- Email: [email protected]
Health Insurance Status and Disparities in Kidney Cancer Care
urologypracticejournal.com
Health Insurance Status and Disparities in Kidney Cancer Care Hung-Jui Tan,* Ryan J. Chuang, Joseph D. Shirk, Aaron A. Laviana and Jim C. Hu From the Veterans Affairs/University of California-Los Angeles Robert Wood Johnson Clinical Scholars Program and Department of Urology, David Geffen School of Medicine (RJC, JDS, AAL, JCH), University of California-Los Angeles (HJT), Los Angeles, California
Abstract Introduction: Through PPACA (Patient Protection and Affordable Care Act) many adults have or will gain health insurance via Medicaid expansion. To understand how this policy change may potentially impact patients with kidney cancer we examined the relationship between insurance status and cancer related outcomes. Methods: Using SEER (Surveillance, Epidemiology and End Results) data we identified 18,632 patients 26 to 64 years old with kidney cancer from 2007 to 2009. For each patient we classified insurance status as no insurance, Medicaid or private insurance. After adjusting for patient and county characteristics we measured the association of insurance status with cancer stage, treatment and 1-year mortality using multinomial logistic regression with clustering or generalized estimating equations as appropriate. Results: In our study cohort 937 (5.0%) and 2,027 patients (10.9%) had no insurance and Medicaid, respectively. These patients were more likely to be younger, nonwhite, unmarried and residing in areas with lower income, education or employment (p <0.001). On adjusted analyses uninsured and Medicaid patients more often presented with advanced disease (21.3% vs 19.6% vs 11.0%) but less frequently received treatment (86.2% vs 87.9% vs 93.4%, each p <0.001) compared with privately insured patients. These adults also died of kidney cancer more often (13.6% vs 12.5% vs 6.4%, p <0.001) likely due to differences in stage and receipt of cancer directed therapy. Conclusions: Uninsured and Medicaid patients suffer disproportionately from kidney cancer with equal magnitude. Given the reliance on Medicaid, even as insurance coverage expands differences in outcomes will likely persist, underscoring the need for additional efforts that address disparities in kidney cancer care. Key Words: kidney neoplasms; quality of health care; insurance, health; Medicaid; Patient Protection and Affordable Care Act
Kidney cancer has increased in incidence in the last 2 decades. Now the seventh most common solid tumor in the United States, kidney cancer accounts for 63,920 new cases
and 13,860 corresponding deaths annually.1e3 Although it is more frequently diagnosed in the elderly population, kidney cancer has become increasingly common in younger adults.4
Submitted for publication March 6, 2015. No direct or indirect commercial incentive associated with publishing this article. The corresponding author certifies that, when applicable, a statement(s) has been included in the manuscript documenting institutional review board, ethics committee or ethical review board study approval; principles of Helsinki Declaration were followed in lieu of formal ethics committee approval; institutional animal care and use committee approval; all human subjects provided written informed consent with guarantees of
confidentiality; IRB approved protocol number; animal approved project number. Supported by Veterans Affairs Office of Academic Affiliations through the Veterans Affairs/Robert Wood Johnson Clinical Scholars Program (HJT) and the H & H Lee Surgical Research Scholars Program (AAL). * Correspondence: Veterans Affairs/University of California-Los Angeles Robert Wood Johnson Clinical Scholars Program, University of California-Los Angeles, 10940 Wilshire Blvd., Suite 710, Los Angeles, California 90024 (telephone: 310794-2206; FAX: 310-794-3288; e-mail address: [email protected]).
2352-0779/16/31-18/0 UROLOGY PRACTICE Ó 2016 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
AND
RESEARCH, INC.
http://dx.doi.org/10.1016/j.urpr.2015.05.009 Vol. 3, 18-24, January 2016 Published by Elsevier
Health Insurance Status and Disparities in Kidney Cancer Care
In fact, in the last decade the cancer incidence has approximately doubled in adults 20 to 40 years old, highlighting a potential new epidemiological trend in kidney cancer.4,5 For these younger adults aggressive surgical treatment remains the standard of care.6 Even so some patients face difficulties in obtaining appropriate and timely treatment due to a lack of health insurance. As many as 18.5% of working age adults in the United States are without health coverage with rates peaking at 23.5% for adults 26 to 34 years old.7,8 Previous population based studies have identified a link between a lack of insurance and more advanced stage disease.9 However the impact of insurance status on care use and outcomes in kidney cancer remains poorly defined. For other major malignancies patients without insurance encounter lower rates of treatment and poorer survival.10,11 Cancer outcomes among patients with Medicaid also appear to lag behind those observed among adults with private insurance.11 Accordingly we used the SEER database to compare kidney cancer severity, treatment and short-term outcomes according to insurance status. In doing so we can begin to anticipate how changes in insurance coverage expected through PPACA may impact the growing segment of young and middle-aged adults now being diagnosed with renal cancer.
