thorac Surg 2013;95:1221 6) 2013 by The Society of Thoracic Surgeons As the number one cause of cancer-specific mortality in the world, lung cancer, whose most common type is non-small cell lung cancer (NSCLC), is a leading public health concern. Of the estimated 226,160 people in the United States who will be diagnosed with lung cancer in 2012 [1], nearly one fifth of them will have early-stage (American Joint Committee on Cancer [AJCC] stage I and II) disease [2]. Without treatment, the 5-year survival for early-stage NSCLC is poor (less than 5%) [3, 4]. With appropriate therapy, however, 5-year survival rates exceeding 65% can be achieved [5]. Therefore, access to and delivery of optimal treatment is essential, especially for malignancies such as NSCLC, where survival is significantly impacted by appropriate, high-quality treatment. Unfortunately, there is a growing body of evidence that racial [6 9], gender [8], and socioeconomic disparities [8, 10] permeate the timeliness of and appropriateness of lung cancer staging and treatment [11]. As a result, such disparities have a negative impact on survival rates [6, 7, 10, 12, 13]. Accepted for publication Oct 31, Address correspondence to Dr Habermann, Mayo Clinic: Division of Health Care Research and Policy, 200 First St SW, Rochester, MN 55905; habermann.elizabeth@mayo.edu. For patients with early-stage NSCLC, the best opportunity for cure remains surgical resection in those patients with sufficient cardiopulmonary reserve. The current standard of care for patients who have sufficient cardiopulmonary reserve is an anatomic lobectomy with mediastinal lymph node dissection or sampling [5, 14]. Unfortunately, health care insurance status often dictates whether or not an operation is performed [10, 13, 15, 16], treatment is conducted at high-volume centers [17, 18], and recommended treatment guidelines are followed [19]. Notably absent from the literature (to the best of our knowledge) are studies examining the extent to which insurance status impacts whether or not a lobectomy is performed for early-stage NSCLC. We hypothesized that insurance status strongly predicts receipt of lung cancer surgery for early-stage NSCLC. Material and Methods Data The California Cancer Registry (CCR) is a statewide population-based cancer dataset that was established by California s Department of Public Health Cancer Surveillance and Research Branch in 1985 and began state-wide 2013 by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc

2 GENERAL THORACIC 1222 GROTH ET AL Ann Thorac Surg INSURANCE STATUS AND LOBECTOMY FOR NSCLC 2013;95: data collection in 1988 [20]. Because the randomized trial that established lobectomy as the current standard of care for treating early-stage NSCLC was published in 1995 [5], we used data collected from January 1, 1996 (to allow for a year of dissemination of the data from that study), through December 31, Insurance status was classified as private insurance, Medicare, Medicaid, no insurance, or unknown. Inclusion Criteria We selected patients from the CCR database for inclusion in our study using the International Classification Disease for Oncology, 3rd Edition (ICD-O3) topography and morphology codes for NSCLC [21]. We only included patients 50 years and older with AJCC (6th edition) clinical stage I or II NSCLC who underwent a lobectomy, a sublobar resection (either an anatomic segmentectomy or wedge resection), or who did not undergo a resection. Though the AJCC staging system was recently revised, we chose to use the 6th edition because surgeons treating patients during the time period of our study would have used the 6th edition to make treatment decisions. Exclusion Criteria Due to concern for confounding, we excluded patients with more than 1 primary tumor and patients who were unlikely to have received aggressive cancer treatment based on their reporting source (a nursing home, coroner s office, or death certificate). To preserve anonymity, patients 95-years old and older were excluded from our study. Patients who underwent pneumonectomy or sleeve resection (who likely had proximal tumors and wouldn t be candidates for lobectomy), local ablation, chest wall resection, and patients whose surgical treatment was unknown were excluded from our analysis. Statistical Analysis Data were analyzed using SAS version 9.1 (SAS Institute, Cary, NC). Between-group comparisons were made using a 2 test for categoric variables. Univariate and multivariate logistic regression models were utilized to assess the association between insurance status (private insurance, Medicare, Medicaid, no insurance, and unknown) and whether or not a lobectomy was performed. For all statistical testing, we used a 2-sided significance level of Results Of the 149,291 patients in the CCR database with invasive lung cancer, 10,854 met our inclusion criteria (Fig 