Better U.S. Health Care at Lower Cost


ARNOLD MILSTEIN

HELEN DARLING

Better U.S. Health Care at Lower Cost

We know what steps must be taken to improve the performance of the health system. Now we must develop the political will.

In the United States, the amount of money spent on health care by all sources, including government, private employers, and individuals, is approximately $7,500 a year per person. In other advanced industrial nations, such as Germany, the bill is roughly one-third less. Yet scores on health care quality measures in the United States are not generally higher than in other wealthy countries and compare poorly on multiple measures. Nor is higher spending buying greater user satisfaction, as chronically ill U.S. patients, who are in most frequent need of care, are generally less satisfied with their care than are their counterparts in other wealthy countries.

This picture can be changed. As the Institute of Medicine’s Roundtable on Evidence-Based Medicine has found through a series of three workshops, there is substantial evidence that the United States can attain better health with less money. Compared with other wealthy nations, the United States’ higher levels of health care spending are primarily attributable to higher prices for products and services rather than to higher service volumes. Nonetheless, opportunities exist to lower both volume and unit price of services without jeopardizing quality. Some methods for cutting excess costs are incorporated into one or another of the health care reform plans that have been proposed by both political parties. But no plan takes full advantage of the range of cost-cutting tools and enabling public policies that the roundtable estimates would lower per capita health care spending by double-digit percentages while protecting or raising quality of care.

Though the need for change cannot be overstated, trends have been running in the wrong direction. Unsustainable growth in U.S. health care spending—growth in spending on health care in excess of the growth in gross domestic product—out-paced other industrialized nations by 30% from 2000 to 2006, without evidence of a proportionally higher health dividend. Such excess growth is crowding out other spending priorities of federal and state governments and of employers. For example, although education is key to maintaining the nation’s standard of living in an increasingly competitive world, average state spending on Medicaid recently eclipsed state spending on public education. In the private sector, rapid growth in health care spending is suppressing growth in wages, employment, and corporate global competitiveness.

Rising health care costs also wreck havoc among non-affluent Americans who do not qualify for Medicaid or assistance through the Children’s Health Insurance Program. A recent Kaiser Family Foundation poll found that 53% of respondents reported their family decreased use of medical care in the past 12 months because of cost concerns. In addition, 19% reported serious financial problems due to medical bills, 13% had depleted all or most of their savings, and 7% were unable to pay for basic necessities such as food, housing, or heat.

Fixing this situation will require the U.S. health industry to more rapidly develop and adopt innovations that improve value without lowering quality of care or slowing biomedical progress. In essence, it must become a learning health care system, drawing continuously on insights from outcomes research and internal organizational performance assessment to more rapidly improve. What new public policies would enable our health care system to meet the implied annual productivity goal of generating more health with fewer dollars?

Inventory of wasted spending

There are five broad categories of waste in health care: providing services that are unlikely to improve health, using inefficient methods to deliver useful services, charging noncompetitive prices for services and products, inducing or incurring excess administrative costs in the health care and health insurance sectors, and missing opportunities to lower net spending via illness and injury prevention. Collectively, these streams of embedded waste represent a double-digit percentage opportunity to reduce per capita health care spending while improving clinical outcomes and patients’ experience of their care. Estimates of savings from policies to address them are included in an upcoming roundtable report.

Among illustrative problems, it is estimated that more than 3 million preventable serious adverse events occur in hospitals annually, with over half attributable to hospital-acquired infections and adverse drug events. Conservatively estimated, avoiding preventable defects in hospital care can produce net annual savings of $16.6 billion.

Medical imaging is ripe for waste reduction. Although extremely beneficial in a growing number of clinical circumstances, imaging too often is used in cases where it’s likely to add no clinical value and to harm patients by radiation exposure—a fact rarely discussed with patients. As an illustration of the short-term gains possible, at Virginia Mason Medical Center, which operates in a market with a long-standing tradition of efficient use of health care services, physicians treating patients with back pain recently achieved a 30% reduction in MRI use while speeding patients’ recoveries.

As with imaging, almost all service categories are marked by excess use, as when laboratory tests are performed without a clinical rationale. Some of these tests lead to further tests, such as cardiac catheterizations, that carry substantial risk of serious complications. What causes excess use of services? Failure to rapidly access prior medical records is one prominent cause. For example, it is estimated that $8.2 billion in annual spending is due to duplicative testing in hospitals, most often because physicians cannot readily obtain prior test results. Another source is that hospitals in some areas have too many beds and too many affiliated medical specialists who, in turn, are more inclined to order services of unproven value in order to fill available capacity. This phenomenon is demonstrated in studies where large variations in service use relative to population size and illness occur among hospitals in the same metropolitan area.

