1

INTRODUCTION AND BACKGROUND

Health services research provides the foundation for progress, effectiveness, and value in health care. Given the widening gap between what should be possible and what is achieved in health and health care, strengthening the pillars of the nation’s capacity to assess and improve health system performance is essential. For more than a half century, health services research has been poised to inform and guide decision-making in health and health care. Building on a historic base of certain path-breaking insight on how care delivery alters results, assessment of health services emerged as a field in the 1960s when federally-funded support for health care and construction of health care facilities grew rapidly, and has played a pivotal role in contributing to health policy and the delivery of health care services in the US. However, in the current policy environment, questions have been raised about the scope, scale, structure, and function of government support for health services research and, as a result, now is a critical time for the field to reflect on its past accomplishments; identify shortfalls, challenges, and future priorities; and investigate ways of organizing to effectively and efficiently address those challenges and priorities.

Health progress is shaped by advances in three dimensions: understanding the causes and processes of disease, disability, or injury (etiologies); developing approaches to ameliorate or eliminate the impact of those causes and processes (interventions); and determining the effectiveness, efficiency, and consequences of alternative interventions and strategies for their delivery (services). Health services research activities are anchored in the service delivery domain, but many of the analytic tools and findings are applicable to each, hence the overlap can be substantial (McGinnis, 2018).

Perhaps the first structured trial of alternative service intervention strategies demonstrates the bridging nature of etiologic, intervention, and service studies: the 1747 experiment by the Scottish naval physician James Lind, using empirically-based dietary approaches to reduce scurvy among sailors (Lind, 1753). Dividing the sailors into six groups receiving standard diets, and each supplemented by

one variable element, he observed that those receiving lemons and limes fared significantly better, even though the basis was unknown and Vitamin C was not identified for more than 150 years.

In the medical care setting, the best known early example of health services research dates to the Hungarian physician Ignaz Semmelweis who, while providing obstetrical care in 1847 in the Vienna General Hospital, proposed routine hand-washing with a chlorinated lime solution after observing that maternal deaths from fever were several times higher on wards for physician deliveries, relative to those for midwife deliveries (Semmelweis, 1861, 1983 English translation). Semmelweis based his proposal on the observations 30 years earlier of A.G. Labarraque, a French chemist, that such a solution could prevent the spread of infection from animal processing facilities and morgues (Labarraque, 1829). It also set the stage for later statistical work on sanitation in medical care, advanced by Florence Nightingale, the British mathematician and nurse who systematically shaped the delivery and assessment of nursing.

Over the nineteenth and early twentieth centuries, as understanding grew regarding the sources and nature of the spread of many infectious diseases, as well as the identification of groups at higher social and environmental risk, so did mandates for the capacity to develop the data and records infrastructure for tracking and studying the spread and behavior of the disease. Health services research began to take shape as a field of scientific inquiry in the United States in the 1960s, when federally funded support for payment for health care and construction of health care facilities grew rapidly. With these increased investments, interest and support for their assessment grew, as did some federally sponsored data capacities to provide a needed infrastructure.

At that time a relatively small group of academic researchers developed key conceptual approaches to studying issues involving the cost, quality, and accessibility of health care (Bindman, 2013), but the numbers began to accelerate, catalyzed by the attention from both the public and private sectors. In 1968, following a 1967 Congressional authorization, the Department of Health, Education, and Welfare established the first National Center for Health Services Research & Development. Shortly thereafter, the Robert Wood Johnson Foundation, established in 1971, made health services research a primary emphasis of its grant-making and field development. Over the past five decades, a substantial field has developed that, while providing critical insights into the costs and quality of personal health services, is the central source of information and insights on how our health system functions as a whole. As will be considered throughout the descriptions in this publication of the historical roots and evolving foci, health services research now extends far beyond the delivery of individual health care

to the assessment of how systems behave and interface to shape health status and outcomes.

The contributions of what we now call health services research have been of seminal importance to health progress on topics such as quality improvement and patient safety and in laying the groundwork for integrative progress in fields ranging from psychology and economics to pharmacoepidemiology, genetic counseling, and personalized medicine. Ironically, at a time in which appreciation has never been higher for both the need and potential from health services research, the political and financial support for sustenance and growth appear to be weakening. Although the 2018 Agency for Healthcare Research and Quality (AHRQ) appropriation increased by $10M since 2017, the AHRQ appropriation has declined by $37M since 2014 (Agency for Healthcare Research and Quality, 2018).

