Advances in medical care in recent decades, such as new and improved surgical techniques and prescription drugs, have greatly benefited the health of many Americans. At the same time, the costs of medical care have risen greatly, and many Americans lack adequate health insurance coverage to meet their needs for care and for financial protection in the event of illness or injury. A new Supplemental Poverty Measure (SPM), which takes account of health insurance premiums and other out-of-pocket medical care costs, became available in November 2011. It showed that, were it not for the cost that families incurred for premiums and other medical expenses not covered by health insurance, the poverty rate in 2010 would have been 3 percentage points lower (Short, 2011:Table 3a).
The SPM was based on the work of a National Research Council (NRC) panel, which in 1995 issued a report, Measuring Poverty: A New Approach, that led to extensive research, culminating in the new measure as a supplement to the outdated official poverty measure. The NRC panel also recommended a separate measure of the economic risk to families because of inadequate health insurance coverage for needed medical care.
With the passage of the Affordable Care Act (ACA) and as new policies are implemented that seek to expand and improve health insurance coverage and to protect against the high costs of medical care relative to income, such a measure will be particularly important to inform policy. The goal of this report is to help move the field forward toward development of a measure of medical care economic risk. This introductory chapter provides historical background on the measurement of poverty in the United States and the role of medical care expenses, summarizes issues in accounting for
the financial burden of medical care, states the charge to the panel and describes the scope and limitations of the study, and outlines the organization of the report.
The U.S. poverty measure is an important indicator of economic well-being that influences public opinion and public policies. The official poverty thresholds are used to determine eligibility for many government assistance programs, and the measure plays a role in planning and evaluating government programs for low-income people and assessing the effectiveness of public policies in alleviating economic deprivation.1
The current official poverty measure was developed in the early 1960s by Mollie Orshansky, staff economist in the Social Security Administration (SSA). It was first used by the Office of Economic Opportunity and then adopted as an official statistic by the U.S. Bureau of the Budget (now the Office of Management and Budget, OMB) in 1969. SSA published the poverty measure until 1967, when the Census Bureau assumed the responsibility of publishing the measure on an annual basis using data from the Current Population Survey (CPS). Over the years, social and economic conditions changed, along with changes in public policies and an overall increase in the standard of living, making the measure less adequate for its intended uses.
The official poverty measure has weaknesses in both the definition of family resources and the specification of the thresholds. These thresholds are set at the same level across the country, without regard to geographic variations in the cost of living, and they have not been updated for real growth in the standard of living, but only to account for inflation. The definition of family resources does not include near-cash in-kind support from such sources as the Supplemental Nutrition Assistance Program (SNAP, formerly the Food Stamp Program), the school meals programs, and other programs for low-income populations. It also fails to deduct federal, state, and payroll taxes paid by families, expenses for work (child care and other work-related expenses), and child support payments to another household.
Most important for this report, the official poverty measure does not take account of the dramatic increases in medical care costs and spending since the measure was first adopted. At that time, national health care spending accounted for only 5 to 6 percent of gross domestic product compared with nearly 18 percent today.2 The rapid growth in medical care costs relative to income, particularly for middle- and low-income families,
1 This, and the next section, draws heavily on National Research Council (1995:Chapter 1).
increasingly competes for resources to cover other basic needs. Taxes, work-related expenses, and child support are not available to cover such basic needs as food, clothing, and shelter, and neither are medical care expenses for insurance premiums, copays, deductibles, or other out-of-pocket costs.
In response to a request from the U.S. Congress, in 1992 the Committee on National Statistics (CNSTAT) at the National Academy of Sciences/National Research Council (NAS/NRC)3 established the Panel on Poverty and Family Assistance: Concepts, Information Needs, and Measurement Methods to address the various concerns about the poverty measure as well as the related conceptual and methodological issues in establishing standards for welfare payments to needy families. The panel concluded that the current measure needs to be revised; it no longer accurately reflects the differences in the extent of economic poverty across population groups and geographic areas or over time.