Materials and Methods Data Source and Study Cohort
We used NCI (National Cancer Institute) SEER data to identify patients diagnosed with incident kidney cancer in the United States from 2007 to 2009. SEER is a nationally representative cancer registry that collects data on incidence, treatment and mortality.12 The SEER program captures cases from 18 registries (ie Alaska, Atlanta, Connecticut, Detroit, Greater California, Greater Georgia, Hawaii, Iowa, Kentucky, Los Angeles, Louisiana, New Jersey, New Mexico, Rural Georgia, San Francisco-Oakland, San JoseMonterey, Seattle-Puget Sound and Utah), encompassing 28% of the American population. Drawing from the entire data set we identified 37,435 patients with primary nonurothelial kidney cancer based on ICD-3 site code C64.9 and ICD-9 clinical modification code 189.0. We excluded from analysis 476 patients identified by death certificate or autopsy and narrowed our sample to adults 26 to 64 years old due to Medicare eligibility and the extension of dependent health coverage, leaving 19,213. We next excluded 23 patients (0.1% of the sample) from the Alaska Native Registry as none were uninsured, having presumably obtained coverage through the Indian Health
19
Service. Finally we removed 520 patients (2.7% of sample) with unknown insurance status and 38 (0.2%) with bilateral disease to produce a final cohort of 18,632 patients. Primary Exposure and Outcomes
Using insurance information available in SEER beginning in 2007 we assigned patients to 1 of 3 categories, including 1) no insurance, 2) Medicaid coverage and 3) private insurance. We then considered certain primary outcomes. 1) We assessed disease severity, classifying stage in accordance with AJCC (American Joint Committee on Cancer) staging groups I to IV.13 2) We created a binary variable for no treatment, including those potentially on active surveillance, vs any treatment, consisting of cancer directed surgery or radiation therapy based on SEER treatment variables. Among patients with stage I disease we constructed a 3-category treatment variable, classifying management as nonoperative, nephron sparing (ie ablation and partial nephrectomy) or radical nephrectomy. 3) We measured all cause and kidney cancer specific mortality 1 year from diagnosis given the available followup for this cohort. Covariates
For each patient we determined age, gender, race/ethnicity, marital status, geographical region and year of diagnosis. In addition to characterizing tumor stage we ascertained tumor histology and grade for our survival analyses. Because SEER does not include conventional measures of socioeconomic status and comorbidity, we used data provided by the Area Health Resource File from the United States Department of Health and Human Services Health Resources and Services Administration.14 These databases contain county level measures of health services access, resource use, socioeconomic indicators and health status. Through Federal Information Processing Standard county codes we linked these measures to kidney cancer cases in SEER. We characterized the local care environment in terms of median household income, nonhigh school education and unemployment; the density of urologists, total physicians, kidney cancer cases, managed care, hospitals and hospital beds; and rates of death from heart disease, diabetes mellitus, chronic obstructive pulmonary disease, liver disease and/or cancer per county population. Rural/urban status and the number of cancer hospitals per county were also identified. Analyses Statistical. We evaluated the association of insurance status with each primary outcome and covariate using the
20
Health Insurance Status and Disparities in Kidney Cancer Care
chi-square test. We then determined the association between insurance status and each outcome measure using multinomial logistic regression or generalized estimating equations as appropriate, adjusting for patient and county factors and county level clustering. To assess treatment we also adjusted for stage and performed stage specific analyses. When comparing mortality, we constructed an additional model that further adjusted for tumor stage, histology, grade and treatment. Finally we determined the model adjusted probability of each outcome according to insurance status and used bootstrapping with replacement for 1,000 replications to obtain 95% CIs for our predicted stage, treatment and mortality. Sensitivity. To assess the robustness of our findings we performed several sensitivity analyses. 1) To ensure that our county level measures substantively contributed to our model we performed likelihood ratio tests comparing models without these covariates to the full model. 2) Because insurance status may relate to age and income, we performed subgroup analyses by patient age and median income in the residing county. 3) We performed secondary survival analysis using Cox proportional hazard models. All statistical testing was 2-sided, completed with STATAÒ, version 13.1 and done at the 5% significance level. The UCLA institutional review board determined that this study did not require formal review, given the publicly available, de-identified nature of the data set.
Results
From 2007 to 2009 we identified 937 patients (5.0%) with no insurance, 2,027 (10.9%) with Medicaid and 15,668 (84.1%) with private insurance. Private insurance was more common among older, white and married patients, and those who resided in urban locales or areas with greater income, education or employment (p <0.001, supplementary table, http://urologypracticejournal.com/). Private insurance was also associated with stage I disease, nonclear cell renal cell carcinoma and well or moderately differentiated tumors (p <0.001). After adjusting for patient and county characteristics patients with no insurance or Medicaid had a 25.7% (p <0.001) and 12.6% (p ¼ 0.032) lower probability of stage I disease than those with private insurance (fig. 1). Uninsured and Medicaid patients had a twofold increase in stage IV disease compared to their privately insured counterparts (p <0.001). With respect to treatment patients with no insurance or Medicaid were less likely to receive an intervention compared to patients with private insurance (adjusted probability 86.2% vs 87.9% vs 93.4%, p <0.001). Treatment
Figure 1. Probability of stage based on AJCC kidney cancer staging manual by insurance status adjusted for patient demographics and county level measures of health with 95% CI obtained from bootstrapping with replacement for 1,000 replications. Dark blue bars indicate stage I. Light blue bars indicate stage II. Gray bars indicate stage III. Yellow bars indicate stage IV.
probabilities for stages I, II and III disease were substantial across all insurance groups but highest in those with private insurance (fig. 2). Nonmetastatic, stage specific treatment probabilities did not differ between Medicaid and uninsured patients (p >0.500). For stage IV disease Medicaid patients had a 16.8% lower probability of nonmedical treatment compared to privately insured patients (p <0.001). Uninsured patients had the lowest likelihood of treatment, that is 30.8% and 16.8% lower than patients with private insurance and Medicaid (p <0.001 and 0.018, respectively). In the subset of patients with stage I disease those with private insurance had a 27.1% higher probability of receiving nephron sparing treatment than Medicaid patients (p <0.001) but they did not differ significantly from uninsured patients (p ¼ 0.193, fig. 3).
Figure 2. Probability of kidney cancer stage specific treatment by insurance status with probabilities adjusted for patient demographics and county level measures of health, and 95% CI obtained from bootstrapping with replacement for 1,000 replications. Dark blue bars indicate no insurance. Light blue bars indicate Medicaid. Yellow bars indicate private insurance.
Health Insurance Status and Disparities in Kidney Cancer Care
Figure 3. Predicted treatment for patients with stage I disease by insurance status with probabilities adjusted for patient demographics and county level measures of health, and 95% CI obtained from bootstrapping with replacement for 1,000 replications. Dark blue bars indicate nonoperative. Light blue bars indicate nephron sparing. Gray bars indicate radical nephrectomy.
The table shows unadjusted and adjusted 1-year mortality. When adjusting for patient and county characteristics only, uninsured and Medicaid patients died more frequently of any cause and of kidney cancer specifically compared to adults with insurance (p <0.001). No difference was observed between patients with Medicaid and no insurance (all cause p ¼ 0.766 and kidney cancer specific p ¼ 0.417). However after further accounting for stage, histopathology and treatment mortality appeared similar across all groups with only Medicaid patients having a slightly higher probability of 1-year all cause mortality compared to uninsured and privately insured patients (p ¼ 0.027 and 0.001, respectively). Our findings did not substantively change on our sensitivity analyses.