As the complexity of care and care delivery increases due to our aging population and the development of more intervention prospects, we need the capacity to embed effectiveness insights and evidence generation directly into the care delivery process. As the costs of care continue to increase beyond sustainability at both the institutional and societal levels, the need is for the real-time, rigorous assessment of costs and outcomes. In the face of the challenges and critical decision points, core stakeholders, with the vital interests of the nation in prominent focus, coordinated to convene a national meeting under the auspices of the National Academy of Medicine.

NATIONAL ACADEMY OF MEDICINE WORKSHOP

This publication presents a summary of the discussion at a meeting called for by the leaders of the field of health services research, planned by a steering group comprised of experts representing those field leaders, and sponsored by stakeholder organizations that have been its stewards—including AcademyHealth, the American Association of Colleges of Nursing, the American Board of Family Medicine, the American Society of Anesthesiologists, the Association of American Medical Colleges, the Federation of American Hospitals, and the Robert Wood Johnson Foundation. The impetus for the meeting was very much in the spirit of a field which, throughout its history, has engaged in study of its own internal dynamics and its relationships with the broader health care community and policy makers (Westfall et al. 2007, Pittman, 2010; Gold, 2016). Within the National Academy of Medicine (NAM), the NAM Leadership Consortium for a Value & Science-Driven Health System provided the coordinating capacity for planning and implementing the workshop, entitled Building the Evidence Base for Improving Health Care: Contributions, Opportunities, and Priorities and held on February 26-27, 2018, at the National Academy of Sciences Building in Washington, DC.

The meeting brought together many of the individuals and organizations that have helped to create, expand, and lead the field of health services research. Workshop presenters and other participants included a balanced and geographically diverse representation of evidence producers, evidence users, and funders of health services research from across the United States to ensure diverse perspectives. Appendix A contains the workshop participant list.

The workshop was designed to acknowledge the past accomplishments of health services research; identify shortfalls, challenges, and future priorities for the field; and investigate ways of organizing the field to address those challenges and future priorities. (Appendix B contains the workshop agenda.) It sought to reexamine the vision for health services research while also engaging in reality testing of the expectations and current status of the field. Box 1-1 presents the focus of the meeting, the core questions it was designed to answer, and the meeting’s intended outcomes.

POLICY CONTEXT

Opening remarks from the NAM and AcademyHealth, representing key organizers of this workshop, emphasized the discussion’s timeliness. In the current changing health care environment, the contributions of health services research have never been more important, NAM President Victor Dzau observed. Existing evidence needs to be widely implemented; the areas in greatest need of new evidence need to be identified; a strong data infrastructure needs to be built; critical issues such as the social determinants of health need to be investigated; the link between primary care and social and community services needs to be explored; and innovations in consumer-driven care and tiered networks need to be examined. Dzau pointed out that transformative approaches to improve patient and physician experiences and outcomes also need to be considered while the current fee-for-service system are giving way to population- and value-based payment models that change provider incentives. Novel data analytics are needed to target and improve care, reduce waste, improve patient outcomes, and yield innovations in evaluating cost and care.

Health services researchers, by studying the quality, accessibility, cost, and outcomes of health care, deliver the information for health care providers, insurers, government, and patients to assist in making the right decisions for better care. But Dzau also pointed to a worrisome softening of the support base, with, for example, the Administration’s fiscal year 2018 budget proposing folding the Agency for Healthcare Research and Quality (AHRQ) under the National Institutes of Health (NIH), which, if it led to reductions in both funds and

priority for the field “could deal a significant blow to the field of health services research.” Although Congress rejected this proposal and continued supporting AHRQ in its current state, a similar proposal was included in the 2019 budget. In concluding his comments, Dzau underscored the importance of working with the federal government, as well as with private philanthropies, to sustain the funding leadership necessary for the field.

Building on these themes, Lisa Simpson, President of AcademyHealth, observed that health services research is at an inflection point, with questions being raised about the scope, scale, structure, and function of government support for health services research. “Our shared mission requires a sustained and sustainable federal investment in health services research and data,” she said, “but, as in other areas of federal support, money is necessary . . . but not sufficient. We also need a functional and efficient structure for federally funded health services research that delivers on its promise of advancing knowledge, informing choices, and improving health and health care.”

Simpson evoked three broad principles from AcademyHealth. First, evidence is essential. The policies and practices affecting health and the performance of the health system should be informed by the best and most relevant evidence. What works for whom in what context? How can the answers be implemented at scale to achieve better health and better health care?

Second, the production of evidence should be a public good, and the development of the health services research workforce and a high-quality data and information infrastructure should be part of the overall federal investment in health research. Simpson noted that “Yes, the private sector needs to be at the table. But if all the evidence is created behind proprietary walls, what will happen? Relying purely on market forces for improvement is likely to perpetuate and deepen the very real disparities and inequities that we have in this country.”

