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.
6 Essential Priorities for a Robust CER Enterprise Abstract: This chapter presents the committeeâs findings and recommenda- tions for ensuring effective implementation of comparative effectiveness research (CER) and its translation into health care delivery. A short-term priority research agenda alone will not fulfill the potential of CER to im- prove the health of Americans and the quality of health care in the United States. The committee strongly recommends that Congress and the Secre- tary of Health and Human Services act to establish a sustainable strategy to coordinate government CER activity. The organizational and scientific challenges of CER are immense and the case for a strong, coordinating authority is compelling. Effective implementation of the CER agenda will involve collaboration among multiple government agencies and numer- ous professional disciplines and areas of expertise. The relevant areas of research encompass the complete continuum of health care services, all age groups, numerous disease conditions and health technologies, diverse health care settings, and the organization of health care delivery itself. No single U.S. research agency or organization possesses the breadth of exper- tise necessary to address this considerable scientific challenge. Four CER program priorities merit high-level attention and coordination: (1) mean- ingful participation of consumers, patients, and caregivers; (2) building of robust data and information systems as well as research and innovation in the methods of CER research; (3) development and support of a highly skilled CER workforce; and (4) vigorous support of research and efforts to translate CER knowledge into everyday clinical practice. Early in its deliberations, the Institute of Medicine (IOM) committee agreed that the nationâs investment in comparative effectiveness research 139
140 INITIAL NATIONAL PRIORITIES FOR CER (CER) should go beyond recommending the individual CER topics sug- gested in the previous chapter. A short-term priority research agenda alone will not fulfill the potential of CER to improve the health of Americans and the quality of health care in the United States. The most important priority of all should be the building of a broad and supportive infrastructure to carry out a sustainable national CER strategy. Congress and the Secretary of Health and Human Services (HHS) must take concerted steps to establish a robust CER enterprise. This chapter refers to this effort, its coordination, and its recommended tasks as the âCER Program.â Rather than develop a comprehensive blueprint, the committee fo- cused on four essential program priorities that the HHS Secretary should embrace: (1) meaningful consumer, patient, and caregiver participation (in addition to other stakeholders); (2) investing in building robust data and information systems and in strengthening the research infrastructure for conducting new prospective CER studies; (3) investing in development, deployment, and support of a highly skilled CER workforce; and (4) sup- porting a vigorous translational effort to help bring CER knowledge into everyday clinical decision making. The objective of this chapter is to describe these essential priorities. It begins with a section outlining the imperative for effective coordination of CER activities. The rest of the chapter reviews the other four program priorities listed above. See Chapter 4 for the committeeâs recommendations for CER priority setting. The Imperative for effective Coordination of the CER Enterprise The committee strongly agreed that Congress should direct the HHS Secretary to implement a sustainable strategy to coordinate government CER activity including the Agency for Healthcare Research and Quality (AHRQ), Centers for Disease Control and Prevention, Centers for Medicare & Medicaid Services (CMS), National Institutes of Health (NIH), Food and Drug Administration (FDA), Department of Defense (DOD), and Veterans Administration. Recommendation 5: The HHS Secretary should establish a mechanismâ such as a coordinating advisory bodyâwith the mandate to strategize, organize, monitor, evaluate, and report on the implementation and im- pact of the CER Program.
ESSENTIAL PRIORITIES 141 Organizational Challenges Broad Research Scope The organizational and scientific challenges of CER are immense and the case for a strong, coordinating authority is compelling. This reportâs recommended research topics encompass not only the complete continuum of health care services (prevention, early detection, diagnosis, treatment, rehabilitation, palliation, end-of-life care), but also the effective organiza- tion of health care delivery itself. The high priority CER topics also span all age groups, from infants to adolescents to young, middle age, and older adults; numerous disease conditions; health care technologies (drugs, imag- ing, surgery, devices); and diverse health care settings. And, the investigative means include an array of complex methods including randomized clinical trials and observational studies. No current research organization in the United States possesses the breadth of expertise necessary to address this considerable scientific chal- lenge. High-level coordination and funding authority across this broad front is of paramount importance. Scientific Rigor An essential component of CER is the study of representative popula- tions in real-world clinical settings. This demands a wide array of study de- signs including systematic reviews and meta-analysis, observational analytic methods, modeling, clinical trials, and others. The field must set uniform quality standards at each phase (i.e., priority setting, design and analysis of observational and experimental studies, interpretation and dissemination) and maintain a highly skilled workforce. Objectivity Objectivity will be central to the publicâs trust and confidence in the integrity of the CER Program. Conflict of interest and bias in clinical researchâpublished in even the most respected medical journalsâis well- documented (IOM, 2009b). Selective reporting or publication bias is com- mon. Positive findings are more likely to be published than negative results (Chan et al., 2004; Dickersin, 2005; Dickersin and Min, 1993; Rising et al., 2008; Turner et al., 2008). In addition, there have been significant in- stances in which leading journals have not sufficiently enforced disclosure â See Chapter 5.
142 INITIAL NATIONAL PRIORITIES FOR CER requirements for authors and reviewers (Schwartz et al., 2008; Weinfurt et al., 2008). CER is as vulnerable to bias and conflict of interest as any other area of medical research. The ultimate value of the CER enterprise will rest, in part, on vigilant attention to these issues. A 2009 IOM report, Conflict of Interest in Medical Research, Education, and Practice, recommends principles to inform the design of policies to identify, limit, and manage conflicts of interest in health care research. The committee urges that the CER Program be constituted and managed in accordance with the recom- mendations of this report. Public-Private Collaboration The U.S. health care system has substantial resourcesâboth public and privateâto contribute to the CER effort. These resources include the private health care organizations that provide care for potential enrollees in CER studies. At present, experience in developing and maintaining such collaborative relationships is very limited within the federal government. To lead this coordinated effort, the committee agreed, will require an organiza- tion that is highly, preferably solely, focused on achieving the goals of CER. A national program of CER must engage the publicâincluding all of the stakeholdersâat all levels of its organization if it is to fulfill its potential to improve health care outcomes and to reduce unnecessary health care costs, which are both urgent needs. Sustainability The CER Program needs sustained and predictable funding beyond the American Recovery and Reinvestment Act (ARRA) of 2009 (P.L. 111-5) to achieve its objectives. To ensure research activities that truly embrace the definition of CER, the ARRA fundsâand subsequent funding to sup- port CERâshould flow through a CER coordinating authority directly to grantees, through federal agencies, or both. MEANINGFUL CONSUMER, PATIENT, and Caregiver ENGAGEMENT In Chapter 4, the committee urged that consumers, patients, and care- givers be active participants in setting research priorities. In this chapter, the committee recommends that consumers also be integrally involved in â In this chapter, the term âconsumerâ is used to represent not only consumers, but also patients and their families and caregivers.