1). We noted significant difference between patients who underwent a lobectomy and those who did not (Table 1). Patients who did not undergo lobectomy were more likely to be older, men, or African American, and were more likely to have larger tumors. They were also less likely to have private insurance. On univariate analysis, we found that patients with Medicare, Medicaid, or no insurance were significantly less likely to undergo lobectomy, as compared with patients with private insurance (Table 2). After adjusting for patient factors (age, race, and gender) and tumor characteristics (histology and tumor size), our results were unchanged; patients with Medicare, Medicaid, or no insurance were significantly less likely to undergo lobectomy than patients with private insurance. We identified several other independent patient (ie, age and race) and tumor-level predictors (ie, histology and size) of whether or not a lobectomy was performed. With increasing age, patients were less likely to undergo lobectomy. African Americans were significantly less likely to undergo lobectomy as compared with Caucasians. Compared with patients with adenocarcinomas, patients with bronchoalveolar carcinomas were more likely to undergo a lobectomy while patients with squamous cell carcinomas and other NSCLC histologies were less likely to undergo lobectomy. Finally, with increasing tumor size, patients were less likely to undergo lobectomy (Table 2). Because non-caucasians may not have the same access to care as Caucasians, we assessed for interaction between race and insurance status and did not find a significant association (p 0.24). This indicates that the effect of insurance status on lobectomy rates does not vary by race. Because age is correlated to insurance status (patients 65 years and older have access to Medicare), we also assessed for interaction between age and insurance status and found a significant association (p ). Consequently, we also stratified our analysis by age (Table 3). For patients under 65 years, those with private insurance were more likely to undergo lobectomy. For patients 65 years and older, however, insurance status did not significantly impact whether or not a lobectomy was performed. Comment There is a growing body of evidence that social disparities permeate NSCLC treatment. In particular, a number of studies indicate that insurance status impacts the quality of care, including whether or not resection (the cornerstone of early-stage NSCLC treatment in patients with sufficient cardiopulmonary reserve) is performed. Using data from over 10,000 patients from California, we found that early-stage NSCLC patients with Medicare, Medicaid, or no insurance were significantly less likely than patients with private insurance to undergo lobectomy. To our knowledge, this is the first study that has specifically examined the association between insurance status and the standard of care resection (ie, lobectomy) for early-stage NSCLC. We believe that our study makes a unique contribution to the literature on the social disparities in lung cancer treatment. Previous studies in the literature that examined the association between insurance status and the receipt of surgery for early-stage NSCLC did not specifically examine the type of resection that was performed (ie, pneumonectomy, lobectomy, anatomic segmentectomy, or wedge resection) [15, 16], yet the choice of operation has a significant impact on local recurrence

3 Ann Thorac Surg GROTH ET AL 2013;95: INSURANCE STATUS AND LOBECTOMY FOR NSCLC CCR 1996 Invasive Lung Cancer N = 149,291 Other Histology N = 30,509 Fig 1. Patients included in study. (CCR California Cancer Registry; NSCLC nonsmall cell lung cancer.) 1223 GENERAL THORACIC NSCLC N = 118,782 Other Stages N = 101,642 Stage I (N = 14,008) Stage II (N = 3,132) > 1 Primary Tumor N = 5,144 One Primary Tumor N = 11,996 Excluded Based On Reporting Source N = 41 Reported Source Included in Study N = 11,955 Age < 50 or > 94 N = 359 Age 50 and < 94 N = 11,596 Included in Study N = Local Excision (N = 66) Sleeve Resection (N = 4) Pneumonectomy (N = 373) En-bloc Resection (N=254) Unknown (N = 45) and overall survival rates [5]. Other studies included other lung cancer histologies [19] and did not specifically provide results for early-stage NSCLC patients [19]. As such, it is difficult to make conclusions regarding disparities in early-stage NSCLC treatment based on insurance status based on their data. In contrast to our study, Potosky and colleagues [22] found no significant difference between patients with private and patients with public insurance with regard to the delivery of guideline therapy (which they defined as either lobectomy or pneumonectomy) for patients with early-stage NSCLC [22]. Using data obtained from the Surveillance Epidemiology and End-Results Database in 1996, they assessed whether