Third, diverse perspectives lead to richer and more nuanced understanding of issues related to health and the performance of the health system. That is one of the reasons why this workshop is so valuable—because the diversity of views represented provide very important input to the continuing conversation. Simpson indicated that it is now time to step back and take a broader look at the federal infrastructure to push ourselves to assess whether and how we will be able to meet the data and evidence needs of policy makers, health care system leaders, patients, consumers, and communities now and in the future.

Finally, as J. Michael McGinnis, NAM Leonard D. Schaeffer Executive Officer, also said in his opening remarks, now is a time “in which our tools and

knowledge for application in health and health care have never been greater, in which expenditures have never been higher, and in which the gap between our accomplishments and our potential has never been clearer.” Perhaps the most pressing policy context relates to the economic unsustainability of the growth in health care costs, and relatedly, the inability to identify ways in which the nation can find effective ways of bridging what happens inside clinic doors to what happens outside that ultimately determines the value of our investment. Given that context, “we’ve gathered here to discuss solutions—and the science of bringing solutions to practice is through health services research.”

ORGANIZATION AND THEMES

The flow of this summary publication follows the meeting structure. Chapter 1 presents the introductory context, and chapter 2 examines the history, funding environment, and status of health services research. Since its formation in the 1960s, health services research has expanded substantially, but it still faces some of the same issues it faced early in its history. Many parts of the federal government fund health services research, but they typically ask different types of questions, and coordination among agencies is sometimes lacking. Foundations also have played an important role in supporting the field, again with a wide range of research interests.

Chapter 3 looks at the impact of health services research on decisions ranging from macro-level policy choices to micro-level implementation decisions. Health services research has made transformative contributions on multiple levels. For example, health services research has improved detection and minimization of health care mistakes in key clinical circumstances, led to the development of patient safety indicators, and informed national policy conversations. However, while health services research has been particularly impactful in motivating the policy changes mentioned, it has had more limited utility in defining service delivery changes that clearly demonstrate improved value—in part because these changes often require broader systemic and cultural reward system changes to be implemented and sustained, and partly because of a mismatch between traditional academic reward structures and the institutional reward systems involved in moving new knowledge to practice.

Chapter 4 considers emerging issues and approaches important to engage in better guiding health system performance. For example, predictive modeling and artificial intelligence have made it possible to target interventions to the people who are most likely to become high-need, high-cost patients. In addition, new ways of organizing, leading, and reforming health care systems, including

large-scale culture change and alternative payment models, can lead to better quality and outcomes while slowing the rate of growth in health care spending.

Chapter 5 examines emerging approaches to improving access to care and bridging health and social services. Ubiquitous forces shape the health of populations, though their effects are difficult to separate out in studies of health problems. Health services research could accelerate its evolving focus of attention beyond academic health centers and hospitals to integrated primary care, community health services, and public health. In the process, it could provide valuable new knowledge on benefits design, caring for patients with complex health needs, and the continuum and coordination of care.

Chapter 6 turns to the data infrastructure for health services research and related quality and care improvement activities. Large amounts of health care data exist and have been used by health services researchers, but many data still are unstructured and/or controlled by others and unavailable to researchers. Challenges include proprietary barriers, the sustainability of data sources, and the dissemination of data and the results derived from data. In addition, the currently available data have serious limitations and are often expensive to access. Novel analytic approaches and innovative data-gathering techniques (such as the use of smart phones) bear significant promise.

Chapter 7 draws on earlier discussions to describe potential research and infrastructure priorities for health services research. Priorities for the field of health services research are considered through the lenses of different stakeholders, including policy makers, clinicians, patient advocates, and payers. The discussion highlighted the importance of additional research on health care financing and the impact on patients, high-need patients, patient preferences regarding data sharing, provider burnout, provider consolidation, and understanding the social determinants of health.

Chapter 8 loops back across key points to identify appropriate mechanisms for organizing the field of health services research in the twenty-first century. A research agenda for the twenty-first century would include not only research topics but how that research will be applied, which constituents are involved, and the value proposition for each constituency. Shared goals could foster public investment in innovation, evaluation, and implementation of what is learned, and a governing structure for investments in health services research could help ensure that decisions are made in an efficient and coherent manner. In addition, the field needs to develop a communication strategy that articulates its value in terms of improving health, health care, and health policy. Transforming health services research will require ongoing involvement from the organizations represented at the meeting, not only on their own behalf, but as recruiters, motivators, and engagers of public and private stakeholders across the nation.