Transforming Insurance Coverage Into Quality Health Care: Title and subTitle BreakVoltage Drops From Potential to Delivered Quality

Americans often assert that the US health care system is the best in the world. T. S. Eliot might have responded, as he wrote in "The Hollow Men" in 1925,

Between the idea
And the reality,
Between the motion
And the act,
Falls the Shadow.

The stream of innovations that reach the US health care system holds enormous potential to improve the quality of care. Yet Americans continue to experience unacceptable disparities in the quality of care they receive—among different population groups, among different regions, and across different institutions and clinicians. This gap, between the capability of medical science to improve health and the care that people actually receive, is the "shadow" that falls between "the motion" and the act of health care.

Policy makers have not been blind to this shadow. The nation has made incremental improvements in financial access to health care by providing wider availability of health insurance coverage. For the most part, quality of care has been treated as a natural outgrowth of better coverage. Given educated licensed health professionals, public-spirited institutions, and a solid base of medical research to provide better diagnosis and treatment, the argument goes, the availability of insurance coverage should lead to high-quality care.

Just as voltage can be lost in an electrical system as the current passes through resistance, there are many voltage drops between the availability of insurance and the delivery of high-quality care and between the capacity to serve and the service that meets people's needs. This article explores the nature of these voltage drops and what is needed to close the gap between the idea and the reality of quality health care. Elimination of these voltage drops will require participation by purchasers, patients, and providers (providers includes clinicians and health care institutions; hereafter referred to as providers).

The cascade from insurance to quality care, summarized in Figure 1, includes 7 transitions in which the potential for quality care may be lost. These 7 transitions are the following:

Whether they are clinicians exercising their professional responsibility, institutions fulfilling their public responsibility, or policy makers leading and addressing the wishes of the people, health care leaders need to seek opportunities to overcome barriers to the delivery of high-quality care at each of these voltage drops.

People can and do get care—even high-quality care—without being insured. Nonetheless, being uninsured increases the likelihood that the opportunities to receive good care and to improve health will be missed. Lack of insurance is associated with inadequate care and poorer health outcomes,^{1 - 2} including higher rates of adverse events³ and higher mortality rates.^{4 - 5}

Experts estimate that more than 42 million Americans—roughly 1 in 6—are without health insurance.⁶ From Medicare and Medicaid in the 1960s through the Health Insurance Portability and Accountability Act and the State Children's Health Insurance Program of the 1990s, the United States has attacked the problem of insurance access. Yet during the past 2 decades, the problem of uninsurance has actually grown. Between 1977 and 1998, the proportion of nonelderly Americans without health insurance increased from 12.3% to 15.8%.^{6 - 7}

The barriers to obtaining health insurance are not evenly distributed. The Medical Expenditures Panel Survey shows that more than 20% of African Americans and nearly 32% of Hispanics lack health insurance, compared with 12% of whites.⁶ Of all uninsured, 23% are Hispanic and 16% are African American, while only 10% of the total population is Hispanic and 11% is African American.⁸ Men are less likely to be insured than women, and young adults are at a higher risk of lacking insurance than other age groups. Nearly 32% of adults aged 19 to 24 years are uninsured, compared with 12.9% aged 55 to 64 years.⁶

In the 1997 Academy Award–winning movie As Good As It Gets, Helen Hunt's character (a waitress in a diner) garnered applause across the nation as she erupted in frustration with her managed care plan. Although the frustrations she expressed reflect real concerns, what was most remarkable about her character was that her child had insurance. Of children in similar families with single working parents, only 55% have private insurance coverage.⁹ Employment-based insurance for workers in small establishments is even rarer; only 48% of all persons working in businesses with 10 or fewer employees are offered employment-related insurance.¹⁰ Other workers with especially low access to employment-based insurance include those younger than 25 years (51% are offered insurance), Hispanic workers (61%), and low-wage workers (43%).¹⁰

Thus, despite considerable effort to improve access to insurance, a substantial proportion of Americans lack even this first basic step in acquiring high-quality care.

Even when insurance is available, people who are eligible do not always enroll. Among those with access to employment-based insurance, only 89% are actually covered by a health care plan.¹⁰ Furthermore, the take-up rate has declined to 89%, down from 93% in 1987. One reason is that insurance can be costly, especially for low-wage workers.