Discussion
Disparities in health services use and outcomes have been a long-standing concern in oncology. In kidney cancer differences in practice patterns and survival have been identified with respect to gender, race and socioeconomic
21
position.15e17 In this study we found additional variation attributable to insurance status. From 2007 to 2009 uninsured and Medicaid covered patients more often presented with aggressive disease compared to privately insured patients. After diagnosis these patients less frequently received treatment and more often died, indicating that younger adults with government or no insurance suffered disproportionately. Despite the relatively short followup differences in the onset, management and outcomes of advanced disease between patients with and without private insurance appear pronounced, consistent with other malignancies.11 Our findings indicate that Medicaid covered and uninsured patients have a much higher frequency of metastatic cancer. Furthermore these patients less often receive cytoreductive nephrectomy, a resource intense procedure that has been associated with improved outcomes in carefully selected patients.18 Although this could reflect increasing use of medical treatments among uninsured and Medicaid covered patients, previous studies have shown decreased use of systemic therapy (eg immunotherapy and/or targeted therapy) in these patients compared to those with private insurance.19,20 Altogether these considerations likely drive the observed mortality differences associated with insurance status, particularly for the uninsured population. Aside from metastatic disease patients without insurance and those with Medicaid otherwise fared comparably. For many other malignancies Medicaid coverage has been linked to intermediate levels of cancer directed therapy and survival, which are higher than those in uninsured patients but less than among those who are privately insured.11 However in kidney cancer uninsured and Medicaid patients have equivalent outcomes for nonmetastatic disease at least in the first year.6 For small renal malignancies uninsured adults show higher levels of nephron sparing surgery, which is one of the primary quality standards in the field. Heterogeneity in the uninsured group likely has a role. Patients without insurance may experience a temporary
Table. One-year mortality probability according to insurance status 1-Yr Mortality* Unadjusted: All cause Ca Adjusted for pt þ county characteristics: All cause Ca Adjusted for pt þ county characteristics, stage, treatment þ histopathology: All cause Ca
% No Insurance (95% CI)
% Medicaid (95% CI)
% Private Insurance (95% CI)
18.1 (15.4e20.7) 13.6 (11.4e15.9)
18.6 (16.7e20.3) 12.5 (11.0e13.9)
9.1 (8.6e9.6) 6.4 (6.0e6.8)
17.3 (14.8e19.9) 13.2 (11.2e15.7)
17.7 (15.6e19.4) 12.3 (10.4e13.8)
9.2 (8.7e9.7) 6.4 (6.0e6.9)
10.3 (9.0e11.8) 7.5 (6.5e8.9)
12.1 (10.6e13.1) 8.1 (7.2e8.9)
10.3 (9.8e10.9) 7.3 (6.8e7.8)
*Probabilities derived from GEEs accounting for clustering at county level with CIs obtained by bootstrapping with replacement for 1,000 replications.
22
Health Insurance Status and Disparities in Kidney Cancer Care
fluctuation in insurance status due to changes in economic or social circumstances. They may also obtain care through charitable means or prefer self-payment. Although it is not well understood, county hospitals and private providers have been found to vary substantially in the treatment of disadvantaged men with prostate cancer. A similar phenomenon may occur with respect to insurance type and nephron sparing treatment in patients with kidney cancer.21 Even so these findings suggest that quality concerns in kidney cancer appear to be equally pressing in patients with Medicaid, which is a population expected to grow through PPACA. Our findings should be considered within the scope of several limitations. 1) We could not adjust for patient comorbidity, which may bias our findings. However, we included county level measures of health in our analytical models and adjusted for socioeconomic indicators, which are viewed as major determinants of health status, especially for under resourced populations.22 Furthermore, differences noted with respect to disease stage and kidney cancer specific mortality are less likely to be attributable to omitted variable bias with respect to comorbidity. 2) SEER does not provide detailed information on insurance plans, including the extent of benefits, provider network and/or co-payments that may also influence kidney cancer care. Nonetheless it is likely that the differences between patients who are insured vs uninsured are much larger than the differences among insured patients with varying policies. 3) As mentioned, heterogeneity likely exists in the uninsured group with some patients unable to afford insurance and others refusing coverage purposely for alternative reasons. However our findings remained consistent even when stratifying our cohort according to age and income areas. 4) The quality of safety net systems and Medicaid benefits can vary by state, which in turn may influence outcomes. However state of residence by and large did not correlate with treatment or survival in our multivariable regression models. 5) SEER does not provide more granular information related to patient underlying tumor anatomy, diagnostic workup, treatment approach and followup, limiting more detailed assessments of insurance based disparities. 6) Our data may not reflect more recent trends in kidney cancer treatment, such as the emergence of active surveillance, ablative therapies and robotic surgery, which increased in popularity after the conclusion of our study interval. Notwithstanding these limitations our study has potential implications for kidney cancer care delivery in the postPPACA era. Early returns indicate that the uninsured rate has already decreased due in large part to Medicaid expansion by participating states.7 To further improve cancer care ASCO (American Society of Clinical Oncology) has urged remaining states to expand Medicaid or offer a
comparable alternative.23 While this is promising, notably previous state level reforms have yielded mixed results. Although not specific to cancer, Medicaid expansion in New York, Arizona and Maine demonstrated small improvements in access, self-reported health and mortality.24 In contrast gaining Medicaid insurance in Oregon did not translate into better health outcomes for newly covered patients with chronic disease.25 In Massachusetts nephrectomy rates remained unchanged following insurance expansion but the use of partial nephrectomy increased in nonwhite and low income populations.26 Extending health coverage remains a key step to improving cancer outcomes but the benefit of insurance expansion alone may be somewhat limited in kidney cancer, given the similarities in stage, treatment and survival between uninsured and Medicaid covered patients. Accordingly further study is needed to address health disparities as they pertain to insurance status and kidney cancer. As noted, patients without private insurance tend to be nonwhite and reside in under resourced areas, which in turn have been associated with disparities in kidney cancer care.15e17 Such social determinants may serve as barriers to health care even when Medicaid benefits are made available. In fact, compared to Medicare beneficiaries those with Medicaid are far less likely to be offered cancer screening or be surgically evaluated for cancer, highlighting issues related to access.27,28 Raising Medicaid reimbursement as advocated by ASCO may help offset these disparities.23 Recent data suggest that greater office visit reimbursements may raise levels of cancer care in accordance with guidelines.29 By understanding and intervening on mechanisms that contribute to inequitable care across insurance types we can more effectively decrease disparities and improve the health of more vulnerable patients with kidney cancer. Conclusions
Uninsured patients with kidney cancer currently present with more aggressive disease and receive treatment less frequently, highlighting a population in need of better care. While insurance expansion through PPACA offers a potential mechanism, many uninsured patients will receive Medicaid coverage, which is also associated with suboptimal quality. These findings underscore the need for additional efforts aimed at improving care in Medicaid beneficiaries, especially patients with kidney cancer. References 1. Hock LM, Lynch J and Balaji KC: Increasing incidence of all stages of kidney cancer in the last 2 decades in the United States:
Health Insurance Status and Disparities in Kidney Cancer Care
an analysis of surveillance, epidemiology and end results program data. J Urol 2002; 167: 57. 2. Siegel R, Ma J, Zou Z et al: Cancer statistics, 2014. CA Cancer J Clin 2014; 64: 9. 3. Chow WH, Dong LM and Devesa SS: Epidemiology and risk factors for kidney cancer. Nature reviews. Urology 2010; 7: 245. 4. King SC, Pollack LA, Li J et al: Continued increase in incidence of renal cell carcinoma, especially in young patients and high grade disease: United States 2001 to 2010. J Urol 2014; 191: 1665.
