1.1 Background
There is a signifi cant disconnect between cancer research discovery/development (i.e., what we know) and the delivery of care to cancer patients (i.e., what we do).6 This disconnect is an important factor contributing to an imbalanced and unjust burden of cancer in our society—the burden falling on some racial and ethnic minority groups, individuals with low socioeconomic status (SES), residents in certain geographic locations, and other medically underserved groups.
Improving the delivery of cancer care to these population groups would help to reduce cancer health disparities in the United States.
Examining and understanding the economic and human costs of cancer health disparities is an important step in eliminating such disparities. Understanding the economic costs and human costs of cancer health disparities may provide guid-ance to policy makers with regard to cancer health care. To address this need, the Center to Reduce Cancer Health Disparities (CRCHD) of the National Cancer Institute (NCI) convened a Think Tank meeting on December 6–7, 2004. This meeting brought together health economists, cancer care providers, insurers, and policy experts to explore the economic costs to the nation resulting from cancer health disparities among certain population groups (including racial and ethnic minority groups and individuals with low SES) and to identify potential interventions to address these disparities. The purpose of this report is to provide a summary of the ideas and discussions that occurred during this meeting and to review the current knowledge on the economics of cancer health disparities.
Economic Costs of Cancer Health Disparities: Summary of Meeting Proceedings 11
1.2 Objectives
The original objectives of the meeting were:
■ To examine the current evidence regarding the costs of cancer health disparities;
■ To assess the currently available cost data and data needs related to costs of cancer health disparities;
■ To explore new and creative ways of examining and estimating the economic costs of cancer health disparities (since there are currently not enough databases containing data of this nature);
■ To strengthen the economic evidence base regarding the cost of cancer health disparities; and,
■ To explore new and creative strategies for reducing and eventually eliminating cancer health disparities.
During the initial discussions among Think Tank participants, it became clear to the participants that the evidence to support the completion of the objectives listed above was not available; therefore, the participants focused on intervention strategies and future research areas which will enable completion of these objectives.
1.3 Think Tank Process
The Think Tank meeting consisted of individual presentations from the interdisciplinary team of experts, as well as group discussions and breakout sessions to explore certain issues in greater depth (see Appendix
A for a list of meeting participants). Participants for the meeting were selected based on their expertise in specifi c areas of relevance to the Think Tank discussions, including clinical epidemiology, health care policy, and cost-effectiveness analyses. Prior to the meeting, all participants received a package of pre-planning documents, including a copy of the notes from the planning meeting convened by CRCHD
( Appendix B) and a background paper on economic costs of cancer health disparities prepared by the CRCHD staff ( Appendix C).
The two-day meeting began with introductions and discussions regarding the purpose and rationale behind convening this Think Tank meeting and the core questions to be discussed and answered. The agenda for Day 1 of the meeting consisted of six individual presentations followed by group discussion (The agenda and brief descriptions of the presentations are provided in Appendix D). Additionally, the six individual presenters participated in a panel discussion, during which both participants and observers asked questions and discussed the major issues from the day-long session. Day 2 began with a breakout session where participants were divided into two groups and given the same set of key questions:
■ What is the total cost of cancer care?
■ What proportion of the total cost of cancer care is related to health disparities?
■ What would be the cost of eliminating cancer health disparities in America?
■ What is the value of reducing cancer health disparities in America?
■ What are the policy implications of reducing cancer health disparities?
■ What is the cost of changing policies?
Deliberations and discussions from the breakout session were summarized and followed by fi nal discussions and wrap-up.
1.4 Organization of Report
This report begins with an overview of the determinants of cancer care disparity and a description of the
“cancer care continuum” (Section 2). Section 3 discusses the total cost of cancer care and limitations of currently available data sources. Section 4 presents the methodological issues related to reducing cancer health disparities, followed by Section 5, a discussion of the costs and cost-effectiveness of implementing interventions to reduce cancer health disparities. Finally, Section 6 summarizes the Think Tank’s recommendations for future research and policy initiatives.
12 Economic Costs of Cancer Health Disparities: Summary of Meeting Proceedings
Overview SECTION 2
2.1 Defi nition of Disparity
There are several different defi nitions of disparities and the conclusions regarding the impact of disparities can differ based on the defi nition used.2 The NCI’s defi nition of cancer health disparities is as follows:
“Disparities, or inequalities, occur when members of some population groups do not enjoy the same health status as other groups.