ESSENTIAL PRIORITIES 143 the governance of the CER Program, the framing of research questions and research protocols, peer review of systematic reviews and monitoring of trials, and interpreting and disseminating the results of CER studies to ensure that new knowledge improves everyday clinical practice. CER will not achieve its basic objectives unless it embracesâand acts uponâa patient-centered mindset. Centering on the patient is fundamental to high-quality health care (IOM, 2001). Patient-centered health care demands that CER be developed and applied with respect to each patientâs unique needs, beliefs, and values. There is strong evidence that many consumers want to be involved in deci- sion making about their care (Presidentâs Commission for the Study of Ethi- cal Problems in Medicine and Biomedical and Behavioral Research, 1982). Manyâbut not allâpatients expect to make their own decisions about diagnosis and treatments and look to their health providers for support in interpreting and assessing the available information (Deber et al., 1996; Degner and Russell, 1988; Guadagnoli and Ward, 1998; Mansell et al., 2000; Mazur and Hickam, 1997). Yet, even the most sophisticated health care consumers often struggle to find the information that is relevant to their specific health-related questions and particular clinical circumstances (IOM, 2008). Recommendation 6: The CER Program should fully involve consum- ers, patients, and caregivers in key aspects of CER, including strategic planning, priority setting, research proposal development, peer review, and dissemination. â¢ The CER Program should develop strategies to reach out to, engage, support, educate, and, as necessary, prepare consumers, patients, and caregivers for leadership roles in these activities. â¢ The CER Program should also encourage broad participation in CER in order to create a representative evidence base that could help identify health disparities and inform decisions by patients in special population groups. Consumersâ Role in Informing and Framing the Research Experts and consumers often have different perspectives on the ques- tions that research should answer. Clinicians and patients do not always consider the same factors when weighing the tradeoffs posed by important health care alternatives (Entwistle et al., 1998). To ensure that the fruits of CER support consumersâ health care decision making, the CER Program should focus on the questions and perspectives of patients as well as their health care providers.
144 INITIAL NATIONAL PRIORITIES FOR CER Many researchers, acknowledging the importance of consumersâ con- tribution to framing research questions, advocate for decision makers to participate directly in formulating research questions (IOM, 2008). Con- sumers can inform investigatorsâ decisions about which patient outcomes to measure, patient populations to study (including important subgroups and relevant comorbidities), and interventions to compare, among other issues (Andejeski et al., 2002a,b; Hubbard et al., 2008; Saunders et al., 2008). Diabetes researchers, for example, have reported that involving patients is particularly helpful at keeping their research relevant and applicable to real-world settings (Lindenmeyer et al., 2007). Researchers responding to a survey on consumer involvement in randomized controlled studies in the United Kingdom reported that involving consumers helped to refine research questions and make the trials more relevant to patientsâ needs (Hanley et al., 2001). Nevertheless, opportunities for public input into clinical research remain rare. In breast cancer, the involvement of consumers at all levels of decision making at the DOD Breast Cancer Research Program including vision setting, and peer and programmatic review, has proven valuable to the research process, resulted in an educated and engaged consumer force, and influenced clinical research beyond the DOD research program and beyond breast cancer (IOM, 1997). Cultivating Consumersâ, Patientsâ, and Caregiversâ Participation in CER Community-based participatory research refers to research that in- volves community members or recipients of interventions in all phases of the research process, starting with a research topic of importance to the community. Numerous researchers have advocated that community-based participatory research is key to improving the relevance of clinical research, especially research on health care disparities (Faridi et al., 2007; Fretheim et al., 2006; Jones and Wells, 2007; Minkler et al., 2003; Omenn, 1999; Zerhouni, 2005). Nevertheless, consumer participation in research is not the norm and there is no agreed-upon model for conducting community- based research effectively (Johnson et al., 2008). Researchers do not know how to meaningfully engage consumers in their work or to whom to turn for advice on a consumer representative. If scientists choose the consumer representatives, the representatives may not represent the consumer view- point. In addition, independent consumer groups face numerous challenges when their members want to contribute to research. The CER Program should identify best practices for consumer involvement and set standards for the key competencies required for consumer participation in CER. Consumers will likely need appropriate information and education about CER to contribute meaningfully (Hubbard et al., 2008; Saunders
ESSENTIAL PRIORITIES 145 et al., 2008). The CER Program must reach out first to engage consumers and then to support, educate, and, as necessary, prepare them for their roles. Several programs have already been developed for this purpose. For example, a special initiative of the U.S. Cochrane Center, Consumers United for Evidence-based Healthcare, has developed a web-based course to help consumers understand the fundamentals of evidence-based health care (United States Cochrane Center Consumers United for Evidence-based Healthcare, 2009). Other programs include the National Breast Cancer Coalitionâs Project LEAD Institute, a science education program for breast cancer advocates, and Quality Care Project LEAD; the DOD Breast Cancer Research Programâs peer review process, which involves consumer advo- cates; the FDAâs Office of Special Health Initiatives, which trains patient representatives to participate on Advisory Panels (FDA, 2009); and the United Kingdomâs national advisory group, which educates the public about involvement in research (INVOLVE, 2009). To achieve meaningful consumer participation, the CER program should incorporate the following: â¢ Substantial consumer representation in program governance â¢ Focus groups, forums, and citizen juries. Public meetings should be well-publicized and held at convenient times and locations â¢ Well-publicized web-based surveys (see Chapter 4 for how the com- mittee solicited public nominations of CER topics) â¢ Educational programs offered through public symposia and semi- nars. Active consumer participants should have formal training opportunities and be compensated for their time Public Trust Some members of the public have voiced concerns that CER research may lead to health care rationing and inappropriate limits on patientsâ treatment choices (Meier, 2009). Engaging consumers in CER, and building the case for the value of CER, could help improve the publicâs trust in the U.S. research enterprise, because the communication is expanded to be in- clusive, rather than exclusive, among the key decision makers (IOM, 2002). In fact, consumers may have the most credibility in conveying information about CER back to the general public and help in explaining health and health care delivery (Oliver et al., 2008). A public that is more informed about the processes and value of CER is likely to have greater enthusiasm and confidence in both the research â See http://www.stopbreastcancer.org/index.php?option=com_content&task=view&id=395 for further information on Project LEAD.
146 INITIAL NATIONAL PRIORITIES FOR CER and the research community (Academy of Medical Sciences, 2006) and may be more likely to participate in CER, either actively or passively as research subjects. The CER Program should, therefore, work to lower bar- riers to active public participation in planning research, such as the lack of adequate financial support to allow for travel and to compensate for the time required to participate (Staniszewska et al., 2007). Robust Data and INFORMATION Systems As noted earlier and described in greater detail in Chapter 2, CER comprises a broad spectrum of established and emerging research methods including systematic reviews of existing evidence, observational research, and experimental studies such as clinical trials (each described in this sec- tion). A critical first step in launching a comparative effectiveness study is to identify the most appropriate design for the type of research question being asked (IOM, 2008). Every study design has limitations and no single method is ideal for addressing questions of comparative effectiveness. Each study should be well-designed to ensure scientific rigor and minimal bias. Systematic Reviews Systematic reviews address a specific research question by identifying, selecting, assessing, and summarizing the existing body of evidence. Indi- vidual research studies often do not provide definitive answers to clinical effectiveness questions (IOM, 2008). If conducted properly, a systematic review should make obvious the gap between what is known about the effectiveness of an intervention and what clinicians and patients want to know. Thus, systematic reviews provide a central link between research evidence and clinical decision making. If the systematic review is both scientific and transparent, researchers and decision makers should be able to interpret the evidence, to know what is not known, and to describe the extent to which the evidence is applicable to clinical practice and particular patients. As such, systematic reviews are integral to framing research ques- tions for future study regarding comparative effectiveness. To date, the quality of systematic reviews has been variable and some published reviews have been unreliable. Criticisms include a confusing array of schemes for grading evidence in the literature, hierarchies of evi- dence that may not account for the true quality of studies, no disclosure of potential bias or conflict of interest, and a failure to use existing standards for reporting methods and results in systematic reviews. In Knowing What Works in Health Care, the IOM recommended that the HHS Secretary only fund systematic reviews that commit to and consistently meet evidence-
ESSENTIAL PRIORITIES 147 based, methodologic standards (IOM 2008). This principle should also be followed in HHS funding of CER studies. Clinical Trials Fundamental questions of comparative effectiveness often require head- to-head comparisons of alternative interventions using randomized assign- ment to the interventions to be compared. Randomized controlled trials (RCTs) are the gold standard for determining effectiveness because they minimize selection bias, that is, the likelihood that study participants will be given a treatment related to their prognosis such as comorbidities. RCTs have answered many important comparative effectiveness questions. The ALLHAT trial, for example, compared the benefits and harms of different forms of antihypertensive therapies (Furberg et al., 2002). Clinical trials, however, cannot address many comparative effectiveness questions because of cost, ethical considerations, or other issues. RCTs are expensive because they involve careful follow-up of study participants as well as multiple clinical centers and investigators and centralized data coordination. Ethical considerations preclude trials of many types of in- terventions. For example, a randomized study comparing prophylactic mastectomy to âwatchful waitingâ in women positive for BRCA1 is very unlikely. Smaller scale trials with small study populations conducted at a single site are often not representative of real-world clinical settings. This is not to say that small single center studies should never be done. For example, entirely new research questions should be addressed using observational studies to begin with, progressing to small scale trials and finally testing in the context of large scale trials and real-world settings. Studies of intervention effectiveness and prognosis often require years of follow-up (e.g., interventions for chronic diseases and interventions in children). As a result, such research is subject to high drop-out rates and missing data. The findings must be interpreted cautiously. Moreover, as time goes by, the technology being studied may change, its use may improve, or, in the case of medications, the indications for use may broaden (Kent and Hayward, 2007). Registration trials conducted for the purpose of FDA approval are unlikely to detect uncommon adverse effects because they typically involve relatively few subjects and often address short-term outcomes. The study âThe Medicare Improvements for Patients and Providers Act of 2008 (P.L. 110-275 Sec. 304) directed the HHS Secretary to contract with the IOM for the purpose of identifying such standards and reporting the results of this effort to Congress. This study is scheduled to begin the summer of 2009.
