association of insurance with cancercare utilization and outcomes

DOI: 10.3322/CA.2007.0011 2008;58;9-31; originally published online Dec 20, 2007; CA Cancer J Clin

Bandi, Rebecca Siegel, Andrew Stewart and Ahmedin Jemal Elizabeth Ward, Michael Halpern, Nicole Schrag, Vilma Cokkinides, Carol DeSantis, Priti

Association of Insurance with Cancer Care Utilization and Outcomes

This information is current as of April 3, 2011

http://caonline.amcancersoc.org/cgi/content/full/58/1/9the World Wide Web at:

The online version of this article, along with updated information and services, is located on

http://caonline.amcancersoc.org/subscriptions/individuals only): , go to (USCA: A Cancer Journal for CliniciansTo subscribe to the print issue of

ISSN: 1542-4863. OnlineAtlanta GA 30303. (©American Cancer Society, Inc.) All rights reserved. Print ISSN: 0007-9235.

is owned, published, and trademarked by the American Cancer Society, 250 Williams Street NW,CAWiley-Blackwell. A bimonthly publication, it has been published continuously since November 1950.

is published six times per year for the American Cancer Society byCA: A Cancer Journal for Clinicians

by on April 3, 2011 (©

Am

erican Cancer S

ociety, Inc.) caonline.am

cancersoc.orgD

ownloaded from

http://caonline.amcancersoc.org/cgi/content/full/58/1/9

http://caonline.amcancersoc.org/subscriptions/

http://caonline.amcancersoc.org:80

9Volume 58 • Number 1 • January/February 2008

CA Cancer J Clin 2008;58:9–31

Association of Insurance with CancerCare Utilization and Outcomes

Elizabeth Ward, PhD; Michael Halpern, MD, PhD; Nicole Schrag, MSPH;Vilma Cokkinides, PhD, MSPH; Carol DeSantis, MPH; Priti Bandi, MS;Rebecca Siegel, MPH; Andrew Stewart, MA; Ahmedin Jemal, DVM, PhD

ABSTRACT Advances in the prevention, early detection, and treatment of cancer have resulted

in an almost 14% decrease in the death rates from all cancers combined from 1991 to 2004

in the overall US population, with remarkable declines in mortality for the top 3 causes of can-

cer death in men (lung, colorectal, and prostate cancer) and 2 of the top 3 cancers in women

(breast and colorectal cancer). However, not all segments of the population have benefited

equally from this progress, and evidence suggests that some of these differences are related

to lack of access to health care. Lack of adequate health insurance appears to be a critical

barrier to receipt of appropriate health care services. This article provides an overview of sys-

tems of health insurance in the United States, demographic and socioeconomic characteris-

tics associated with health insurance coverage, and economic burdens related to health care

among individuals and families. This article also presents data on the association between

health insurance status and screening, stage at diagnosis, and survival for breast and colorec-

tal cancer based on analyses of the National Health Interview Survey and the National Cancer

Data Base. Although this article focuses on associations between health insurance and can-

cer care utilization and outcomes, it is important to recognize that barriers to receipt of optimal

cancer care are complex and involve patient-level, provider, and health system factors. Evidence

presented in this paper suggests that addressing insurance and cost-related barriers to care

is a critical component of efforts to ensure that all Americans are able to share in the progress

that can be achieved by access to high-quality cancer prevention, early detection, and treat-

ment services. (CA Cancer J Clin 2008;58:9–31.) © American Cancer Society, Inc., 2008.

INTRODUCTION

Over the last decade, there have been major advances in the prevention, earlydetection, and treatment of cancer. However, these advances have not been expe-rienced equally by all segments of the population. Evidence is increasing that forsome Americans, lack of health insurance or inadequate heath insurance is a majorbarrier to preventive health services and adequate treatment. Early detection throughscreening has been demonstrated to decrease mortality from breast and colorectalcancer, yet people without health insurance are half as likely as those with private insur-ance to receive such screening.1 As a result of lower screening rates and limited accessto primary care for prompt evaluation of symptoms, individuals with no health insurance are much more likely thanthose with private health insurance to be diagnosed with advanced stages of cancer.2–5 Furthermore, many cancerpatients with health insurance face severe financial hardships due to the costs of treatment that are not covered bytheir insurance. For some, this financial burden requires choices between health care and other necessities, thereby con-tributing to poor health outcomes. These insurance and cost-related barriers to high-quality cancer prevention, earlydetection, and treatment are growing due to declines in coverage by employer-sponsored health insurance; increases

Dr. Ward is Managing Director, Sur-veillance Research, Department of Epi-demiology and Surveillance Research,American Cancer Society, Atlanta, GA.

Dr. Halpern is Strategic Director,Health Services Research, Departmentof Epidemiology and SurveillanceResearch, American Cancer Society,Atlanta, GA.

Ms. Schrag is Epidemiologist, HealthServices Research, Department ofEpidemiology and Surveillance Re-search, American Cancer Society,Atlanta, GA.

Dr. Cokkinides is Strategic Director,Cancer Risk Factors and Screening,Department of Epidemiology and Sur-veillance Research, American CancerSociety, Atlanta, GA.

Ms. DeSantis is Epidemiologist,Cancer Surveillance, Departmentof Epidemiology and SurveillanceResearch, American Cancer Society,Atlanta, GA.

Ms. Bandi is Epidemiologist, CancerRisk Factors and Screening, Depart-ment of Epidemiology and SurveillanceResearch, American Cancer Society,Atlanta, GA.

Ms. Siegel is Manager, SurveillanceInformation Services, Department ofEpidemiology and Surveillance Research,American Cancer Society, Atlanta, GA.

Mr. Stewart is Senior Manager, Na-tional Cancer Data Base, Commissionon Cancer, American College of Sur-geons, Chicago, IL.

Dr. Jemal is Strategic Director, CancerSurveillance, Department of Epidemi-ology and Surveillance Research, Amer-ican Cancer Society, Atlanta, GA.

Published online throughCA First Lookat http://CAonline.AmCancerSoc.org.

DOI: 10.3322/CA.2007.0011


Am

erican Cancer S


cancersoc.orgD

ownloaded from


in health insurance premiums, deductibles, andcopayments; and rising costs of medical care.

This article provides an overview of systemsof health insurance in the United States, demo-graphic and socioeconomic characteristics asso-ciated with health insurance coverage, andeconomic burdens related to lack of insuranceand underinsurance in the US population. Thisarticle also presents data on the association betweenhealth insurance status and screening, stage atdiagnosis, and survival for breast and colorectalcancer based on analyses of the National HealthInterview Survey (NHIS) and the National CancerData Base (NCDB).

Although this article focuses on associationsbetween health insurance and cancer care uti-lization and outcomes, it is important to recog-nize that barriers to receipt of optimal cancercare are complex and involve patient-level,provider, and health system factors.6 Even amongindividuals with adequate health insurance,variation in receipt of optimal care and out-comes can be observed by race and ethnicity, sex,age, income, education, urban versus rural areaof residence, and other sociodemographic andgeographic factors. African Americans in par-ticular are at increased risk of developing anddying from cancer and often receive poorer-quality treatment and have lower survival ratesthan their White counterparts who are diag-nosed with cancers of similar disease stage.6

Addressing insurance and cost-related barriersto high-quality prevention, early detection, andtreatment is not the only measure that will beneeded to eliminate cancer disparities, but thedata that will be presented in this article suggestthat it is an important one.

MATERIALS AND METHODS

Data Sources

Information on insurance status and relation-ships with access to health care, preventive serv-ices, and cancer screening was obtained byanalysis of data from the NHIS conducted in2005 and 2006. The NHIS is a survey of theCenters for Disease Control and Prevention’sNational Center for Health Statistics. The sur-vey is designed to provide national prevalence

estimates on personal, socioeconomic, demo-graphic, and health characteristics of US adults.Data are gathered through a computer-assistedpersonal interview of adults aged 18 years andolder living in households in the United States.7

Data from the NCDB were used to examinethe relationship between insurance status at thetime of diagnosis and cancer survival for all can-cers combined and for breast and colorectal can-cer. The NCDB is a joint project of the Commis-sion on Cancer of the American College ofSurgeons and the American Cancer Society thatcollects information on demographic and clinicalcharacteristics and first course of treatment forcancer patients diagnosed at approximately 1,500Commission on Cancer-approved hospitals, rep-resenting almost 70% of all cancer patients treatedin the United States.8 We selected cancer patientsaged 18 to 64 years reported to the NCDB dur-ing 1999 and 2000, the most recent years for which5-year follow up is available. Patients were furtherrestricted to those with private insurance, Medicaidinsurance, and no insurance. Among the 719,915patients who met these criteria, 7,886 wereexcluded because the time variable could not becalculated, and 113,394 were excluded becausethey had other or unknown race (only White,African American, and Hispanic patients wereincluded) or had missing area socioeconomic sta-tus (SES) data. A total of 598,635 cases were avail-able for analysis, including 129,644 female breastcancer patients and 44,898 male and female col-orectal cancer cases. Cox regression analysis (pro-portional hazards analysis) was used to model 5-yearsurvival by insurance status, controlling for age,race, sex, and zip code-based income. The pro-portional hazards assumption was tested beforeanalysis, and none of the variables included in themodel violated the assumption when analyseswere stratified by age group and site. Results foroverall and stage-specific survival were plotted byinsurance status.