Health service and product prices that are not determined by robust market competition cause substantial wasted spending. In one study, introducing competitive bidding for durable medical equipment, such as wheelchairs and oxygen equipment, lowered prices offered to Medicare by more than 25% for many of the products. Mergers among insurers, among hospitals, and among physician groups—a growing trend—more often than not boost prices due to monopoly or oligopoly pricing power. Noncompetitive pricing resulting from hospital mergers is now estimated to account for approximately 0.5% of annual health care spending.

Administrative waste imposes excess direct and indirect costs on health care consumers, clinicians, and health plans alike. Patients waste time in repetitious completion of paper forms or in waiting for doctors with poorly managed schedules. This, in turn, lowers U.S. workforce productivity. Health system productivity losses accrue from the excessive time that physicians and their staffs spend on valueless paperwork, much of it the result of a failure to standardize billing and insurance-related activities. These activities consume roughly 43 minutes a day per physician on average—more than three hours per week, or nearly three weeks per year—and the value of this time translates into approximately $31 billion per year nationwide. The amount of time that physicians and staff members spend on various administrative tasks results, in large measure, from requirements imposed by third-party payers, often insurance companies. But variation in payers’ requirements of providers has been shown to add little or nothing to health care value. Indeed, roughly $26 billion of the total spent annually on administrative costs is attributable to differences in payer billing rules that do not add any value, according to estimates by the Massachusetts General Hospital Physicians Organization.

These examples illustrate the diversity and magnitude of waste that could be trimmed without loss of health or reductions in the quality of patients’ experience of their care. Fortunately, there is an extensive inventory of tools available for trimming this waste.

Waste-trimming tools

Electronic health records (EHRs) can be both a waste-trimming tool and an enabler of other tools. If implemented successfully nationwide, EHRs can yield savings of $77 billion annually, while simultaneously improving health outcomes. An EHR provides an easily accessible view of a patient’s health information generated by all encounters in any care-delivery setting. Such information typically includes the patient’s demographics, past medical history, physical examination findings, progress notes, medications, immunizations, laboratory data, and radiology reports. Giving clinicians access to full information enables them to order and provide safer, faster, and better-coordinated care.

More important to perpetual gains in the efficiency of U.S. health care, EHRs also enable clinicians to apply the tools of systems engineering to continuously improve how they provide care on a routine basis. Systems engineering applications can lower annual U.S. health care spending by a conservatively estimated $62 billion over the next ten years and greatly improve the safety and quality of care. For example, 40 million people are hospitalized each year. Reengineering the discharge process, by better educating patients about what they need to do after leaving the hospital and by improving communication between inpatient and outpatient clinicians and at-home care-givers can dramatically reduce readmissions—yielding savings of nearly $400 per hospitalization. Exemplars such as Virginia Mason and Thedacare in Appleton, Wisconsin report average reductions of over 30% in the cost of each clinical service that they systematically re-engineer. Similarly, reengineering health insurer administrative activities by, for example, instituting electronic funds transfers to providers and standardizing payer credentialing of clinical providers could yield savings of $332 billion over the next decade.

Expanding the availability of palliative care for serious advanced illness and quality end of life care would save at least $6 billion annually. Such care brings patients, families, and treating physicians together to discuss pain control and other quality of life issues as well as the likely outcome from available treatment options. When patients and family members are given objective information about the likely benefits and risks of available options, they more often choose less invasive and less costly treatments, often avoiding great patient suffering from treatments offering little or no additional longevity.

What would happen if all health providers were strongly motivated to attain the levels of quality and cost that are now generally accepted as benchmarks of high performance? In one estimate, hospitals’ production cost per admission (a common standard of comparison) and mortality rates would both drop by approximately 15%. Two plausible methods of motivating such attainment have been demonstrated: tying the amount of consumer sharing of the cost of health insurance or of health care at time of service to the comparative cost-effectiveness of the health insurer or clinicians that consumers’ select; and tying the amount of payment to clinicians or clinician organizations to the comparative cost-effectiveness of the care that they deliver. An example of the latter method is bundled payment methods that are subject to meeting high quality of care standards. In bundling, a clinician or clinician organization such as a medical group or hospital would be paid an all-inclusive amount for treating a patient with a given illness or injury rather than being paid a fee for each service provided. Such a payment method shifts clinician focus from service volume to service value. According to one projection based on serving commercially insured patients, if hospitals would agree to accept bundled payments geared to high-value care benchmarks for 13 common types of treatment, such as year-long care for asthma, diabetes, and heart disease, spending might be reduced nationally by up to $167.5 billion. There is evidence that payment methods geared to service value rather than service volume can improve health outcomes as well. In a Medicare study of 225 hospitals that committed to this form of payment for patients with heart attacks, hospitals reduced mortality by an estimated 4,200 deaths and increased use of proven methods to prevent hospital-acquired infections to 92.6% from 69.3%.

