National Academies Press: OpenBook
« Previous: 4 Medical Product Regulatory Issues
Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×

5

Infrastructure Needs and Opportunities

KEY SPEAKER THEMES

Platt

•   Large simple trials (LSTs) should make every attempt to not interfere with the normal work flow of a clinical operation.

•   Electronic health records (EHRs) are useful and possibly essential for obtaining the benefits of LSTs but are not a panacea.

•   Organizational consortia, often required to conduct LSTs, are expensive and have complicated governance challenges.

Ferguson

•   It is possible to do an LST at the point of care.

•   The questions being asked by LSTs should be driven by the information needs of clinical practice.

•   The greater use of LSTs will require a rethinking of the relationship between research and clinical care.

Kush

•   Technologies and resources that would allow the conduct of regulated clinical research from electronic sources, in particular, EHRs, without the use of paper exist.

Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×

•   The use of these approaches could lead to great advances in research efficiency, quality, and cost. They are particularly suited to use with LSTs.

•   Given the maturity of these resources, guidance from the U.S. Food and Drug Administration, and the support of vendors, the only thing missing is a sponsor willing to be the first to conduct a trial with data from an EHR.

Lannon

•   Reusable networks can be a way to engage patients, families, clinicians, and researchers and for clinicians and institutions to learn from a patient population much larger than the one that they serve.

•   These networks can effectively use technologies to collect data once and serve many purposes, including population management, quality reporting, and research.

•   Advances in information technology hold much promise for the future of data networks that pull data from many diverse sources.

INTRODUCTION

Large simple trials (LSTs) are attractive because they can answer certain questions about the effectiveness of drugs and other interventions at less cost or in less time, or both, than the standard randomized clinical trial (RCT). In this session of the workshop, presenters addressed the infrastructure needs for the greater adoption of LSTs and the opportunities for and benefits of relying on electronic health records (EHRs) and other information systems and organizational arrangements to conduct LSTs.

Richard Platt, professor and chair of the Harvard Medical School Department of Population Medicine at the Harvard Pilgrim Health Care Institute, addressed the issue of aligning care and research for greater integration. Ryan E. Ferguson, acting director of the U.S. Department of Veterans Affairs (VA) Cooperative Studies Program Coordinating Center in the VA Boston Healthcare System and program director of VA’s Point of Care Research Institute shared VA’s experiences with carrying out trials with EHR platforms at the point of care. Rebecca Daniels Kush, president and chief executive officer of the Clinical Data Interchange Standards Consortium, discussed opportunities to get research-quality data from EHRs to greatly improve the efficiency of clinical trials. Carole M. Lannon, director of the Learning Networks Core within the James Anderson Center

Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×

for Health Systems Excellence at Cincinnati Children’s Hospital Medical Center and professor of pediatrics at the University of Cincinnati, shared her experiences with building and maintaining reusable research networks.

ALIGNING CARE AND RESEARCH TO REDUCE
BURDENS AND IMPROVE INTEGRATION

Richard Platt focused his presentation on three points that resonated with him from earlier sessions of the workshop. First, he noted that LSTs should not interfere with the normal work flow of a clinical operation. Second, he echoed the observation that EHRs are useful and possibly essential for obtaining the benefits of LSTs. Finally, he drew from his own experiences and those of other presenters to state that organizational consortia are often required to conduct LSTs.

LSTs Should Not Interfere with Normal Clinical Work Flow

Platt noted that, by definition, a clinical trial, even an LST, involves a change in the way in which things are usually done and necessarily affects normal clinical operations. Engagement of the leadership of the health care system involved in the trial is therefore crucial to the success of research done at the point of care, as investment of managerial time and systems support are needed to minimize the impact of a trial on frontline health care providers. Getting support from frontline clinicians can be further facilitated, he suggested, if the trial is testing something that is of interest to them and that can be helpful to their work.

Platt offered the example of a recent LST that he and his colleagues conducted involving 75,000 patients in 43 hospitals of the Hospital Corporation of America (HCA). Although the trial required only minor modifications to regular care, it required significant time and effort from a wide range of other HCA employees, including the vice president for clinical operations, the chief nursing officer, the quality improvement staff, the infection prevention team, the intensive care unit directors, pharmacy staff, supply chain management, and the information technology department. He estimated that this involvement cost a total of $1 million (provided in kind by HCA), which was in addition to the $2 million provided by the study’s federal sponsor.