148 INITIAL NATIONAL PRIORITIES FOR CER population for FDA pre-approval and marketing trials is often younger and healthier than the target population of the health care intervention. Comparator interventions in these trials may not reflect the comparisons of interest to clinicians and patients because the comparator is often a placebo or an atypical dose of a competing drug. Increasingly, trialists are applying methods to adapt clinical trials to real-world conditions. Methods to recruit and follow patients efficiently and to adapt trial designs to accommodate changing technologies are being incorporated in the design of randomized trials (Berry, 2003; Godwin et al., 2003). These methodologies should be refined and applied to meet the need for stronger, more applicable comparative effectiveness trials. Observational Research Observational studies can address gaps in the evidence when a random- ized trial design is not practical. Observational research includes prospec- tive and retrospective cohort studies, case-control studies, and case series analyses. In observational studies, the researcher does not intervene in pa- tient care but observes the process of patient care and its outcomes as they occur in everyday life. Well-characterized cohort studies are particularly useful. In the Womenâs Health Initiative, for example, this method was used to identify predictors of disease, medication-related outcomes, and factors associated with health disparities in women ages 50 to 79 years old (National Heart, Lung, and Blood Institute, 2009). Case-control studies are useful for identifying risk factors for rare events such as deep venous thrombosis during long-distance travel or harm from interventions (Aryal and Al-khaffaf, 2006). Observational studies are typically most appropriate for answering questions related to prognosis, diagnostic accuracy, incidence, prevalence, and etiology (Chou and Helfand, 2005; Tatsioni et al., 2005). They have the potential to address gaps in randomized trial evidence by including larger, more representative populations to identify rare or long-term adverse effects. Observational studies that link process of care datasets (such as admin- istrative claims data) to outcomes datasets (such as national death indexes) provide excellent opportunities to study both health services utilization and health outcomes, as discussed in the next section. Despite their potential advantages,Â however, observational studies are more subject to bias than randomized trials,Â and the decision to rely on data from observational studies must be weighed against the possibility of misleading results. The main form of bias (selection bias) occurs when the factors causing a person to experience the intervention are associated with the patientâs prognosis.
ESSENTIAL PRIORITIES 149 Innovation Is Needed in CER Methods Recommendation 7: The CER Program should devote sufficient re- sources to research and innovation in the methods of CER, including the development of methodological guidance for CER study design such as the appropriate use of observational data and more informa- tive, practical, and efficient clinical trials. There is a significant need for new and better research methods for studying comparative effectiveness (IOM, 2007; McClellan and Benner, 2009; Rawlins, 2008; Tunis, 2009). Current study designs, both experimen- tal and nonexperimental, must be further refined if CER is to be scientifically valid, efficient, and credible. In systematic reviews, for example, research is needed on how to identify and use evidence from observational studies on intervention effectiveness, and also on how to assess a heterogeneous body of evidence (IOM, 2008). New analytic techniques are needed to evaluate the effects of bias due to confounding when assessing comparative effec- tiveness using large observational datasets. Many fundamental questions of comparative effectiveness relate to small but clinically important differences in treatment effects that cannot be detected by current nonexperimental methods (Tunis, 2009). Clinical trials will always be essential to CER, but more efficient, larger, simpler, and pragmatic designs are needed. The Potential of Existing Data CER may also draw from analyses of existing data, such as that held by payers, health care delivery systems, and electronic health records. ARRAâs $40 billion support for advancing health information technology and implementing an interoperable electronic health record system with compatible data definitions and formats can help make these ambitious aspirations a reality (Office of National Coordinator for Health Informa- tion Technology, 2009). Claims data from large national insurers, electronic health records maintained by large integrated health systems, data collected through practice-based research networks, and patient registry data hold tremen- dous potential for CER. Harnessing these sources of existing data could markedly enhance the timeliness and value of CER. Existing data sources can be used for many research purposes: to study prognosis, risks and harms, and etiology of disease (Cupples et al., 1988); to analyze trends over time and capture long-term outcomes (Fung et al., 2004); to examine the causes of geographic variation (Wennberg and Fisher, 2008); to analyze racial and ethnic disparities in both access to and outcomes of health care (Peterson and Yancy, 2009); to study low prevalence conditions (many
150 INITIAL NATIONAL PRIORITIES FOR CER occurring in pediatric populations) (Merlini et al., 2008); to assess clinical effectiveness in populations and subpopulations, such as minority groups and people with comorbidities (Bell et al., 2009); and to generate hypoth- eses for experimental research (de Simone et al., 2009). Such data sources can also provide efficient sampling frames for recruitment to experimental studies, such as large practical individual-level RCTs, cluster randomized trials, and other prospective studies (Sabin et al., 2008). Researchers can use existing data from larger clinical populations to assess whether the benefits of interventions suggested in smaller clinical trials persist in the broader populations to which treatments are applied in practice. For example, non-Whites have been underrepresented in clinical research compared to white Americans (Braunstein et al., 2008; Brown et al., 2000; Farmer et al., 2007; Giuliano et al., 2000; Wendler et al., 2006; Williams and Corbie-Smith, 2006). Older patients, patients with many co- morbid conditions and/or disabilities, children, and other subpopulations are also underrepresented in most clinical trials (Van Spall et al., 2007). Existing data sources can also help generate hypotheses about the charac- teristics of patients who are most likely to benefit from a therapy, as well as the characteristics of patients who are more likely to have adverse reactions to otherwise safe and effective drugs. Administrative Claims Data Administrative claims data comprise data obtained to support claims for reimbursement from insurers for services rendered. They include infor- mation on diagnoses, treatments (both medications and procedures), as well as many outcomes from millions of insured members. Claims data typically lack detailed information on clinical variables, such as laboratory results, lifestyle factors, and other physiological measures (e.g., height, weight, blood pressure, health status). The nationâs largest and most representative claims database is held by CMS. Medicare now covers more than 45 mil- lion Americans, mostly over age 64. Information on drugs paid for under Medicare Part D, available since 2006, covers nearly 27 million people (CMS, 2009). Medicaid data and data from the Childrenâs Health Insur- ance Program (CHIP) include younger Americans of lower socioeconomic status and individuals with significant comorbid conditions (The Henry J. Kaiser Foundation, 2009). In 2005, Medicaid provided coverage to 29.4 million children and 15.2 million adults (primarily poor working parents), and CHIP covered an additional 7 million children by 2007 (The Henry â Cluster randomized trials are RCTs in which the participants are assigned to the ex- perimental or comparison groups (clusters) defined by a common feature, such as the same physician or health plan.
