 |
 |
|
 |
||
Print 
Download Reader 
|
|
|
|
||
Print Download Reader
 |
|
|
Secretary’s Advisory Committee on Human Research Protections March 27 and 28, 2008 Arlington, VA Minutes THURSDAY, MARCH 27 Welcome and Opening Remarks Dr. Tilden, SACHRP Chair, read the charter for SACHRP. He recognized two new ex officio members, Elaine Gilby from the Social Security Administration (SSA) and Marjorie Bailey from the Department of Homeland Security. He told the committee that three SACHRP members were unable to attend due to family issues. The Chair acknowledged Francine Romero’s service, noting that her term will expire before the next SACHRP meeting. He said SACHRP had benefitted from her tenure in many ways, in particular from her training and background in both anthropological and public health research, her long-time experience with human subjects protection, and her deep commitment to the ethical conduct of research in Native American populations. He extended gratitude and best wishes. Minutes for the previous meeting (October 29 and 30, 2007) were approved unanimously. However, an editorial note on p. 11 will be removed in the final minutes. Report on Issues Dr. Pritchard expressed appreciation to Dr. Wright, Principal Deputy Assistant Secretary for Health, for devoting his morning to attending the meeting. He also thanked SACHRP members, ex officio members, and members of the public for their input. Dr. Pritchard highlighted the following recent OHRP news and activities:
The Acting Director told SACHRP that a New York Times editorial (Gawande, 2007) raised questions about how quality improvement activities should be handled by the Federal oversight system. Since then, the entire office has been extremely busy and has engaged in conversations with people within and outside of health and human services. Dr. Genel asked about the status of the project cited in the editorial. Dr. Pritchard explained that the case is closed and Johns Hopkins took actions to address the original problem, which related to decisions about exempt and nonexempt research activities. A current followup study now in progress no longer meets the criteria for human subjects research. Remarks Dr. Wright represented the new Assistant Secretary for Health (ASH), Dr. Joxel Garcia, who hopes to meet SACHRP members at a future meeting. Regarding the current controversy related to quality improvement, Dr. Wright assured SACHRP that while the Department is strongly committed to quality assurance and quality improvement in hospitals, it is equally committed to ensuring human subjects are protected. He said the Department was fortunate to have SACHRP’s expertise and recommendations, adding that he anticipated that SACHRP’s charter would be reauthorized in October of 2008. Quality Assurance, Quality Improvement and Health Services Activities Panel
The Chair read the charge given to the panel, which follows:
Note: The SACHRP meeting of October 4-5, 2004 included a panel discussion of issues related to the subject of what should and should not be considered research in reference to human subjects protection. The panel included remarks by representatives of the National Committee for Quality Assurance (NCQA) and the Agency for Healthcare Research and Quality (AHRQ). All SACHRP minutes are available at the OHRP Web site: www.hhs.gov/ohrp/sachrp/index.html. Remarks by Donald Berwick: The Nature of Quality Improvement Dr. Berwick noted that there have been significant societal changes since the quality improvement (QI) function was established. Activity in the area of QI has increased for a variety of reasons, including the attention given to the importance of QI in two critical reports: the National Academy of Sciences report, To Err Is Human: Building a Safer Health System (Kohn, Corrigan, & Donaldson, 2000), and a 2001 report by the Institute of Medicine (IOM), Crossing the Quality Chasm: A New Health System for the 21st Century (Committee on Quality of Health Care in America, 2001). The former report called attention to the sometimes fatal consequences of errors in patient care, while the IOM report expanded on specific areas in need of improvement. These include:
The speaker stressed that the source of problems is not a defective or uncaring work force; rather, health professionals are trapped in defective systems. A relatively simple example is the use of similarly designed bottles of racemic epinephrine and Vitamin E, a design error that resulted in the death of infants. However, in most cases, the causes of errors are less obvious. Much as a golfer improves his or her swing by making changes and observing the results, continual learning is needed to improve health care. Process improvement includes:
The basic steps of “plan, do, study, act” encapsulate the process through which continual learning occurs. All improvement is “local” and must be adapted to a specific context. However, the speaker argued, it does not make sense for every hospital and system to have to duplicate the same learning process. When a hospital learns how to remedy a common defect and improve performance, it is clearly in society’s interest for this result to be shared by publishing findings. The alternative is avoidable harm and continued ignorance. Examples of QI efforts cited by the speaker that illustrate this principle include a 70-hospital system in the Midwest, Ascension Health, that was able to reduce cases of ventilator pneumonia to nearly zero; the Organ Donation Collaborative, a network of over 100 hospitals that has saved thousands of lives by increasing donation rates; the Indian Health Service (IHS), which is succeeding in reducing blood sugar rates among Alaskan natives and American Indians; and the Institute for Health Care Improvement, which has been able to prevent Central Line Associated Bloodstream Infections by applying specific evidence-based techniques. The speaker argued that such attempts to improve quality of care are not research, but rather represent the “stewardship of human systems.” Healthcare providers must engage in these activities or be in dereliction of their duties; they are a professional obligation that is part of the day-to-day work of responsible organizations. A key question is the issue of “informed consent” and what it means in the context of health care. It is important to consider whether patient care is being manipulated for the benefit of someone other than the patient. He suggested that every hospital that uses QI strategies should have a sign in its lobby that notifies the patient