Public policy opportunities

After baseline waste has been trimmed via methods such as those described above, continuous improvement in the health industry’s ability to deliver valuable health care services would then offer a perpetual means of improving health and bringing growth in health care spending into closer alignment with growth in the nation’s gross domestic product. To attain this vision, the nation will need a coordinated set of new policies across a range of fronts. Many of these policy options were described almost a decade ago in the Institute of Medicine’s pioneering report Crossing the Quality Chasm. But even as its recommendations found fertile soil in some quarters—for example, in the evolution of the National Quality Forum—most remain unimplemented. Three unexploited broad policy options are pivotal.

First, the federal government should strengthen antitrust policies to ensure that no health industry participants are able to opt for noncompetitive price increases and tepid annual gains in quality and cost-efficiency. Second, stronger policies are needed to promote comprehensive health-relevant societal changes. These include enabling chronically ill people to easily access condition-specific and treatment-specific performance comparisons of providers and treatment options; including nutrition education and health-promoting life styles in public schools; and assuring that people in all communities have safe, accessible places to play and exercise. The government should give special attention to reducing the burden of obesity on health care spending by prioritizing programs likely to reduce obesity. In the near term, the president should issue an executive order requiring that an obesity impact statement be developed whenever federal funding is being considered for a project that might significantly impact obesity. As with environmental impact statements, these documents could help encourage all recipients of federal funds to support obesity reduction.

However, the greatest opportunity to improve health system efficiency probably lies in the enactment of federal policies to harmonize the influence that the nation’s health care payers have on the health care decisions of patients and their clinicians. Five facets of harmonization are especially important:

Standardizing measurements of comparative performance. Units of useful performance comparisons include multi-component health care systems, hospitals, physician groups, individual clinician-led care teams, treatment options, and treatment delivery methods. Measurement efforts should include clinical outcomes, patients’ experience of care, and combined consumer and payer spending per treatment episode or per year of chronic illness care. Sweden knows how many of its citizens are able to walk without pain 5 years after hip-joint replacement by each hospital and surgeon. The United States doesn’t have such information for the vast majority of the care that it is buying.

Fortunately, national momentum to standardize clinical performance measurement across payers is rising, and collaboratives of multiple stakeholders are converging on standardized performance measurements for public reporting and performance improvement. Groups such as the National Quality Forum, the National Committee on Quality Assurance, and the National Priorities Partners have begun to act, but much faster progress is needed. Once performance measurements are trustworthy and easily accessible, all payers should use the same standardized set to assess providers’ performance.

To help drive evolution of clinical performance measurements, payers should be encouraged to test new measures. But when doing so, payers should fully disclose to consumers and their providers the specifications of the measure being assessed, the rationale behind the measure, and the expected duration of the test. These stipulations have been endorsed in a 2008 “patient charter” by groups such as the American Medical Association and AARP, as well as by multiple other stakeholders. If new measurements require providers to collect new data, the payer should in some cases offer providers an incentive as a temporary bridging step. Past experience suggests the benefits of this approach. Several years ago, when the Centers for Medicare and Medicaid Services tied public quality reporting to its hospital payment, there was an immediate positive hospital response, substantially advancing standardized comparisons of hospital performance. The resulting comparisons have been incorporated in Hospital Compare, a national performance comparison tool that is now widely used by clinicians, hospital managers and their boards, payers, and consumers.

Standardizing payer methods for administrative interactions with providers. Existing multi-stakeholder efforts in administrative simplification provide a solid foundation for standardizing payer interactions with health care providers. Promoting spread of these efforts, which ultimately can ease providers’ administrative burdens, is a ripe opportunity for reducing waste.

For example, the Committee on Operating Rules for Information Exchange, a collaboration of more than 100 industry stakeholders, has been developing and promulgating operating rules and national standards for electronically exchanging data that enable providers to access administrative information before or at the time of service. The types of data covered include such things as patient eligibility and benefits verification, patient financial liability for various services, patient deductibles, and co-pays. In addition, the Workgroup on Electronic Data Interchange, through its Strategic National Implementation Process, has defined standards for Health ID Cards, and several payers have implemented these standards. The payers have produced millions of magnetic stripe ID cards to enable electronic eligibility determination and provide accurate co-payment information at the point of care. However, adoption is lagging badly in physician offices where photocopying of magnetic strip ID cards remains a common practice. Also, the Council for Affordable Healthcare, a nonprofit alliance of health plans, has developed the Universal Provider Datasource as a Web-based electronic service for collecting provider data used in credentialing, claims processing, quality assurance, and emergency response, and in providing such member services as directories and referrals. Approximately 760,000 physicians and other health care professionals in more than 500 organizations now use the system. Providers can enter information free-of-charge into a central, secure database, then authorize health care organizations to access it, greatly reducing or eliminating redundant paperwork.