RESULTS

Systems of Health Insurance Coveragein the United States

The major systems of health insurance cover-age in the United States are employer-sponsored


10 CA A Cancer Journal for Clinicians


Am

erican Cancer S


cancersoc.orgD

ownloaded from


health insurance, Medicare insurance, Medicaidand State Children’s Health Insurance Program(SCHIP) insurance, and private nongroup healthinsurance. In addition to these broad categoriesof health insurance, which cover 95% of theinsured US population under age 65 years, thereare other forms of federal insurance, includingcoverage through the Veteran’s Administrationbenefits and Indian Health Service, and stateinsurance programs, including high-risk pools,which we do not discuss further.

Employer-sponsored Health Insurance. MostAmericans under age 65 years receive their healthinsurance coverage through their own employeror the employer of a family member (Figure 1).Nearly all companies with more than 200employees offer health insurance coverage.9 In2007, the average costs to employers and employ-ees, respectively, were $3,785 and $694 per year

for individual coverage and $8,824 and $3,281per year for family coverage.10 The employer-based system of health insurance has severaladvantages, most notably the creation of work-based risk pools in which healthy low-risk par-ticipants subsidize the health costs of sick andhigh-risk participants.11 However, there are someserious disadvantages to this system. Not all com-panies offer health benefits, not all workers areeligible for coverage, and not all employees chooseto participate or can afford their share of thehealth premium.12 Moreover, the cost of healthinsurance premiums has been rising much fasterthan the rate of overall inflation and workers’earnings (Figure 2).10 An important disadvan-tage of employer-sponsored health insurance isthat people who develop a serious illness, suchas cancer, may not be able to keep their employ-ment and may lose access to their insurance.



FIGURE 1 Health Insurance Coverage Among Individuals Under Age 65 Years, 2006 (in Millions).Reference: Collins SR, White C, Kriss JL.9

Source of Data: Current Population Survey, March 2007.


Am

erican Cancer S


cancersoc.orgD

ownloaded from


While the Consolidated Omnibus BudgetReconciliation Act (COBRA) allows them topay to receive health insurance benefits after theyleave their job,13 without employment, the costof maintaining insurance may be prohibitive.Another disadvantage of employer-sponsoredhealth insurance is that if an individual changesemployment or his or her employers changeinsurance plan options, it may not be possibleto continue to use the same health care providers,resulting in discontinuity of care.

Medicare. Most Americans aged 65 years andolder and about 2% of those under age 65 yearsreceive health insurance through Medicare.US citizens and permanent residents are eli-gible for Medicare if they or their spouse paidinto Social Security for 40 quarters (10 years).

Individuals eligible for Social Security benefitsare automatically enrolled in Medicare Part A(hospital insurance) when they turn age 65years. Medicare Part B provides other types ofmedical insurance coverage, including cover-age for physician’s services (inpatient or outpa-tient); administration of drugs that are notusually self-administered by the patient; out-patient hospital services; diagnostic tests; andspecific preventive services, including mam-mograms, Pap tests, and colorectal cancerscreening. Beneficiaries must enroll in MedicarePart B and pay a monthly premium based ontheir income. Medicare Part A is financed pri-marily through payroll taxes, while Part B isfinanced by beneficiary premiums and by fed-eral general revenues. Medicare Part C, also



FIGURE 2 Cumulative Changes in Health Insurance Premiums, Overall Inflation, and Workers’ Earnings, 2000 to 2007.Note: Data on premium increases reflect the cost of health insurance for a family of 4.Reference: Adapted from Economic challenges facing middle class families: Hearing before the Committee on Ways and Means of the US Houseof Representatives, 110th Cong. (January 31, 2007). (Testimony of Diane Rowland: Health care: squeezing the middle class with more costsand less coverage).Source of Data: Claxton G, Gabel J, DiJulio B, et al.10


Am

erican Cancer S


cancersoc.orgD

ownloaded from


known as Medicare Advantage, was establishedin 1997 to allow beneficiaries to enroll in pri-vate health insurance plans, and Medicare PartD was enacted in 2003 to provide prescriptiondrug coverage through private drug plans.14

Health care premiums and out-of-pocket costsfor Medicare beneficiaries are set on an annualbasis. In 2008, Medicare beneficiaries are respon-sible for paying a $1,024 deductible for the first1 to 60 days of inpatient hospital care. For stayslonger than 60 days, beneficiaries pay an increas-ing percentage of the cost. Part B premiums areset at $96.40 per month for most beneficiaries,with a sliding scale up to $238.40 for those withhigh incomes. Care at skilled nursing facilitiesis not covered by Medicare for the first 20 days;in days 21 to 100, Medicare will cover $256 perday. Medicare beneficiaries must also pay 20%of the Medicare allowable costs for services cov-ered under Part B, which can be considerablein the case of a major illness such as cancer.15

Medicaid (Title XIX). Medicaid is a federallyaided, state-operated and administered programthat provides benefits for certain indigent or low-income persons in need of health and medicalcare. The program, authorized by Title XIX of theSocial Security Act, is basically for the poor. Itdoes not cover all of the poor, however, but onlypersons who meet specified eligibility criteria.16

Eligible groups include low-income children,families, and pregnant women; elderly and dis-abled people who need long-term care services;and low-income elders who need assistance withthe costs of Medicare coverage. Within broad fed-eral guidelines, states establish their own eligibil-ity standards; determine the type, amount, duration,and scope of services; set the payment rate forservices; and administer their own programs. Thus,each state’s Medicaid program is unique.14

In 1997, the SCHIP was established by Con-gress to expand coverage to uninsured low-income children. States were allowed to expandincome-eligibility levels and receive enhancedmatching funds for children by either expand-ing their Medicaid programs or creating newprograms separate from Medicaid. All 50 statesand the District of Columbia have implementedSCHIP programs, although the extent of cov-erage varies.17 As a result of SCHIP and otherprograms, as of 2001, almost all children from

families with incomes below 200% of the fed-eral poverty level are eligible for either Medicaidor SCHIP. Medicaid and SCHIP insurance areimportant sources of coverage for children withcancer. Based on the NCDB, approximately25% of children under age 18 years diagnosedwith cancer are covered by Medicaid and SCHIPprograms (A.S., written communication, October31, 2007).

Consistent with the emphasis of the Medicaidprogram on providing health care to eligiblechildren and families with children, the proba-bility of having Medicaid coverage is highest forchildren under age 18 years and higher for womenthan men (Figure 3). The proportion of adultsaged 45 to 64 years with Medicaid coverageranges from 5% for White men to 15% amongAfrican American and Hispanic women.18 Arecent study found that only 8% of uninsuredchildless adults were eligible for Medicaid orMedicare assistance.18

In most states, people who develop seriousillnesses, including cancer, can qualify for Medicaideven if their income is higher than the stateMedicaid eligibility limit. To qualify for Medicaidas medically needy, individuals or families maybe required to “spend down” to Medicaid eligi-bility by offsetting their excess income with med-ical and/or remedial care expenses.14

The Breast and Cervical Cancer Preventionand Treatment Act, enacted in February 2000,permits states to provide medical assistance throughMedicaid to eligible women who are screenedthrough the National Breast and Cervical CancerEarly Detection Program. All of the states are par-ticipating in this program.19 However, it is estimatedthat only 13.2% of eligible women received amammogram in 2002/200320 due in part to inad-equate funding of the program. There are alsodifferences in implementation of the Breast andCervical Cancer Prevention and Treatment Actbetween states, which may limit opportunitiesfor some women to benefit from the program.21

Private, Nongroup Health Insurance. Individualsand families who do not have health insurancecoverage through their employers or public pro-grams may seek coverage under the individual(nongroup) insurance market. Less than 5% ofUS adults under age 65 years have this type ofinsurance, in part because the premiums are much




Am

erican Cancer S


cancersoc.orgD

ownloaded from


higher than those for employer-sponsored insur-ance.9 A survey of older adults (aged 50 to 70years) in 2004 found that more than half (54%)of people with private nongroup insurance paidmore than $3,600 per year for individual poli-cies, and 26% paid more than $6,000 per year.22

Private nongroup insurance can be difficult toobtain or extremely costly for individuals withpre-existing health conditions and, therefore, isnot a viable option for many Americans wholack employer-sponsored coverage.