23
16. Vaishampayan UN, Do H, Hussain M et al: Racial disparity in incidence patterns and outcome of kidney cancer. Urology 2003; 62: 1012. 17. Becker A, Roghmann F, Trinh QD et al: Sociodemographic disparities in the treatment of small renal masses. BJU Int 2013; 111: E274. 18. Flanigan RC, Salmon SE, Blumenstein BA et al: Nephrectomy followed by interferon alfa-2b compared with interferon alfa-2b alone for metastatic renal-cell cancer. N Engl J Med 2001; 345: 1655.
5. Tan HJ, Filson CP and Litwin MS: Contemporary, age-based trends in the incidence and management of patients with earlystage kidney cancer. Urol Oncol 2015; 33: 21.e19.
19. Saigal CS, Deibert CM, Lai J et al: Disparities in the treatment of patients with IL-2 for metastatic renal cell carcinoma. Urol Oncol 2010; 28: 308.
6. Novick AC, Campbell SC, Belldegrun A et al: Guideline for Management of the Clinical Stage 1 Renal Mass. Available at http://www.auanet.org/education/guidelines/renal-mass.cfm. Accessed January 25, 2015.
20. Smaldone MC, Handorf E, Kim SP et al: Temporal trends and factors associated with receipt of systemic therapy among patients undergoing cytoreductive nephrectomy: an analysis of the National Cancer Database. J Urol 2015; 193: 1108.
7. Sommers BD, Musco T, Finegold K et al: Health reform and changes in health insurance coverage in 2014. N Engl J Med 2014; 371: 867.
21. Parsons JK, Kwan L, Connor SE et al: Prostate cancer treatment for economically disadvantaged men: a comparison of county hospitals and private providers. Cancer 2010; 116: 1378.
8. Smith JC and Medalia C: Health Insurance Coverage in the United States: 2013. US Census Bureau, Current Population Reports P60250. Washington, D.C.: U.S. Government Printing Office 2014.
22. Marmot M: Social determinants of health inequalities. Lancet 2005; 365: 1099.
9. Halpern MT, Ward EM, Pavluck AL et al: Association of insurance status and ethnicity with cancer stage at diagnosis for 12 cancer sites: a retrospective analysis. Lancet Oncol 2008; 9: 222. 10. Niu X, Roche LM, Pawlish KS et al: Cancer survival disparities by health insurance status. Cancer Med 2013; 2: 403. 11. 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. 12. Hankey BF, Ries LA and Edwards BK: The surveillance, epidemiology, and end results program: a national resource. Cancer Epidemiol Biomarkers Prev 1999; 8: 1117.
23. Polite BN, Griggs JJ, Moy B et al: American Society of Clinical Oncology policy statement on Medicaid reform. J Clin Oncol 2014; 32: 4162. 24. Sommers BD, Baicker K and Epstein AM: Mortality and access to care among adults after state Medicaid expansions. N Engl J Med 2012; 367: 1025. 25. Baicker K, Taubman SL, Allen HL et al: The Oregon experimentdeffects of Medicaid on clinical outcomes. N Engl J Med 2013; 368: 1713. 26. Ellimoottil C, Miller S, Wei JT et al: Anticipating the impact of insurance expansion on inpatient urological surgery. Urol Pract 2014; 1: 134.
13. Edge SB, Byrd DR, Compton CC et al: AJCC Cancer Staging Manual, 7th ed. New York: Springer 2010.
27. Bradley CJ, Dahman B and Given CW: Inadequate access to surgeons: reason for disparate cancer care? Med Care 2009; 47: 758.
14. Area Health Resources Files. Rockville: Bureau of Health Professions, Health Resources and Services Administration, United States Department of Health and Human Services 2012-2013.
28. Schuur JD, Shah A, Wu Z et al: The impact of Medicaid coverage and reimbursement on access to diagnostic mammography. Cancer 2009; 115: 5566.
15. Stafford HS, Saltzstein SL, Shimasaki S et al: Racial/ethnic and gender disparities in renal cell carcinoma incidence and survival. J Urol 2008; 179: 1704.
29. Halpern MT, Romaire MA, Haber SG et al: Impact of statespecific Medicaid reimbursement and eligibility policies on receipt of cancer screening. Cancer 2014; 120: 3016.
Editorial Commentary
Unfavorable socioeconomic status and health insurance status negatively impact patient access to health care and treatment options for cancer.1,2
Despite the limitations of this study pivotal observations regarding health care disparities in kidney cancer in the United States demonstrated conspicuous benefits
24
Health Insurance Status and Disparities in Kidney Cancer Care
for privately insured patients compared to other groups. Undeniably universal health care is a moral imperative. Can the PPACA repair these disparities in health care? We do not know. What we know is that 63% of physicians are recruited by hospitals vs by physicians.3 Ultimately if PPACA is fully implemented, the input of health care access may improve but services rendered and output issues may suffer unintended consequences. Proactive measures with PPACA must address the decreased number of specialists, decreased physician reimbursement and compensation and increased cost of medical education as well as policies that may interfere with physician fiduciary obligations with patients. Soon the Supreme Court will decide the fate of PPACA. It is our hope that these disparities seen only with our patients will not create new disparities among providers.
Fernando J. Kim Department of Urology Denver Health Medical Center Minimally Invasive Urological Oncology University of Colorado Cancer Center Denver Department of Surgery University of Colorado Denver Denver, Colorado References 1. Adler NE and Newman K: Socioeconomic disparities in health: pathways and policies. Health Affairs 2002; 21: 60. 2. Denberg TD, Beaty BL and Kim FJ: Marriage and ethnicity predict treatment in localized prostate carcinoma. Cancer 2005; 103: 1819. 3. Kirchhoff SM: Physician Practices: Background, Organization, and Market Consolidation. Washington, D.C.: Congressional Research Service 2013.