A separate study found that the average monthly contribution for workers earning less than $7 per hour, in the least expensive plans they were offered, was $27 per month, compared with $17 per month for workers earning more than $15 per hour. The rate of enrollment among low-wage workers was 78%, compared with 89% for high-wage workers. The overall take-up rate in this study was 87%. Two thirds of the uninsured workers who had access to an employer-sponsored policy cited cost as the main reason they did not enroll.¹¹ Although nearly two thirds of the workers who turn down employer-sponsored insurance are covered through another family member,¹² there is no evidence that the decrease in enrollment is because of employees being covered more often by a family member's policy than in previous years. As with single coverage, the demand for dual coverage is sensitive to a couple's out-of-pocket premium costs and their wages.¹³ The majority (74%) of the uninsured say they do not have health insurance because it is too expensive. Only 19% say that they do not think they need insurance.⁸

Barriers to enrollment exist in the public sector as well. Older persons covered by Medicare who are poor are eligible to enroll in Medicaid, which covers their Medicare co-payments and some services that Medicare does not cover (such as outpatient prescription drugs). Yet 16% of poor elderly persons have no insurance other than Medicare.¹⁴ Similarly, of an estimated 11 million children younger than 18 years who lacked health insurance during all of the first half of 1996, 4.7 million were eligible for Medicaid but were not enrolled.¹⁵ Although roughly 2.7 million children were eligible for the State Children's Health Insurance Program in 1999 under state plans,^{16 - 17} only about half of that number had actually enrolled in the program by mid-year.¹⁸

Some of the gap between the numbers of eligible and enrolled individuals in public programs is because of differences in eligibility policies across states, such as maximum family income. But other barriers exist as well, ranging from the stigma of registering for a welfare program to literacy, language, and cultural barriers. Administrative barriers alone can be powerful deterrents. For instance, in a 21-state study of Medicaid coverage among families moving from welfare to work, researchers found that transitional Medicaid participation ranged from 4% in Idaho to 94% in Connecticut.¹⁹ Barriers to participation included lack of knowledge of the benefit and substantial documentation requirements for continued eligibility.

As some states with "buy-in" components to their public programs have found, cost can also be a barrier to enrollment in public insurance. In Oregon, for instance, low-income persons not otherwise eligible for Medicaid can enroll by paying premiums averaging only $11 per month.²⁰ Nonetheless, about 700 households per month are terminated from the program for nonpayment of premiums.²⁰

It is often easy to forget that being offered access to insurance is not the same as being insured. As these examples show, individuals have many reasons for not enrolling in a health insurance plan.

Of course, being insured does not guarantee access to all services and all clinicians and institution providers. Certain services may not be covered; certain physicians and hospitals may not be included among those participating in a plan or contracting with it; a provider may be unwilling to accept payment rates. Cost-sharing requirements may deter patients from seeking care. While these limitations may result from cost-cutting efforts to make insurance more affordable, collectively they represent another potential voltage drop in the conversion of available insurance into quality care.

Medicare, for example, was established as an insurance program primarily to cover hospital expenses, and it did not cover preventive services or outpatient drugs. Cervical cancer screening has been covered only since 1990, influenza vaccination since 1993, and colorectal cancer screening since 1998. Most outpatient pharmaceuticals are still excluded from the list of Medicare benefits, and 35% of elderly persons covered by Medicare have no source of prescription drug coverage.²¹ Average out-of-pocket prescription drug spending for Medicare beneficiaries in 2000 is projected to be $525 per year and $800 per year for those with no supplemental drug coverage.²²

Patient cost-sharing for covered services is a form of coverage policy, and lack of coverage for a service is tantamount to a 100% patient co-payment. Evidence suggests that lower cost-sharing leads to more medical care but increases the use of ineffective as well as effective services.^{23 - 24} Lower cost-sharing is associated with better health outcomes in low-income persons in poor health²⁵ and with greater use of preventive services.²⁶

Other cost controls also limit coverage and influence access to care. For example, when New Hampshire Medicaid instituted a 3-drug payment cap in the 1980s, use of several important medications was reduced, including insulin and diuretics, and nursing home admissions doubled for those using medications.^{27 - 28} Researchers have found increases in readmissions of children and adolescents following the institution of utilization management programs.²⁹

Recent public debates over coverage and its potential effect on access to high-quality care have highlighted 2 major aspects of this voltage drop. First, a number of discussions have centered on the coverage of particular technologies and services. Congress recently debated the issue of medical necessity as a basis for coverage denial, for instance, and mental health care advocates have decried the lack of parity between mental health coverage and coverage for other health care services.