In its 1995 report, Measuring Poverty: A New Approach, the CNSTAT panel proposed an approach that separates the measurement of economic poverty from the measurement of medical care needs and the adequacy of resources to meet those needs. The proposed concept for the poverty thresholds includes such budget categories as food and housing but not medical care. For consistency, the panel proposed that medical insurance benefits not be added to income and that out-of-pocket medical care expenses (including health insurance premiums) be subtracted from income as part of determining families’ disposable income that is available for nonmedical basic necessities (National Research Council, 1995:51-52). Because the proposed revised poverty measure would not directly address the availability of affordable medical care, the panel further recommended that the federal government develop a separate measure of medical care risk that would estimate the economic risk to families and individuals lacking adequate health insurance coverage (National Research Council, 1995:69).
The issuance of the CNSTAT report prompted numerous meetings at which policy analysts and researchers considered ways to implement the panel’s recommendations for a new and improved poverty measure. The Census Bureau and the Bureau of Labor Statistics (BLS) collaborated on extensive research to develop and evaluate experimental NRC-based poverty measures, which have been published on the Census Bureau’s website.4 These measures incorporate technical improvements to the proposed NRC measure; they also vary one or more aspects of the proposed measure when there was not agreement on the best implementation (e.g., including medical care premiums and other out-of-pocket expenditures in the threshold
3 The National Research Council is the operating arm of the National Academy of Sciences.
versus deducting such expenditures from resources). However, there was no movement to replace the official poverty measure.
In the late 2000s, there was renewed interest in revisiting the 1995 panel’s recommendations with the goal of agreeing on a revised poverty measure that would supplement rather than replace the official measure. The House Committee on Ways and Means developed draft legislation, introduced in 2008 and again in 2009 as the Measuring American Poverty Act of 2009, which incorporated the NRC recommendations; the official measure would have been termed the “historical measure.”
More recently, an Interagency Technical Working Group (ITWG) on Developing a Supplemental Poverty Measure, under the leadership of OMB, directed the Census Bureau in cooperation with BLS to calculate a new Supplemental Poverty Measure from the CPS Annual Social and Economic Supplement (ASEC) and to publish it concurrently with the official measure beginning in September 2011 (Interagency Technical Working Group, 2010).5 (Failure to obtain needed funding delayed the publication of the SPM until November 2011.)
This new supplemental measure adopts the NRC recommendation to deduct medical insurance and other out-of-pocket expenses from resources prior to determining poverty status. By design, it does not fully address the economic risk to the population in terms of the adequacy of their health insurance coverage to pay for their expected health care needs, which the CNSTAT panel proposed would be covered by a separate measure. However, such a measure has yet to be developed.
STATEMENT OF THE PROBLEM
As aptly stated by Meier and Wolfe (in Part III of this report), the challenge in poverty measurement with respect to medical care need and economic risk has not been identifying the problem, but rather determining the best methods to resolve it. Prior to the 1995 NRC report, research focused on a single measure of economic poverty that would account for medical care needs and resources. Yet achieving agreement on what would constitute a conceptually sound and operationally feasible approach proved stubbornly intractable. The 1995 panel observed (National Research Council, 1995:223):
The issue of how best to treat medical care needs and resources in the poverty measure had bedeviled analysts since the mid-1970s, when rapid growth in the Medicare and Medicaid programs (and in private health
5 The ITWG included representatives from BLS, the Census Bureau, the Council of Economic Advisers, the Department of Commerce, the Department of Health and Human Services, and OMB.
insurance) led to a concern that the official measure was overstating the extent of poverty among beneficiaries because it did not value their medical insurance benefits. Yet after almost two decades of experimentation, there is still no agreement on the best approach to use.