Such policies to speed transaction automation and electronic connectivity are essential to moving all stakeholders from today’s cumbersome and error-prone administrative processes to a world of standardized workflows that will enable more attention to the care of patients. Congress now needs to legislatively mandate a timetable by which all parties will adopt uniform standards to support more efficient administrative transactions.

Standardizing payment methods that give providers robust incentives to improve the value of the care they deliver. Experts widely agree that “quality-blind” fee-for-service payments encourage services that may have little or no likelihood of improving health and that can sometimes cause harm. Payer attempts to dissuade or withhold payment for such services in individual cases have often proved unsatisfactory, due to the sparseness of comparative effectiveness research and limitations on applying such research to specific patients. Promising solutions include providing higher payments for primary care, bundling payments into a single all-inclusive payment, having payers share with providers savings that result from more efficient care, and conditioning provider access to opportunities for higher payment on the formation by providers of better-organized forms of care delivery and management. In order not to repeat the managed care backlash of the 1990s, all such proposed provider payment reforms should include methods of safeguarding and/or improving quality of care.

One simple example of such a better-organized form of care are “medical homes,” in which health care is organized around the relationship between the patient and a single clinician-led team, and when appropriate, the patient’s family. A more complex form is an accountable care organization (ACO). A typical ACO might include a hospital, primary care physicians, specialists, and potentially other service providers. Services would still be billed on a fee-for-service basis, but the ACO’s clinicians would coordinate care for their shared patients with the goal of meeting targeted reductions in quality flaws and total annual spending per patient . Because all components of an ACO assume joint accountability for the value of care, they would share in any cost savings if quality of care also improves. An example is an organization such as Kaiser Permanente that integrates care delivery with health insurance, thereby transferring to clinicians accountability for improvements in patient health, customer service, and the total annual cost of patient care.

The power of such value-based provider payment methods in speeding clinical performance improvements and cost reductions hinges on their adoption by all or most payers. Absent such standardization, the strength of the signal to clinicians and other health industry participants is lost. The most important place to begin may be standardization of provider payment methods and improvement incentives used by Medicare, state Medicaid and Children’s Health Insurance Programs, and large commercial insurers, since together they provide the majority of the health industry’s revenue.

One way to entice care providers to support a shift to more performance-dependent payment methods would be to offer national tort reform to clinical service providers in exchange for their acceptance of a major revenue-neutral shift to performance-sensitive payment methods. This societal tradeoff also would serve to reduce “defensive medicine,” the ordering of tests, imaging, and follow-up visits solely to minimize accusations of not doing “everything possible” to protect patients’ health. It is estimated that at least $20 billion could be saved each year by implementing well-crafted tort reform that ensures faster and more reliable payments to patients suffering harm, especially when causal factors are difficult to pinpoint.

Standardizing payer incentives for patients to improve the value of the care they receive. Payers have a number of avenues for encouraging patients to seek high-value care. Providing patients with information on the comparative value of choices that they can make will be key since many patients are instinctively wary of “cheap” care. For example, payers can structure their reimbursements to reward patients who choose high-value providers, identified as those clinicians, hospitals and/or health systems that offer higher quality care at a relatively low total cost of care per episode of acute care or per year of chronic illness and preventive care. Payers also can structure their plans to encourage consumers to choose higher-value treatment options or to adhere to physician-recommended treatment. In one estimate, if “value-tiered” provider networks induced all consumers with commercial health insurance to select providers who rank favorably on both low total cost of care and quality, per capita spending on health care would decline by roughly 10%. The size and direction of such incentives would need to be tailored to fit different beneficiary populations, since large negative financial penalties for selecting a low value care option would be unfairly coercive for non-affluent patients. These include incentives for consumers to select a health insurance plan that includes only higher-value providers, to select higher-value providers participating in their health insurance plan, and to select higher-value treatment options such as generic drugs.