Demographic Determinants of Insurance Status

Almost everyone is at some risk of being unin-sured. However, the risk of being uninsuredvaries by age, gender, race/ethnicity, and povertystatus, as well as by other characteristics. Amongindividuals under age 65 years, those under age18 years have the lowest and those aged 18 to24 years have the highest probability of being

uninsured (Figure 4).23 Fourteen percent of peo-ple aged 45 to 64 years are uninsured. The prob-ability of being uninsured varies inversely toincome, but increased from 2001 to 2005 at allincome levels (Figure 5).24 African Americans,Hispanics, Asian American/Pacific Islanders, andAmerican Indian/Alaska Natives are much morelikely to be uninsured than non-Hispanic Whites(Figure 6). The most important reason whyworking individuals are uninsured is that theiremployers do not offer them health insurancebenefits.25 Lack of employer-based health insur-ance is common for workers in small compa-nies, low-wage workers, and part-time workers,as well as the self-employed.24 When employ-ees are offered coverage by their own employer,uptake rates are generally over 80%.25

There are numerous ways in which individ-uals or families can lose their health insurance.For example, an individual may lose or leave ajob where insurance was offered; lose Medicaid



FIGURE 3 Medicaid Coverage of the Nonelderly by Age, Sex, and Race/Ethnicity, 2005.Reference: Kaiser Family Foundation.18



Am

erican Cancer S


cancersoc.orgD

ownloaded from


eligibility when they or their children growup; lose insurance through their spouse due toseparation, divorce, or death; or be priced outof the market when the cost of premiumsbecomes unaffordable.12 Parental health insur-ance coverage of children who are not studentsends at age 18 years, as does coverage for manychildren insured under Medicaid/SCHIP.Employer-based coverage sometimes fails toprotect families from large medical expensesbecause illness may lead to job loss and the con-sequent loss of coverage.26

Health insurance generally does not providetotal dollar coverage of health care costs. Coveredservices, deductibles, copays, and yearly or life-time caps can vary considerably among the types

of insurance that are available. Caps on totallifetime coverage or disease-specific coverage(eg, $1,000,000) may be exceeded if protracted,expensive medical care is needed. Almost every-one is at risk of being underinsured in the eventof a major illness, but many individuals and fam-ilies are underinsured even without experienc-ing a major illness. The underinsured includepeople who have some form of health insur-ance but lack coverage for certain proceduresor cannot afford the cost sharing associated withcovered benefits or both.27 One common def-inition is that a person or family is underinsuredif they would have to spend more than 10%of family income on out-of-pocket medicalexpenses in the event of a catastrophic illness.28



FIGURE 4 Percentage of Persons Under Age 65 Years Without Health Insurance Coverage at the Time of Interview byAge Group and Sex, January to March, 2007.Reference: Cohen RA, Martinez ME.23

Source of Data: Family Core component of the 2007 National Health Interview Survey. Data are based on household interviews of a sampleof the civilian noninstitutionalized population.


Am

erican Cancer S


cancersoc.orgD

ownloaded from


A recent study analyzed data from the MedicalExpenditure Panel Surveys sponsored by theAgency for Health Care Research and Qualityfor 1996 and 2003.28 The Medical ExpenditurePanel Surveys household survey collects detailedinformation on health insurance coverage, healthcare utilization, and expenditures by sources ofpayment and additional data on health status,medical conditions, and other sociodemographicvariables. Household reports are supplementedwith information on third-party payments andbilling codes from medical provider billingrecords. Sample sizes for persons under age 65years were 19,022 persons in 1996 and 28,970persons in 2003.28 According to this study, the

prevalence of being underinsured (having out-of-pocket heath care expenditures, excludinghealth insurance premiums, greater than 10%of after-tax family income) increased from 6.7%of the nonelderly population in 1996 to 8.5% by2003. When the costs of insurance premiumswere included, the percentage of the popula-tion with health care expenditures exceeding10% of after-tax income rose from 15.8% in1996 to 19.2% in 2003. As a result of highpremiums, copayments and deductibles, andlifetime-maximum insurance payment limits,persons with private nongroup plans face sig-nificantly greater risk of high financial burdenthan individuals with private employment-related



FIGURE 5 Uninsured Rates Among Adults Aged 19 to 64 Years by Income Level, 2001 to 2005.Note: Income refers to annual income. In 2001 and 2003, low income is <$20,000, moderate income is $20,000-$34,999, middle income is$35,000-$59,999, and high income is $60,000 or more. In 2005, low income is <$20,000, moderate income is $20,000-$39,999, middleincome is $40,000-$59,999, and high income is $60,000 or more.Reference: Collins SR, Davis K, Doty MM, et al.24

Source of Data: The Commonwealth Fund Biennial Health Insurance Surveys (2001, 2003, and 2005).


Am

erican Cancer S


cancersoc.orgD

ownloaded from


insurance. In 2003, 53.4% of nonelderly per-sons with nongroup coverage incurred totalhealth care burdens exceeding 10% of familyincome, a percentage nearly 3 times greater thanamong those with employment-related cover-age.28 Nearly one-quarter (24%) of the poor(family income �100% of federal poverty line,which was $18,400 for a family of 4 in 2003)and 10% of the near-poor (family income 100%to �200% of the federal poverty line) reportedtotal health care burdens exceeding 20% of fam-ily income. The out-of-pocket burden was par-ticularly great among individuals with seriousillness. Among people with cancer, 28.8% hadtotal burdens exceeding 10% of family income,and 11.4% had total burdens exceeding 20% offamily income.28

Even among the elderly population withMedicare insurance, out-of-pocket health carecosts can be considerable. In 2003, among the eld-erly population, 39.7% had private group cov-erage and Medicare (14.6 million), 16.3% hadprivate nongroup coverage and Medicare (6 mil-lion), 10.2% had Medicare and other public cov-erage (3.7 million), and 33.8% had Medicareonly or Medicare HMO coverage (12.4 mil-lion). Of all elderly persons, 29.3% had out-of-pocket spending on medical care in excess of$5,000, and 7.3% had out-of-pocket spendingon medical care in excess of $10,000. Thosewith Medicare and private nongroup coveragewere the most likely to have high out-of-pocketspending. In this group, 46.4% had family-levelspending exceeding $5,000 compared with 35.1%



FIGURE 6 Health Insurance Coverage of the Nonelderly by Race/Ethnicity, 2005.Note: Nonelderly includes individuals up to age 65 years. “Other public” includes Medicare and military-related coverage; SCHIP is includedin Medicaid.Reference: Kaiser Family Foundation.18



Am

erican Cancer S


cancersoc.orgD

ownloaded from


of individuals with Medicare and private groupcoverage.29

Medical debt is an important cause of bank-ruptcy filing in the United States. A study ofcauses of bankruptcy among 931 people whofiled for bankruptcy in the United States in 2001found that about half cited medical causes as animportant reason for bankruptcy. Three-fourthsof those with medical debt were insured at theonset of the bankrupting illness; 60.1% had pri-vate coverage, 5.7% had Medicare, 8.4% hadMedicaid, and 1.6% had veterans/military cov-erage. About one-third of individuals who hadprivate insurance at the onset of their illness lostcoverage during the course of their illness. Onaverage, the mean out-of-pocket expenditurefor all debtors citing medical expenses for bank-ruptcy was $11,854. For debtors citing canceras the medical condition associated with thebankruptcy, it was $35,878.26 Compounding thefinancial consequences for individuals and fam-ilies without health insurance are pricing poli-cies in which uninsured patients are chargedmore for services. In 2004, a survey found thatthe rates charged to uninsured and other “self-pay” patients for hospital services were often 2.5times what most health insurers actually paidand more than 3 times the hospital’s Medicare-allowable costs.30 Even the very poor are at riskof medical debt and aggressive debt-recoverypractices. A cross-sectional study of patients beingseen at 10 safety-net provider sites in Baltimore,Maryland, found that 42% reported that theycurrently had a medical debt (average $3,409),and 39.4% reported ever having been referredto a collection agency for a medical debt. Themean annual income of the patients interviewedwas $7,864, and 47.2% reported that they werehomeless. Among individuals who had currentmedical debt or had been referred to a collectionagency in the past, 24.5% no longer went to thatsite for care, 18.6% delayed seeking care whenneeded, and 10.4 % reported “only going toemergency rooms now.”31