As noted by the panel (see National Research Council, 1995:224), two problems make it very difficult to arrive at a single solution that both achieves the necessary consistency between the threshold concept and the resource definition of a poverty measure and is feasible to implement. The first problem is that medical care benefits are not very fungible; they may free up resources to some extent, but by no means do they have the fungibility of, say, SNAP benefits. SNAP benefits are essentially interchangeable with money, both because virtually all households spend at least some money for food, so the receipt of SNAP benefits frees up money income for consumption of other goods and services, and because the maximum SNAP allowance is low enough that it is unlikely that households would receive more benefits than the amount they would otherwise choose to spend on food. Neither of these conditions holds for medical care benefits; not all families have medical care needs during a year, and, although medical care benefits for low-cost services (e.g., a prescription drug or a doctor visit) may free up money income for other consumption, the “extra” benefits received from insurance (or free care) to cover expensive services (e.g., surgery) are not likely to free up money income to the same degree. Moreover, individual and small group insurance premiums tend to increase with age and illness because older or disabled populations on average have higher levels of health care spending due to poor health.6 At the same time, with any cost-sharing, older and sicker people will have higher out-of-pocket spending for medical care even if they have exactly the same insurance policy as younger, healthier people. This means that simply adding the insurance value of health care services to families’ resources would make the sicker and the older population look “rich” when, in fact, they might have inadequate resources for food, clothing, and shelter.
The panel further noted (National Research Council, 1995:224-225) that any attempt to develop thresholds that appropriately recognize needs for medical care runs into the second problem: such needs are highly variable across the population, much more variable than needs for such items as food and housing. Everyone has a need to eat and be sheltered throughout the year, but some people may need no medical care at all, and others may need very expensive treatments. One would have to develop a large
6 Beginning in 2014, the Affordable Care Act will eliminate a number of techniques for adjusting or “rating” insurance premiums on the basis of such characteristics as preexisting medical conditions or health status (see http://101.communitycatalyst.org/aca_provisions/setting_premiums).
number of thresholds to reflect different levels of medical care need, thereby complicating the poverty measure. Moreover, the predictor variables used to develop the thresholds (e.g., age, self-reported health status) may not properly reflect an individual’s medical care needs during any one year: some people in a generally sicker group may not be sick that year and vice versa for people in a generally healthier group. As a result, it would be very easy to make an erroneous poverty classification.
Another issue is how to account for out-of-pocket medical care costs. Even groups with medical insurance coverage, such as the elderly, pay some of their medical expenses directly, and the dollar amounts for such expenses as health insurance premiums, deductibles, copayments, and payments for uncovered services can be high. Yet little thought had been given prior to the 1995 NRC report as to how to adjust the poverty thresholds or the family resource definition to appropriately account for these costs.
The publication of the 1995 NRC report stimulated extensive research on various elements of the NRC-proposed measure of economic poverty, but it prompted only a few studies on developing a separate measure of medical care economic risk—including those by Doyle (1997); Doyle, Beauregard, and Lamas (1993); Moon (1993); and Short and Banthin (1995). Since that time, hardly any more work has been done, despite increasingly high medical care costs and spending, including increases in insurance premiums and other out-of-pocket expenses, that have put families at increasing financial risk.
Recent preliminary estimates of the financial burden of medical care among the U.S. population based on National Health Interview Survey data collected from January 2011 through June 2011 show that, in the first 6 months of 2011, 20 percent of people—or 1 in 5—were in a family having problems paying medical bills in the past 12 months; 26 percent of people—or 1 in 4—were in a family paying their medical bills over an extended period of time; and almost 11 percent of people—or 1 in 10—were in a family that had medical bills they were unable to pay at all. Overall, 32 percent of people, rising to 41 percent of poor people and 46 percent of near-poor people, were in a family experiencing one or more of these kinds of problems in paying for medical care (Cohen, Gindi, and Kirzinger, 2012).
Renewed interest in a measure of medical care economic risk has come about from the enactment of the Patient Protection and Affordable Care Act of 2010 (ACA), which is designed to significantly extend health insurance coverage in the United States and reduce the financial burden of premiums and other out-of-pocket expenditures for low- and middle-income families. Its passage underlines the potential usefulness of a measure of medical care economic risk that could monitor the effects of various ACA provisions, as well as changes in other medical care programs such as Medicare, on the economic well-being of the U.S. population. The measure would provide
policy makers with a targeted indicator of the level of financial risk faced by Americans due to medical care costs.
