We are living in an extraordinary time in health care. Amid unsustainable rising health care costs that did not deliver commensurate value or quality of care, a remarkable transformation in health care payment was put in motion over the past decade. A proliferation of new payment models is now underway, with an estimated 25 percent of payments across commercial, Medicaid, and Medicare Advantage plans meeting the Health Care Payment Learning and Action Network’s (HCP-LAN) definition of value-based alternative payment models (APM) in 2016. There are at least 840 accountable care organization (ACO) contracts across the nation (more than half established by the Centers for Medicare and Medicaid Services [CMS]), 11,000 primary care practice sites earning National Committee for Quality Assurance (NCQA) patient-centered medical home (PCMH) recognition, and a proliferation in the design and deployment of episode-based (bundled) payment models.
As reported by Catalyst for Payment Reform, efforts continue to track the breadth and depth of APMs. However, it is worth remembering that new payment models themselves do not inherently reduce costs or improve outcomes. They create a business context for providers to change the way they deliver care and an added incentive for payers to support a population health infrastructure. For example, in some PCMHs, providers are no longer paid for how many services they perform but for how well they manage to reduce health care costs and improve patient outcomes. Such adjustments in the traditional fee-for-service payment model have enabled primary care practices to adopt service delivery models that provide more coordinated and comprehensive care, improve individual and population health outcomes, and sometimes produce savings. Among the practices with NCQA PCMH recognition, at least 50 percent have pursued or participated in some type of APM with payers.
Expansive growth in new payment and service delivery models was stimulated by the Center for Medicare and Medicaid Innovation (authorized by the Affordable Care Act) through design and testing of innovative health care delivery and financing models. Another impetus has been the CMS demonstration programs, which promote changes in health care clinical and business models. The federal government created the signal and momentum, and the private sector responded. The 2016 HCP-LAN survey estimates that population-based accountability models account for 22 percent of commercial health care spending and 41 percent of spending by Medicare Advantage plans.
But we are now in a bit of a holding pattern. Signals from the federal government are less clear as the new administration establishes its priorities. Yet, there is little doubt that the health care system will deliver higher-value care if the incentives for providers are aligned with the goals of better care for individuals, better health for populations, and lower per capita costs. Providers and payers must continue to drive toward those goals.
In the short term, providers and payers must lead by learning and diffusing those approaches that are working best. We must emphasize rapid identification of the most effective payment models and delivery strategies. We need to develop quicker and broader ways of spreading these models. In fact, this is an opportune time to study and improve on the science of that process—the diffusion of innovations.
The growing field of implementation science, also known as implementation and dissemination research, offers exciting opportunities to improve the integration of new and promising models and interventions into real-world practice and policy. While the desire to narrow the knowledge-to-practice gap is universal, doing so has been particularly challenging in the health care delivery system, where the dynamics that govern the adoption of innovations are unusually complex. Many factors affect the diffusion process in health care—varying perceptions of any given innovation across multiple diverse stakeholders; different characteristics, actions, and interactions of those who are called upon to adopt innovations; and a range of contextual factors such as leadership, management, communications, and incentives.
For example, in a CMS study on bundled payment models, the effectiveness of episode-based payments has been interpreted quite differently depending on the perspective of the stakeholder, slowing the rate of diffusion. On a per-episode basis, bundled payments have been shown to reduce costs. On the other hand, the volume of episodes also has been shown to increase, raising questions about whether bundled payments can reduce total cost of care. Payers that have little leverage on procedural gatekeeping may be reluctant to implement bundled payments for elective procedures. Providers responsible for total cost of care, such as those in ACO contracts and provider-led health plans, may be better suited to implement bundled payment models because they serve as gatekeepers for their populations and can ensure that only needed procedures are performed. Moreover, complexities in payer ability to attribute savings to both proceduralists and generalists limits the simultaneous use of bundled payments and total cost of care models. Such complexities in context can meaningfully delay diffusion of models that have great promise.
As a result of the complexities of health care, taking what we know about improving health and putting it into practice successfully in one organization does not guarantee similar success in another organization. To better understand which strategies are most effective for diffusion of different types of innovation, we must ask some fundamental questions:
- When should we create positive incentives versus negative ones?
- How can we best engage providers and payers to change?