Health Insurance Status and Disparities in Kidney Cancer Care Hung-Jui Tan,* Ryan J. Chuang, Joseph D. Shirk, Aaron A. Laviana and Jim C. Hu From the Veterans Affairs/University of California-Los Angeles Robert Wood Johnson Clinical Scholars Program and Department of Urology, David Geffen School of Medicine (RJC, JDS, AAL, JCH), University of California-Los Angeles (HJT), Los Angeles, California
Abstract Introduction: Through PPACA (Patient Protection and Affordable Care Act) many adults have or will gain health insurance via Medicaid expansion. To understand how this policy change may potentially impact patients with kidney cancer we examined the relationship between insurance status and cancer related outcomes. Methods: Using SEER (Surveillance, Epidemiology and End Results) data we identified 18,632 patients 26 to 64 years old with kidney cancer from 2007 to 2009. For each patient we classified insurance status as no insurance, Medicaid or private insurance. After adjusting for patient and county characteristics we measured the association of insurance status with cancer stage, treatment and 1-year mortality using multinomial logistic regression with clustering or generalized estimating equations as appropriate. Results: In our study cohort 937 (5.0%) and 2,027 patients (10.9%) had no insurance and Medicaid, respectively. These patients were more likely to be younger, nonwhite, unmarried and residing in areas with lower income, education or employment (p <0.001). On adjusted analyses uninsured and Medicaid patients more often presented with advanced disease (21.3% vs 19.6% vs 11.0%) but less frequently received treatment (86.2% vs 87.9% vs 93.4%, each p <0.001) compared with privately insured patients. These adults also died of kidney cancer more often (13.6% vs 12.5% vs 6.4%, p <0.001) likely due to differences in stage and receipt of cancer directed therapy. Conclusions: Uninsured and Medicaid patients suffer disproportionately from kidney cancer with equal magnitude. Given the reliance on Medicaid, even as insurance coverage expands differences in outcomes will likely persist, underscoring the need for additional efforts that address disparities in kidney cancer care. Key Words: kidney neoplasms; quality of health care; insurance, health; Medicaid; Patient Protection and Affordable Care Act
Kidney cancer has increased in incidence in the last 2 decades. Now the seventh most common solid tumor in the United States, kidney cancer accounts for 63,920 new cases
and 13,860 corresponding deaths annually.1e3 Although it is more frequently diagnosed in the elderly population, kidney cancer has become increasingly common in younger adults.4
Submitted for publication March 6, 2015. No direct or indirect commercial incentive associated with publishing this article. The corresponding author certifies that, when applicable, a statement(s) has been included in the manuscript documenting institutional review board, ethics committee or ethical review board study approval; principles of Helsinki Declaration were followed in lieu of formal ethics committee approval; institutional animal care and use committee approval; all human subjects provided written informed consent with guarantees of
confidentiality; IRB approved protocol number; animal approved project number. Supported by Veterans Affairs Office of Academic Affiliations through the Veterans Affairs/Robert Wood Johnson Clinical Scholars Program (HJT) and the H & H Lee Surgical Research Scholars Program (AAL). * Correspondence: Veterans Affairs/University of California-Los Angeles Robert Wood Johnson Clinical Scholars Program, University of California-Los Angeles, 10940 Wilshire Blvd., Suite 710, Los Angeles, California 90024 (telephone: 310794-2206; FAX: 310-794-3288; e-mail address: [email protected]).
2352-0779/16/31-18/0 UROLOGY PRACTICE Ó 2016 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
AND
RESEARCH, INC.
http://dx.doi.org/10.1016/j.urpr.2015.05.009 Vol. 3, 18-24, January 2016 Published by Elsevier
Health Insurance Status and Disparities in Kidney Cancer Care
In fact, in the last decade the cancer incidence has approximately doubled in adults 20 to 40 years old, highlighting a potential new epidemiological trend in kidney cancer.4,5 For these younger adults aggressive surgical treatment remains the standard of care.6 Even so some patients face difficulties in obtaining appropriate and timely treatment due to a lack of health insurance. As many as 18.5% of working age adults in the United States are without health coverage with rates peaking at 23.5% for adults 26 to 34 years old.7,8 Previous population based studies have identified a link between a lack of insurance and more advanced stage disease.9 However the impact of insurance status on care use and outcomes in kidney cancer remains poorly defined. For other major malignancies patients without insurance encounter lower rates of treatment and poorer survival.10,11 Cancer outcomes among patients with Medicaid also appear to lag behind those observed among adults with private insurance.11 Accordingly we used the SEER database to compare kidney cancer severity, treatment and short-term outcomes according to insurance status. In doing so we can begin to anticipate how changes in insurance coverage expected through PPACA may impact the growing segment of young and middle-aged adults now being diagnosed with renal cancer.