THE PANEL STUDY
Contract Charge to the Panel
The U.S. Department of Health and Human Services (HHS) is responsible for carrying out the provisions of the ACA. To monitor the effectiveness of health care reform in providing coverage for low-income families and children, a new SPM became available to HHS in fall 2011; the new measure subtracts health insurance premiums and other out-of-pocket expenses for medical care from income in determining a family’s resources for basic needs (see Short, 2011). To the extent that provisions of the ACA or changes in other health care programs affect premiums and other out-of-pocket expenses, the SPM poverty rate will be higher or lower than otherwise. However, the SPM will not directly assess the extent to which population groups are likely to incur medical care needs that put them at financial risk.
HHS would also find useful a companion measure of medical care economic risk, which estimates the proportion of families and children who are at risk of incurring high out-of-pocket medical care expenses, including health insurance premiums, in relation to their resources, for monitoring the effectiveness of health care reform. Such a measure would enable HHS to answer such questions as which groups face a greater likelihood of economic insecurity due to lack of or inadequate health insurance coverage.
In fall 2010, the Office of the Assistant Secretary for Planning and Evaluation in HHS requested the NRC and the Institute of Medicine (IOM) to convene an ad hoc panel of experts to
organize, commission papers for, and conduct a public workshop to critically examine the state of the science in the development and implementation of a new measure of medical care risk as a companion measure to the new Supplemental Poverty Measure. An agenda for the workshop will be developed by the panel to examine retrospective and prospective measures of medical care risk, defined as the risk of incurring high out-of-pocket medical care expenses (including insurance premiums) relative to income. It will consider the variability of risk across populations and the vulnerability of population groups, including the insured, underinsured, and uninsured and those with chronic health conditions, acute but not catastrophic conditions, catastrophic conditions, and other relevant issues. Based on the workshop and its deliberations, the panel will prepare a report with findings and recommendations that will help the field to move forward to implement a new measure of medical care risk that will
be valuable for monitoring the implementation of health care reform. The report will include a summary of the workshop and commissioned papers.
As expressed by the sponsor, much work has been done on a new income poverty measure, the Supplemental Poverty Measure, but the medical care economic risk measure is a separate measure and needs to move forward.
In response to this request, the NRC’s Committee on National Statistics, in collaboration with the IOM’s Board on Health Care Services, appointed a panel of nine members representing a range of expertise related to the scope of the study.
Developing a measure of medical care economic risk presents many difficult issues relating to defining risk, resources, and financial burden. Such issues include
- considering basic concepts, such as prospective versus retrospective measures, the difference between measures of incurred financial burden and expected financial risk, and the unit of analysis (family or individual);
- categorizing and estimating health risks (acute, chronic, catastrophic) for population groups and the associated costs;
- establishing thresholds for medical care affordability;
- determining adequacy of health insurance benefit plans;
- determining if and how to adjust components of the measure for people living in different geographic areas or other factors;
- deciding on what to count as resources to meet medical care expenditure needs (specifically, whether and how to count assets in addition to income); and
- determining how best to achieve operational feasibility, data quality, and timeliness of the resulting medical care economic risk measure.
The panel executed its charge to “critically examine the state of the science in the development and implementation of a new measure of medical care economic risk as a companion measure to the new Supplemental Poverty Measure” through the conduct of a workshop, panel meetings, and background research. The panel’s goal was to advance the development of a measure to inform policy that is feasible to collect and estimate and that will monitor changes in medical care economic risk as health
care reform is implemented and other relevant public- and private-sector changes occur.
The panel met face to face three times. The first meeting focused on planning the workshop as called for in the study contract, including development of an agenda and identification of potential participants. To avail ourselves of expert and detailed analysis of key issues beyond the time and resources of our members and as called for in the study contract, the panel commissioned three background papers from experts in the subject areas, which appear in Part III:
- “Conceptual Framework for Measuring Medical Care Economic Risk” by Sarah Meier and Barbara Wolfe;
- “Incorporating Data on Assets into Measures of Financial Burdens of Health” by Jessica S. Banthin and Didem Bernard; and
- “An Assessment of Data Sources for Measuring Medical Care Economic Risk” by John L. Czajka.
The workshop on Developing a Measure of Medical Care Economic Risk was held on September 8, 2011, and a summary prepared by the rapporteur is included in Part II. The second meeting of the panel was held immediately following the workshop to deliberate on the workshop discussions and to reach agreement on a preliminary outline of the final report. The third and final meeting was devoted to reviewing the draft chapters and reaching consensus on the panel’s findings and recommendations.
