Building a Bridge to Somewhere Better: Linking Health Care Research and Health Policy

Of all the ways to influence health policy, using research to inform change has the dual appeal of sounding both straightforward and rigorous. However, reality intrudes in many forms to disrupt the otherwise common-sense connection between what we know, what we consider as policy and what we are doing. Different languages (academic versus political), disparate timeframes (deliberate versus opportunistic), and contrasting priorities (most rigorous versus good enough) often make translating research into policy an exercise in frustration. Just as basic scientists and clinical researchers grapple with the challenge of bringing innovations from bench to bedside, health services researchers must wrestle to make their work most relevant to policy realities in communities near and far.

Yet, this is a worthy—and potentially rewarding—pursuit. Evidence for such a claim is reflected in this supplement, which includes an array of approaches to bridging the gap between research and policy taken by current scholars and alumni of the Robert Wood Johnson Foundation Clinical Scholars program (RWJFCSP). Traditionally focused on training physicians in rigorous methods in biostatistics, clinical epidemiology, and health services research, as well as in other fields such as social sciences, the RWJFCSP has purposefully shifted over the past decade to emphasize the value of actionable research. Whether partnering with community members to conduct participatory research, or with policy makers to ensure that timely and relevant questions are being addressed, scholars are encouraged to consider how the results of their research will lead to short-term change and long-term progress. These research efforts encompass a broad variety of research methods, including qualitative and quantitative approaches, local and national data, and cutting-edge work involving community-based participatory research, multi-level hierarchical analyses, and dynamic multi-state modeling.

Given its leadership in the realm of research and policy, the RWJFCSP is the source of the lead or senior authors of all of the articles in this volume; the authors are either current scholars or alumni of the program. Of note, these articles do not reflect the entire sum of policy-relevant research conducted by RWJFCSP-connected investigators. Rather, these articles comprise a fraction of those submitted for inclusion in this supplement, and policy-relevant research by these authors and others with connections to the RWJFCSP appear in other journals as well.

The authors addressed important barriers to bridging the gap between research and policy. First, new evidence informs policy if it makes it into the hands of individuals who have a stake in what the data is measuring, and who are engaged in acting on the findings. Even the most rigorously conducted, widely cited manuscripts will have only marginal impact unless brought to life by potential change agents. Second, the evidence being generated may not be relevant to community members or policy makers. That is, time-sensitive policy needs frequently exceed the reach of available research, which may be well designed and carefully conducted but fails to answer questions in ways that policymakers find useful. In this issue, authors addressed this challenge using a range of creative, yet rigorous approaches. They not only identified important patient and system-level barriers to health that should inform policy, but they also prospectively evaluated the potential impact of a new health policy before it is implemented—to identify potential gaps and quantify the relative impact of potential obstacles to successful implementation as well as unintended consequences.

Working in close partnership with community members, academic researchers can help to ensure that relevant research is integrated into policy discussions and solutions at local and national levels. For instance, Spatz et al. (2012) have taken a proactive role in translating research that demonstrated barriers to access to specialty care into policy-based solutions for those barriers. They describe the formation of a team in New Haven, which worked with Project Access to create a partnership of providers, advocates, and researchers, to identify and overcome challenges to obtaining specialty care. They describe the steps in forming this partnership, rooted in principles of community-based participatory research. They enrolled numerous specialty providers and involved several hospitals. The resulting collaborative was successful not only in obtaining funding, but also in modifying the internal policies of numerous stakeholders in a local health care system to facilitate access.

Similarly, Grudzen et al. (2012) describe their partnership between University of California-Los Angeles and the Los Angeles County Emergency Medical Services Agency. Their partnership arose from a foundation of research suggesting that emergency medical technicians frequently try to resuscitate all patients, even when there is very low likelihood of benefit and a legitimate risk of physical harm as well as violating patient preferences. To translate these concerns into actionable policy change, the team developed and implemented specific clinical indicators, related to issues such as asking family members about or honoring written and verbal do-not-resuscitate requests.

A different story of community is told by Pollack et al. (2012), who used a combination of Medicare data and cancer registry data to examine the distinct practice patterns of urologists. Perhaps without knowing it, patients become a part of a community when they receive care from a given physician or group of physicians, and Pollack et al. underscore how different practice patterns are in these communities—in ways that matter at aggregate levels to payers such as Medicare and affect utilization and outcomes for populations. Ultimately, such profound differences at the level of the physician and group level prompt questions about the potential value of standardizing clinical approaches in ways that incorporate comparative effectiveness evidence.

Articles in this issue from the RWJCSP alumni also illustrate evaluating and influencing policy at the national level. For example, value-based insurance design (VBID) is a relatively new idea, based upon the longstanding concept of price elasticity of demand, that is, that utilization of a service or product is inversely related to cost. Applied to health care for insured patients, services which are more effective can be incentivized by decreasing patients’ out-of-pocket costs, while less effective or less efficient interventions can be discouraged with higher costs faced by the patients themselves (e.g., through cost-sharing mechanisms such as higher copays or co-insurance). In work spanning the past decade, Fendrick et al. have demonstrated an effective approach to translating research into policy: sustained interaction with purchasers, policy makers, promoting, and explaining the concept. Their piece (Fendrick, Weiss, and Martin 2012) describes the efforts to enhance dissemination (by 2008 as many as 30 percent of employers used some form of VBID), culminating with incorporation of this concept into national policy by its inclusion in the Patient Protection and Affordable Care Act (PPACA).

Also at the national level, Downey et al. (2012) take an exhaustive look at changes in inpatient patient safety indicators from 1998 to 2009—an era with unprecedented policy and institutional attention to improving patient quality and safety. In their analysis of more than 69 million hospitalizations, Downey et al. do find measures of quality that improved over time but also find worsening measures. The latter may inform focused improvement efforts in the inpatient setting.