Impact of Health Insurance Statuson Access to Care

Individuals who are uninsured, underinsured,or insured by government programs may face

significant barriers to obtaining health care. Someprivate physicians do not accept new patientsunless they have private insurance or are able topay the full cost at the time of the visit. For exam-ple, a recent national survey of office-based physi-cians found that although 96% were acceptingnew patients, 40.3% did not accept “no charge”or charity patients, 25.5% did not accept Medicaidpatients, and 13.9% did not accept patients cov-ered by Medicare (Figure 7).32 Patients who areunable to afford outpatient care in private prac-tice settings often seek care in emergency depart-ments, which are required by law only to examinepatients to determine whether a medical emer-gency exists.33 Consequently, many patients ini-tially seen in emergency departments are referredto outpatient providers for follow-up care, butuninsured or Medicaid-insured patients may beexcluded from care by the system. A recent studyemployed scripted interviewers to contact clin-ics stating that they had been seen in a commu-nity emergency room the previous night andwere seeking a follow-up appointment for a seri-ous medical condition such as pneumonia or sus-pected ectopic pregnancy.34 Callers claiming tohave private insurance were almost twice as likelyto receive prompt appointments as those statingthat they had Medicaid insurance (63.6% versus34.2%). Uninsured callers who said that theycould pay cash for the entire charge at the timeof the visit were equally likely to receive anappointment as those with private insurance,while only 25.1% of uninsured individuals whooffered to pay $20 at the time of the visit wereoffered appointments.

As more Americans go without health insur-ance and access to affordable health care decreases,millions of Americans turn to the health care“safety net” for their health care needs. At thecore of the safety net are health centers, public hos-pital systems, and local health departments. Inaddition, some communities are served by school-and church-based health clinics, private physi-cians, and nonprofit hospitals committed to serv-ing vulnerable patients. Although such programsprovide lifesaving services, some are understaffed,have inadequate resources, and are unable to pro-vide specialty care. Recent studies suggest thatthe resources available in the health care “safetynet” are declining even as the need is growing.33




Am

erican Cancer S


cancersoc.orgD

ownloaded from


In addition, although poverty is increasing inboth urban and suburban neighborhoods, par-ticularly in Midwestern and Southern metropol-itan areas, there has been a shift in where thelargest concentrations of poverty exist. The poorare increasingly moving to suburban and ruralareas to find jobs and affordable housing as eco-nomic forces make cities less affordable. Thisdemographic shift may lead to a disconnectbetween where poor people live and where theyare able to find safety-net health clinics and hos-pitals because these services are disproportion-ately concentrated in central-city neighborhoods.35

Impact of Health Insurance Status on Cancer

Lack of access to health care can adverselyaffect cancer incidence and mortality through-out the continuum from cancer prevention and

early detection to treatment, survivorship, and pal-liative care. Lack of health insurance, even forintermittent periods, is associated with lowerlikelihood of having a “medical home” or usualsource of health care. Such individuals are lesslikely to have preventive care and to have ade-quate management for chronic conditions. Basedon NHIS 2006, 53.6% of uninsured individu-als aged 18 to 64 years had no usual source ofhealth care compared with 9.9% of privatelyinsured and 10.8% of individuals with Medicaidinsurance. Among individuals who had beenuninsured for �12 months, 58.7% had no usualsource of care (Table 1). Individuals who wereuninsured at the time of the interview were morelikely than insured individuals to report that theydid not get care due to cost, delayed care due tocost, did not get prescription drugs due to cost,and had no health care visits in the past 12 months



0 10 20 30 40 50

FIGURE 7 Percentage of Office-based Physicians Not Accepting New Patients According to Method of Payment,2003 to 2004.Source: Hing E, Burt CW.32


Am

erican Cancer S


cancersoc.orgD

ownloaded from


due to cost (Table 1). Patients with Medicaidwere more likely than privately insured patientsto have no usual source of care but less likely toreport no health care visits in the past 12 months.Medicaid-insured patients were more likely toreport that they did not get care due to cost,delayed care due to cost, or did not get prescrip-tion drugs due to cost compared with privatelyinsured patients. However, for all of these meas-ures of access to care, patients with Medicaidinsurance reported much greater access thanthose who were uninsured.

Cancer Prevention. Smoking, poor nutrition,and physical inactivity are important risk factorsfor cancer. Health care encounters provide anopportunity for providers to counsel individu-als on tobacco use, nutrition, and physical activ-ity and provide support for tobacco cessationand weight loss. However, individuals who areuninsured are much more likely to report nohealth care encounters in the past year than thosewho are Medicaid-insured or privately insured(Table 1). Among those who did have a healthcare encounter, the uninsured were less likelythan privately or Medicaid-insured individuals tobe advised to quit smoking or to lose weight(Table 1). An analysis of data from an earlier(2000) NHIS survey found that individuals with

no insurance or with Medicaid insurance wereless likely to use tobacco cessation aids in a quitattempt during the past year.36

Early Detection and Screening. Analyses of theNHIS and the Behavioral Risk Factor Surveil-lance Survey have consistently found that indi-viduals without health insurance have lower ratesof cervical, breast, and colorectal cancer screen-ing than individuals with health insurance.37–39

A few studies reported screening rates forMedicaid-insured patients that were lower thanthose for privately insured patients, but higherthan for uninsured patients.40,41 Studies of indi-viduals aged 65 years and over, using other sur-veys and data sources, found that individualswho were dually insured by Medicare and Med-icaid or uninsured were less likely to receive can-cer screening tests than comparison groups (thosewith Medicare alone or those with Medicareplus supplemental private insurance, dependingon the study).41–43

Analyses of the NHIS 2005 survey also findthat the likelihood of receiving recommendedcancer screening tests varies markedly by insur-ance status (Table 2). About three-quarters (74.5%)of women aged 40 to 64 years who had privatehealth insurance had received a mammogram inthe past 2 years compared with 56.1% of women



All Private Medicaid Uninsured Uninsured for >12Proportion (%) (at Time of Interview) Months

Have no usual source of care 18.9 9.9 10.8 53.6 58.7

Did not get care due to cost 8.4 3.8 10.3 22.8 24.4

Delayed care due to cost 10.7 6.1 11.1 25.8 27.1

Did not get prescription drugs due to cost 9.3 4.4 15.2 22.9 23.1

Had no health care visits in the 21.6 16.6 12.5 43.2 49.0past 12 months

Counseling by a health care provider*

Smokers advised to quit† 58.2 58.1 67.0 50.4 48.2

Obese adults (BMI>30) advised 51.7 53.9 51.2 40.3 35.6to lose weight‡

*Among individuals with at least one health care visit in the past 12 months.†Adults who reported that they were advised to quit using tobacco by a health care provider in the past 12 months; informationavailable only in NHIS 2005.‡Adults who reported that they were advised to control or lose weight by a doctor or health professional in the past 12 months.Source: National Health Interview Survey Public Use Data File 2005, 2006, National Center for Health Statistics, Centers forDisease Control and Prevention, 2006, 2007.

TABLE 1 Access to Healthcare and Preventive Services by Health Insurance Status in Adults Aged 18 to64 Years, 2006


Am

erican Cancer S


cancersoc.orgD

ownloaded from


with Medicaid insurance and 38.1% of uninsuredwomen. Similarly, 87.9% of women who hadprivate health insurance had a Pap test in the past3 years compared with 82.5% of women withMedicaid insurance and 68% of uninsured women.Among men and women aged 50 to 64 yearswith private insurance, 48.3% had had a recom-mended colorectal cancer screening test in thepast 10 years compared with 39.6% of individu-als with Medicaid insurance and only 18.8% ofthose who were uninsured. The percentage ofmen aged 50 to 64 years who had a prostate-specific antigen test for prostate cancer followeda similar pattern: 37.1% among the privatelyinsured, 20.8% among Medicaid-insured, and14% among the uninsured.

Given that health insurance status is associ-ated with other sociodemographic characteris-tics, such as race/ethnicity, immigration status/country of birth, and level of education, it ispossible that differences in screening rates reflectdifferences in knowledge about cancer preven-tion, culture, or other barriers to care. However,when data from the NHIS 2005 are analyzed toestimate the likelihood of receiving mammog-raphy and colorectal cancer screening by race/ethnicity (non-Hispanic White, non-HispanicAfrican American, and Hispanic), level of edu-cation, and insurance status (Figure 8 and Figure9), it is apparent that having health insurance isan important predictor of screening across allmajor racial and ethnic populations. Moreover,

at every level of education, individuals withhealth insurance are about twice as likely as thosewithout health insurance to have had mammog-raphy or colorectal cancer screening.

