Improving quality of care in the United States is one of three pressing problems in health care today. Access for all and reductions in the rate of cost growth are the other two. But, as in most areas of health, these challenges are interrelated. Relationships between quality and cost have dominated the research and public literature of late. A brief historical review is instructive as we think through key interventions for the future.
In the mid 1960s, the physician Avedis Donabedian provided a model for evaluating and improving the health care system. The Donabedian Model indicated that the structure of delivery organizations influenced the processes of care employed, and that the processes of care influenced resulting outcomes for patients. Since then, multiple organizations and groups have focused on several areas in a sequential manner. Each of these activities has involved herculean efforts with sometimes modest success and with frequent negative or neutral results. A few comments follow about each method.
Enumerating optimal processes of care - This method can be fairly straightforward on a one-by-one basis, such as using beta blockers or aspirin after a heart attack. What becomes more complicated is the creation of composite measures of quality, and the jury is still out on the best way to achieve this. How should each of the processes be "weighted" in a composite index? The obvious answer is according to their impact on outcomes, but information on the individual effect of several processes on outcomes is hard to come by. Hence, individuals typically combine all the effects. The Centers for Medicare and Medicaid Services (CMS) used such an approach in the Health Quality Demonstration Project (HQI) .
Disseminating the use of known processes of care - This has been at the heart of the work of the Institute for Health Improvement (IHI) for decades. IHI differentiates between pure diffusion and active dissemination, which is their ultimate goal. In addition to the need for an improvement "campaign," the IHI emphasizes involvement of top leadership at all institutions, the evaluation of results by qualitative and quantitative approaches, an understanding of the variation in measures across sites and a determination of its causes, and a review of improvement over time.
Improving health care outcomes - While this is clearly the goal of the health care system, quantifying results is easier said than done. On a global level, recent work  has indicated that the United States has made great progress in improving health over time. Documenting this fact took years and considerable resources. However, most health care insurers, systems, providers, and patients are interested in more granular information, such as how outcomes for disease x or procedure y differ across delivery sites or providers. The data to obtain accurate measures of these outcomes tend to be insufficient in most situations, with the notable exception of cardiology and cardiac surgery.
Public reporting - Building on the approaches to measuring processes and outcomes of care, public reporting has been used in several states and by several insurers and private groups such as Castlight Health. One common approach to public reporting provides data on outcomes as related to volume, or so-called "volume-outcome" relationships.
Pay for performance - Originally, this was a promising approach to improving patient care. However, its use in the United Kingdom, one of the first implementation sites, proved ineffective when evaluated for its effect on process measures for several chronic conditions . One of the largest U.S. studies on the impact of pay for performance on 30-day mortality after an acute myocardial infarction showed no effect .
Bundled or budgeted payments - This method has been introduced as a way of linking cost-conscious behavior with improved outcomes. One recent study reviewed a prototypical example-the impact of the Alternative Quality Contract in Massachusetts on outcomes . It showed a modest effect on underlying growth in medical spending, while slightly improving quality of care for chronic care management, adult preventive care, and some aspects of pediatric care. In one study, a national program, Accountable Care Organizations, was associated with lower spending but not with consistently improved quality .
Where do these issues and data lead us? We have a long way to go before fully understanding how to improve quality in its multiple dimensions, while simultaneously restraining cost growth. Learning from multiple disciplines will help us on this journey. At the least, we need to emphasize the second imperative: "Promoting Novel Approaches to Process Improvement." We must understand what process measures to develop and emphasize, and then implement them in innovative ways that ensure quality of care, cost-effectiveness, and accessibility for all.
What process improvements do you think have really moved the needle on costs, quality or access?
2 Members of the U.S. Burden of Disease Collaborators. "The state of U.S. health, 1990-2010: Burden of diseases, injuries, and risk factors." JAMA 310 (2013): 591-608.
3 Campbell, S.M. and Others. "Effects of pay for performance on the quality of primary care in England." N Engl J Med 2009; 261: 368-78.
4 Jha, A.K. and Others. N Engl J Med 344 (2012): 1606-15.
5 Song, Z. and Others. "The 'alternative quality contract' based on a global budget lowered spending and improved quality." Health Aff 31 (2012): 1885-94.
6 McWilliams, J.M. and Others. "Changes in health care spending and quality for Medicare beneficiaries associated with a commercial ACO contract." JAMA 310 (2013): 829-36.