A second set of policy debates has included controversies about which health professionals are covered. Sometimes the debate is over the absence of coverage for certain clinicians (eg, those unwilling to accept the plan's fee schedule or those a managed care plan perceives would have high utilization rates). In other instances, whole classes of providers may be excluded from coverage. As of 1997, for example, chiropractic care still was not covered by 15% of private fee-for-service plans, and routine dental care was covered by only one third of such plans.³⁰

Using scientific evidence to ground decisions has gained credence as a tool for addressing the often contentious issue of coverage (although interest in evidence of effectiveness does not necessarily lead to consistent coverage policies).³¹ The Health Care Financing Administration recently announced plans to obtain evidence-based advice on which services should be covered by Medicare to enhance the scientific rationale underlying coverage decisions.³² In addition, it has appointed expert panels to review the technologies and services that are being considered for coverage.³³

Because coverage issues are so frequently in the public eye, they are a well-recognized voltage drop on the way to quality. Less recognized is that coverage is synonymous with neither choice of the source of care nor access to that care. Nor does coverage guarantee that the service itself, even if received, will be of high quality.

The ability to choose among health care plans and providers with different prices, locations, covered services, and other characteristics is an important component of a market-driven health care system. Individuals and purchasers acting on their behalf should be able to exercise their preference for quality by choosing plans and providers that offer the best value—the highest quality care that, from their perspective, is worth the cost. Yet, in the current US health care system, most people do not have this choice or do not have the information that allows them to exercise it in a meaningful way.

Employers express interest in selecting health plans for their employees on the basis of overall value rather than only cost,³⁴ but most of them currently make decisions based principally on price.³⁵ Obstacles are partly attitudinal (eg, an overriding concern about cost reduction), but the lack of credible information on which to base quality purchasing decisions is also a barrier.³⁵

Individuals face an even larger voltage drop in exercising choice. Fewer than half (45%) of workers who have insurance through their employers report that they are offered more than 1 plan.³⁶ Many of those who do have a choice work for a relatively small number of large employers. Sixty percent of large employers offer a choice of plans,³⁷ but only 17% of all private employers who offered insurance offered more than 1 plan.³⁸

When employees do have a choice of plan, they historically have lacked information that would help them make an informed choice about the plan that would offer the best-quality care for their needs. In a 1996 survey, only 39% of respondents said they had seen information comparing health plans.³⁶

People value having a choice of physicians, hospitals, and other providers. Patients with a choice of physician, for example, are more satisfied with their care than those assigned a physician, even within the same set of physicians.³⁹ "Report cards" that summarize patient outcomes by practitioner have occasionally been used to focus attention on potential quality problems.⁴⁰ Current efforts, however, seem to arouse more interest among purchasers than among individuals trying to choose a physician. Existing methods for creating practitioner-specific report cards also suffer from difficulties in distinguishing among patients whose outcomes reflect different health risks.⁴¹

Markets need both information and choice to function at their best, and the lack of both hinders the health care market from identifying and supplying the quality of care that patients and purchasers want to buy. This voltage drop hampers market efficiency; it also impedes the health care system from achieving better health outcomes for patients overall. Addressing this voltage drop requires better information tools for choice, such as a version of the Consumer Assessment of Health Plans survey, which was sponsored by the Agency for Healthcare Research and Quality⁴² (now available to help consumers evaluate health plans) to provide similar information on providers. It also requires continued policy attention to the issue of choice itself—the extent to which a choice of providers and plans exists and the extent to which individuals and purchasers can exercise that choice.

Individuals can choose a plan and be covered for a service but still have difficulty finding a place where they can receive care regularly. People with a regular source of care are much more likely to receive preventive care services^{43 - 44} and may be more likely to receive care for acute and chronic conditions as well.⁴⁵

Insurance conveys only partial protection against an inconsistent source of care. More than 14% of nonelderly adults with private insurance lack a regular source of care, as do 13% of those with public insurance.⁴⁶ Reasons may be as simple as patient choice (people in good health may forgo contact with health care providers) or as complex as disruptions in available sources of care stemming from hospital closures, changes in managed care contracts, or an employer's shift from one insurer and provider network to another. Research suggests that people who change health insurance plans are more likely to delay seeking needed care.⁴⁷

Among racial and ethnic groups, Hispanic Americans are especially likely to lack a regular provider. The proportion of Hispanics without a regular source of care increased from 20% in 1986 to 30% in 1996.⁷ Declines in health insurance coverage explained only one fifth of this change.