Stage at Diagnosis and Survival. Informationon the relationship between stage at diagnosisand insurance status is quite limited becausepopulation-based cancer incidence registries donot collect information on insurance status.Several studies have examined the relationshipbetween Medicaid enrollment status and stageat diagnosis by matching cancer registry datawith state-based Medicaid records. One suchstudy based on linkage of State of MichiganMedicaid and cancer registry records found thatMedicaid-insured patients under age 65 yearswho were diagnosed with cancer during 1996 to1998 were more likely to be diagnosed with late-stage cancer of the breast, uterus, cervix, lung,and prostate than patients without Medicaidcoverage (including uninsured and privatelyinsured).44 However, this study could not dif-ferentiate among patients who were enrolled inMedicaid before their diagnosis from thoseenrolled as a result of diagnosis. Later stage atdiagnosis among patients enrolled as a resultof diagnosis does not reflect the extent to whichMedicaid insurance provides access to care,including prevention and early detection. A sub-sequent study in the Michigan registry Medicaid-linked data found that for cancer cases diagnosedin 1996 and 1997, 64% were enrolled before



All Private Medicaid Uninsured Uninsured for >12Proportion (%) (at Time of Interview) Months

Women aged 40 to 64 years who had 67.9 74.5 56.1 38.1 32.9a mammogram in the past 2 years

Women aged 18 to 64 years who had a 83.6 87.9 82.5 68.0 62.7Pap test in the past 3 years

Adults aged 50 to 64 years who had a 44.2 48.3 39.6 18.8 14.9colorectal cancer screening test*

Men aged 50 to 64 years who had a 33.5 37.1 20.8 14.0 11.5prostate-specific antigen test in the past year

*Had a fecal occult blood test in the past year or an endoscopy in the past 10 years.Source: National Health Interview Survey Public Use Data File 2005, National Center for Health Statistics, Centers for DiseaseControl and Prevention, 2006.

TABLE 2 Cancer Screening by Health Insurance Status in Adults, 2005


Am

erican Cancer S


cancersoc.orgD

ownloaded from




FIGURE 8 Mammogram Within the Last Year, Women Aged 40 to 64 Years, by Race/ Ethnicity, Years of Education, andInsurance Status, 2003 to 2005.*Groups have been combined (years of education 13+) due to small sample sizes.Source: National Health Interview Survey 2003 and 2005, National Center for Health Statistics, Centers for Disease Control and Prevention,2006.

FIGURE 9 Colorectal Cancer Screening*, Ages 50 to 64 Years, by Race/Ethnicity, Years of Education, and InsuranceStatus, 2003 to 2005.*Either a fecal occult blood test within the past year or an endoscopy within the past 10 years. †Groups have been combined (years of education 13+) due to small sample sizes.Source: National Health Interview Survey 2003 and 2005, National Center for Health Statistics, Centers for Disease Control and Prevention,2006.


Am

erican Cancer S


cancersoc.orgD

ownloaded from


being diagnosed with cancer; just over one-thirdof the Medicaid sample enrolled the same monthor after diagnosis. A higher percentage of col-orectal and lung cancer patients were enrolledthe same month or after diagnosis (46% and 42%,respectively).45 In this study, the odds of laterstage at diagnosis were higher among individu-als who were enrolled in Medicaid during themonth of or after diagnosis compared with thosewith longer-term enrollment before diagnosis,but the latter group had increased odds of latestage compared with those without Medicaidinsurance (the majority of whom would beexpected to be privately insured).45 A furtherstudy of the same population found that bothpre-enrolled and late-enrolled Medicaid patientswere at substantially increased risk of dying within8 years of diagnosis compared with patients whowere not Medicaid-enrolled; although survivalwas somewhat poorer in the late-enrolled com-pared with the pre-enrolled, this difference wasnot statistically significant.46 A study of stage atdiagnosis for cervical cancer patients diagnosedin California in 1996 to 1999 found that womeninsured by Medicaid were significantly morelikely than women without Medicaid coverage(including uninsured and privately insured) tobe diagnosed at late stage.47 However, when riskswere analyzed by duration of Medicaid enroll-ment, increased risk of late-stage diagnosis wasconfined to those enrolled at the time of or lessthan 12 months before diagnosis and was notapparent for those who had been enrolled inMedicaid for 12 or more months. A study link-ing data from the Florida State cancer registrywith inpatient and outpatient discharge abstractsto ascertain insurance status found that personswho were uninsured were more likely to be diag-nosed with late-stage breast, colorectal, andprostate cancer, and melanoma and that patientswho were Medicaid-insured were more likelyto be diagnosed with late-stage breast cancer andmelanoma. This study could not examine dura-tion of Medicaid enrollment before diagnosis.2

The NCDB, a registry containing informa-tion about cancer patients treated at over 1,500Commission on Cancer-approved facilities inthe United States, has collected information onpatient insurance status at the time of diagnosissince 1996. Several recent studies have used this

database to examine the relationship betweeninsurance status and stage at diagnosis. Patientsdiagnosed with oropharyngeal and laryngealcancer from 1996 to 2003 who were uninsuredor covered by Medicaid were significantly morelikely to be diagnosed with late-stage and largertumors.3,4 Another study of breast cancer patientsdiagnosed from 1998 to 2003 and included inthe NCDB found that women who were unin-sured or had Medicaid insurance were about 1.5times more likely to be diagnosed with Stage IIversus Stage I and 2.5 times more likely to bediagnosed with Stage III/IV versus Stage I dis-ease than those with private insurance.5

Data from the NCDB were also used to inves-tigate the relationship between insurance status,stage at diagnosis, and survival. These analyseswere restricted to patients diagnosed in 1999and 2000, the most recent years of diagnosis forwhich at least 5 years of vital status follow up isavailable. Survival analyses controlled for age atdiagnosis, race/ethnicity, sex, and area-basedincome. In addition, for cancer sites whereAmerican Joint Committee on Cancer stagingis used, analyses were performed with and with-out control for stage at diagnosis to better under-stand how much of the survival differences byinsurance status could be explained by differ-ences in stage at diagnosis.

In analyses of cancer survival for all cancersites combined, patients who were uninsured andthose who were Medicaid-insured at the time ofdiagnosis were 1.6 times as likely to die in 5 yearsas those with private insurance (Figure 10). Moredetailed analyses were done for breast and col-orectal cancers, 2 important cancers for whichboth early detection and quality of treatment areknown to influence survival.

Figure 11 shows the stage distribution of breastcancer cases diagnosed among White, AfricanAmerican, and Hispanic women in 1999/2000.In each racial/ethnic group, patients with pri-vate insurance were more likely to be diagnosedwith Stage I breast cancer and less likely to bediagnosed with Stage III and IV cancer thanthose who were uninsured or had Medicaidinsurance. Breast cancer survival for all stagescombined was also associated with insurance sta-tus (Figure 12). Among White patients with pri-vate insurance, 89% survived 5 years compared




Am

erican Cancer S


cancersoc.orgD

ownloaded from


with 76% of patients who were uninsured orhad Medicaid insurance; among African Americanwomen with private insurance, 81% survived5 years compared with 65% of uninsured patientsand 63% of Medicaid-insured patients. AmongHispanic patients with private insurance, 86%survived 5 years compared with 83% who were

uninsured and 76% of those with Medicaid insur-ance. Stage-specific survival for breast cancerpatients is depicted in Figure 13, with solid linesrepresenting the survival experience of patientswith private insurance and dotted lines repre-senting the survival experience of patients whowere uninsured or had Medicaid insurance in each



FIGURE 10 Cancer Survival by Insurance Status*.*Patients aged 18 to 64 years diagnosed from 1999 to 2000; excluded from the analysis: unknown stage; race/ethnicity otherthan White, African American, or Hispanic; missing information on stage, age, race/ethnicity, or zip code. Covariates includedin the model are age, race, sex, and zip code-based income.Data Source: National Cancer Data Base.


Am

erican Cancer S


cancersoc.orgD

ownloaded from


Sharriet

Sticky Note

This chart needs to be distributed as widely as possible.

racial and ethnic group (Figure 13). Althoughinsurance status was an important predictor ofsurvival within stage, in each stage and insur-ance group, African American patients had lowersurvival than White patients.