ESSENTIAL PRIORITIES 151 J. Kaiser Foundation, 2009). However, there is significant variation in the quality of Medicaid claims data and the difficulties that researchers face in securing these dataâoften requiring inquiries to multiple states and their Institutional Review Boards (IRBs)âhave greatly constrained the use of Medicaid and CHIP data for research. Electronic Health Records Several large health delivery systems have used their electronic health records to build clinical datasets for research. These include the Veterans Health Administrationâs VISTA system, Kaiser Permanente, Group Health Cooperative of Puget Sound, Intermountain Healthcare, and Geisinger Health System. The clinical data in electronic health records collected by these organizations are more comprehensive than claims data and can be used for detailed management of chronic disease and some prospective studies of treatment outcomes. Large health care systems, however, may have restricted drug formularies that do not cover newer, non-evidence- based medications, which limits the data available for analysis and reduces the generalizability of findings. To date, the adoption of fully functional electronic health records has been slow (Ford et al., 2009); however, the incentives provided under ARRA should spur more rapid uptake, which could benefit the conduct of CER if the deployment is undertaken with systems capable of interoperability and connectivity for data sharing with repositories and if the electronic health records have the capability of clini- cal decision support and opportunities for patients to add relevant informa- tion into their records via secure web portals. Without these considerations, a major public investment will fall short of meeting the needs of CER and our nation will miss an opportunity to enhance its capacity to dramatically improve system performance through the benefit of CER. Practice-Based Research Networks Practice-based research networks are designed for research on clinical practices and quality improvement activities. These networks generate both primary and specialty care data, often using data gathered prospectively for the purpose of research (in contrast to most existing data from practice, which document routine clinical care and may have important limitations for research purposes). These data may thus provide detailed clinical infor- mation from settings not captured in large integrated systems (Westfall et al., 2007). A limitation of most current practice-based research networks is that they typically enroll relatively small numbers of patients and may not be representative of the population as a whole, a problem that could be overcome with larger networks and/or integration across networks.
152 INITIAL NATIONAL PRIORITIES FOR CER Disease- and Treatment-Specific Registries Patient registries can be a valuable data source for research on the real-world effectiveness of health care interventions. A patient registry is an organized system, designed for a predetermined purpose (e.g., clini- cal or policy research), that employs observational research methods to collect standardized clinical and other data in order to evaluate specific outcomes for a population defined by a specific disease, condition, or expo- sure (Gliklich and Dreyer, 2007). The data may be exclusively drawn from existing electronic sources or may also include primary data provided by clinicians. Many registries have preexisting IRB review and approvals that can facilitate data sharing and reduce concerns related to patient consent and privacy (Gliklich and Dreyer, 2007). At present, there are hundreds of patient registries in the United States. Many are designed around specific diseases; others are product registries maintained for post-marketing surveillance (IOM, 2007). The Clozapine registry, for example, was mandated by the FDA to detect adverse events associated with use of the drug (Teva Pharmaceuticals, 2008). The Cystic Fibrosis Foundation has sponsored a national patient registry for more than 40 years to enable clinicians and researchers to observe trends in the health of people with cystic fibrosis, create clinical care guidelines, design clinical trials to test new therapies, and improve the delivery of care (Cystic Fibrosis Foundation, 2009). Registries have two major drawbacks: first, their restriction to patient populations who undergo a particular intervention (e.g., immunization reg- istries) or who have a condition of interest (e.g., cystic fibrosis) and, second, they may not collect all the data needed to answer specific comparative ef- fectiveness questions. An advantage of registries is that they can centralize data collection for rare conditions or procedures in order to improve the likelihood of observing trends. Challenges to Using Existing Data The CER Program will have to overcome three critical hurdles to reap the potential contribution of existing datasets to CER: (1) linking patient- level data from multiple sources, (2) protection of the privacy and security of patient data, and (3) ensuring that holders of large datasets actively participate in the CER enterprise. These are described below. Data Linkage CER often requires data to be linked at the individual patient level from multiple large-scale, clinical research networks. Datasets may be
ESSENTIAL PRIORITIES 153 duplicative or complementary. A single database may not provide a large enough patient cohort or a sufficiently complete picture of a patientâs con- dition or health history, whereas several complementary datasets may be needed to meet the needs of CER. In order to link data across databases, the data must have standardized definitions and be electronically compatible. Moreover, different individual patient identifier codes in different datasets must recognize each otherâ requirements that also apply to linkage and pooling of data from two or more similar observational cohorts or the complementary linkage of a clini- cally defined cohort with a payerâs data. Patient identifiers in pharmaceuti- cal dispensing, hospital discharges, and diagnosis and procedure codes are standardized across most systems. However, information from laboratory results, enrollment, and utilization data usually require significant harmo- nization of patient identifiers to link the information from multiple sources into an analyzable, single patient record. The FDAâs 2008 Sentinel Initiative is the most ambitious linkage pro- posal to date. Its ultimate goal is the creation and implementation of the Sentinel Systemâa national, integrated, electronic database to detect ad- verse effects of drugs and other medical products. The system, which will eventually monitor as many as 100 million individuals, will be built from participating electronic health record or claims databases. The Sentinel Initiative system could also be used to study questions of comparative ef- fectiveness. Because of concerns related to patient privacy and to health care systemsâ proprietary interests, the Sentinel System has proposed to use the tools and processes of distributed network analysis. In a distributed network, all clinical data remain with the source systemsâ databases. Cen- tralized software is used to query each networked system, provided that system has approved the query. Transfer of identifiable data from the source health care systems to a central location is eliminatedâa large advantage for adherence to the Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule regulations and protection of privacy (FDA, 2008). Data Privacy and Security Individuals are more likely to participate in research and to support using medical records in research if they are convinced that their personal health information will remain confidential. Highly publicized privacy and security breaches undermine public trust in the research community, hin- der recruitment of research participants, and threaten the overall research enterprise (IOM, 2009a). Under the HIPAA Privacy Rule, researchers must obtain an informed consent for every use of an individualâs protected health information. How- ever, the Privacy Rule acknowledges that obtaining informed consent from
154 INITIAL NATIONAL PRIORITIES FOR CER every research participant in every research situation is not always feasible. Thus, the Privacy Rule specifies several situations, including using âdeiden- tifiedâ patient data, in which researchers can use protected health informa- tion without each patientâs consent. A recent IOM committee concluded that the HIPAA Privacy Rule does not protect privacy as well as it should and, as currently implemented, it impedes important health research (IOM, 2009a). That committeeâs prin- cipal recommendation proposed congressional authorization of HHS and other relevant federal agencies to develop a comprehensive, new approach to ensuring privacy while facilitating health research. Recognizing that this ambitious recommendation might be controversial and difficult to imple- ment, that committee also provided a set of more limited recommendations that addressed particular provisions of the HIPAA Privacy Rule, involving issues such as accounting for disclosures, authorizations for specified future research, activities preparatory to research, and mechanisms for linking multisource data, among others (Box 6-1). This committee views these recommendations as crucial to enabling robust CER. Recommendation 8: The CER Program should help to develop large- scale, clinical and administrative data networks to facilitate better use of data and more efficient ways to collect new data to inform CER. â¢ The CER Program should ensure that CER researchers and institu- tions consistently adhere to best practices to protect privacy and maintain security. â¢ The CER Program should support the development of methodolo- gies for linking patient-level data from multiple sources. â¢ The CER Program should encourage data holders to participate in CER and provide incentives for cooperation and maintaining data quality. The committee also agreed that the federal government should support the development of privacy-enhancing technologies for sharing health in- formation for CER, including methods that minimize or eliminate transfer of protected health information. Distributed research networks, as noted earlier, are designed to keep clinical data within and under the control of the source data systems participating in the network (Brown et al., 2009). With the permission of a system authority, researchers can extract deidenti- fied data from the source system and export it to a central site where the data are pooled with data from other network participants. While this technology does require further refinement before it can be fully exploited â Deidentified data are stripped of information that could be used to identify individuals.