EHRs Are Necessary, But Not Sufficient

Platt noted that although EHRs hold much promise, in practice, EHRs can be difficult to use for research. One reason for this is that they are usually different in each health care system, even if they are obtained from the

Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×

same commercial vendor. One way around this lack of interoperability is to extract information from the EHR system on a regular basis. This information can then be transformed and analyzed separately and in a secure setting behind the organization’s firewall, which addresses Health Insurance Portability and Accountability Act and other privacy concerns. However, he noted that although data can be removed from EHRs, the placement of information back into EHRs, which would be needed to run a clinical trial, can be even more difficult.

Platt commented that the information from EHRs is often insufficient for use in clinical studies because EHRs cannot provide information about the care received by participants outside of the organization. He suggested that administrative data from health care insurers are an often undervalued source of information, especially if they can be linked to EHR data. Administrative data provide information about the care delivered across the entire spectrum of health care locations, are available for large populations, and are more standardized than most EHR data.

Organizational Consortia Are Often Required to Conduct LSTs

Many of the important topics that LSTs could address will require consortia of organizations to make reasonable progress, but the formation of consortia raises issues concerning governance and data sharing. Consortia are also expensive to build and maintain, Platt said. He discussed the benefits of distributed information networks, in which aggregate information, rather than individual patient data, can be shared and queries can be run behind organizational firewalls. In these systems, such as the Mini-Sentinel system of the U.S. Food and Drug Administration (FDA), the cancer Biomedical Informatics Grid (caBIG) system of the National Institutes of Health, the Scalable Partnering Network for Clinical Effectiveness Research of the Agency for Healthcare Research and Quality, and the Health Maintenance Organization Research Network, researchers can submit a question through a secure portal and receive the answer without having to access protected health information about individuals.

POINT-OF-CARE TRIALS USING EHR PLATFORMS

Ryan E. Ferguson presented an example of an LST that VA is carrying out using its EHR system. He described the problems that VA faced with the inefficiency of evidence creation and the failure of the research enterprise to meet the information needs of the health care system.

VA’s solution was to create a learning health care system in which important clinical questions were identified and the answers could be determined through studies using VA’s EHR system: the Veterans Health

Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×

Information Systems and Technology Architecture (VistA). The concept for the program is that in situations in which clinical equipoise exists between drug or intervention choices, a substantial portion of the operations of a trial to assess the question being asked could be conducted by the clinical staff as part of providing regular health care to VA beneficiaries. These LSTs could be implemented entirely within VistA, including participant identification and enrollment, consent and randomization, and data capture and management. The results learned from the trial would then become part of the decision support included in VistA.

Ferguson detailed how VA investigated the feasibility of an EHR-based point-of-care approach by conducting a pilot study comparing the use of sliding-scale versus weight-based protocols for insulin administration in diabetic patients. The study’s primary endpoint was length of stay, and the secondary endpoints were rates of inpatient glycemic control and readmission within 30 days. Ferguson reported that the pilot study demonstrated the feasibility of conducting point-of-care trials by use of the EHR by finding high rates of acceptance by the providers and patients participating in the study and rates of participation higher than those usually seen in RCTs.

Ferguson closed his presentation by reflecting on VA experiences to draw broader implications for LSTs. He highlighted that it is important that the questions being asked be driven by the information needs of clinical practice and pointed out that this approach is particularly well suited to answer certain questions. He characterized these questions as ones that are asked when options between approved products and interventions with well-described toxicity are being considered, questions whose answers can be provided by measurement of objectively identifiable endpoints, and questions that can be answered with a minimal need for study-specific visits.

Looking ahead to the requirements and priorities for the generation of more evidence at the point of care, he noted that it will be important to rethink the relationship between research and clinical care, that buy-in from providers and the leaders of health care systems is key, and that a rational approach to regulatory oversight will be crucial.