Age and sex can have a striking influence on access to services. In a 1997 survey of adolescent girls by The Commonwealth Fund, one fourth of insured girls reported that they did not get needed care at least once, and 12% had no regular physician.⁴⁸

Geography results in limited services and available health care providers in some rural and inner-city areas. In 1998, for instance, more than 22 million rural Americans lived in designated Primary Care Health Professional Shortage Areas.⁴⁹ Differences in geographic location can translate directly into differences in the use of effective services. Women in rural areas are screened for cervical and breast cancer approximately 10% less often than women in urban areas, and urban women living in Health Professional Shortage Areas have lower screening rates.⁵⁰ Even insurance coverage cannot substantially improve the quality of care if people do not reach health services.

Individuals can be insured and have access to a usual source of primary care but still lack access to referral and specialty services that are important to high-quality care. When 7700 Californians enrolled in managed care plans were surveyed in 1997, 75% to 91% (depending on the medical problem) stated that they preferred going to their primary care physician first. They expected their personal physician to refer them to specialty care when needed, and most of them trusted their primary care physician to do so, but some (15%) did not.⁵¹ A recent international survey asking a more general question found that 39% of US patients said it was difficult to see a specialist when they needed to.⁵²

Administrative limitations on access to specialty care are a management tool that health plans use to restrict overuse, but concerns that these limitations also lead to underuse are frequently voiced. In a survey by the Kaiser Family Foundation, nearly 9 out of 10 physicians claimed that some of their patients had been denied coverage for services they had ordered during the past 2 years.⁵³ Sixty-five percent of denials for a mental health service and 50% of denials for referrals to specialists were rated by the physicians as resulting in a serious health decline for the patient.

Disparities in access to specialty services across different populations have become a source of public policy concern. Researchers have documented, for example, that African Americans undergo fewer coronary revascularization procedures than do whites,⁵⁴ and African Americans with early-stage lung cancer are less likely than white patients to undergo surgery (a factor that may explain much of the difference in lung cancer survival rates between these groups).⁵⁵ Disparities are not limited to differences across racial and ethnic groups. Elderly women with breast cancer, for instance, are less likely than younger women to undergo chemotherapy or radiation treatment.⁵⁶

Finding the appropriate balance between management tools to prevent overuse on the one hand and access to needed specialty care on the other hand requires better scientific evidence comparing the outcomes of different administrative tools. Eliminating disparities in access to services also requires better scientific evidence, but in this case, the need is for research that explores why these disparities occur. A study by Schulman et al,⁵⁷ for example, showed that physicians' decision making is one important factor and that decisions about whether to refer a patient are influenced by the race and sex of the patient, not just by clinical or economic factors. Such studies to understand why differences in referral rates occur across groups are the necessary first step in being able to address them.

Even when all of the voltage drops in achieving insurance coverage, choice, and access are addressed, evidence abounds that the care delivered is not as good as it can be.

Schuster and colleagues⁵⁸ highlighted the many well-documented examples of quality gaps in a detailed review of the evidence of the overuse, underuse, and misuse of health care in the United States. These include the following:

A report by the Institute of Medicine drew particular attention to the importance of medical errors to the quality gap. Extrapolating from findings of previous studies, the Institute of Medicine estimated that medical errors cause 44,000 to 98,000 deaths per year and would rank as at least the eighth leading cause of death in the United States.⁵⁹ Errors are gaining recognition as a means of targeting opportunities for quality improvement rather than as a negative consequence that requires assigning blame.

Moving away from blame does not imply moving away from accountability, especially for ensuring that at least a minimum level of quality is provided. Health care decision makers continue to rely on basic measures of quality to establish a minimum level below which a provider may not be allowed to deliver services, such as whether a health plan or hospital is accredited or a physician is licensed or board certified. Nonetheless, ensuring a minimum acceptable level of care does not ensure that the care is as good as it should or can be. These measures of accountability are like the sign at the entrance to a ride at the amusement park, "You must be taller than this sign to go on this ride." If you are taller, it does not matter whether you are 1 in or 2 ft taller than the qualifying level. To enhance the quality of care, health care providers must be able to identify the current outcomes of the patients they serve, introduce evidence-based interventions to improve care, and then assess whether outcomes did improve.

Identifying that a technology or service is effective is an important part of the evidence base for quality improvement interventions, but it is not enough to overcome problems of underuse, overuse, or misuse.^{60 - 61} For example, evidence has existed for years that administering corticosteroids to women in preterm labor reduced postnatal morbidity and mortality in the newborns. Nonetheless, confusion about which women would benefit and when to administer the therapy led to the underuse of this intervention. A focused quality improvement intervention to address this problem, centered on the dissemination of a consensus report on treatment, led to doubled use of the technology within 2 years.⁶²

A fundamental element in delivering quality health care is being able not only to perform a technical procedure or to do a clinical test well but also to know when to perform it and for whom. In the year 2000, no one would argue that performing tonsillectomies on asymptomatic children—once an accepted medical practice—is good quality care, no matter how technically perfect the procedure. Current practices are equally deserving of scrutiny. As the late sharp-tongued critic and literary wit Dorothy Parker is said to have quipped: "If it's not worth doing, it's not worth doing well."