ESSENTIAL PRIORITIES 155 in CER, work is under way in groups like the HMO Research Network and similar emerging research networks. Widespread Participation by Data Holders To develop the concepts of shared data research networks to their po- tential for advancing health care, the CER Program must address both data holderâs proprietary interests and the costs that they incur in order to share data. Patterns of medication use, rates of use of various procedures, organi- zational performance with respect to quality, and complications of therapy are all subjects of intense proprietary interest to competing health care organizations. These organizations may not share their information if do- ing so threatens a competitive advantage. Privacy-enhancing technologies, such as distributed networks, may successfully address these proprietary concerns because users cannot identify the source of the patient data. Holders of data will incur costs setting up and maintaining databases that are usable for CER research. The CER Program will need to create financial incentives that effectively offset these costs to encourage health systems to participate in CER research. The Medicare programâs âcover- age with evidence developmentâ initiative is an example and adaptation to Medicaid and CHIP programs that could be explored. Other types of incentives to share data may also be helpful. For instance, organizations might provide data and analysis in return for periodic reportsâsuitably deidentifiedâcomparing their data (e.g., on drug utilization or performance metrics) with that of other participants. develop, deploy, and Support a CER Workforce Health interventions are inherently complex and often involve multiple systems and diverse patients, providers, health care organizations, financing mechanisms, communities, and sociodemographic factors. Research that addresses the functions of complex systemsâsuch as CERârequires collab- oration between many disciplines and also continual methods innovation. CER researchers come from a range of professional disciplines, includ- ing clinical medicine, epidemiology, biomedical informatics, biostatistics, health services research, economics, methods research, decision and cog- nitive sciences, genomics, proteomics, library science, communications, as well as other areas (IOM, 2009c). They may have medical or other clinical degrees, doctoral degrees in public health specialties, specific training in systematic reviews and clinical trials, and/or post-doctoral or masterâs-level training. The CER workforce needs individuals with expertise in designing and conducting trials, statistical modeling, conducting systematic reviews and meta-analysis, quasi-experimental design and other observational
156 INITIAL NATIONAL PRIORITIES FOR CER BOX 6-1 IOM Recommendations for Changes to the HIPAA Privacy Rule and Associated Guidance Beyond the HIPAA Privacy Rule: Enhancing Privacy, Improving Health through Research A. HS should reduce variability in interpretations of the HIPAA Privacy H Rule in health research by covered entities, Institutional Review Boards (IRBs) and Privacy Boards through revised and expanded guidance and harmonization. 1. HHS should develop a dynamic, ongoing process to increase empirical knowledge about current âbest practicesâ for privacy protection in respon- sible research using protected health information (PHI), and promote the use of those best practices. 2. HHS should encourage greater use of partially deidentified data called âlimited datasetsâ and develop clear guidance on how to set up and comply with the associated data use agreements more efficiently and effectively, in order to enhance privacy in research by expanding use and usability of data with direct identifiers removed. 3. HHS should clarify the distinctions between âresearchâ and âpracticeâ to ensure appropriate IRB and Privacy Board oversight of PHI disclosures for these activities. 4. HHS guidance documents should simplify the HIPAA Privacy Ruleâs provi- sions regarding the use of PHI in activities preparatory to research and harmonize those provisions with the Common Rule, in order to facilitate appropriate IRB and Privacy Board oversight of identification and recruit- ment of potential research participants. B. HS should develop guidance materials to facilitate more effective use H of existing data and materials for health research and public health purposes. m Â ethods, use and analysis of large datasets, cost-effectiveness analysis, clinical prediction rules, measurement of patient-reported and clinical outcomes, and communicating research findings to patients, providers, and others. The CER Program will have to ensure the participation of indiÂviduals with a sound foundation in these areas. Current Workforce Capacity The significant increase in CER activity will create a substantial need for the types of expertise just described. Gauging the capacity of the current CER workforce is difficult because so many disciplines are involved and so
ESSENTIAL PRIORITIES 157 1. HHS should develop guidance that clearly states that individuals can au- thorize use of PHI stored in databases or associated with biospecimen banks for specified future research under the HIPAA Privacy Rule with IRB/Privacy Board oversight, as is allowed under the Common Rule, in order to facilitate use of repositories for health research. 2. HHS should develop clear guidance for use of a single form that permits individuals to authorize use and disclosure of health information in a clinical trial and to authorize the storage of their biospecimens collected in conjunc- tion with the clinical trial, in order to simplify authorization for interrelated research activities. 3. HHS should clarify the circumstances under which DNA samples or se- quences are considered PHI, in order to facilitate appropriate use of DNA in health research. 4. HHS should develop a mechanism for linking data from multiple sources so that more useful datasets can be made available for research in a manner that protects privacy, confidentiality, and security. C. HS should revise provisions of the HIPAA Privacy Rule that entail heavy H burdens for covered entities and impede research without providing substantive improvements in patient privacy. 1. HHS should reform the requirements for the accounting of disclosures of PHI for research. 2. HHS should simplify the criteria that IRBs and Privacy Boards use in mak- ing determinations for when they can waive the requirements to obtain au- thorization from each patient whose PHI will be used for a research study, in order to facilitate appropriate authorization requirements for responsible research. SOURCE: IOM (2009a). many educational pathways to the field exist. For these reasons, no one has yet analyzed the current workforce to see if it is sufficient to respond to the ARRA mandate for expanding CER. Nonetheless, ARRAâs infusion of $1.1 billion into CER will clearly stress the limits of the current CER workforce. ARRA appropriations increased AHRQâs CER budget tenfold. Aggregate current NIH spending on CER is not known, but the Institutes will receive at least an additional $400 million to conduct CER. Recommendation 9: The CER Program should develop and support the workforce for CER to ensure the nationâs capacity to carry out the CER mission. Important next steps include the following:
158 INITIAL NATIONAL PRIORITIES FOR CER â¢ Development of a strategic plan for research workforce development â¢ Long-term, sufficient funding for early career development including expanding grants for graduate and postgraduate training opportuni- ties in comparative effectiveness methods as well as career develop- ment grants and mid-career merit awards Ensuring a Highly Skilled CER Workforce The committee agreed that, at the outset, the CER Program should de- velop a strategic plan for research workforce development. The plan should include assessments of both the capacity of the current workforce to carry out the Programâs research agenda and the capacity and effectiveness of current training programs for producing researchers with the relevant skills. Developing an adequate CER workforce will involve the training, deploy- ment, and collaboration of a significant number of professional disciplines. Data on education paths and training programs for CER investigators are scarce. The NIH Roadmap for Medical Research, together with the Clini- cal and Translational Science Consortium are two mechanisms by which workforce development can be efficiently achieved (National Center for Research Resources, 2009; NIH, 2009). Training grants, such as K12, K30, and T32, should incorporate concepts of CER in their curricula exposing young Âscientists to CER and expanding the opportunities for participation in CER. CER is a fast-growing field that has experienced changes over time. At the present state of development of CER, it appears to be growing as a cohesive discipline. However, the career path is ill defined, and other areas of clinical research compete for the best and the brightest investigators. To be attractive to them, the field needs sustainable research funding and must adhere to high standards of research quality and scientific integrity, be open to new ideas and people, and provide excitement about the potential to contribute to health research and health care practice overall. The CER Program should secure long-term, sufficient funding for career development including expanding grants for graduate and postgraduate training oppor- tunities in comparative effectiveness methods, as well as career development grants and mid-career merit awards. Without adequate training and secure, stable financial support, talented investigators are likely to pursue other areas of research. Undoubtedly, a stable funding stream for CER will attract investigators to CER, as will a sense that the nation places a high priority on CER as a partial but important part of paying for health care reform and improving the quality of care.