“Disparities are determined and measured by three health statistics: incidence (the number of new cancers), mortality (the number of cancer deaths), and survival rates (length of survival following diagnosis of cancer). Health disparities occur when one group of people has a higher incidence or mortality rate than another, or when survival rates are less for one group than another.3”
The Minority Health and Health Disparities Act of 2000 provides the following defi nition of “disparity population”:
“A population is a health disparity population if there is a signifi cant disparity in the overall rate of disease incidence, prevalence, morbidity, mortality, or survival rates in the population as compared to the health status of the general population. In addition, ... [the defi nition may include] populations for which Economic Costs of Cancer Health Disparities: Summary of Meeting Proceedings 13
there is a considerable disparity in the quality, outcomes, cost, or use of health care services or access to, or satisfaction with such services as compared to the general population.”7
2.2 Determinants of Disparities
Determinants of cancer health disparities are underlying factors that may have an effect on individual outcome measures. Disparities are most often identifi ed along racial and ethnic lines, i.e., African Americans, Hispanics, Native Americans/Alaska Natives, Asian Americans/Pacifi c Islanders, and whites have different disease rates and survival rates. However, factors contributing to disparities extend beyond race and ethnicity though.3 They include factors within the health care delivery system (e.g., access to health care, insurance coverage, health care network disconnects) as well as factors outside of the system (e.g., education, SES, geographic differences).
Factors Within the Health Care Delivery System
Within the health care delivery system, lack of insurance coverage is a major contributor to cancer health disparities. Furthermore, the stability and quality of insurance coverage is even more important than simply being insured. The U.S. Census Bureau reports that, in 2005, approximately 16% of the population (46.6
million people) had no health insurance coverage.5 The percentage of persons without health insurance was higher in certain racial groups and in groups with lower SES. Of Americans under 65 years of age who are diagnosed with cancer, 20% of Hispanics, 14% of African Americans, and 10% of whites do not have health insurance.8
An Institute of Medicine (IOM) report “Care Without Coverage: Too Little, Too Late, ” found that uninsured patients with breast, colorectal, or prostate cancer are in poorer health and more likely to die prematurely than their insured counterparts, primarily due to delayed diagnosis.9 For example, an uninsured FIGURE 1 Ratio of the Probability of Diagnosis of Cancer at Late Stage,Uninsured Compared with Insured,* 1994
3
Ratio of the Probability of Late vs. Early
3
Ratio of the Risk of Death,
Stage Cancer, Uninsured/Insured
Uninsured/Insured
2.6
2
2
1.7
1.7
1.5
1.4
1
1
0
0
Colorectal
Melanoma
Breast
Prostate
Colorectal
Cancer
Cancer
Cancer
Cancer
* Privately insured all had commercial indemnity plans. ** Among cancer cases identifi ed in 1994; mortality followup through 1997.
All differences are statistically signifi cant after adjusting for age, sex, race/ethnicity, co-morbidity, marital status (when appropriate), smoking status, socioeconomic status, education, stage at diagnosis, and treatment.
SOURCES: The Kaiser Commission on Medicaid and the Uninsured, 2003.10
14 Economic Costs of Cancer Health Disparities: Summary of Meeting Proceedings woman with breast cancer faces a 30%-50% higher risk of dying compared with her insured counterpart, and an uninsured person with colorectal cancer has a 50% greater chance of dying compared with someone who has private insurance.9 This fi nding is also true for other cancers, as shown in Figure 1.
Over time, high rates of persons without insurance coverage lead to unstable connections to care, disrup-tions in care, and greater costs. Lack of adequate insurance coverage limits access to care, partially due to cost-related issues12 and partially due to the lack of a primary care provider.13 Finally, even in persons with low incomes who are insured, cost-sharing and out-of-pocket expenses compromise receipt of effective medical care.11
Factors such as availability of appropriate providers and services at an affordable cost and access to appropriate referral services are important to receiving high quality health care and, if compromised, can contribute to health care disparities. For instance, in rural areas there may be a lack of specialists which can lead to delays in diagnosis of cancer and treatment, or language barriers can result in non-English speakers not seeking or receiving appropriate care.9 Disruption of traditional community-based care can also lead to disparities as providers familiar with individuals in a particular locality may no longer be available.14
Factors External to the Health Care Delivery System
Factors external to the health care delivery system which contribute to disparities include gender, race, ethnicity, SES (income, education), and geographic location. According to Freeman (2004)1, the three principal determinants of cancer disparities ( Figure 2) are:
■ Poverty (low SES);
■ Culture; and
■ Social Injustice.