Materials and Methods Data Source and Study Cohort
We used NCI (National Cancer Institute) SEER data to identify patients diagnosed with incident kidney cancer in the United States from 2007 to 2009. SEER is a nationally representative cancer registry that collects data on incidence, treatment and mortality.12 The SEER program captures cases from 18 registries (ie Alaska, Atlanta, Connecticut, Detroit, Greater California, Greater Georgia, Hawaii, Iowa, Kentucky, Los Angeles, Louisiana, New Jersey, New Mexico, Rural Georgia, San Francisco-Oakland, San JoseMonterey, Seattle-Puget Sound and Utah), encompassing 28% of the American population. Drawing from the entire data set we identified 37,435 patients with primary nonurothelial kidney cancer based on ICD-3 site code C64.9 and ICD-9 clinical modification code 189.0. We excluded from analysis 476 patients identified by death certificate or autopsy and narrowed our sample to adults 26 to 64 years old due to Medicare eligibility and the extension of dependent health coverage, leaving 19,213. We next excluded 23 patients (0.1% of the sample) from the Alaska Native Registry as none were uninsured, having presumably obtained coverage through the Indian Health
19
Service. Finally we removed 520 patients (2.7% of sample) with unknown insurance status and 38 (0.2%) with bilateral disease to produce a final cohort of 18,632 patients. Primary Exposure and Outcomes
Using insurance information available in SEER beginning in 2007 we assigned patients to 1 of 3 categories, including 1) no insurance, 2) Medicaid coverage and 3) private insurance. We then considered certain primary outcomes. 1) We assessed disease severity, classifying stage in accordance with AJCC (American Joint Committee on Cancer) staging groups I to IV.13 2) We created a binary variable for no treatment, including those potentially on active surveillance, vs any treatment, consisting of cancer directed surgery or radiation therapy based on SEER treatment variables. Among patients with stage I disease we constructed a 3-category treatment variable, classifying management as nonoperative, nephron sparing (ie ablation and partial nephrectomy) or radical nephrectomy. 3) We measured all cause and kidney cancer specific mortality 1 year from diagnosis given the available followup for this cohort. Covariates
For each patient we determined age, gender, race/ethnicity, marital status, geographical region and year of diagnosis. In addition to characterizing tumor stage we ascertained tumor histology and grade for our survival analyses. Because SEER does not include conventional measures of socioeconomic status and comorbidity, we used data provided by the Area Health Resource File from the United States Department of Health and Human Services Health Resources and Services Administration.14 These databases contain county level measures of health services access, resource use, socioeconomic indicators and health status. Through Federal Information Processing Standard county codes we linked these measures to kidney cancer cases in SEER. We characterized the local care environment in terms of median household income, nonhigh school education and unemployment; the density of urologists, total physicians, kidney cancer cases, managed care, hospitals and hospital beds; and rates of death from heart disease, diabetes mellitus, chronic obstructive pulmonary disease, liver disease and/or cancer per county population. Rural/urban status and the number of cancer hospitals per county were also identified. Analyses Statistical. We evaluated the association of insurance status with each primary outcome and covariate using the
20
Health Insurance Status and Disparities in Kidney Cancer Care
chi-square test. We then determined the association between insurance status and each outcome measure using multinomial logistic regression or generalized estimating equations as appropriate, adjusting for patient and county factors and county level clustering. To assess treatment we also adjusted for stage and performed stage specific analyses. When comparing mortality, we constructed an additional model that further adjusted for tumor stage, histology, grade and treatment. Finally we determined the model adjusted probability of each outcome according to insurance status and used bootstrapping with replacement for 1,000 replications to obtain 95% CIs for our predicted stage, treatment and mortality. Sensitivity. To assess the robustness of our findings we performed several sensitivity analyses. 1) To ensure that our county level measures substantively contributed to our model we performed likelihood ratio tests comparing models without these covariates to the full model. 2) Because insurance status may relate to age and income, we performed subgroup analyses by patient age and median income in the residing county. 3) We performed secondary survival analysis using Cox proportional hazard models. All statistical testing was 2-sided, completed with STATAÒ, version 13.1 and done at the 5% significance level. The UCLA institutional review board determined that this study did not require formal review, given the publicly available, de-identified nature of the data set.
Results
From 2007 to 2009 we identified 937 patients (5.0%) with no insurance, 2,027 (10.9%) with Medicaid and 15,668 (84.1%) with private insurance. Private insurance was more common among older, white and married patients, and those who resided in urban locales or areas with greater income, education or employment (p <0.001, supplementary table, http://urologypracticejournal.com/). Private insurance was also associated with stage I disease, nonclear cell renal cell carcinoma and well or moderately differentiated tumors (p <0.001). After adjusting for patient and county characteristics patients with no insurance or Medicaid had a 25.7% (p <0.001) and 12.6% (p ¼ 0.032) lower probability of stage I disease than those with private insurance (fig. 1). Uninsured and Medicaid patients had a twofold increase in stage IV disease compared to their privately insured counterparts (p <0.001). With respect to treatment patients with no insurance or Medicaid were less likely to receive an intervention compared to patients with private insurance (adjusted probability 86.2% vs 87.9% vs 93.4%, p <0.001). Treatment
Figure 1. Probability of stage based on AJCC kidney cancer staging manual by insurance status adjusted for patient demographics and county level measures of health with 95% CI obtained from bootstrapping with replacement for 1,000 replications. Dark blue bars indicate stage I. Light blue bars indicate stage II. Gray bars indicate stage III. Yellow bars indicate stage IV.
probabilities for stages I, II and III disease were substantial across all insurance groups but highest in those with private insurance (fig. 2). Nonmetastatic, stage specific treatment probabilities did not differ between Medicaid and uninsured patients (p >0.500). For stage IV disease Medicaid patients had a 16.8% lower probability of nonmedical treatment compared to privately insured patients (p <0.001). Uninsured patients had the lowest likelihood of treatment, that is 30.8% and 16.8% lower than patients with private insurance and Medicaid (p <0.001 and 0.018, respectively). In the subset of patients with stage I disease those with private insurance had a 27.1% higher probability of receiving nephron sparing treatment than Medicaid patients (p <0.001) but they did not differ significantly from uninsured patients (p ¼ 0.193, fig. 3).
Figure 2. Probability of kidney cancer stage specific treatment by insurance status with probabilities adjusted for patient demographics and county level measures of health, and 95% CI obtained from bootstrapping with replacement for 1,000 replications. Dark blue bars indicate no insurance. Light blue bars indicate Medicaid. Yellow bars indicate private insurance.
Health Insurance Status and Disparities in Kidney Cancer Care
Figure 3. Predicted treatment for patients with stage I disease by insurance status with probabilities adjusted for patient demographics and county level measures of health, and 95% CI obtained from bootstrapping with replacement for 1,000 replications. Dark blue bars indicate nonoperative. Light blue bars indicate nephron sparing. Gray bars indicate radical nephrectomy.
The table shows unadjusted and adjusted 1-year mortality. When adjusting for patient and county characteristics only, uninsured and Medicaid patients died more frequently of any cause and of kidney cancer specifically compared to adults with insurance (p <0.001). No difference was observed between patients with Medicaid and no insurance (all cause p ¼ 0.766 and kidney cancer specific p ¼ 0.417). However after further accounting for stage, histopathology and treatment mortality appeared similar across all groups with only Medicaid patients having a slightly higher probability of 1-year all cause mortality compared to uninsured and privately insured patients (p ¼ 0.027 and 0.001, respectively). Our findings did not substantively change on our sensitivity analyses.