ESSENTIAL PRIORITIES 159 Bringing Knowledge into Practice Many stakeholders and members of the public asked the committee to prioritize CER topics related to the comparative effectiveness of methods for bringing proven health care interventions into everyday clinical practice (see Chapter 5). Dougherty and Conway have proposed that three steps in knowledge translation must occur before research can improve health care quality and value: (1) translation of basic biomedical science into clinical ef- ficacy knowledge, (2) translation of clinical efficacy knowledge into clinical effectiveness knowledge, and (3) translation of clinical effectiveness knowl- edge into health system improvement (Dougherty and Conway, 2008). Biomedical research has traditionally focused on steps one and two. The Clinical and Translational Science Consortium is now beginning to expand research networks and emphasize community engagement. But, the health care system will not benefit from CER without the third translational step, and more effort can be made by the Consortium to assess the integration of new findings into practice and their impact on health outcomes. The CER Program should require researchers to publish all federally funded CER studies and make the research readily available to the pub- lic. Health care professionals and patients must use CER results to make informed decisions that integrate the best available evidence, the patientsâ preferences, and specific characteristics of the patient (Mattews, 2009; Weinstein et al., 2007). Recommendation 10: The CER Program should promote rapid adop- tion of CER findings and conduct research to identify the most effective strategies for disseminating new and existing CER findings to health care professionals, consumers, patients, and caregivers and for helping them to implement these results in daily clinical practice. The American health research infrastructure lacks a systematic way to translate knowledge from research to practice. The translation of research findings into practice is slow and incomplete. Many barriers exist: perverse reimbursement incentives, physician perceptions about patientsâ expecta- tions, and patientsâ concerns about denials of care or reluctance to question clinicians (Shojania and Grimshaw, 2005). These barriers and others should be addressed and, insofar as possible, overcome. Knowledge translation research must be a high priority. conclusion In summary, the HHS Secretaryâs CER agenda will fall far short of its potential without effective coordination and governance of the enterprise.
160 INITIAL NATIONAL PRIORITIES FOR CER The research agenda will involve a broad array of study designs, the full range of health care services, and an extensive corps of experts in diverse professional disciplines. However, an ambitious research enterprise alone will not improve health care in the United States without the Secretaryâs attention to high fidelity translation of knowledge into practice. Moreover, consumers, patients, and caregivers as well as their health care providers must be involved in all aspects of CER to ensure its relevance to everyday health care delivery. The $1.1 billion ARRA investment in CER is an unprecedented vote of confidence in patient-centered research. The CER program should be held accountable to its mission. Sustained program evaluation and continuous quality improvement must be a bedrock feature of the enterprise. references Academy of Medical Sciences. 2006. Personal data for public good: Using health information in medical research. http://www.acmedsci.ac.uk/images/project/Personal.pdf (accessed August 28, 2008). Andejeski, Y., I. T. Bisceglio, K. Dickersin, J. E. Johnson, S. I. Robinson, H. S. Smith, F. M. Visco, and I. M. Rich. 2002a. Quantitative impact of including consumers in the scientific review of breast cancer research proposals. Journal of Womenâs Health & Gender-Based Medicine 11(4):379-388. Andejeski, Y., E. S. Breslau, E. Hart, N. Lythcott, L. Alexander, I. Rich, I. Bisceglio, H. S. Smith, and F. M. Visco. 2002b. Benefits and drawbacks of including consumer reviewers in the scientific merit review of breast cancer research. Journal of Womenâs Health & Gender-Based Medicine 11(2):119-136. Aryal, K. R., and H. Al-khaffaf. 2006. Venous thromboembolic complications following air travel: Whatâs the quantitative risk? A literature review. European Journal of Vascular and Endovascular Surgery 31(2):187-199. Bell, C. L., J. Davis, R. C. Harrigan, E. Somogyi-Zalud, M. K. G. Tanabe, and K. H. Masaki. 2009. Factors associated with place of death for elderly Japanese American men: The Honolulu heart program and Honolulu-Asia aging study. Journal of the American Geri- atrics Society 57(4):714-718. Berry, D. A. 2003. Statistical innovations in cancer research. In Cancer Medicine 6th Edition, Edited by J. Holland, E. Frei, D. W. Kufe, R. E. Pollock, R. R. Weichselbaum, R. C. Bast and T. S. Gansler. Hamilton: BC Decker. Pp. 465-478. Braunstein, J. B., N. S. Sherber, S. P. Schulman, E. L. Ding, and N. R. Powe. 2008. Race, medi- cal researcher distrust, perceived harm, and willingness to participate in cardiovascular prevention trials. Medicine 87(1):1-9. Brown, D. R., M. N. Fouad, K. Basen-Engquist, and G. Tortolero-LUna. 2000. Recruitment and retention of minority women in cancer screening, prevention and treatment trials. Annals of Epidemiology 10:S13-S21. Brown, J., J. Holmes, J. Maro, B. Syat, K. Lane, R. Lazarus, and R. Platt. 2009. Developing a distributed research network and cooperative to conduct population-based studies and safety surveillance. In Report 1: Design Specifications for Network Prototype and Research Cooperative.