Scope and Limitations
The scope of the study is complex, covering a wide range of issues from concepts and definitions to issues of thresholds and resources, sources of needed data, methods, implementation, and application in assessing program performance. Many other issues relevant to the broad subject areas of health and health care exist. Although the panel recognizes their importance, their discussion is beyond the scope of this study. Within the constraints of time and available resources, the panel did not address every issue but covered those areas specifically called for in the contract charge. For example, we addressed medical care and not all aspects of health and health care; we addressed issues of financial risk of medical care and not all medical care risks; and we focused on survey data rather than on modeling issues. The panel also did not examine the impact of various insurance plans on out-of-pocket spending or address the appropriate treatment of medical care benefits and costs in measures of inequality.
More broadly, time and resources did not permit the panel to carry out the original analysis that will be necessary to construct and refine spe-
cific measures of medical care economic risk and burden. The panel has endeavored to provide as much guidance as possible for needed research and implementation, acknowledging that there are many issues that can be resolved only on the basis of empirical work.
Regarding available survey data, the panel notes that the CPS ASEC, which is the basis for the official poverty measure and the SPM and the most feasible source for producing timely measures of medical care financial burden and risk, excludes institutionalized populations, most members of the armed forces, and the homeless. (This is true of most major federal household surveys.) This limitation places constraints on measuring medical care economic risk for two reasons: (1) the definition of the survey universe excludes respondents who are institutionalized residents at the time of the survey; and (2) the cross-sectional design does not capture transitions into or out of nursing homes and similar long-term care facilities.
To fully capture nursing home transitions and associated costs, one would need a longitudinal survey design that included both the institutionalized and noninstitutionalized components of the population, with a mortality follow-back instrument to fill in the information lost due to participant nonresponse after the time of death. With such data, prospective measures could be developed for the component of the population that was noninstitutionalized at the start of, say, a 1-year follow-up period, to capture an array of medical care and long-term care costs that are currently unmeasured. The downside of a longitudinal design is the time delay in getting a measure needed to monitor the implementation of a policy or program.
In one sense, it does not matter that data on transitions are lacking, because the official poverty measure and the SPM both exclude the institutionalized, so that people who move into nursing homes move out of the universe for poverty measurement. However, this is a particularly significant limitation for the measurement of prospective risk, in that the biggest health-related economic risk for many elderly must be excluded.
ORGANIZATION OF THE REPORT
The panel used three criteria to guide the development of the report and its recommendations. First, the subject areas examined must be relevant to and within the scope and purview of the panel’s contract charge. Second, the evidence and analysis should be sufficient to support and justify the panel’s findings, conclusions, and recommendations. Third, recommendations should be clearly stated and attainable at reasonable cost.
The report is organized in a manner responsive to the contract charge. Part I contains the panel’s review, conclusions, and recommendations. Parts II and III contain the resources obtained and used by the panel to assist
Following this introduction, Chapter 2 describes the conceptual difference between medical care economic burden due to actual out-of-pocket medical expenses and medical care economic risk, discusses why both measures may be needed to inform national and state policies and assess trends, and outlines why it is important to keep the measures conceptually distinct. It also recommends an approach to measuring burden. Chapter 3 addresses concepts of resources and what should be included in a measure of financial resources for assessing medical care economic risk and burden.
Chapter 4 focuses on developing the concept of medical care economic risk as distinct from burden. It considers various methods, including retrospective and prospective approaches, to constructing a measure of medical care economic risk and outlines the panel’s proposed approach.
Chapter 5 covers data sources for estimating the components of a medical care economic risk index. Finally, Chapter 6 highlights the panel’s key conclusions and recommendations for moving forward to develop and implement a measure of medical care economic risk.
Although the principal intent of this report is to address the specific concerns of the sponsor as defined in the contract charge, the panel hopes that the report will provide guidance to a wider audience responsible for the implementation of relevant policies and programs. The panel also hopes the report will provide the basis for further research on the broader issues of measuring the benefits and costs of medical care for the U.S. population.