or not patient characteristics (ie, age, race, gender, comorbidity score, and smoking history) or socioeconomic factors (ie, marital status, household income, urban versus rural residence, and insurance) influenced receipt of guideline therapy for stage I to IV NSCLC. Similar to our study, they found that patients with private insurance were more likely than patients with public insurance (74% vs 68%) to receive recommended treatment for early-stage NSCLC on univariate analysis. However, after adjusting for other factors, they found that the receipt of guideline therapy was not significantly different between patients with private and patients with public insurance. The discrepancy between their multivariate analysis and ours is likely due (at least in part) to differences in the study designs. For instance, they created an aggregate variable for public insurance (though patients with Medicare, Medicaid, Veterans Affairs, or other forms of public insurance may be different), and they did not include patients without insurance in their analysis. Furthermore, our definition of guideline therapy (receipt of lobectomy) was different than theirs (receipt of lobectomy or pneumonectomy). The reasons behind the disparity in cancer care based on insurance status are likely multifactorial, including both health care system and patient factors. Several studies that utilized statewide administrative databases have demonstrated that Medicaid and uninsured patients are more likely to be treated at low-volume centers as compared with patients with private insurance or Medicare [13, 17]. Lung cancer patients who are treated at high-volume centers have higher survival rates as com-

4 GENERAL THORACIC 1224 GROTH ET AL Ann Thorac Surg INSURANCE STATUS AND LOBECTOMY FOR NSCLC 2013;95: pared with patients who are treated a low-volume centers [13, 17, 23]. Though the factors that underlie the association between hospital volume and outcomes have not been fully delineated, adherence to recommended cancer treatment guidelines may be important. Similar to other studies, we found that age and race were significant predictors of whether or not patients undergo lobectomy for NSCLC [8, 9, 24]. Not surprisingly (as older patients are more likely to be enrolled in Medicare), we found a significant interaction between age and insurance status. However, when we stratified by age, our results were unchanged. For patients younger than 65 years, those without private insurance were significantly less likely to undergo lobectomy as compared with patients with private insurance. For patients 65 years and older, however, this difference was not significant. Of note, however, the trend in the odds ratios was unchanged for patients 65 years and older the odds of undergoing a lobectomy were lower in patients without private insurance. Possible explanations for a lack of Table 1. Demographics of Patients Who Underwent Lobectomy Versus Patients Who Did Not Number of Patients No Lobectomy 5,633 No Resection (n 4,642) Sublobar Resection (n 991) Age 50 to 59 years 6.7% 10.1% 15.1% 60 to 69 years 19.8% 27.3% 34.2% 70 to 79 years 41.3% 43.4% 38.9% 80 to 94 years 32.2% 19.2% 11.8% Gender Men 50.3% 45.0% 46.1% Women 49.7% 55.0% 53.9% Race White 84.1% 88.7% 85.9% Black 7.6% 4.7% 5.3% Other 8.3% 6.6% 8.8% Tumor size 2 cm 16.3% 53.6% 28.4% 2.1 to 5 cm 50.8% 40.3% 57.7% 5 cm 18.0% 3.5% 12.9% Unknown 14.9% 2.6% 1% Histology Adenocarcinoma 27.3% 42.7% 49.8% Squamous cell 34.2% 25.3% 24.3% Bronchoalveolar 3.0% 17.4% 13.7% Other 35.6% 14.6% 12.2% Insurance Private 33.3% 36.2% 45.2% Medicare 59.1% 58.3% 49.7% Medicaid 4.6% 3.9% 3.1% None 1.1% 0.3% 0.8% Unknown 1.9% 1.3% 1.2% Lobectomy 5,221 p Value Table 2. Univariate and Multivariate Analyses Assessing the Association Between Insurance Status and Whether or Not a Lobectomy Was Performed Univariate Analysis Multivariate Analysis OR 95% CI OR 95% CI Age 50 to 59 years 1.00 Reference 60 to 69 years to 79 years to 94 years Gender Men 1.00 Reference Women Race White 1.00 Reference Black Other Tumor Size 2 cm 1.00 Reference 2.1 to 5 cm cm Unknown Histology Adenocarcinoma 1.00 Reference Squamous cell BAC Other Insurance Private 1.00 Reference 1.00 Reference Medicare Medicaid None Unknown BAC bronchoalveolar carcinoma; CI confidence interval; OR odd ratio. statistical significance in the older age group include the following: (1) a difference in the strength of the effect of insurance status on older patients; (2) a lack of power (as Table 3. Multivariate Analyses Assessing the Association Between Insurance Status and Whether or Not a Lobectomy Was Preformed Stratified by Age a Insurance 50 to 64 Years Old 65 Years and Older OR 95% CI OR 95% CI Private 1.00 Reference 1.00 Reference Medicare Medicaid None Unknown a Odds ratios (OR) adjusted for age, gender, race, tumor size, and histology. CI confidence interval.