Coordinating methods for assisting patients and providers to improve health care value. Payers, especially private payers, should be given incentives to pool their efforts to assist patients and providers to improve the value of health care. Examples of possible approaches include:

Private and public payers and physicians in their networks should jointly communicate to consumers the value of having a personal physician who is accountable for coordinating care and approaching regional benchmark performance on measures of quality, service, and low total cost of care.

All payers should contribute to supporting joint obesity prevention efforts, given the evidence about the mounting economic and health burden of this national epidemic.

Payers should collaborate to ensure that all patients with advanced illness have access to and coverage for accredited palliative care programs in all communities.

Payers should converge on a common method to provide clinicians and smaller institutional providers with information on how to improve the value of their clinical services. This should include joint funding by all payers to support effort by the nation’s highest-value providers to coach their colleagues on how to replicate their success most rapidly.

Within federal privacy guidelines, payers should agree to standards for sharing with providers the identity of other providers involved in their patients’ care to enable providers to better coordinate the care they provide.

Building public-private partnerships

Although most health reform debate in health care spending reduction has focused on government-sponsored health benefits programs, more than half of all national health expenditures originate in the private sector. The private sector therefore has a central role to play in lowering per capita health care spending and raising quality of care. Private groups, especially employer or union-managed self-insured health plans, have the ability to make management decisions about health benefit plan policies without the rancor that public payers and large insurers often face. Anyone who doubts the seriousness of this problem for public payers need only examine the current public debate around health care reform. Private health benefits sponsors’ unique ability to rapidly test and adopt successful methods make them a natural leader in payer innovation. But the private sector needs help in more rapidly disseminating their successes. Medicare and Medicaid, because of their concentrated purchasing power, are uniquely positioned to accelerate the spread and impact of successful private sector innovations in purchasing. Now is the moment to legislate an explicit partnership that combines the strengths of both purchasing sectors in a better harmonized pursuit of their shared goal of more health for fewer dollars.

To accelerate such cross-payer harmonization, Congress should create safe harbors from antitrust challenges for employers, unions, health insurers, and providers who collaborate to attain one or more of the five key facets of harmonization described above. Congress also should direct the Centers for Medicare and Medicaid Services to develop explicit all-payer harmonization plans. If nonfederal payers are reluctant to collaborate, Congress should then consider inducements, such as changes in federal tax policy.

Given the complexity of the nation’s health care system and how many factors can influence policy success or failure, payers around the country should be given fairly broad latitude—at least initially—in their efforts to move toward greater harmonization. Payer harmonization carries risks and benefits, so the more the various parties can test diverse ideas on a modest scale, the better. Action should be expected first in the pooling of health insurance claims data to facilitate publicly reporting on standardized all-payer measurements of clinician and clinician organization performance on measurements of quality and total cost of care. Such information will also supply the information required to evaluate the effects of payer harmonization initiatives and pinpoint which harmonization methods generate the greatest net benefit under various circumstances. In 2007, Senators Judd Gregg (R-NH) and Hillary Clinton (D-NY) jointly introduced the Medicare Quality Enhancement Act, which detailed a set of public policies that can enable this key initial step. The Medicare Quality Enhancement Act would have made Medicare claims data available to enable consumers and clinicians to understand the relative total cost of care and quality of individual health care providers, while also safeguarding patient privacy. The timing for this bill, coming in the last year of a Congress with little appetite for significant changes, was 2 years premature. The concept has been supported in the current reform debate by a bipartisan collection of legislators including Senators Warner (DVA), Cornyn (R-TX), Specter (D-PA), Collins (R-ME), and Lieberman (I-CT). Public comparisons of the value of care by providers is the oxygen of value-improvement efforts.

With national health care reform under intense debate and with an unequivocal need for quick and dramatic action to control costs and improve quality, we face an unprecedented opportunity to accelerate the evolution of a learning health care system that attains ever higher levels of population health at much lower growth rates in health care spending. Compared with the potential for progress, current reform proposals from both political parties appear under-powered. Although fear of vilification by advocates of the status quo understandably inhibits legislative efforts to attain more health with much less money, the impending exhaustion of the Medicare Trust Fund and rising middle class anxiety about exposure to uninsurance may provide the voter pressure to overcome the inertia.

The IOM’s series of roundtables provided ample evidence that the biggest barrier to progress is not lack of effective waste-trimming tools. Rather, it is the vulnerability of elected officials to accusations of impairing a service that is both consciously and unconsciously equated with protection from death and suffering. We can and must navigate through this political minefield so that we can accelerate efforts to produce better U.S. health care at a lower cost.


Arnold Milstein is medical director of the Pacific Business Group on Health in San Francisco and Chief Physician at Mercer Health & Benefits. Helen Darling is president of the National Business Group on Health in Washington, DC.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>