Poverty is generally correlated with lack of information, risk-promoting behaviors, and reduced access to appropriate health care. The percentage of individuals living in poverty is disproportionately high among African Americans, Hispanics/Latinos, Native Americans, Pacifi c Islanders, and Native Hawaiians compared with white Americans. The poor are at greater risk of being diagnosed and treated for cancer at late stages of disease and are less likely to survive a diagnosis of cancer. Among the three main determinants, poverty contributes to health disparities more than the other two factors.1 A study of colorectal cancer screening among Medicare benefi ciaries concluded that much of the disparities in screening rates can be explained by differences in socioeconomic status. Disparities in socioeconomic status decreased but remained signifi cant even after adjustment for personal and health system factors. 15
FIGURE 2 Causes of Cancer Health Disparities
Causes of Health Disparities
Poverty/Low
Economic Status
Social Injustice
Culture
Prevention
Early
Diagnosis/
Treatment
Post Treatment/
Survival
Detection
Incidence
Quality of Life
and Mortality
SOURCE: Freeman, Adapted from Cancer Epidemiology Biomarkers & Prevention, April 2003.
Economic Costs of Cancer Health Disparities: Summary of Meeting Proceedings 15
A recent report published by CRCHD on cervical cancer (2005) and research by Singh and colleagues16
highlights the correlation of SES and disparities in the incidence and mortality of cervical cancer at the county-level in the United States. The study found substantial inequalities in both the incidence and mortality of cervical cancer, with rates of disparities becoming higher with increasing poverty and decreasing education levels. Patients living in lower SES census tracts were also signifi cantly more likely to be diagnosed at late stages of the disease and were less likely to survive. Figure 3 highlights these disparities.
Although poverty is considered a primary determinant of cancer health disparities, much of the statistics on disparity are also related to racial differences. Race itself does play a role in determining cancer disparities.17, 18 A recent IOM report (2002) found that racial and ethnic disparities in health care exist in the context of broader-scale social and economic inequalities. The report concluded that these disparities persist even after controlling for SES (i.e. income and education). Selected statistics from the NCI and the National Center for Health Statistics (NCHS) indicate several racial disparities. 19
■ African Americans have the highest cancer incidence and cancer-related death rates overall;
■ African American males have the highest incidence and mortality rates for colon, prostate, and lung cancers;
■ While Caucasian American females have the highest incidence of breast cancer, African American females have the highest death rates for breast cancer;
FIGURE 3 U.S. Cervical Cancer Mortality by Race and Poverty Level, 1996-2000
Percent of County Population Below Poverty Level in 1999
8
< 10%
10%-19.99%
20%
6.6
6.0
6
4.8
4.5
4.2
4.2
4
3.6
3.3
3.1
3.1
3.2
2.7
2.7
2.5
2.5
2.3
2.1
2.0
2
verage Annual Age-Adjusted Death Rate per 100,000 PopulationA
0
All Races
Non-Hispanic
Black
American
Asian/Pacific
Hispanic
White
Indian
Islander
Race
SOURCE: Singh et al., 2004.16
16 Economic Costs of Cancer Health Disparities: Summary of Meeting Proceedings FIGURE 4 Critical Disconnect Between Research/Discovery and Delivery of Care Discovery Development Delivery Critical Disconnect
SOURCE: Freeman, 2000.6
FIGURE 5 Cancer Care Continuum
➔
➔
➔
➔
➔
➔
➔
Risk Assessment Primary PreventionScreening/Detection Diagnosis
Treatment
Recurrence Surveillance
End-of-Life Care
➔
➔
➔
➔
➔
➔
➔
Failure to
Primary
Failure in
Failure during
Failure of
Failure in
Failure in
Identify Need
Prevention
Detection
Diagnostic
Treatment
Surveillance
Surveillance
to Screen or
Failure
Evaluation
Counsel
Failure in
Failure to
Failure
Failure During
Failure to
Failure to
Access to
Screen
During
Follow up of
Follow up
Access
Care
Follow up of
Diagnostic/
Surveillance
Care
Abnormal
Treatment
Plan
Results
Plan
Potential Failures During Process of Care
SOURCE: Zapka et al., 2003.20
Economic Costs of Cancer Health Disparities: Summary of Meeting Proceedings 17
■ Hispanic/Latina females have the highest incidence rates among all racial groups for cervical cancer, and Vietnamese females [a subset of Asian-Pacifi c Islanders] have the highest mortality rates among all racial groups for cervical cancer; and,
■ Asian/Pacifi c Islanders have the highest incidence rates of liver and stomach cancers for both sexes.