Figure 14 shows the stage distribution of col-orectal cancer cases diagnosed among White,African American, and Hispanic patients in 1999and 2000. In each racial/ethnic group, patientswith private insurance were more likely to bediagnosed with Stage I colorectal cancer and lesslikely to be diagnosed with Stage IV colorectalcancer than those who were uninsured or hadMedicaid insurance. Survival for all stages com-bined was also associated with insurance status(Figure 15). Among White patients with privateinsurance, 66% survived 5 years compared with50% of patients who were uninsured and 46%of those with Medicaid insurance; among AfricanAmerican patients with private insurance, 60%survived 5 years compared with 41% of unin-sured patients and Medicaid-insured patients;among Hispanic patients, 63% of those with pri-vate insurance survived 5 years compared with57% of those who were uninsured and 53% ofthose with Medicaid insurance. Stage-specificsurvival for colorectal cancer patients is depictedin Figure 16, with solid lines representing thesurvival experience of patients with private insur-ance and dotted lines representing the survivalexperience of patients who were uninsured or hadMedicaid insurance in each racial and ethnicgroup. For both White and African Americanpatients, privately insured patients with Stage IIdisease had better survival than patients whowere Medicaid-insured or uninsured who hadStage I disease, and privately insured patientswith Stage III disease had similar survival toMedicaid-insured or uninsured patients withStage II disease. Although insurance status was animportant predictor of survival within stage, ineach stage and insurance group, African Americanpatients had lower survival than White patients.

The results of the analysis of breast and col-orectal cancer survival by insurance status amongpatients diagnosed in 1999 and 2000 and reportedto the NCDB were similar to those of a previousstudy that examined 3-year cancer survival byinsurance status among patients diagnosed inKentucky from 1995 to 1998 and followed through1999.48 The latter study found that 3-year relativesurvival among breast cancer patients was 90.6%for privately insured patients, 75.5% for patientswith Medicaid insurance, and 77.7% among theuninsured. For colorectal cancer patients, 3-year



Private Uninsured/Medicaid

0%

10%

20%

30%

40%

50%

60%

I II III IV

Stage at Diagnosis

White

African American

I II III IV

Stage at Diagnosis

0%

10%

20%

30%

40%

50%

60%

I II III IV

Stage at Diagnosis

Hispanic

0 %

10 %

20 %

30 %

40 %

50 %

60 %

FIGURE 11 Breast Cancer Stage Distribution byRace and Insurance Status*.*Patients aged 18 to 64 years diagnosed from 1999 to 2000;excluded from the analysis: unknown stage; race/ethnicityother than White, African American, or Hispanic; missing infor-mation on stage, age, race/ethnicity, or zip code.Data Source: National Cancer Data Base.


Am

erican Cancer S


cancersoc.orgD

ownloaded from


0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 10 20 30 40 50 60

lavivruS

detsujdA

etairavoC

White

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 10 20 30 40 50 60

lavivruS

detsujdA

etairavoC

African American

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 10 20 30 40 50 60

lavivruS

detsujdA

etairavoC

Survival Time in Months

Hispanic

Stage 1, Private

Stage 2, Private

Stage 3, Private

Stage 4, Private

Stage 1, Uninsured/Medicaid






0.75

0.76

0.77

0.78

0.79

0.80

0.81

0.82

0.83

0.84

0.85

0.86

0.87

0.88

0.89

0.90

0.91

0.92

0.93

0.94

0.95

0.96

0.97

0.98

0.99

1.00

0 10 20 30 40 50 60

lavivruS

detsujdA

etairavoC

White

Private Insurance

Medicaid

Uninsured

0.62

0.64

0.66

0.68

0.70

0.72

0.74

0.76

0.78

0.80

0.82

0.84

0.86

0.88

0.90

0.92

0.94

0.96

0.98

1.00

0 10 20 30 40 50 60

lavivruS

detsujdA

etairavoC

African American

Private Insurance

Medicaid

Uninsured

0.76

0.77

0.78

0.79

0.80

0.81

0.82

0.83

0.84

0.85

0.86

0.87

0.88

0.89

0.90

0.91

0.92

0.93

0.94

0.95

0.96

0.97

0.98

0.99

1.00

0 10 20 30 40 50 60

lavivruS

detsujdA

etairavoC


Hispanic

Private Insurance

Medicaid Uninsured

FIGURE 13 Breast Cancer Survival by Stage and Insurance Status*.*Patients aged 18 to 64 years diagnosed from 1999 to 2000; excluded fromthe analysis: unknown stage; race/ethnicity other than White, African American,or Hispanic; missing information on stage, age, race/ethnicity, or zip code.Data Source: National Cancer Data Base.

FIGURE 12 Breast Cancer Survival by Race and Insurance Status*.*Patients aged 18 to 64 years diagnosed from 1999 to 2000; excluded fromthe analysis: unknown stage; race/ethnicity other than White, African American,or Hispanic; missing information on stage, age, race/ethnicity, or zip code.Data Source: National Cancer Data Base.


Am

erican Cancer S


cancersoc.orgD

ownloaded from


survival was 70.9% for those with private insur-ance, 53% for those with Medicaid insurance, and52.8% for those who were uninsured.

Although neither the NCDB analyses nor theKentucky Registry study were able to control

for sociodemographic factors other than race/ethnicity, sex and age, or for the presence ofother health conditions that might impact sur-vival, both studies were able to control for stage,and the NCDB analysis controlled for zip codelevel of income. In addition, when survival byinsurance status was examined using the NCDBfor a cancer with very high survival (Stage I andII thyroid cancer), the largest difference in pre-dicted 5-year survival based on differences ininsurance status was only 2%. Thus, it does notappear likely that the large differences in sur-vival between insurance groups are accountedfor by factors other than those related to diagno-sis and treatment of their cancer.

Possible Reasons for Associations BetweenInsurance Type, Stage at Diagnosis, and Survival

Later stage at diagnosis for cervical, breast,colorectal, and prostate cancer among patientswho are uninsured or have Medicaid insurancecan be explained in part by lower access toand/or utilization of cancer screening. Analysesof NHIS 2005 data presented in this report, aswell as prior studies, found that screening rateswere substantially lower among uninsured com-pared with privately insured individuals and thatMedicaid-insured patients consistently hadscreening rates that were lower than those forthe privately insured and higher than those forthe uninsured. In addition, later stage at diag-nosis may be associated with lack of follow upor delay in follow up of abnormal screening testresults. A review of studies evaluating follow-up care for an abnormal cancer screening resultfound that less than 75% of patients receivedsuch care and identified barriers to follow upat the provider, patient, and health care systemlevels.49 Appropriate follow up of an abnormalscreening test requires a number of critical stepswhere the process can break down. The pri-mary care provider and/or patient must beinformed of the abnormal result, the appropri-ate diagnostic evaluation must be recommended,a provider and site for the diagnostic evaluationmust be identified, and the patient must makeand keep the appointment. Patients withouthealth insurance and those whose health insur-ance is not widely accepted face additional cost,



African American

I II III IV

I II III IV

I II III IV

FIGURE 14 Colorectal Cancer Stage Distributionby Race and Insurance Status*. *Patients aged 18 to 64 years diagnosed from 1999 to 2000;excluded from the analysis: unknown stage; race/ethnicityother than White, African American, or Hispanic; missing infor-mation on stage, age, race/ethnicity, or zip code.Data Source: National Cancer Data Base.


Am

erican Cancer S


cancersoc.orgD

ownloaded from




0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 10 20 30 40 50 60

lavivruS

detsujdA

etairavoC

White

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 10 20 30 40 50 60

lavivruS

detsujdA

etairavoC

African American

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 10 20 30 40 50 60

lavivruS

detsujdA

etairavoC


Hispanic

Stage 1, Private

Stage 2, Private

Stage 3, Private

Stage 4, Private





0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 10 20 30 40 50 60

lavivruS

detsujdA

etairavoC

White

Private Insurance

Medicaid

Uninsured

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 10 20 30 40 50 60

lavivruS

detsujdA

etairavoC

African American

Private Insurance

Medicaid

Uninsured

0.5

0.6

0.7

0.8

0.9

1.00

0 10 20 30 40 50 60

lavivruS

detsujdA

etairavoC


Hispanic

Private Insurance

Medicaid

Uninsured

FIGURE 16 Colorectal Cancer Survival by Stage andInsurance Status*.*Patients aged 18 to 64 years diagnosed from 1999 to 2000; excluded fromthe analysis: unknown stage; race/ethnicity other than White, African American,or Hispanic; missing information on stage, age, race/ethnicity, or zip code.Data Source: National Cancer Data Base.

FIGURE 15 Colorectal Cancer Survival by Race and InsuranceStatus*.*Patients aged 18 to 64 years diagnosed from 1999 to 2000; excluded fromthe analysis: unknown stage; race/ethnicity other than White, African American,or Hispanic; missing information on stage, age, race/ethnicity, or zip code.Data Source: National Cancer Data Base.


Am

erican Cancer S


cancersoc.orgD

ownloaded from


administrative, and access barriers that may beinsurmountable for many patients.