5 Ann Thorac Surg GROTH ET AL 2013;95: INSURANCE STATUS AND LOBECTOMY FOR NSCLC most [66%] patients 65 years and older in our study were classified as Medicare recipients); or (3) insurance misclassification. For example, patients who have Medicare hospitalization insurance may also have a private policy (a secondary insurer from a former employer), which pays for premiums or other health care expenses. Such a patient may be classified in a dataset as having either Medicare or private insurance. In fact, data from the Centers for Medicare and Medicaid Services suggest that over 94% of patients 65 years and older receive Medicare services (rather than 66% of patients in our dataset) [25]. This discrepancy may be due to difficulty classifying such patients with dual enrollment. We also found that African Americans were significantly less likely to undergo lobectomy as compared with Caucasians, but did not observe a significant interaction between race and insurance status. This indicates that the effect of insurance status on lobectomy rates does not vary by race. Other factors are involved. In particular, social factors may be important. For example, African Americans are less likely to be offered surgical treatment options (even when it is not contraindicated) and are more likely to decline surgery when it was offered [9]. Other investigators have found that this disparity has significant implications - African Americans have lower lung cancer survival rates, which are largely explained by lower rates of tumor resection [6, 24]. Of note, for patients who undergo resection, survival rates are similar between ethnic groups, suggesting that there is a disparity in the delivery of guideline therapy to appropriate candidates [7]. The reasons behind this disparity have yet to be elucidated. We acknowledge a number of limitations of our study. Our study was a retrospective review of a prospectively maintained administrative database. As such, our study has the inherent limitations of all observational studies (ie, selection bias). There are also several particular limitations of our dataset. Insurance misclassification in elderly patients (as previously discussed) may be important. In addition, there are a number of important factors in the delivery of optimal lung cancer treatment that we were unable to assess or adjust for with our data, including hospital volume, treatment by a multidisciplinary team (including a board certified thoracic surgeon) [26], pretreatment staging [26], and patient comorbidities (including underlying pulmonary function) [24]. Socioeconomic status is also an important predictor of the receipt of recommended lung cancer treatment [8, 16] and likely is significantly associated with insurance status. Unfortunately, we were unable to assess the impact of socioeconomic status using our dataset. It is essential to identify impediments to the delivery of quality cancer treatment to optimize the outcomes of our patients, especially in the setting of health payer systems that demand increasingly efficient, high-quality care. It is well known that the treatment of choice for patients with early-stage NSCLC and sufficient cardiopulmonary reserve is surgery [26]. However, in addition to simply offering all appropriate candidates an operation, the type of resection has significant implications. In the only published randomized trial on surgery for early-stage NSCLC, lobectomy was associated with a 2.4-fold (versus segmentectomy) to threefold (versus wedge resection) reduction in local recurrence rates, even for tumors smaller than 1 cm. Furthermore, compared with sublobar resections, lobectomy was associated with a 50% reduction in cancer-specific mortality [5]. Given the impact of the extent of resection on outcomes based on this level I evidence, we believe that our study (which identified disparities in the receipt of lobectomy for early-stage NSCLC) has important implications. Though social disparities in NSCLC treatment clearly exist, much has yet to be learned in order to narrow the gap in the treatment of and survival from lung cancer. One of the central issues in the national health care reform debate has been the cost and scope of health insurance coverage. Certainly, improving access to costeffective health care (through expanded insurance coverage) is an important goal in the pursuit of health care reform; other issues are likely important as