ESSENTIAL PRIORITIES 161 Chan, A. W., A. Hrobjartsson, M. T. Haahr, P. C. GÃ¸tzsche, and D. G. Altman. 2004. Em- pirical evidence for selective reporting of outcomes in randomized trials: Comparison of protocols to published articles. JAMA 291:2457-2465. Chou, R., and M. Helfand. 2005. Challenges in systematic reviews that assess treatment harms. Annals of Internal Medicine 142(12 [Part 2]):1090-1099. CMS (Centers for Medicare & Medicaid). 2009. Prescription drug coverage general informa- tion: Overview. http://www.cms.hhs.gov/PrescriptionDrugCovGenIn/ (accessed May 9, 2009). Cupples, L., R. DâAgostino, and D. Kiely. 1988. The Framingham heart study. Section 35. An Epidemiological Investigation of Cardiovascular Disease Survival Following Cardiovas- cular Events: 30 Year Follow-up. Cystic Fibrosis Foundation. 2009. Patient registry report. http://www.cff.org/treatments/Care- CenterNetwork/PatientRegistryReport/ (accessed June 16, 2009). de Simone, G., R. B. Devereux, M. Chinali, M. J. Roman, E. T. Lee, H. E. Resnick, and B. V. Howard. 2009. Metabolic syndrome and left ventricular hypertrophy in the prediction of cardiovascular events: The strong heart study. Nutrition, Metabolism and Cardiovascular Diseases 19(2):98-104. Deber, R. B., N. Kraetschmer, and J. Irvine. 1996. What role do patients wish to play in treat- ment decision making. Archives of Internal Medicine 156:1414-1420. Degner, L. F., and C. A. Russell. 1988. Preferences for treatment control among adults with cancer. Research in Nursing & Health 11:367-374. Dickersin, K. 2005. Publication bias: Recognizing the problem, understanding its origins and scope, and preventing harm. In Publication Bias in Meta-analysis: Prevention, Assess- ment, and Adjustments, Edited by H. Rothstein, A. Sutton and M. Borenstein. London, UK: John Wiley and Sons, Ltd. Dickersin, K., and Y.-I. Min. 1993. Publication bias: The problem that wonât go away. In Doing More Good than Harm: The Evaluation of Health Care Interventions. Edited by K.S. Warren and F. Mosteller. New York Academy of Sciences. 135-148. Dougherty, D., and P. H. Conway. 2008. The â3Tâsâ road may transform US health care: The âhowâ of high-quality care. JAMA 299(19):2319-2321. Entwistle, V. A., M. J. Renfrew, S. Yearley, J. Forrester, and T. Lamont. 1998. Lay perspectives: Advantages for health research. BMJ 316:463-366. Faridi, Z., J. Grunbaum, B. S. Gray, A. Franks, and E. Simoes. 2007. Community-based participatory research: Necessary next steps. Preventing Chronic Disease: Public Health Research, Practice and Policy 4(3):1-5. Farmer, D., S. A. Jackson, F. Camacho, and M. A. Hall. 2007. Attitudes of African American and low socioeconomic status White women toward medical research. Journal of Health Care for the Poor and Underserved 18:85-99. FDA (Food and Drug Administration). 2008. FDAâs Sentinel Initiative. http://www.fda.gov/oc/ initiatives/advance/sentinel/ (accessed May 5, 2009). âââ. 2009. Information for patient advocates http://www.fda.gov/oashi/home.html (ac- cessed May 13, 2009). Ford, E. W., N. Menachemi, L. T. Peterson, and T. R. Huerta. 2009. Resistance is futile: But it is slowing the pace of EHR adoption nonetheless. Journal of the American Medical Informatics Association 16(3):274-281. Fretheim, A., H. J. Schanemann, and A. D. Oxman. 2006. Improving the use of research evi- dence in guideline development: 3. Group composition and consultation process. Health Research Policy and Systems 4. Fung, M. M., R. Bettencourt, and E. Barrett-Connor. 2004. Heart disease risk factors predict erectile dysfunction 25 years later: The rancho bernardo study. Journal of the American College of Cardiology 43(8):1405-1411.
162 INITIAL NATIONAL PRIORITIES FOR CER Furberg, C. D., J. T. Wright Jr, B. R. Davis, J. A. Cutler, M. Alderman, H. Black, W. Cushman, R. Grimm, L. J. Haywood, F. Leenen, S. Oparil, J. Probstfield, P. Whelton, C. Nwachuku, D. Gordon, M. Proschan, P. Einhom, C. E. Ford, L. B. Piller, I. K. Dunn, D. Goff, S. Pressel, J. Bettencourt, B. DeLeon, L. M. Simpson, J. Blanton, T. Geraci, S. M. Walsh, C. Nelson, M. Rahman, A. Juratovac, R. Pospisil, L. Carroll, S. Sullivan, J. Russo, G. Barone, R. Christian, S. Feldman, T. Lucente, D. Calhoun, K. Jenkins, P. McDowell, J. Johnson, C. Kingry, J. Alzate, K. L. Margolis, L. A. Holland-Klemme, B. Jaeger, J. Wil- liamson, G. Louis, P. Ragusa, A. Williard, R. L. S. Ferguson, J. Tanner, J. Eckfeldt, R. Crow, and J. Pelosi. 2002. Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: The antihypertensive and lipid-lowering treatment to prevent heart attack trial (ALLHAT). JAMA 288(23):2981-2997. Giuliano, A. R., N. Mokuau, C. Hughes, G. Tortelero-Luna, B. Risendal, R. C. S. Ho, T. E. Prewitt, and W. J. McCaskill-Stevens. 2000. Participation of minorities in cancer research: The influence of structural, cultural, and linguistic factors. Annals of Epide- miology 10:S22-S34. Gliklich, R. E., and N. A. Dreyer. 2007. Registries for evaluating patient outcomes: A userâs guide. Rockville, MD: Agency for Healthcare Research and Quality. Godwin, M., L. Ruhland, I. Casson, S. MacDonald, D. Delva, R. Birtwhistle, M. Lam, and R. Seguin. 2003. Pragmatic controlled clinical trials in primary care: The struggle between external and internal validity. BMC Medical Research Methodology 3:1-7. Guadagnoli, E., and P. Ward. 1998. Patient participation in decision-making. Social Science & Medicine 47(3):329-339. Hanley, B., A. Truesdale, A. King, D. Elbourne, and I. Chalmers. 2001. Involving consumers in designing, conducting and interpreting randomised controlled trials: Questionnaire survey. BMJ 322:519-523. The Henry J. Kaiser Foundation. 2009. Medicaid & CHIP. http://www.statehealthfacts.org/ comparecat.jsp?cat=4 (accessed May 9, 2009). Hubbard, G., L. Kidd, and E. Donaghy. 2008. Involving people affected by cancer in research: A review of literature. European Journal of Cancer Care 17(3):233-244. INVOLVE. 2009. Promoting public involvement in NHS, public health and social care re- search. http://www.invo.org.uk/ (accessed May 13, 2009). IOM (Institute of Medicine). 1997. A review of the Department of Defenseâs program for breast cancer research. Washington, DC: National Academy Press. âââ. 2001. Crossing the quality chasm: A new health system for the 21st century, National Academy Press. http://www.nap.edu/catalog/10027.html (accessed April 20, 2009). âââ. 2002. Responsible research: A systems approach to protecting research participants. Washington, DC: The National Academies Press. âââ. 2007. Learning what works best: The nationâs need for evidence on comparative effectiveness in health care. http://www.iom.edu/ebm-effectiveness (accessed April 15, 2009). âââ. 2008. Knowing what works in health care: A roadmap for the nation. Edited by J. Eden, B. Wheatley, B. J. McNeil and H. Sox. Washington, DC: The National Academies Press. âââ. 2009a. Beyond the HIPAA privacy rule: Enhancing privacy, improving health through research. Edited by Nass, S. J., L. A. Levit and L. O. Gostin. Washington, DC: The Na- tional Academies Press. âââ. 2009b. Conflict of interest in medical research, education, and practice. Edited by B. Lo, and M. J. Field. Washington, DC: The National Academies Press.