The finding that patients with Medicaid cov-erage experience later stage at diagnosis thanpatients with private insurance must be inter-preted with caution for several reasons. Mostimportantly, many individuals who are not oth-erwise eligible for Medicaid based on incomeand other characteristics qualify for coveragewhen they are diagnosed with a serious med-ical condition such as cancer. There is no pub-licly available source of information on thepercentage of patients who are enrolled inMedicaid after diagnosis. A study in Michiganfound that 36% of cancer patients with Medicaidcoverage were enrolled in Medicaid after beingdiagnosed with cancer,45 but it is not knownhow proportion may vary by state or cancer site.

A review of the impact of health insurancecoverage on health, which was published by theInstitute of Medicine in 2002, noted that manystudies of overall health status, cancer outcomes,and hospital-based care have found that adultswith Medicaid coverage frequently fare no bet-ter, and sometimes fare worse, than uninsuredpatients in their health-related outcomes.50

According to this report, 2 factors contributeto poor outcomes for Medicaid enrollees. Onefactor mentioned earlier is that patients maybecome eligible for Medicaid as a result of poorhealth. The second is the structure and opera-tion of Medicaid as an insurance program. Theprogrammatic features of Medicaid that con-tribute to worse health-related outcomes includeprovider participation and payment levels andlimited coverage periods. Payment rates belowthe cost of the care delivery reduce access tohealth care services for Medicaid enrollees inmany states and localities. Thus, Medicaidenrollees often find themselves limited to thesame set of overtaxed safety-net providers asuninsured adults, with concomitant delays ingetting appointments and referrals to special-ists. Medicaid’s limited coverage periods alsoweaken any positive effects of insurance. Onestudy based on a federal survey found that themedian length of time that adults under age65 years maintained Medicaid enrollment was just5 months50; Medicaid requires eligibility certi-fications as frequently as monthly, and some

people lose coverage simply because they didnot meet administrative requirements. As a con-sequence of the intermittency of Medicaid cov-erage, adults identified as covered by Medicaidat one point in time may not achieve the ben-efits that continuous health coverage can provide.

Health insurance status may be associated withsurvival through a variety of mechanisms. Laterstage at diagnosis observed for Medicaid-insuredand uninsured patients would lead to lower over-all survival even if quality of treatment and suc-cess of treatment were equivalent to the privatelyinsured. However, analyses of NCDB data forbreast and colorectal cancer find that even withinstage at diagnosis, survival is poorer for patientswith no health insurance or with Medicaid insur-ance. Lower survival within cancer stage mayresult from a variety of factors related to accessto care and quality of care, including adequacyof staging (leading to understaging); differencesin tumor size, grade, or other prognostic factorswithin stage groupings; delays in initiation oftreatment; differences in receipt of treatmentconsistent with recommended guidelines; qual-ity and outcome of specific treatments, such ascompleteness of surgical resection; differencesin provision of supportive care; and completionof the full course of therapy.

Limitations of Existing Data

Data are extremely limited on the relation-ship between insurance status and variations incancer treatment. This limitation is due in part tothe incompleteness of certain types of treatmentinformation in cancer registry records, whichmakes it difficult to study treatment patterns orconcordance with treatment guidelines usingregistry data alone. The most commonly useddata resource for studying cancer treatment is theSEER-Medicare database, which by definitionincludes only insured patients. Among the lim-ited number of studies conducted, one studyfound that insurance status and poverty levelwere predictors of having delays of greater than3 months from initial diagnosis to start of treat-ment among women with invasive breast can-cer.51 Studies of variations in treatment amongpatients with breast and colorectal cancer havenot found consistent variations in treatmentand concordance with treatment guidelines by




Am

erican Cancer S


cancersoc.orgD

ownloaded from


insurance status.52–56 However, there is consid-erable variation between studies in insurancegroups and treatments considered. One studyreported that patients who are uninsured or whohave Medicaid insurance are less likely to receivesurgery for lung and pancreatic cancer at high-volume facilities57; another found that the like-lihood of initial presentation of colon cancer asa surgical emergency due to bowel perforation,peritonitis, or obstruction was 2.1 times higheramong Medicaid enrollees and 2.6 times higheramong uninsured patients than among privatelyinsured patients under the age of 65 years.58

Although variations in health insurance cov-erage likely contribute to racial and ethnic dis-parities in cancer outcomes, disparities persistfor several outcomes even when differences ininsurance status are accounted for. Racial andethnic disparities in health and health care occurin the context of broader historic and contem-porary social and economic inequality and evi-dence of persistent racial and ethnic discriminationin many sectors of American life.59 Even in theabsence of financial barriers to care, factors thatmay impact receipt of optimal care for racial andethnic minority patients include cultural andlanguage differences between providers andpatients that may result in poorer communica-tion, undermining informed decision-makingand patient’s adherence to treatment regimens,experiences of discrimination that may directlyaffect both access to care and health and affect trustin the health care system and the doctor-patient

relationship.50 Even if health insurance and finan-cial barriers can be overcome, further research andinterventions will be needed to address thesebarriers.

Although there is substantial evidence thatinsurance status is an important factor in accessto and utilization of cancer care, there is littleinformation on how economic issues impacttreatment choices at the level of the individualpatient. For example, to what extent do indi-viduals forego treatment or select less than opti-mal treatment because they are unable to find ahealth care provider who is willing to provideit or because they are afraid of the level of med-ical debt that they would incur? As the cost ofsome new cancer therapies can exceed $100,000a year, to what extent will availability and typeof insurance coverage, as well as individual finan-cial resources, determine who has access to themost effective therapies?

CONCLUSION

There is substantial evidence that lack of ade-quate health insurance coverage is associatedwith less access to care and poorer outcomes forcancer patients. As our nation’s investments incancer research provide greater understandingof how to prevent cancer, detect it early, andtreat it effectively, access to health care becomeseven more important to the American CancerSociety’s goal of eliminating cancer as a majorpublic health problem.



REFERENCES

1. American Cancer Society. Cancer Prevention& Early Detection Facts & Figures. Atlanta, GA:American Cancer Society; 2007.

2. Roetzheim RG, Pal N, Tennant C, et al. Effectsof health insurance and race on early detection ofcancer. J Natl Cancer Inst 1999;91:1409–1415.

3. Chen AY, Schrag NM, Halpern MT, Ward EM.The impact of health insurance status on stage atdiagnosis of oropharyngeal cancer. Cancer2007;110:395–402.

4. Chen AY, Schrag NM, Halpern M, et al. Healthinsurance and stage at diagnosis of laryngeal can-cer: does insurance type predict stage at diagno-sis? Arch Otolaryngol Head Neck Surg 2007;133:784–790.

5. Halpern MT, Bian J, Ward EM, et al. Insurancestatus and stage of cancer at diagnosis among womenwith breast cancer. Cancer 2007;110:403–411.

6. Shavers VL, Brown ML. Racial and ethnic dis-parities in the receipt of cancer treatment. J NatlCancer Inst 2002;94:334–357.

7. National Center for Health Statistics, Centersfor Disease Control and Prevention. National HealthInterview Survey. Available at: http://www.cdc.gov/nchs/nhis.htm. Accessed November 6, 2007.

8. Stewart AK, Bland KI, McGinnis LS Jr, et al.Clinical highlights from the National Cancer DataBase, 2000. CA Cancer J Clin 2000;50:171–183.

9. Collins SR, White C, Kr iss JL. Whitheremployer-based health insurance? The current andfuture role of U.S. companies in the provision andfinancing of health insurance. New York, NY: TheCommonwealth Fund; 2007.

10. Claxton G, Gabel J, DiJulio B, et al. EmployerHealth Benefits 2007 Annual Survey. Menlo Park,CA: Kaiser Family Foundation and Health Researchand Educational Trust; 2007.

11. Blumenthal D. Employer-sponsored healthinsurance in the United States—origins and impli-cations. N Engl J Med 2006;355:82–88.

12. Committee on the Consequences of Uninsur-ance, Institute of Medicine. Insuring America’sHealth: Principles and Recommendations. Wash-ington, DC: National Academies Press; 2004.

13. U.S. Department of Labor, Employee BenefitsSecurity Administration. FAQs About COBRAContinuation Health Coverage. Available at: http://www.dol.gov/ebsa/faqs/faq_consumer_cobra.html.Accessed November 6, 2007.

14. Bishop H, Clark P, Leopold B, et al. 2007CCH Medicare and Medicaid Benefits. Chicago,IL: Wolters Kluwer; 2007.

15. Centers for Medicare & Medicaid Services,U.S. Department of Health and Human Services.Medicare FAQs. Available at: http://questions.medicare.gov. Accessed November 6, 2007.


Am

erican Cancer S


cancersoc.orgD

ownloaded from




16. AcademyHealth. Glossary of terms commonlyused in health care. Washington, DC: AcademyHealth; 2004.

17. Gilmer T, Kronick R, Rice T. Children wel-come, adults need not apply: changes in publicprogram enrollment across states and over time.Med Care Res Rev 2005;62:56–78.

