Creating a Business Case for Quality Improvement Research: Expert Views: Workshop Summary (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

2 The Role of Federal Funders “Knowing is not enough; we must apply. Willing is not enough; we must do.” —Goethe P olicy makers recently have shown significant interest in try- ing to influence health care quality improvement, said Denise Cardo of the Centers for Disease Control and Prevention. The spectrum of policies varies widely among the federal, state, and, likely soon, consumer levels. The movement toward consumers is fueled by efforts to enhance transparency of outcomes and reim- bursement policies. Although most policies are well intentioned, many are not evidence based. An opportunity therefore exists for evidence to improve implemented policies. Panelists were asked by the planning committee to use the fol- lowing questions as a guide for their remarks: • What are the fundamental public policy features and objec- tives that will lead to a transformational improvement in the quality and economic viability of our health care system? • What role does transparency of outcomes and cost data play in driving quality improvement? How can quality improvement research better support these efforts? What public policy features might help to unleash its potential? • What public policy features are essential to help finance, promote, and reward relevant research into quality improvement sciences? 24 

THE ROLE OF FEDERAL FUNDERS 25 Agency for Healthcare Research and Quality The federal government plays numerous roles in health care. It funds more than half of U.S. health care spending through various departments, such as the Department of Health and Human Ser- vices, the Department of Defense, and the Department of Veterans Affairs, said Carolyn Clancy of the Agency for Healthcare Research and Quality (AHRQ). Parts of the government are responsible for the actual provision of care, while others are responsible for inform- ing health care decision makers. Although multiple factors influence health care quality and safety, policy initiatives must give organi- zations incentives to improve quality and share their experiences. Therefore, policy can be very helpful in shaping the environment in which care is delivered. For the past decade, quality improvement has been a movement for health care leaders, but only recently has it become a movement for those on the front line of care delivery. In 2003 the Medicare Modernization Act required hospitals to report on selected measures of health care quality in order to receive their full reimbursements. Since then, the number of measures has grown, and beginning in 2008 hospitals will report on patient experiences of care through the Hospital CAHPS survey, Clancy said. As recognized throughout the workshop, health care is a local enterprise. To build on this, Clancy introduced President Bush’s four cornerstones of value-driven health care: (1) transparency of quality standards, (2) transparency of price standards, (3) informa- tion technology interoperability, and (4) incentives for providing high-quality care. To support the coordination of these cornerstones, regional and local public–private collaborations, or chartered value exchanges, have been developed. In support of this effort, AHRQ is developing a learning network to produce public reports, foster pay for performance, and thereby improve quality. Value exchanges will have some access to data at physician group levels, aggregated to distribute data on physician performance. Evidence is used in making many types of policy decisions, from product approval to practice guidelines, from program financ- ing to priority setting. But, Clancy asked, can a case be made for strengthening quality improvement research? The field is relatively new and the current evidence base is mixed about what works to improve quality, but it is becoming better understood that different research designs are needed for different methods. Randomized controlled trials are clearly helpful at times, but may not always be the best method. A large opportunity exists to use other methods, such as quasi-experimental methods. Context is also important to 

26 CREATING A BUSINESS CASE FOR QIR capture, but it remains unclear how context should best be concep- tualized and measured. If the connection between research on qual- ity improvement and health care spending was better understood, policy makers could do a lot to help build the science, Clancy said, such as quickening the development of theories, better research designs, and setting of priorities. However, strengthening quality improvement research presents many challenges. First, the ambiguous nature of quality makes it difficult to understand. Second, the nature of funding for quality improvement research poses a barrier. Third, the discovery of inno- vation almost always seems to be valued more than the use of innovation, which may not be the correct view, Clancy said, for the health care system cannot necessarily handle all health care innova- tions. To face these challenges, AHRQ is supporting a number of activities, including programs such as the National Research Service Awards, which focuses in part on funding quality improvement research with explicit evaluation components for graduate and post- doctoral research. To garner more support for quality improvement research, researchers must market the successes of individual quality improve- ment interventions. Conservative estimates of cost savings from quality improvement (direct medical costs) exist (see Table 2-1), but they are just the tip of the iceberg because they do not include avoid- able sick days and other indirect costs, Clancy said. Communicat- TABLE 2-1 Cost Savings from Quality Improvement Topic QI Focus Cost Savings Cost Type Diabetes Ambulatory care $2.5 billion (2001) Hospital costs Hypertension Ambulatory care $292–$708 million Hospital costs Asthma Ambulatory care $600 million (2003) Hospital costs (pediatric) Waste Efficiency Up to $1 trillion National health expenditures Health care– Hospital care $5 billion (2000) Hospital and associated other infections NOTE: These cost savings can be found in the AHRQ Closing the Quality Gap series. QI = quality improvement. 