ESSENTIAL PRIORITIES 163 âââ. 2009c. A framework for the workforce required for comparative effectiveness re- search. In Learning What Works Infrastructure Required to Learn Which Care is Best (in press). Washington, DC: The National Academies Press. Johnson, B., M. Abraham, J. Conway, L. Simmons, S. Edgman-Levitan, P. Sodomka, J. Schlucter, and D. Ford. 2008. Partnering with patients and families to design a patient- and family-centered health care system: Recommendations and promising practices. http://www.ihi.org/NR/rdonlyres/C810CCBB-2DEB-4678-994A-57D9B703F98D/0/Â PartneringwithPatientsandFamiliesRecommendationsApr08.pdf (accessed June 12, 2009). Jones, L., and K. Wells. 2007. Strategies for academic and clinician engagement in community- participatory partnered research. JAMA 297(4):407-410. Kent, D. M., and R. A. Hayward. 2007. Limitations of applying summary results of clinical trials to individual patients. JAMA 298:1209-1212. Lindenmeyer, A., H. Hearnshaw, J. Sturt, R. Ormerod, and G. Altchison. 2007. Assessment of the benefits of user involvement in health research from the Warwick Diabetes Care Research User Group: A qualitative case study. Health Expectations 10:268-277. Mansell, D., R. M. Poses, L. Kazis, and C. A. Duefield. 2000. Clinical factors that influence patientsâ desire for participation in decisions about illness. Archives of Internal Medicine 160:2991-2996. Mattews, S. C. 2009. Physician autonomy and informed decision making: Finding the balance for patient safety and quality. JAMA 300(24):2913-2915. Mazur, D. J., and D. H. Hickam. 1997. Patientsâ preferences for risk disclosure and role in decision making for invasive medical procedures. Journal of General Internal Medicine 12:114-117. McClellan, M., and J. Benner. 2009. Comparative effectiveness research: Will it bend the health care cost curve and improve quality? In Implementing Comparative Effectiveness Research: Priorities, Methods, and Impact. Washington, DC: Brookings. Meier, B. 2009. New effort reopens a medical minefield The New York Times, B1. Merlini, L., P. Kishnani, B. Byrne, W. Maller-Felber, L. Case, and A. v. d. Ploeg. 2008. The Pompe registry: Centralized data collection to track the natural course of Pompe disease. Clinical Therapeutics 30(SUPPL. 1). Minkler, M., A. G. Blackwell, M. Thompson, and H. Tamir. 2003. Community-based par- ticipatory research: Implications for public health funding. American Journal of Public Health 93(8):1210-1213. National Center for Research Resources. 2009. Clinical and translational science awards. http://www.ncrr.nih.gov/clinical_research_resources/clinical_and_translational_science_ awards/ (accessed June 16, 2009). National Heart, Lung, and Blood Institute. 2009. Womenâs health initiative: Observational fact sheet. http://www.nhlbi.nih.gov/whi/os.htm (accessed May 23, 2009). NIH (National Institutes of Health). 2009. NIH roadmap for medical research. http://nihroad- map.nih.gov/ (accessed June 16, 2009). Office of National Coordinator for Health Information Technology. 2009. Information re- lated to the Economic Recovery Act of 2009. www.HealthIT.hhs.gov (accessed May 11, 2009). Oliver, S. R., R. W. Rees, L. Clarke-Jones, R. Milne, A. R. Oakley, J. Gabbay, K. Stein, P. Buchanan, and G. Gyte. 2008. A multidimensional conceptual framework for analysing public involvement in health services research. Health Expectations 11:72-84. Omenn, G. S. 1999. Caring for the community: The role of partnerships. Academic Medicine 74(7):782-789. Peterson, E., and C. W. Yancy. 2009. Eliminating racial and ethnic disparities in cardiac care. New England Journal of Medicine 360(12):1172-1174.
164 INITIAL NATIONAL PRIORITIES FOR CER Presidentâs Commission for the Study of Ethical Problems in Medicine and Biomedical and B Â ehavioral Research. 1982. Making health care decisions: The ethical and legal impli- cations of informed consent in the patient-provider relationship. http://bioethics.gov/Â reports/past_commissions/making_health_care_decisions.pdf (accessed June 25, 2009). Rawlins, M. 2008. De testimonio: On the evidence for decisions about the use of therapeutic interventions. The Lancet 372(9656):2152-2161. Rising, K., P. Bacchetti, and L. Bero. 2008. Reporting bias in drug trials submitted to the food and drug administration: Review of publication and presentation. Public Library of Science Medicine 5(11):e217. Sabin, J. E., K. Mazor, V. Meterko, S. L. Goff, and R. Platt. 2008. Comparing drug effective- ness at health plans: The ethics of cluster randomized trials. Hastings Center Report 38(5):39-48. Saunders, C., A. Girgis, P. Butow, S. Crossing, and A. Penman. 2008. From inclusion to independenceâTraining consumers to review research. Health Research Policy and Systems 6(3). Schwartz, R. S., G. D. Curfman, S. Morrissey, and J. M. Drazen. 2008. Full disclosure and the funding of biomedical research. New England Journal of Medicine 358(17):1850-1851. Shojania, K. G., and J. M. Grimshaw. 2005. Evidence-based quality improvement: The state of the science. Health Affairs 24(1):138-150. Staniszewska, S., N. Jones, M. Newburn, and S. Marshall. 2007. User involvement in the development of a research bid: Barriers, enablers and impacts. Health Expectations 10:173-183. Tatsioni, A., D. A. Zarin, N. Aronson, D. J. Samson, C. R. Flamm, C. Schmid, and J. Lau. 2005. Challenges in systematic reviews of diagnostic technologies. Annals of Internal Medicine 142(12 Part 2):1048-1055. Teva Pharmaceuticals. 2008. Teva clozapine patient registry. https://clozapineregistry.com/ AboutRegistry/GeneralOverview.aspx (accessed June 16, 2009). Tunis, S. 2009. Strategies to improve comparative effectiveness research methods and data infrastructure. In Implementing Comparative Effectiveness Research: Priorities, Methods, and Impact. Washington, DC: Brookings. Turner, E. H., A. M. Matthews, E. Linardatos, R. A. Tell, and R. Rosenthal. 2008. Selective publication of antidepressant trials and its influence on apparent efficacy. New England Journal of Medicine 358(3):252-260. United States Cochrane Center Consumers United for Evidence-based Healthcare. 2009. Understanding evidence-based healthcare: A foundation for action. http://apps1.jhsph. edu/cochrane/CUEwebcourse.htm (accessed May 23, 2009). Van Spall, H. G. C., A. Toren, A. Kiss, and R. A. Fowler. 2007. Eligibility criteria of random- ized controlled trials published in high-impact general medical journals: A systematic sampling review. JAMA 297(11):1233-1240. Weinfurt, K. P., D. M. Seils, J. P. Tzeng, L. Lin, K. A. Schulman, and R. M. Califf. 2008. Consistency of financial interest disclosures in the biomedical literature: The case of coronary stents. PLoS ONE 3(5). Weinstein, J. N., K. Clay, and T. S. Morgan. 2007. Informed patient choice: Patient-centered valuing of surgical risks and benefits. Health Affairs 26(3):726-730. Wendler, D., R. Kington, J. Madans, G. Van Wye, H. Christ-Schmidt, L. A. Pratt, O. W. Brawley, C. P. Gross, and E. Emanuel. 2006. Are racial and ethnic minorities less willing to participate in health research? PLoS Medicine 3(2):201-210. Wennberg, J. E., and E. S. Fisher. 2008. Tracking the care of patients with severe chronic ill- ness: The Dartmouth Atlas of health care 2008. Dartmouth Institute for Health Policy and Clinical Practice, Center for Health Policy Research. http://www.dartmouthatlas. org/atlases/2008%5FChronic%5FCare%5FAtlas.pdf (accessed June 8, 2009).
ESSENTIAL PRIORITIES 165 Westfall, J. M., J. Mold, and L. P. Fagnan. 2007. Practice-based researchââBlue highwaysâ on the NIH roadmap. JAMA 297:403-406. Williams, I. C., and G. Corbie-Smith. 2006. Investigator beliefs and reported success in recruit- ing minority participants. Contemporary Clinical Trials 27:580-586. Zerhouni, E. A. 2005. Translational and clinical scienceâTime for a new vision. New England Journal of Medicine 353(15):1621-1623.