By now, the frequently cited statistics on health care spending and its growth have contributed to an all too familiar litany of fiscal and public policy concerns. U.S. health care spending represents 17.6% of the gross domestic product (GDP); it is projected to grow to a fifth of the GDP by 2020, and for the last three decades, its annual increase has exceeded that of the GDP by nearly three percentage points. The growth of U.S. health care spending and its per capita levels are considerably higher than those of other developed countries and do not appear to yield an advantage in terms of standard measures of health or system quality and performance. Characterizations of U.S. medical care as "flat of the curve," wasteful and uncoordinated, and prone to costly medical errors have become commonplace in policy discussions, leading to calls for evidence-based practice guidelines, improved information flows among providers, and changes in the incentives that drive our health care system. Finally, a large aging population spurred by a generation of baby boomers and accompanying the implementation of health reform threaten to place new demands on our health care system that will sorely test its current capabilities, as well as our ability to maintain future commitments to provide essential services to our most vulnerable citizens.
The fiscal imperative to control health care spending and to obtain "value" for our health care dollar is a pervasive concern within both the public and private sectors of our economy. In the public sector, federal spending on the Medicare and Medicaid programs alone represents over a fifth of the federal budget and the growth in these programs is frequently cited as a primary contributor to our large budget deficit. While public health care spending continues to compromise our ability to secure resources for other social needs, proposals to gain control of such spending have become the perennial political football. Growing Medicaid costs continue to plague the fiscal health of states and have contributed to cuts in other pressing social obligations such as educational funding. In the private sector, rising health care costs have fueled the increase in health insurance premiums, leading to sustained increases in employee premium contributions and cost-sharing, thus compromising the affordability and value of health coverage during a period of stagnant earnings and mounting income inequality. Arguably, such cost increases have contributed to the rise in the uninsured population and, along with the costs of other non-pecuniary benefits, have helped to make the use of short-term and tenuous employment contracts common, which has added to our worries about job insecurity.
With the Patient Protection and Affordable Care Act (ACA) to be close to full implementation in 2014, concern over health care spending has reached a critical juncture. Coverage expansions are expected to affect some 32 million people over the next decade; they will be accompanied by new spending in the form of federal tax credits and subsidies to defray premium costs and out-of-pocket spending incurred by people with moderate incomes. Additionally, public funding will be required to support the ACA's Medicaid expansions for those with incomes below 133% of the federal poverty line. Unless health care spending is reined in over this period, the ACA's promised contribution to deficit reduction may be compromised. Federal policymakers are well aware of this implication. In devising the ACA, Department of Health and Human Services Secretary Kathleen Sebelius has noted that every cost-containment idea advanced by health economists has been included in the reform law. But despite this assurance, uncertainty remains as to whether the law will effectively contribute to cost containment or further exacerbate the growth in health care spending.
This special issue of Inquiry focuses on the compelling imperative for cost containment and includes five papers to help inform the issue of how to bend the cost curve from its persistent upward trajectory. The contributions were selected from an open call for papers and subject to editorial scrutiny and peer review. The papers encompass issues specific to ACA initiatives as well as more general approaches toward controlling spending. These include the role of alternative delivery systems and payment incentives; the promise of information technology for costs and quality; the optimal use of "doughnut" holes as a cost-containment strategy; and the role of taxing health plans with generous benefits that are believed to contribute to excessive health care spending.
Overview of the Papers
In "Payment Incentives and Integrated Care Delivery: Levers for Health System Reform and Cost Containment," Holly Korda and Gloria N. Eldridge survey the literature on these two interrelated and essential components of health reform. Their paper provides a valuable description of the mechanisms through which these two policy levers can potentially achieve cost savings and enhance quality; while their assessment offers promise for certain initiatives, it also reveals some important gaps in our knowledge. In particular, Korda and Eldridge find evidence that patient-centered medical homes and delivery models that integrate primary and behavioral health care can yield cost reductions and improved quality. While they acknowledge that demonstration projects funded by the Centers for Medicare and Medicaid Services support such findings for accountable care organizations, they note there is insufficient corroborating evidence to draw broader conclusions. Their review also endorses the potential of bundled payment systems, especially those applied to an episode of illness, but they are less sanguine about the potential for pay-for-performance incentives. Finally, they recognize the potential role of information technology (IT) systems in supporting integrated delivery and achieving the delivery goals of payment incentives; they acknowledge the costs associated with IT implementation, and are cautious in drawing conclusions regarding the effectiveness of IT. The authors point to the need for more empirical evaluations of these measures as health reform evolves, both through the use of natural experiments and lessons from the field.
The question of the effectiveness of IT systems is considered in two papers: one, "Health Information Technology and Its Effects on Hospital Costs, Outcomes, and Patient Safety," by William Encinosa and Jaeyong Bae; and another, "The Impact of Accelerating Electronic Prescribing on Hospitals' Productivity Levels: Can Health Information Technology Bend the Curve?" by Eric W. Ford, Timothy R. Huerta, Mark A. Thompson, and Roland Patry. Using medical claims data from MarketScan and IT data from the American Hospital Association for 2007, Encinosa and Bae provide a rigorous econometric analysis of the role played by electronic medical records (EMRs) in assuring patient safety and reducing hospital costs, controlling for the fact that the propensity to adopt such technology is likely to vary across hospitals. Their analysis provides a rather striking insight. While they find little evidence that application of EMRs reduces the rate of patient safety events, they do find that EMRs yield significant reductions in patient deaths, readmissions, and costs once an event occurs. Their findings also indicate that this technology is quite cost effective and validates an important role for such hospital IT.