OBTAINING RESEARCH-QUALITY DATA FROM EHRs

Rebecca Daniels Kush began her presentation by highlighting one of the major challenges to the efficiency of current clinical trials: the continued use of paper records and the multiple varied systems used across clinical sites. She described a number of resources currently available to streamline research studies, highlighting the fact that although these resources have the potential to be disruptive, most have not been widely implemented.

She focused her comments on the concept of eSource, which allows the collection of clinical research information entirely by electronic means

Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×

without the use of paper forms while meeting existing clinical research regulations. The use of eSource, she noted, would make it easier for clinicians to conduct clinical research and would enable the extensive use of data that are collected only once.

Kush described the process used by the eSource Data Interchange (eSDI) Initiative in partnership with FDA. A multidisciplinary working group identified 12 requirements that would allow the use of eSource while still following all international regulatory rules.

The next step, Kush detailed, was part of the FDA’s Critical Path Initiative, specifically, development of a minimal core data set with data in 18 categories (domains) common to all research protocols. This standard was published in 2008 and is called Clinical Data Acquisition Standards Harmonization (CDASH). An interoperability specification (specifying three standards: the continuity of care document, the retrieve form for data capture [RFD], and CDASH) was then developed through the Office of the National Coordinator for Health Information Technology of the U.S. Department of Health and Human Services as a means to obtain a common set of core research data elements to be readily exchanged between EHRs and clinical research systems. This specification was published in 2010.

Another important development was the establishment of a transport standard, called ODM, which contains audit trail information in the metadata (to support regulations described in the Code of Federal Regulations [21 CFR 11]). This audit trail information includes who entered the data and, if they were changed, what data were changed and who changed them, why, and when. ODM is an XML standard that carries CDASH content along with the audit trail information. The RFD standard integration profile streamlines the work flow for the population of electronic case report forms from EHRs.

As an example she highlighted the streamlining of reporting of adverse drug events (ADEs) from EHRs evaluated in the ADE Spontaneous Triggered Event Reporting project conducted by Pfizer and Harvard. That study used RFD to facilitate adverse event reporting and found that the time that it takes to make such a report was greatly decreased (from ~34 minutes to less than 1 minute), thus resulting in the reporting of many more adverse events by clinicians.

Kush explained that RFD, ODM, and CDASH do not depend on the particular EHR being used and that all of these pieces are available to support the establishment of a paperless eSource system for adverse event reporting and research.

In 2012, FDA issued guidance to the industry on data from electronic sources for clinical research, or eSource, citing that such guidance would help ensure the reliability, quality, integrity, and traceability of data from electronic sources. Kush noted that despite the maturity of the resources

Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×

that she described, guidance from FDA, support from EHR vendors, and demonstrations done at industry meetings, a clinical research study of eSource has still not been done with EHRs.

She concluded her presentation with a call for sponsors to step up and take advantage of these disruptive advances that have been shown to increase research efficiency and data quality.

BUILDING REUSABLE RESEARCH NETWORKS

Carole M. Lannon’s presentation covered reusable research networks, which are collaborative research and quality improvement arrangements among health care organizations that can be used for multiple purposes by different stakeholders. These networks are evolving toward the learning health care system model. Her organization, the Cincinnati Children’s Hospital Medical Center, is involved in five national networks: one focused on patient safety, one focused on perinatal health, and three focused on disease.

Lannon explained that the concept behind these networks is that clinicians and researchers can learn from a much larger base of patients than they would encounter in their home institution alone. They can examine the outcomes of widely varying diagnostic and treatment practices and identify factors that lead to better outcomes. They can conduct clinical trials with much larger populations, which is especially important for pediatric care and research on diseases in the pediatric population, because the incidence of disease in pediatric populations is relatively low. Network members can undertake quality improvement efforts on the basis of research and real-time feedback as well as maintain certification credit for their participation.