In addition to performing the right service in the right way, health care quality also includes addressing the patient's needs. Failure to satisfy the patient's (and the family's) desire for information, confidentiality, comfort, and other expectations is another way in which a voltage drop can occur in the delivery of high-quality health care.^{42 ,63 - 65}

The gaps between the health care that is possible and the health care that is delivered represent failures to translate scientific advances and improvements in education and the organization of health care into better health. The voltage drops from making insurance available to getting high-quality care for those who need it—the right care, at the right time, for the right person, in the right way—can be addressed once they are understood.

Health care leaders cannot be complacent with just providing access to basic insurance coverage, important as that is. Nor can they be satisfied with just a bill of rights that ensures individual access to the care that insurance should cover. High-quality health care will be achieved only if the barriers at every voltage drop are addressed. This challenge can be met, and the voltage drops can be reduced or even eliminated, with attention and commitment to an active role by both the government and the private sector. They can be the light that will shine on "the Shadow" that T. S. Eliot described "between the idea and the reality, between the motion and the act." The President's Advisory Committee on Consumer Protection and Quality in the Health Care Industry emphasized this need for strong national leadership, better ways to measure quality, and more research-based evidence on both quality problems and strategies to address them.⁶⁶

The challenge of delivering quality care to all Americans will confront financial constraints. Sometimes better quality care actually costs less, such as eliminating unnecessary surgery or avoiding adverse drug events that result in prolonged hospitalizations and extended treatment. But other efforts to improve quality will involve expanding access to care or providing services to those who now are not receiving them. It is unavoidable that there will be tough policy choices between providing some services for more people and providing more services for some people. In either case, policy makers will seek better quality for the money being spent. This question of where to invest in quality raises fundamental philosophical (and political) questions about whether to seek allocative efficiency (spreading the available money around to cover a larger number of people to achieve fairness) or technical efficiency (spending money so that the greatest amount of positive health outcomes are achieved regardless of who receives them). Will policy makers seek distributive justice or utilitarian cost-effectiveness?

Recent national policy debates have mainly addressed the reduction of voltage drops in 3 areas: incremental additions to the rolls of the insured, such as the State Children's Health Insurance Program; adding benefits, such as prescription drug benefits for Medicare beneficiaries; and eliminating barriers to receiving desired care, such as access to emergency services and specialist care in bills designed to ensure patients' rights. In addition to determining what needs to be done to transform the potential of available insurance into the reality of quality care, policy debates also have focused on who ought to do it. There are legitimate and necessary roles for federal and state government, purchasers, the public as patients and consumers, health care organizations and institutions, and health professionals. These decision makers may be driven by their public responsibility for improving population health, their professional interest in providing better care, their financial interest in getting value for the money spent on health care, or their personal interest in improving individual health. In every case, they will need better information to understand where the gaps are and what measures can be taken to effectively close those gaps.

Information to help these decision makers is gradually emerging as research-based measures of quality are developed. Development of useful, reliable, understandable information becomes even more pressing as the nation moves toward developing national measures of health care quality. As the President's Quality Commission recognized, national tracking of quality and consistent improvement across the nation in the quality of care is not possible without "comparative information on health care quality [that] is valid, reliable, comprehensive, and widely available in the public domain."⁶⁶ Consistent with this goal, 1999 legislation reauthorizing the Agency for Healthcare Research and Quality requires the agency to report annually on national trends in the quality of health care provided to the American people.

We have done well to develop some valid measures—a few ways to test the voltage of quality that is being delivered to patients in the health care system. We now need to progress to more consistent and broadly applicable measures that allow us to track the quality of care and to focus on areas for improvement, especially for vulnerable populations. Alongside this endeavor, we can continue the efforts to improve the evidence basis for health care decision making and to develop more effective ways to translate that evidence into practice.

Preventing these voltage drops between insurance and quality of care will require a multipronged effort to ensure not only that insurance is available but also that it is taken, not only that appropriate services are covered but also that informed choices can be made, and not only that primary care and specialty services are accessible but also that quality care is delivered. Doing so will not only prevent voltage drops but also will transform the nation's health care. The nation's attention on providing health insurance for all Americans is essential, but insurance alone will not deliver high-quality care. Attention to access and quality of care will help close the gap.