18. Kaiser Family Foundation. Key Facts: Race,Ethnicity and Medical Care. Menlo Park, CA:Kaiser Family Foundation; 2007. Available at http://www.kff.org/minorityhealth/6069.cfm.

19. Breast and Cervical Cancer Prevention andTreatment Act of 2000, Pub L No. 106-354 (2000).

20. Tangka FK, Dalaker J, Chattopadhyay SK,et al. Meeting the mammography screening needsof underserved women: the performance of theNational Breast and Cervical Cancer Early DetectionProgram in 2002–2003 (United States). CancerCauses Control 2006;17:1145–1154.

21. French C, True S, McIntyre R, et al. Stateimplementation of the Breast and Cervical CancerPrevention and Treatment Act of 2000: a collabo-rative effort among government agencies. PublicHealth Rep 2004;119:279–285.

22. Collins S, Schoen C, Doty M, et al. Payingmore for less: older adults in the private insurancemarket. Findings from the Commonwealth Surveyof Older Adults. New York, NY: The Common-wealth Fund; 2005. Available at: http://www.commonwealthfund.org/publications_show.htm?doc_id=282104.

23. Cohen RA, Martinez ME. Health InsuranceCoverage: Early Release of Estimates From theNational Health Interview Survey, 2006. Atlanta,GA: Centers for Disease Control and Prevention;2007. Available at: http://www.cdc.gov/nchs/data/nhis/earlyrelease/insur200612.pdf.

24. Collins S, Davis K, Doty M, et al. Gaps inhealth insurance: an all-American problem. Findingsfrom the Commonwealth Fund Biennial HealthInsurance Survey. New York, NY: The Com-monwealth Fund; 2006.

25. Clemens-Cope L, Garrett B. Changes inEmployer-Sponsored Health Insurance Sponsorship,Eligibility, and Participation: 2001–2005. Wash-ington, DC: Kaiser Commission on Medicaid andthe Uninsured; 2006.

26. Himmelstein DU, Warren E, Thorne D,Woolhandler S. Illness and injury as contributorsto bankruptcy. Health Aff (Millwood) 2005;(suppl):W5-63–W65-73.

27. Siminoff LA, Ross L. Access and equity tocancer care in the USA: a review and assessment.Postgrad Med J 2005;81:674–679.

28. Banthin JS, Bernard DM. Changes in financialburdens for health care: national estimates for thepopulation younger than 65 years, 1996 to 2003.JAMA 2006;296:2712–2719.

29. Banthin JS, Bernard D. Medical ExpendituresPanel Survey Statistical Brief #122. Out-of-Pocket

Expenditures on Health Care and InsurancePremiums among the Elderly Population, 2003.Washington, DC: Agency for Health Care Researchand Quality; 2006.

30. Anderson GF. From ‘soak the rich’ to ‘soak thepoor’: recent trends in hospital pricing. Health Aff(Millwood) 2007;26:780–789.

31. O’Toole TP, Arbelaez JJ, Lawrence RS. Medicaldebt and aggressive debt restitution practices: preda-tory billing among the urban poor. J Gen InternMed 2004;19:772–778.

32. Hing E, Burt C. Characteristics of office-basedphysicians and their practices: United States,2003–04. Series 12, No. 164. Hyattsville, MD:National Center for Health Statistics; 2007.

33. Hoffman C, Sered S. Threadbare: Holes inAmerica’s Health Care Safety Net. Washington,DC: Kaiser Commission on Medicaid and theUninsured; 2005.

34. Asplin BR, Rhodes KV, Levy H, et al. Insurancestatus and access to urgent ambulatory care follow-up appointments. JAMA 2005;294:1248–1254.

35. Kuehn BM. Poverty shift may burden healthsystem. JAMA 2007;297:1047–1048.

36. Cokkinides VE, Ward E, Jemal A, Thun MJ.Under-use of smoking-cessation treatments: resultsfrom the National Health Interview Survey, 2000.Am J Prev Med 2005;28:119–122.

37. Coughlin SS, Uhler RJ, Bobo JK, Caplan L.Breast cancer screening practices among womenin the United States, 2000. Cancer Causes Control2004;15:159–170.

38. Sambamoorthi U, McAlpine DD. Racial, eth-nic, socioeconomic, and access disparities in theuse of preventive services among women. PrevMed 2003;37:475–484.

39. Potosky AL, Breen N, Graubard BI, Parsons PE.The association between health care coverage andthe use of cancer screening tests. Results from the1992 National Health Interview Survey. Med Care1998;36:257–270.

40. Ioannou GN, Chapko MK, Dominitz JA.Predictors of colorectal cancer screening partici-pation in the United States. Am J Gastroenterol2003;98:2082–2091.

41. Parker J, Gebretsadik T, Sabogal F, et al.Mammography screening among California Medi-care beneficiaries: 1993–1994. Am J Prev Med1998;15:198–205.

42. Chen JY, Diamant A, Pourat N, Kagawa-SingerM. Racial/ethnic disparities in the use of preven-tive services among the elderly. Am J Prev Med2005;29:388–395.

43. Koroukian SM, Xu F, Dor A, Cooper GS.Colorectal cancer screening in the elderly popula-tion: disparities by dual Medicare-Medicaid enroll-ment status. Health Serv Res 2006;41:2136–2154.

44. Bradley CJ, Given CW, Roberts C. Disparitiesin cancer diagnosis and survival. Cancer 2001;91:178–188.

45. Bradley CJ, Given CW, Roberts C. Late stagecancers in a Medicaid-insured population. MedCare 2003;41:722–728.

46. Bradley CJ, Gardiner J, Given CW, RobertsC. Cancer, Medicaid enrollment, and survival dis-parities. Cancer 2005;103:1712–1718.

47. O’Malley CD, Shema SJ, Clarke LS, et al.Medicaid status and stage at diagnosis of cervicalcancer. Am J Public Health 2006;96:2179–2185.

48. McDavid K, Tucker TC, Sloggett A, ColemanMP. Cancer survival in Kentucky and health insur-ance coverage. Arch Intern Med 2003;163:2135–2144.

49. Yabroff KR, Washington KS, Leader A, et al.Is the promise of cancer-screening programs beingcompromised? Quality of follow-up care afterabnormal screening results. Med Care Res Rev2003;60:294–331.

50. Committee on the Consequences of Uninsur-ance, Board on Health Care Services, Institute ofMedicine. Care Without Coverage: Too Little, TooLate. Washington, DC: National Academy Press;2002.

51. Gwyn K, Bondy ML, Cohen DS, et al. Racialdifferences in diagnosis, treatment, and clinicaldelays in a population-based study of patients withnewly diagnosed breast carcinoma. Cancer 2004;100:1595–1604.

52. Harlan LC, Greene AL, Clegg LX, et al. Insur-ance status and the use of guideline therapy in thetreatment of selected cancers. J Clin Oncol 2005;23:9079–9088.

53. Parviz M, Cassel JB, Kaplan BJ, et al. Breastconservation therapy rates are no different in med-ically indigent versus insured patients with earlystage breast cancer. J Surg Oncol 2003;84:57–62.

54. Roetzheim RG, Gonzalez EC, Ferrante JM,et al. Effects of health insurance and race on breastcarcinoma treatments and outcomes. Cancer2000;89:2202–2213.

55. Richardson LC, Tian L, Voti L, et al. The rolesof teaching hospitals, insurance status, and race/eth-nicity in receipt of adjuvant therapy for regional-stage breast cancer in Florida. Am J Public Health2006;96:160–166.

56. Wu X, Chen VW, Andrews PA, et al. Treatmentpatterns for stage III colon cancer and factors relatedto receipt of postoperative chemotherapy inLouisiana. J La State Med Soc 2004;156:255–261.

57. Liu JH, Zingmond DS, McGory ML, et al.Disparities in the utilization of high-volume hos-pitals for complex surgery. JAMA 2006;296:1973–1980.

58. Diggs JC, Xu F, Diaz M, et al. Failure to screen:predictors and burden of emergency colorectalcancer resection. Am J Manag Care 2007;13:157–164.

59. Smedley B, Stith A, Nelson A. Unequal Treat-ment: Confronting Racial and Ethnic Disparitiesin Health Care. Washington, DC: NationalAcademies Press; 2003.


Am

erican Cancer S


cancersoc.orgD

ownloaded from


association of insurance with cancercare utilization and outcomes

Documents

cancer journal

colorectal cancer

treatment of cancer

prostate cancer

cancer care utilization

tems of health insurance

health insurance coverage

american cancer society