Using national data regarding childhood asthma, Coker, Kaplan, and Chung (2012) examine the ways in which insurance coverage and quality of asthma care are related, in terms of diagnosis and appropriateness of controller medications (e.g., inhaled corticosteroids). This approach has implications at the national policy level (regarding insurance coverage initiatives) and also at the clinical level (regarding practice patterns). They find that, for children with more severe asthma, coverage is not as strong a determinant of medication use as it is for children with milder forms of asthma. Stated another way, children with more severe asthma are more likely to already have coverage and be on appropriate medications than children with milder forms of the condition.

Similarly at the national level, but in a different policy arena, Tarini, Brooks, and Bundy (2012) examine a policy implemented in 2010 by the National Collegiate Athletic Association (NCAA), to screen all Division I athletes for sickle cell trait. Their policy-impact study—a mathematical model informed by the latest scientific and collegiate data—was prompted by new NCAA policy which runs counter to prior research about screening for sickle cell trait. The controversial policy, implemented in 2010, mandated screening for all student-athletes at Division 1 schools to identify sickle cell trait. While proponents of the program argue that it will identify athletes at risk for life-threatening events, critics charge that the program may lead to discrimination and little certain benefit. With mathematical models, Tarini et al. address the proponents and critics, to inform the NCAA with much of its own data about the wisdom of this program—the likelihood of seven lives saved over 10 years of screening.

Other RWJFCSP scholars and alumni present their work that attempts to inform policy prior to its implementation. The PPACA addresses affordability for health care, and has the potential to remove financial barriers to care for millions of American. In an excellent example of how research precede policy, Kullgren et al. (2012) assessed the prevalence of important barriers to care that will likely remain during after the expected implementation of the major coverage provisions of the PPACA in 2014. They found that about one in five Americans may experience nonfinancial barriers such as availability of providers and access to health care facilities. These findings reinforce the importance of tracking and improving access throughout implementation of various components of PPACA, and ensuring a population-based perspective is maintained in these evaluation strategies.

Hollingsworth et al. (2012) provide another example of a population-based perspective that can inform efforts to evaluate the feasibility and potential effectiveness of implementing a key area of health reform: the patient-centered medical home. Using the National Ambulatory Medical Care Survey, the authors assessed whether medical practices have adequate infrastructure to provide medical homes. They found that a substantial proportion of practices lacked important components such as patient registry/tracking, self-management support, care management, and performance improvement programs. These findings will have important implications for efforts to assure that medical homes live up to their promise.

A third example of evaluation-before-implementation comes from Jacobs, Fu, and Rathouz (2012), who evaluated the incremental value of changing from face-to-face and telephonic interpreting for Spanish-speaking patients to a video-interpretation platform in the emergency department setting. The rationale for their study was that face-to-face and telephonic interpretation methods offer contrasting merits in terms of timeliness of language-concordant care (telephone ≫ face-to-face) and nonverbal cues (face-to-face ≫ telephone), and that video-based interpretation might provide a valuable combination of strengths of the current modes and permit more timely and effective interpretation services for patients in emergent care settings. Their rigorous evaluation of patients presenting with chest or abdominal pain is based on measures of hospital admissions, time in the emergency department, and tests ordered compared with a concurrent control group. Although their study finds no advantage of video-based services versus telephone, such a “negative” study can be instructive for policymakers deciding about how to proceed and what norms to establish for interpreter services.

While care management is an important aspect of the medical home model, some vulnerable patients require significantly more intense management. In a thoughtful description of a chronic care management program for homeless adults, Basu et al. (2012) add an important economic perspective to the conversation. Previous work by this team demonstrated that a housing and care management program was associated with a substantial decrease in emergency room visits and hospital admissions. In this issue, the authors report that the intervention was associated with societal cost-savings, providing a strong rationale for future research as well as policy designed to further disseminate. This work also reinforces the contributions that human services programs can make to improving health.

Research can also generate important data to guide policy-making, providing a new perspective on important health issues. Bharmal et al. (2012) explored racial disparities in life expectancy, and found marked variation across states in the magnitude of racial differences in health. They found that in some states, smaller disparities were not due to better outcomes in black patients, but worse outcomes in white patients. Policies that bluntly aim to reduce disparities can miss opportunities not only to effectively improve health of minorities, but also to elevate the health status of all. It is particularly germane to consider disparities within and across states, given that health reform efforts will vary across states.

Another way in which RWJFCSP scholars and alumni bring novel voices to the effort to translate research into policy is by incorporating an interdisciplinary perspective to the research effort. Diamond et al. (2012) used a validated measure of non-English language proficiency from a nonmedical source—the U.S. Foreign Service, which has an obvious interest in measuring language proficiency of its staff. They re-evaluated the non-English proficiency of physicians in a large clinical organization who had previously self-reported their proficiency on less broadly vetted measures originating in the medical world alone. The investigators found that self-ratings of proficiency fell for many physicians using the widely validated Foreign Service measure, which has implications for how clinical institutions in the United States must address the language needs of an increasingly diverse patient population.

In aggregate, the studies included in this volume represent an illuminating cross-section of approaches designed to inform policy through research. As scholars and alumni of the RWJFCSP, the authors and their colleagues demonstrate an ability not just to “think outside of the box,” but to recognize when the conventional wisdom represented by the “box” itself may need to be dissolved and replaced with a different paradigm. In fact, one of the greatest insights we gained from the process of compiling this volume is that the work featured here is not just about research informing policy. Instead, it is clear that the best researchers in this arena inform policy and then immediately look for ways that policy will again prompt research. This “virtuous cycle” of research to policy and on again to research can be an effective means of advancing evidence-based policy in ways that improve patient and population health.