Ford and colleagues offer a somewhat different perspective on IT adoption. Using American Hospital Association data for the period 2005 through 2008 and focusing on computerized prescriber order entry (CPOE) systems for more than 1,800 hospitals, they apply data envelopment analysis to estimate total factor productivity "best practice" frontiers. They find that hospitals that implemented CPOE on a small scale and then moved to full implementation had the greatest positive gains in efficiency and productivity, while hospitals that implemented such technology systemwide within a given year experienced declines in these measures compared to other facilities. Their findings speak to a more conservative implementation approach to reap the benefits from IT adoption, one that allows for a process whereby medical staff can provide constructive feedback on their experience with such technology and obtain relevant educational opportunities before full implementation. Thus, their analysis provides a caution to those institutions seeking a payoff to quick IT adoption and also to public policy that would encourage such rapid implementation.
Although coinsurance in health insurance plans remains a standard tool for addressing problems of moral hazard, recent policy interest has shifted to the use of "doughnut holes," which expose enrollees to the full costs of medical care after a specific level of spending has been attained. Although use of doughnut holes is most prominent in Medicare Part D prescription drug coverage, this tool also appears as a quasi-deductible in a variety of consumer-driven health plans. Such plans require consumers to pay the full costs of coverage from tax-preferred accounts up to a pre-determined level of spending. As Richard C. van Kleef, Wynand P. M. M. van de Ven, and Rene C. J. A. van Vliet note in "Risk-Adjusting the Doughnut Hole to Improve Efficiency and Equity," a key issue for policymakers concerns the appropriate level of medical spending to locate the doughnut hole in order to obtain the strongest incentives for cost containment.
To inform this issue, the authors provide a theoretical discussion to demonstrate that the optimal location of the doughnut hole depends on an individual's probability of incurring health care expenditures, and hence upon the person's health status and whether he/she has a chronic illness. Based upon this analysis, the authors conclude that incentives for cost containment can be improved by implementing a risk-adjusted starting point for the doughnut hole that is located near the center of a condition-specific expenditure distribution. Applying risk-adjustment methodology to data from a Dutch health insurer, they also demonstrate that such a starting point also can achieve greater equity across health condition categories by reducing differences in out-of-pocket spending, While the authors clearly articulate the challenges to this approach, their analysis points to both gains in cost containment and equity from using a risk-adjusted approach to implement cost-sharing in the form of a doughnut hole.
Eliminating or capping the tax preference for employer-sponsored health insurance has been a long-standing proposal to contain health care costs. Proponents of this approach argue that the tax deductibility of employer contributions to health insurance creates incentives for workers to purchase more generous health benefits, thus contributing to moral hazard and inefficient spending, and also resulting in a substantial drain on federal revenues. In the policy debates leading up to the enactment of the ACA, this politically unpopular approach was rejected. In its place, despite considerable political opposition, the ACA will impose a 40% excise tax on insurers offering employment-based "Cadillac" health plans--plans whose values (premiums and any amounts deposited in tax-preferred accounts) are in excess of $27,500 for family coverage and $10,200 for single coverage. This provision is scheduled to go into effect in 2018.
In "What Can We Expect from the Cadillac Tax in 2018 and Beyond?" Bradley Herring and Lisa Korin Lentz explore the expected effects of this cost-containment initiative during its initial decade of implementation. Using data from the Kaiser Family Foundation-Health Research and Educational Trust's Employer Health Benefits Survey, they project the number and types of health plans likely to cross the value thresholds established by the ACA, the extent to which employers and employees are likely to select plans of lower value in response to the excise tax, and the implications of the tax for revenue enhancement and deficit reduction. Their findings reveal that a substantial number of health plans will likely be subject to this cost-containment measure over the next decade, and they explore the mechanisms likely to govern switching to plans of lower actuarial value as a result of the tax. Their revenue projections from direct and indirect effects of the "Cadillac" tax (i.e., the revenue attributable directly to the excise tax and indirectly from increases in federal income and payroll taxes resulting from employer and employee adjustments after the tax) indicate a cumulative increase in revenues of more than $931 billion over the first decade of implementations, representing a significant contribution to deficit reduction. In making these projections, the authors are careful to alert readers as to the sometimes tenuous nature of their estimates.
In conclusion, implementing effective health care cost containment represents perhaps the most important challenge for restoring our fiscal health, for attaining the promise of the ACA, and for addressing other social needs. We hope that you, our readers, find the issues raised by these papers, the evidence from their evaluations, and the frank recognition of the limitations of these tools and analyses to be informative and valuable in shaping your own thinking regarding the imperative for "bending the cost curve."
Alan C. Monheit, Ph.D.