THE ROLE OF FEDERAL FUNDERS 27 ing the economic benefits of interventions is not in the forefront of researchers’ minds—a perspective that must change. In changing the health care system, it cannot merely be insisted that changes be made; instead, actions must be taken, such as imple- menting smarter quality metrics as the basis for payment incentives so that the wrong behaviors are not rewarded up front. There is a need to understand more evidence-based management approaches, and to adjust policies to support them. We are just starting to look at the whole environment and need to become better at considering evidence and quality improvement research, Clancy said. National Institutes of Health The National Institutes of Health (NIH) is primarily a medical and behavioral research agency, not a health policy or health policy research agency, said Barnett Kramer of the NIH, voicing his per- sonal opinions and not those of the federal agency. Although there is some overlap and support, quality improvement is not the main focus. As mentioned, there is a lack of coordination within the fed- eral government regarding quality improvement in health care. Paraphrasing the NIH mission statement, Kramer said the NIH’s mission is science in the pursuit of fundamental knowledge about the nature and behavior of living systems and the application of that knowledge to extend healthy life and reduce the burdens of illness and disability. Many goals relate to achieving this mission, each with its own constituency and relevant budget. The primary focus of the NIH is the development of basic knowledge and interventions to improve health. Less focus exists on optimizing the delivery of those interventions, which some would categorize as health services. Quality improvement is an obvious yet relatively small component of a larger mission. However, a substantial amount of NIH-funded research informs quality of care, Kramer said. For example, research studying the efficacy of screenings for prostate cancer and new tech- nologies provide data to determine whether these procedures should be implemented or whether the harms outweigh the benefits. Health services research has constituted 3 percent of the NIH’s annual budget, a percentage that has held relatively constant in the overall budget in recent years (see Table 2-2). The percentage of health services research that is devoted to quality improvement research is not specifically reported as a budget item. Of the 215 NIH study sections, only one focuses on health services and qual- ity improvement: Health Services Organization and Delivery. This section reviews approximately 270 applications per year, largely 

28 CREATING A BUSINESS CASE FOR QIR TABLE 2-2 Health Services Research as a Percentage of NIH Annual Budget by Fiscal Year (millions of dollars) FY 2003 FY 2004 FY 2005 FY 2006 FY 2007 FY 2008 (actual) (actual) (actual) (actual) (est.) (est.) HSR 873 887 940 929 921 920 Total NIH 27,066 27,887 28,495 28,461 28,578 28,858 budget % HSR 3% 3% 3% 3% 3% 3% in NIH budget NOTE: HSR = health services research; NIH = National Institutes of Health. focused on health services research such as community person- nel, economic issues, and utilization. Research on quality is only a minority of submitted applications, Kramer said. Health services researchers may apply to numerous agencies; this discussion focused specifically on applications to either the NIH or AHRQ. Research proposals may be divided among agencies in a variety of ways. Applications may be “preassigned,” where either AHRQ or the NIH is requested by an investigator. Applications may also be “reviewed and referred,” a method in which a division within the NIH assigns applications to either agency. Finally, every application for more than $300,000 in the area of health services research is automatically sent to the NIH, Kramer explained. Discussion Research Budget Clancy explained that resources in AHRQ’s budget for grant applications outside priority areas (e.g., patient safety, health infor- mation technology, care management/prevention, and comparative effectiveness) are severely limited; applications for the priority areas are capped at $300,000 (total cost) per year. Additionally, there are clear expectations that much of AHRQ’s budget should be invested in key areas such as patient safety and information technology. In response to a question about why the NIH’s health services research funding has stayed relatively flat, Kramer said that although 3 percent of the NIH budget (equating to approximately $920 mil- lion) is a substantial amount of money, about 80 percent of grants are 