Discussion
Disparities in health services use and outcomes have been a long-standing concern in oncology. In kidney cancer differences in practice patterns and survival have been identified with respect to gender, race and socioeconomic
21
position.15e17 In this study we found additional variation attributable to insurance status. From 2007 to 2009 uninsured and Medicaid covered patients more often presented with aggressive disease compared to privately insured patients. After diagnosis these patients less frequently received treatment and more often died, indicating that younger adults with government or no insurance suffered disproportionately. Despite the relatively short followup differences in the onset, management and outcomes of advanced disease between patients with and without private insurance appear pronounced, consistent with other malignancies.11 Our findings indicate that Medicaid covered and uninsured patients have a much higher frequency of metastatic cancer. Furthermore these patients less often receive cytoreductive nephrectomy, a resource intense procedure that has been associated with improved outcomes in carefully selected patients.18 Although this could reflect increasing use of medical treatments among uninsured and Medicaid covered patients, previous studies have shown decreased use of systemic therapy (eg immunotherapy and/or targeted therapy) in these patients compared to those with private insurance.19,20 Altogether these considerations likely drive the observed mortality differences associated with insurance status, particularly for the uninsured population. Aside from metastatic disease patients without insurance and those with Medicaid otherwise fared comparably. For many other malignancies Medicaid coverage has been linked to intermediate levels of cancer directed therapy and survival, which are higher than those in uninsured patients but less than among those who are privately insured.11 However in kidney cancer uninsured and Medicaid patients have equivalent outcomes for nonmetastatic disease at least in the first year.6 For small renal malignancies uninsured adults show higher levels of nephron sparing surgery, which is one of the primary quality standards in the field. Heterogeneity in the uninsured group likely has a role. Patients without insurance may experience a temporary
Table. One-year mortality probability according to insurance status 1-Yr Mortality* Unadjusted: All cause Ca Adjusted for pt þ county characteristics: All cause Ca Adjusted for pt þ county characteristics, stage, treatment þ histopathology: All cause Ca
% No Insurance (95% CI)
% Medicaid (95% CI)
% Private Insurance (95% CI)
18.1 (15.4e20.7) 13.6 (11.4e15.9)
18.6 (16.7e20.3) 12.5 (11.0e13.9)
9.1 (8.6e9.6) 6.4 (6.0e6.8)
17.3 (14.8e19.9) 13.2 (11.2e15.7)
17.7 (15.6e19.4) 12.3 (10.4e13.8)
9.2 (8.7e9.7) 6.4 (6.0e6.9)
10.3 (9.0e11.8) 7.5 (6.5e8.9)
12.1 (10.6e13.1) 8.1 (7.2e8.9)
10.3 (9.8e10.9) 7.3 (6.8e7.8)
*Probabilities derived from GEEs accounting for clustering at county level with CIs obtained by bootstrapping with replacement for 1,000 replications.
22
Health Insurance Status and Disparities in Kidney Cancer Care
fluctuation in insurance status due to changes in economic or social circumstances. They may also obtain care through charitable means or prefer self-payment. Although it is not well understood, county hospitals and private providers have been found to vary substantially in the treatment of disadvantaged men with prostate cancer. A similar phenomenon may occur with respect to insurance type and nephron sparing treatment in patients with kidney cancer.21 Even so these findings suggest that quality concerns in kidney cancer appear to be equally pressing in patients with Medicaid, which is a population expected to grow through PPACA. Our findings should be considered within the scope of several limitations. 1) We could not adjust for patient comorbidity, which may bias our findings. However, we included county level measures of health in our analytical models and adjusted for socioeconomic indicators, which are viewed as major determinants of health status, especially for under resourced populations.22 Furthermore, differences noted with respect to disease stage and kidney cancer specific mortality are less likely to be attributable to omitted variable bias with respect to comorbidity. 2) SEER does not provide detailed information on insurance plans, including the extent of benefits, provider network and/or co-payments that may also influence kidney cancer care. Nonetheless it is likely that the differences between patients who are insured vs uninsured are much larger than the differences among insured patients with varying policies. 3) As mentioned, heterogeneity likely exists in the uninsured group with some patients unable to afford insurance and others refusing coverage purposely for alternative reasons. However our findings remained consistent even when stratifying our cohort according to age and income areas. 4) The quality of safety net systems and Medicaid benefits can vary by state, which in turn may influence outcomes. However state of residence by and large did not correlate with treatment or survival in our multivariable regression models. 5) SEER does not provide more granular information related to patient underlying tumor anatomy, diagnostic workup, treatment approach and followup, limiting more detailed assessments of insurance based disparities. 6) Our data may not reflect more recent trends in kidney cancer treatment, such as the emergence of active surveillance, ablative therapies and robotic surgery, which increased in popularity after the conclusion of our study interval. Notwithstanding these limitations our study has potential implications for kidney cancer care delivery in the postPPACA era. Early returns indicate that the uninsured rate has already decreased due in large part to Medicaid expansion by participating states.7 To further improve cancer care ASCO (American Society of Clinical Oncology) has urged remaining states to expand Medicaid or offer a
comparable alternative.23 While this is promising, notably previous state level reforms have yielded mixed results. Although not specific to cancer, Medicaid expansion in New York, Arizona and Maine demonstrated small improvements in access, self-reported health and mortality.24 In contrast gaining Medicaid insurance in Oregon did not translate into better health outcomes for newly covered patients with chronic disease.25 In Massachusetts nephrectomy rates remained unchanged following insurance expansion but the use of partial nephrectomy increased in nonwhite and low income populations.26 Extending health coverage remains a key step to improving cancer outcomes but the benefit of insurance expansion alone may be somewhat limited in kidney cancer, given the similarities in stage, treatment and survival between uninsured and Medicaid covered patients. Accordingly further study is needed to address health disparities as they pertain to insurance status and kidney cancer. As noted, patients without private insurance tend to be nonwhite and reside in under resourced areas, which in turn have been associated with disparities in kidney cancer care.15e17 Such social determinants may serve as barriers to health care even when Medicaid benefits are made available. In fact, compared to Medicare beneficiaries those with Medicaid are far less likely to be offered cancer screening or be surgically evaluated for cancer, highlighting issues related to access.27,28 Raising Medicaid reimbursement as advocated by ASCO may help offset these disparities.23 Recent data suggest that greater office visit reimbursements may raise levels of cancer care in accordance with guidelines.29 By understanding and intervening on mechanisms that contribute to inequitable care across insurance types we can more effectively decrease disparities and improve the health of more vulnerable patients with kidney cancer. Conclusions
Uninsured patients with kidney cancer currently present with more aggressive disease and receive treatment less frequently, highlighting a population in need of better care. While insurance expansion through PPACA offers a potential mechanism, many uninsured patients will receive Medicaid coverage, which is also associated with suboptimal quality. These findings underscore the need for additional efforts aimed at improving care in Medicaid beneficiaries, especially patients with kidney cancer. References 1. Hock LM, Lynch J and Balaji KC: Increasing incidence of all stages of kidney cancer in the last 2 decades in the United States:
Health Insurance Status and Disparities in Kidney Cancer Care
an analysis of surveillance, epidemiology and end results program data. J Urol 2002; 167: 57. 2. Siegel R, Ma J, Zou Z et al: Cancer statistics, 2014. CA Cancer J Clin 2014; 64: 9. 3. Chow WH, Dong LM and Devesa SS: Epidemiology and risk factors for kidney cancer. Nature reviews. Urology 2010; 7: 245. 4. King SC, Pollack LA, Li J et al: Continued increase in incidence of renal cell carcinoma, especially in young patients and high grade disease: United States 2001 to 2010. J Urol 2014; 191: 1665.