well. In particular, we must identify specific factors that impede the delivery of quality care and strive to find solutions to the foundational problems that contribute to these social disparities. We would like to thank Dr Anasooya Abraham for her advice regarding use of the California Cancer Registry. This study was supported with a grant from the National Institutes of Health s National Center on Minority Health and Health Disparities. References Siegel R, Naishadham D, Jemal A. Cancer statistics, CA Cancer J Clin 2012;62: Surveillance, Epidemiology, and End Results (SEER) Program ( SEER*Stat Database: Incidence - SEER 17 Regs Public-Use, Nov 2009 Sub ( varying), National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch, released April 2010, based on the November 2009 submission. 3. Salomaa ER, Liippo K, Taylor P, et al. Prognosis of patients with lung cancer found in a single chest radiograph screening. Chest 1998;114: Flehinger BJ, Kimmel M, Melamed MR. The effect of surgical treatment on survival from early lung cancer. Implications for screening. Chest 1992;101: Ginsberg RJ, Rubinstein LV. Randomized trial of lobectomy versus limited resection for T1 N0 non-small cell lung cancer. Lung Cancer Study Group. Ann Thorac Surg 1995; 60: Bach PB, Cramer LD, Warren JL, Begg CB. Racial differences in the treatment of early-stage lung cancer. N Engl J Med 1999;341: Wisnivesky JP, McGinn T, Henschke C, Hebert P, Iannuzzi MC, Halm EA. Ethnic disparities in the treatment of stage I non-small cell lung cancer. Am J Respir Crit Care Med 2005;171: 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: 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: GENERAL THORACIC

6 GENERAL THORACIC 1226 GROTH ET AL Ann Thorac Surg INSURANCE STATUS AND LOBECTOMY FOR NSCLC 2013;95: Greenberg ER, Chute CG, Stukel T, et al. Social and economic factors in the choice of lung cancer treatment. A populationbased study in two rural states. N Engl J Med 1988;318: Groth SS, D Cunha J. Lung cancer outcomes: the effects of socioeconomic status and race. Semin Thorac Cardiovasc Surg 2010;22: McDavid K, Tucker TC, Sloggett A, Coleman MP. Cancer survival in Kentucky and health insurance coverage. Arch Intern Med 2003;163: Cheung MC, Hamilton K, Sherman R, et al. Impact of teaching facility status and high-volume centers on outcomes for lung cancer resection: an examination of 13,469 surgical patients. Ann Surg Oncol 2009;16: Darling GE, Allen MS, Decker PA, et al. Randomized trial of mediastinal lymph node sampling versus complete lymphadenectomy during pulmonary resection in the patient with N0 or N1 (less than hilar) non-small cell carcinoma: results of the American College of Surgery Oncology Group Z0030 Trial. J Thorac Cardiovasc Surg 2011;141: Bradley CJ, Dahman B, Given CW. Treatment and survival differences in older Medicare patients with lung cancer as compared with those who are dually eligible for Medicare and Medicaid. J Clin Oncol 2008;26: 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: Liu JH, Zingmond DS, McGory ML, et al. Disparities in the utilization of high-volume hospitals for complex surgery. JAMA 2006;296: Neighbors CJ, Rogers ML, Shenassa ED, Sciamanna CN, Clark MA, Novak SP. Ethnic/racial disparities in hospital procedure volume for lung resection for lung cancer. Med Care 2007;45: Harlan LC, Greene AL, Clegg LX, Mooney M, Stevens JL, Brown ML. Insurance status and the use of guideline therapy in the treatment of selected cancers. J Clin Oncol 2005; 23: California Cancer Registry. Available at: ccrcal.org. Accessed February 1, Fritz AG, Percy C, Jack A, Sobin LH. International Classification of Diseases for Oncology, 3rd. Ed. Geneva: World Health Organization; Potosky AL, Saxman S, Wallace RB, Lynch CF. Population variations in the initial treatment of non-small-cell lung cancer. J Clin Oncol 2004;22: Birkmeyer JD, Siewers AE, Finlayson EV, et al. Hospital volume and surgical mortality in the United States. N Engl J Med 2002;346: Cykert S, Dilworth-Anderson P, Monroe MH, et al. Factors associated with decisions to undergo surgery among patients with newly diagnosed early-stage lung cancer. JAMA 2010;303: DeLew N. Medicare: 35 Years of Service. Health Care Financ Rev 2000;Fall; 22: The NCCN Clinical Practice Guidelines in Oncology Non- Small Cell Lung Cancer (Version ). National Comprehensive Cancer Network, Inc. Available at: NCCN.org. Accessed January 4, 2011.