2.3 Screening, Diagnosis, and Treatment Disparities:
The Cancer Care Continuum
Freeman has hypothesized that there is a critical disconnect between cancer research discovery/development and the delivery of care to cancer patients6 as illustrated in Figure 4.
Even for those who have access to care, the Cancer Care Continuum ( Figure 5) illustrates that disparities in cancer care can occur at any stage of screening, diagnosis, or treatment. In the continuum from risk assessment through end-of-life care, a patient can fail to receive adequate care during any or all steps of the process. For example, a patient may not be screened appropriately, may not receive adequate treatment, may not be able to access end-of-life care, or may experience all three as well as other failures during the process.
18 Economic Costs of Cancer Health Disparities: Summary of Meeting Proceedings
Total Cost of Cancer Care SECTION 3
otential critical disconnects in the cancer care continuum may result in additional costs to both the health care system in terms of potentially Phigher costs of treating late-stage cancers and to society as a whole due to premature mortality. Several presentations and numerous discussions were held during the Think Tank meeting to better understand the costs related to cancer care and the challenges of measuring these cost impacts. In this section, the key themes from these presentations and discussions are summarized. The section begins with a background on the cost domains, followed by a summary of the overall cost associated with cancer care, and fi nally an in-depth discussion of the challenges of measuring the costs related to cancer health disparities.
Economic Costs of Cancer Health Disparities: Summary of Meeting Proceedings 19
3.1 Overview of Cost Domains
Economic costs of cancer include all resources required and used to provide a service—and the value of foregone opportunities to use these resources for a different service. The economic costs of cancer care and control include a wide range of factors: expenditures for cancer health care services; costs associated with time and effort spent by patients and their families and by cancer treatment providers; and costs associated with lost productivity due to cancer-related disability and premature death. Disparities in cancer care may increase the costs for individuals, families, employers, governments, and society.
Health care costs can be divided into direct, indirect, and intangible costs.
■ Direct costs are related to expenditures for goods, services, and other resources used in the direct provision of a service. Both direct medical (e.g., cost of medications) and direct non-medical (e.g., paid child care) costs are categorized as direct costs.
TABLE 1 Specifi c Cost Elements Required for Measuring
Total Cost of Cancer Care
Core Direct Costs:
■ Screening
■ Hospitalization
■ Outpatient clinical care
■ Physician visits
■ Rehabilitation/ home health care
■ Prescription and non-prescription drugs
■ Medical devices (walkers, wheel chairs, etc.)
■ Nursing home/long-term care
■ Hospice care
Other Direct Costs:
■ Transportation to health care providers
■ Child care related to obtaining health care services
■ Special diets
■ Lodging for remote treatment facilities
Core Indirect Costs (impact on patient):
■ Reduced productivity
■ Job loss/Shift to lower-wage employment
■ Loss of promotion opportunities
■ Lost wages due to premature death
Other Related Indirect Costs (impact on family/friends):
■ Time lost from work and housekeeping by family members
or friends
■ Loss of volunteers/caregivers to the community
Intangible Costs:
■ Pain and suffering
■ Bereavement
■ Psycho-social impairment
■ Familial health
SOURCE: Gold et al., 199623 and Fryback et al., 2004.24
20 Economic Costs of Cancer Health Disparities: Summary of Meeting Proceedings
■ Indirect costs are generally resources related to days lost from work (i.e., loss of productivity). Medical or health-related indirect costs are generally broken down into morbidity (e.g., lost productivity due to work disability) and mortality (e.g., lost productivity due to premature death).
■ Intangible costs are those related to adverse health effects for which there are no market prices (e.g., reduction in quality of life due to physical pain, emotional problems, and lifestyle changes). A reduction of intangible costs does not free up resources that could be used to produce other goods and services. This makes it diffi cult to estimate the impact of these costs, which can also extend beyond the patient to relatives who experience grief, bitterness, or depression.21,22
Understandably, economic studies often focus only on direct and indirect costs due to the diffi culty in assessing intangible costs. In presenting direct and indirect costs, an additional distinction is often made in economic studies between costs primarily within the health care system (core costs) and costs outside of the health system (non-core costs). The costs used in assessing the cost of cancer are shown in Table 1.
3.2 Overall Cost of Cancer Care
Costs related to cancer care contribute signifi cantly to the overall health care costs in the United States.
Hence, reliable and timely estimation of cancer-specifi c costs can help to assess the following:25