- Who can best purvey information about innovation?
- When do we need changes in health care policies and procedures?
- When do we need clinical guidelines? To whom should we look to drive change—the provider, the payer, or the policy maker—and under what circumstances?
- What is the role of organizational culture change?
- And what are the strengths and limitations of learning collaboratives?
Health care stakeholders who are frustrated with the pace of adoption of new models and interventions need support in understanding methods for spreading innovations and how to affect the pace and style of this spread in their own organizations. These activities can be viewed along a continuum that moves from the passive, untargeted, and unplanned spread of new practices (diffusion) to the more active spread of new practices to a target group using planned strategies (dissemination) and finally to the process of adoption, integration, and use of new practices within a particular setting (implementation).
As the science in the field is maturing, we are better able to address these issues. Early implementation research was largely empirically driven, produced mixed results, and was criticized for lacking a theoretical basis for understanding how and why implementation succeeds or fails. One prominent example is the AFFECT trial, which attempted to use administrative data feedback to improve hospital performance on key indicators of cardiac care. The intervention failed because it did not consider the use of opinion leaders as influencers or the mechanisms of action through which the intervention would be likely to succeed. Consequently, this effort to implement evidence-based practice in various settings was critically viewed as “an expensive version of trial and error.” The research was not guided by nor did it generate underlying universal knowledge for guiding future successful implementation strategies.
Today, implementation studies apply a variety of theoretical approaches from psychology, sociology, behavioral economics, and organizational theory to understand the behavior of health care professionals and other stakeholders as a key variable in the diffusion process. For example, the use of behavioral economics is now widely considered as a way to encourage specific healthy behaviors in patients. A wide array of approaches, ranging from positive reinforcement, loss aversion, social incentives, the use of lotteries, and punishment have been tested for behaviors such as weight loss, smoking cessation, and medication adherence. Increasingly, researchers are attempting to compare these behavioral economic strategies, using empirical evidence to determine the best way to broadly encourage behaviors that have been shown to improve health and reduce health care costs. Moreover, researchers are examining how different strategies affect population subgroups, to create targeted interventions to most efficiently and effectively influence behavior. By couching this research in a behavioral economics framework, researchers are enhancing the ability to scale interventions across populations while increasing the likelihood of a beneficial return on investment. Similarly, other implementation studies must identify factors that predict the likelihood of implementation success and inform the development of strategies to repeat successful implementations. We see implementation science researchers increasingly focusing on context to better understand how different strategies vary in effectiveness in different settings.
As health care constantly evolves, we need to move faster, and the science must map onto the rapidly changing landscape. Now recognized as a national priority, in 2012, the National Institutes of Health has created the National Center for Advancing Translational Sciences with the goal of transforming the translational science process so that new treatments and cures for disease can be delivered to patients faster. In addition, the Agency for Healthcare Research and Quality, the Department of Veterans Affairs, and the Patient-Centered Outcomes Research Institute sponsor ongoing funding opportunities for identifying strategies that support the adoption, implementation, sustainability, and ongoing improvement of evidence-based interventions and the optimal application of scientific knowledge to improve health and health care outcomes. The Center for Medicare and Medicaid Innovation institutionalizes learning and diffusion as a core feature in every payment model it entertains.
To further propel this crucial field forward, Health Affairs will publish a theme issue to be released in February 2018 that will provide a broad look at diffusion of innovations in the health care delivery system. The issue will explore the concept of diffusion and its theoretical underpinnings, how innovative ideas are spread, why diffusion is an important discipline and considered separately from innovation, and the influence of public policies and private-sector developments on diffusion. Additional empirical analyses of diffusion initiatives of varying scales and origins, from both the private and public sectors, and from the United States and abroad, will fill out the issue. In addition, Health Affairs will run a Blog series on diffusion of innovation, for which this article is the kick-off. We encourage the community to submit Blog posts on diffusion topics. Together, the theme issue and Blog series should help to foster dialog and advance evidence and knowledge about learning, diffusion, and implementation science.
Authors’ Note
William Shrank and Donna Keyser are employees of the UPMC Health Plan.
Diffusion Of Innovations In Health Care—Obtaining Evidence To Move Faster posted first on http://ift.tt/2lsdBiI
No comments:
Post a Comment