THE ROLE OF FEDERAL FUNDERS 29 awarded to investigator-initiated awards. Kramer viewed the steady percentage of funding, especially in a time of budget restraints, as a sign of hope for quality improvement research. A question was asked about whether either agency would con- duct the same activities it does now if its budget was 3 times greater or if there were specific areas in which further investments would be made. Although current resources cannot fund certain areas, Clancy said, more needs to be done than continuously collecting examples of great work. Clancy would invest in information technology net- works, finding ways to allow information technology infrastructures to be reused. Another area needing attention is the extension of efforts to include vulnerable populations and institutions. Kramer noted that the NIH constantly thinks of expansion, citing the NIH Roadmap, which identified underfunded, crosscutting areas to help achieve the NIH mission. One idea from the Roadmap was the Clini- cal Translational Science Awards, which are granted to networks of interlocking academic health centers focused on translational science. Funders’ Roles in Research Clancy posed the question of whether funders of health care ser- vices should support the development of quality measures because such support potentially could be viewed as a conflict of interest. Agencies such as the Centers for Medicare & Medicaid Services are currently involved in both roles; it is unclear what the relationship should be because it has been shown to have both positive and nega- tive consequences. The right place for this nexus between research- ers and policy makers should be discussed, said Raynard Kington of the NIH. In response to a question about what roles the NIH and AHRQ should play in prioritizing research and the criteria for doing so, Clancy noted that resources in quality improvement should be pri- oritized with public input. Money should be invested where the biggest problems are. Only 10 percent or less should focus on emerg- ing challenges and innovations. Kramer agreed, adding that there must be a compromise; it is unclear where the line should be drawn because both play a role. Prioritization should not be simply top down. 

30 CREATING A BUSINESS CASE FOR QIR Transparency The recent focus on transparency has been both a hindrance and an ally, Clancy said. It has been a hindrance because it pro- motes thinking that quality improvement and implementation are not sciences. Transparency has been an ally in that it helps people understand what health care managers are confronting on a daily basis. However, without science, managers do not know where to start or which interventions to employ to improve quality. Good theories and frameworks to unify concepts in ways that fit together are missing. Although significant advances have been made, much more needs to be done. O’Neill noted that the focus should not be on management, but rather on leadership. Data Collection O’Neill cited a recent study that found that 47 percent of chil- dren do not receive medically indicated care (Mangione-Smith et al., 2007). This research, based on medical records, begs the questions of whether medical records are legitimate bases for generalized results and whether research findings need follow-up. In response, Kramer noted that data from medical records do indeed limit research because they only approximate what actually occurred. However, these records are critical to quality improvement research and health services research, which are dominated by retrospective looks at charts and medical records. Fields are always enhanced if prospectively collected information can be interjected, Kramer said. Clancy stated that what this says about the field of research is that knowledge—not application—is prized. Answering a question about initiatives for the secondary use of data, Clancy noted that partnerships with integrated delivery systems and physician practice networks have been developing, although common data definitions are still needed, both for research and for quality improvement. Relationship Between Quality Improvement and Research One audience member said his impression from the panel was that quality improvement has been divorced from research. If the pur- pose is to improve health, quality improvement research and clinical research could be viewed as a continuum, with quality improvement research conducted before clinical research begins. If so, should health care delivery research design trials to answer questions important to advancing biology? Kramer responded that the times and contexts in 

THE ROLE OF FEDERAL FUNDERS 31 which health care is practiced are changing and that therefore quality improvement should also be changing. One impediment is the belief among individual physicians that clinical judgment dominates. The notion that the two have been “divorced” is a depiction of perspec- tives from 10 to 20 years ago. Kramer now believes the fields are moving in the right direction, albeit slowly. Another person asked about the relationship between cost- effectiveness and quality of life. Kramer responded that the two are separate but connected. Although quality of life remains difficult to capture, increasing attention is being paid to quality of life and more economists and quality-of-life experts are being incorporated into study teams. The IOM Forum on the Science of Health Care Quality Improve- ment and Implementation should focus on the urgent need to build a science base for quality improvement, said Clancy, in response to a question about what this group can do to really change health care.