Lannon presented results from ImproveCareNow, the network that has been in existence the longest that focuses on improving care for inflammatory bowel disease (IBD) for children and adolescents. She showed how, over a 4-year period, the rate of remission for IBD among some 50 children’s hospitals has increased from less than 50 percent to nearly 80 percent. In another network, the 20 largest maternity hospitals in Ohio were able to reduce the rate of scheduled deliveries at between 36 and 38 completed weeks without a medical indication, shifting more than 25,000 births from preterm to term over 4 years and saving an estimated 500 neonatal intensive care unit admissions and between $15 million and $20 million. Eight Ohio children’s hospitals were able to reduce the rate of surgical site infections from 4.4 to 1.7 percent, reducing the number of children harmed by about 31 a year, with a cost savings of about $680,000 a year.

Lannon presented three prerequisites for maintaining a stable learning network able to support multiple projects over time. First was a focus on outcomes. She explained that this engages patients, clinicians, and research-

Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×

ers and united them in their efforts to achieve something meaningful. The second focus was building a community of engaged patients, families, clinicians, and researchers through a variety of communication mechanisms and by inclusion of patients and families in governance and research. She described the learning process that they are going through to build a community, including the use of social network sites for patients to connect and share stories and for patients, families, clinicians, and researchers to collaborate effectively. The third factor is the effective use of technology, especially for more efficient data collection and use. Lannon described efforts that they are undertaking so that data entered once can be fed into automated processes to determine the proper clinical care for patients with chronic conditions, processes of measurement for quality improvement and learning, and data quality improvement processes and so that data can be used for research at population and individual levels.

Lannon discussed some of the challenges that these networks face. They include variations in institutional review boards across sites, which have led to the use of a federated institutional review board model, as well as time challenges for clinicians. Lannon also pointed to a number of trends that could improve the effectiveness of learning networks. She mentioned trends in health information technology that will enable large aggregate sets of data to be pulled from EHRs and even the use of patient sensors, as well as the exploitation of opportunities for distributed and collaborative production, in which patients and clinicians can work together to quickly tests what approaches work.

Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×
Page 35
Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×
Page 36
Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×
Page 37
Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×
Page 38
Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×
Page 39
Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×
Page 40
Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×
Page 41
Suggested Citation:"5 Infrastructure Needs and Opportunities." Institute of Medicine. 2013. Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/18400.
×
Page 42
Next: 6 Ethical and Privacy Policy Issues »
Large Simple Trials and Knowledge Generation in a Learning Health System: Workshop Summary Get This Book
×
Buy Paperback | $38.00 Buy Ebook | $30.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Randomized clinical trials (RCTs) are often referred to as the "gold standard" of clinical research. However, in its current state, the U.S. clinical trials enterprise faces substantial challenges to the efficient and effective conduct of research. Streamlined approaches to RCTs, such as large simple trials (LSTs), may provide opportunities for progress on these challenges. Clinical trials support the development of new medical products and the evaluation of existing products by generating knowledge about safety and efficacy in pre- and post-marketing settings and serve to inform medical decision making and medical product development. Although well-designed and -implemented clinical trials can provide robust evidence, a gap exists between the evidence needs of a continuously learning health system, in which all medical decisions are based on the best available evidence, and the reality, in which the generation of timely and practical evidence faces significant barriers.

Large Simple Trials and Knowledge Generation in a Learning Health System is the summary of a workshop convened by the Institute of Medicine's Roundtable on Value & Science-Driven Health Care and the Forum on Drug Discovery, Development, and Translation. Experts from a wide range of disciplines--including health information technology, research funding, clinical research methods, statistics, patients, product development, medical product regulation, and clinical outcomes research--met to marshal a better understanding of the issues, options, and approaches to accelerating the use of LSTs. This publication summarizes discussions on the potential of LSTs to improve the speed and practicality of knowledge generation for medical decision making and medical product development, including efficacy and effectiveness assessments, in a continuously learning health system.

Large Simple Trials and Knowledge Generation in a Learning Health System explores acceleration of the use of LSTs to improve the speed and practicality of knowledge generation for medical decision making and medical product development; considers the concepts of LST design, examples of successful LSTs, the relative advantages of LSTs, and the infrastructure needed to build LST capacity as a routine function of care; identifies structural, cultural, and regulatory barriers hindering the development of an enhanced LST capacity; discusses needs and strategies in building public demand for and participation in LSTs; and considers near-term strategies for accelerating progress in the uptake of LSTs in the United States.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  7. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!