23
16. Vaishampayan UN, Do H, Hussain M et al: Racial disparity in incidence patterns and outcome of kidney cancer. Urology 2003; 62: 1012. 17. Becker A, Roghmann F, Trinh QD et al: Sociodemographic disparities in the treatment of small renal masses. BJU Int 2013; 111: E274. 18. Flanigan RC, Salmon SE, Blumenstein BA et al: Nephrectomy followed by interferon alfa-2b compared with interferon alfa-2b alone for metastatic renal-cell cancer. N Engl J Med 2001; 345: 1655.
5. Tan HJ, Filson CP and Litwin MS: Contemporary, age-based trends in the incidence and management of patients with earlystage kidney cancer. Urol Oncol 2015; 33: 21.e19.
19. Saigal CS, Deibert CM, Lai J et al: Disparities in the treatment of patients with IL-2 for metastatic renal cell carcinoma. Urol Oncol 2010; 28: 308.
6. Novick AC, Campbell SC, Belldegrun A et al: Guideline for Management of the Clinical Stage 1 Renal Mass. Available at http://www.auanet.org/education/guidelines/renal-mass.cfm. Accessed January 25, 2015.
20. Smaldone MC, Handorf E, Kim SP et al: Temporal trends and factors associated with receipt of systemic therapy among patients undergoing cytoreductive nephrectomy: an analysis of the National Cancer Database. J Urol 2015; 193: 1108.
7. Sommers BD, Musco T, Finegold K et al: Health reform and changes in health insurance coverage in 2014. N Engl J Med 2014; 371: 867.
21. Parsons JK, Kwan L, Connor SE et al: Prostate cancer treatment for economically disadvantaged men: a comparison of county hospitals and private providers. Cancer 2010; 116: 1378.
8. Smith JC and Medalia C: Health Insurance Coverage in the United States: 2013. US Census Bureau, Current Population Reports P60250. Washington, D.C.: U.S. Government Printing Office 2014.
22. Marmot M: Social determinants of health inequalities. Lancet 2005; 365: 1099.
9. Halpern MT, Ward EM, Pavluck AL et al: Association of insurance status and ethnicity with cancer stage at diagnosis for 12 cancer sites: a retrospective analysis. Lancet Oncol 2008; 9: 222. 10. Niu X, Roche LM, Pawlish KS et al: Cancer survival disparities by health insurance status. Cancer Med 2013; 2: 403. 11. 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. 12. Hankey BF, Ries LA and Edwards BK: The surveillance, epidemiology, and end results program: a national resource. Cancer Epidemiol Biomarkers Prev 1999; 8: 1117.
23. Polite BN, Griggs JJ, Moy B et al: American Society of Clinical Oncology policy statement on Medicaid reform. J Clin Oncol 2014; 32: 4162. 24. Sommers BD, Baicker K and Epstein AM: Mortality and access to care among adults after state Medicaid expansions. N Engl J Med 2012; 367: 1025. 25. Baicker K, Taubman SL, Allen HL et al: The Oregon experimentdeffects of Medicaid on clinical outcomes. N Engl J Med 2013; 368: 1713. 26. Ellimoottil C, Miller S, Wei JT et al: Anticipating the impact of insurance expansion on inpatient urological surgery. Urol Pract 2014; 1: 134.
13. Edge SB, Byrd DR, Compton CC et al: AJCC Cancer Staging Manual, 7th ed. New York: Springer 2010.
27. Bradley CJ, Dahman B and Given CW: Inadequate access to surgeons: reason for disparate cancer care? Med Care 2009; 47: 758.
14. Area Health Resources Files. Rockville: Bureau of Health Professions, Health Resources and Services Administration, United States Department of Health and Human Services 2012-2013.
28. Schuur JD, Shah A, Wu Z et al: The impact of Medicaid coverage and reimbursement on access to diagnostic mammography. Cancer 2009; 115: 5566.
15. Stafford HS, Saltzstein SL, Shimasaki S et al: Racial/ethnic and gender disparities in renal cell carcinoma incidence and survival. J Urol 2008; 179: 1704.
29. Halpern MT, Romaire MA, Haber SG et al: Impact of statespecific Medicaid reimbursement and eligibility policies on receipt of cancer screening. Cancer 2014; 120: 3016.
Editorial Commentary
Unfavorable socioeconomic status and health insurance status negatively impact patient access to health care and treatment options for cancer.1,2
Despite the limitations of this study pivotal observations regarding health care disparities in kidney cancer in the United States demonstrated conspicuous benefits
24
Health Insurance Status and Disparities in Kidney Cancer Care
for privately insured patients compared to other groups. Undeniably universal health care is a moral imperative. Can the PPACA repair these disparities in health care? We do not know. What we know is that 63% of physicians are recruited by hospitals vs by physicians.3 Ultimately if PPACA is fully implemented, the input of health care access may improve but services rendered and output issues may suffer unintended consequences. Proactive measures with PPACA must address the decreased number of specialists, decreased physician reimbursement and compensation and increased cost of medical education as well as policies that may interfere with physician fiduciary obligations with patients. Soon the Supreme Court will decide the fate of PPACA. It is our hope that these disparities seen only with our patients will not create new disparities among providers.
Fernando J. Kim Department of Urology Denver Health Medical Center Minimally Invasive Urological Oncology University of Colorado Cancer Center Denver Department of Surgery University of Colorado Denver Denver, Colorado References 1. Adler NE and Newman K: Socioeconomic disparities in health: pathways and policies. Health Affairs 2002; 21: 60. 2. Denberg TD, Beaty BL and Kim FJ: Marriage and ethnicity predict treatment in localized prostate carcinoma. Cancer 2005; 103: 1819. 3. Kirchhoff SM: Physician Practices: Background, Organization, and Market Consolidation. Washington, D.C.: Congressional Research Service 2013.