that the hospital makes “continual, informed changes in its processes of care, based on current and new science, to improve safety, effectiveness, patient-centeredness, timeliness, efficiency, and equity.” The sign should also inform patients that the hospital continually measures its results and compares them to results achieved by others. The hospital should also display data on results achieved over time. Dr. Berwick saw QI as primarily a component of proper management, not the creation of new knowledge from research. He stressed that unlike researchers, clinicians and health care organizations have an obligation to improve quality, and he held that the review mechanisms used for human subjects research should not replace the function of proper organizational management. He emphasized that the difference between QI and research does not lie in publication and sharing results, which are essential to both kinds of activity. Similarly, both require measurement. The level of risk of the proposed activity may be relevant in determining the level and type of oversight required; however, hospitals and clinics generally can be trusted to manage QI endeavors as well as they manage any other endeavor in which they are engaged. While this effort toward continual improvement must be ethically managed, and while there will be borderline cases in which QI and research must be distinguished, QI in general should not be treated as human subjects research. If QI efforts come under the purview of IRBs, the speaker claimed, the requirement would “slow and potentially stop health care improvement in a nation that badly needs it.” This, he said, would amount to asking permission to learn. Patients would surely ask, “why did you do this in the name of protecting us?” Dr. Berwick suggested that OHRP do the following:
Remarks by Brent C. James: Managing for Ethical Healthcare Delivery Dr. James explained that Intermountain Health Care, an integrated delivery system for which he is the chief quality officers, includes 23 hospitals that employ 550 physicians. Dr. James explained that Intermountain interacts with patients in three fundamental ways. These include:
He argued that core ethical principles – autonomy, beneficence, non-maleficence, and justice – cross all these categories of activities and apply equally within each. The speaker emphasized that organizations that deliver health care are required by professional standards (for example, those of the Joint Commission on Accreditation of Healthcare Organizations [JCAHO]) and explicit regulatory requirements (for example, those of the Centers for Medicare and Medicaid Services [CMS]) to oversee and manage ethical practice in all these areas through quality assurance and quality improvement – an obligation that has been extended over the last 20 years. The speaker maintained that the traditional belief that an ethical clinician-patient relationship guarantees "best care" is scientifically untenable in today's complex care delivery environment. We now have over 40 years of research, he said, that shows huge variation in care delivery, high rates of inappropriate care, unacceptable rates of preventable injury and death, and massive waste (which Dr. James, going further than Dr. Berwick, estimated as over 50 percent). Waste reduces access to care, and the speaker believes that “we wouldn’t have an insurance crisis if we dealt with it.” The speaker cited the definition of health care operations cited in the Health Insurance Portability and Accountability Act (HIPAA):
Dr. James distinguished between the functions of quality assurance and quality improvement. Quality assurance, he explained, uses implicit or explicit criteria to assess ethical conduct and outcome performance one case at a time through case-by-case peer review. Classic examples include credentialing and privileging of health professionals, surgical case review, mortality and morbidity conferences, infection control, and informed consent, including treatment-specific informed consent for high-risk surgical procedures or medical interventions. In contrast, quality improvement tracks empiric care delivery performance across groups of similar cases. It is also known as clinical epidemiology because it uses measurement techniques derived from traditional epidemiology but applied in a different setting. Quality improvement (QI), he said, is primarily about care delivery management, not the creation of new The speaker stressed that it is not possible to tell whether an activity is research by the investigator’s stated intent, by its funding source, by whether or not findings are published, or by the measurement techniques used. He believes that the only differentiator that works consistently is the presence or absence of patient conflict of interest. The focus is usually on implementing what is known through expert consensus rather than discovering new procedures. "Clinical epidemiology" uses epidemiologic measurement methods to implement evidence-based practice systematically. However, its methods take into account not only the level of evidence backing the use of specific procedures but also factors related to the physical and information environment in which health care is delivered and the many complex social structures that impact care. QI efforts do not attempt to step between the clinician and patient, but focus on improving patient care by implementing “what we know works best.” QI, he said, is primary about care delivery management – not the creation of new scientific knowledge. Both research and QI do generate new knowledge – though on a different scale, and with different ethical considerations. Turning to the issue of how QI efforts can best be overseen, Dr. James distinguished between “prevent” controls and “detect” controls:
The speaker argued that “detect” controls are appropriate for QI. Sophisticated mechanisms can be developed to monitor performance. For example, the Intermountain Healthcare system holds about 40 investigations and takes 2-4 actions in response to violations each month. He emphasized the need to adapt the understanding of QI vs. HSR to meet the circumstances of a new generation of care. Remarks by Carolyn M. Clancy: Quality Improvement Research The Agency for Healthcare Research and Quality (AHRQ) has the following mission: to improve the quality, safety, efficiency, and effectiveness of health care for all Americans. Created in response to research on the variations in the delivery of medical care by Rand Corporation and others, the agency has no “silver bullet” to accomplish this critical aim. Currently, health care quality is estimated to be increasing by a “modest” – some would say “glacial” – pace of 2.3 percent annually across all settings and populations, while health care costs are rising by approximately 6.7 percent annually. Because of this disconnect between price and quality, some employers are moving their work out of the country. Dr. Clancy defined “quality” as the right care for the right person at the right time. The AHRQ’s National Healthcare Disparities Report (2007) documents, however, that disparities in health care quality and access are staying the same or increasing. The agency uses identified measures to track the status of disparities over time. The 2007 report found improvements in six of the core measures used but no change in the remaining nine. She stressed that concerns about equity in healthcare should be considered “urgent” and addressing them is well worthy of public investment. Describing the confluence of research and QI as intersecting circles, she emphasized the critical need to learn “what works.” Previously, QI campaigns were “patched on” affairs (largely for the benefit of JCAHO) with relatively little meaning. Few hospitals had the capacity to collect meaningful data. Now, however, many systems have made investments in quality. Much, but not all, of QI research should be considered outside the domain of the Common Rule, but some clearly does fall within it. QI research is a complex topic. Until 20 years ago, most researchers thought that if they published something saying “this works,” it would then happen. We now understand this is not the case. QI efforts gained urgency after studies demonstrated substantial and important variations in the delivery of medical care, with subgroups such as the poor, less educated, and racial and ethic minorities at particular risk of receiving poor quality care. To be effective, practitioners not only need to be aware of best practices, but also to adapt them for local use. QI interventions may change over time and from site to site. The speaker described QI research as “not an intervention in the way we understand clinical interventions.” She said that such research:
The Federal government has a major stake in the process of translating innovative practices that may improve health care into interventions that will work in specific contexts. Translational steps include (1) translation to patient-level interventions; (2) testing through human subject trials; and (3) taking the procedures to scale so the right thing is done by default rather than exception. The time frame for the translation process is unacceptably long. For example, Lee (2007) shows that it has taken 25 years since a landmark trial to make the provision of beta blockers routine. Since 1993, AHRQ (often in partnership with NIH, VA and others) has generated research on such topics as treatment for cancer, diabetes, asthma; IT applications in health care; patient safety; and models of chronic care. A recent review found that that we have much to learn about translating what we know into routine practice. (See the following site for findings: http://www.ahrq.gov/clinic/epcindex.htm#quality.) Provisionally, the speaker defined a QI intervention as:
In contrast, she proposed this definition of implementation research:
The speaker cited numerous examples of the promise of QI and the dramatic results it can achieve. She pointed to the previously cited study by Pronovost et al. (2006), which was able to reduce Central Venous Catheter (CVC) Associated Bloodstream Infections (CABSIs). In addition, she cited the Northern New England Cardiovascular Disease Study Group, in which participating hospitals pay dues to a nonprofit organization to help them maintain registries, databases, and data collection tools to track outcomes and help develop risk-adjusted models of care. Many professional organizations now develop registries to help practitioners improve work. Partners in Care (PIC), a collaborative QI effort, was able to document “dramatic” results in the treatment of depression that eliminated disparities in care (Wells et al., 2004). To move forward in QI , she said, there is a need for better theoretical definitions that relate the intersecting spheres of QI and other research. The field needs appropriate constructs, frameworks, and methods appropriate to address QI questions; methods for synthesizing results; resources, researchers, and research participants (including policymakers, delivery systems, and providers); and funding. She stressed that the area of overlap between QI and research is important and potentially fruitful. To make progress, however, it is important to consider the need to improve practice in small hospitals that do not ordinarily participate in research, but do want to improve the quality of care they give their patients. Dr. Clancy stressed that what is at issue here is of importance not only to the research community, but to small hospitals that may not know what “IRB” stands for. She concluded:
Remarks by Christine Grady: QA, QI, and Health Services Activities: Ethical Challenges Dr. Grady explored the question of how QI is similar to or different from human subjects research (HSR) from an ethical perspective. She observed that QI and HSR are similar in that both involve systematic investigations that aim to improve care and/or understanding:
As QI has become increasingly sophisticated, apparent similarities between the two activities have increased. Examples include:
She pointed to the claim that there are differences related to the “moral imperative” underlying the two activities. Some have claimed that institutions have a moral obligation to improve quality. She noted, however, that even if there is an imperative to improve quality, this does not mean that every QI activity is necessarily good — simply that it is part of the aim of each QI project to improve the quality of care. At the same time, she acknowledged that HSR is not “totally optional,” but is also driven by a societal moral imperative to improve the understanding of human health and how to prevent, diagnose, and treat human illness. Dr. Grady also considered distinctions in the area of patients’ interests. She noted that some argue that the goal of QI — improved care — is consonant with the interests of patients in receiving quality care. The need for oversight in HSR arises from a recognition of the potential for the overall goals of research to diverge from the interests of the subjects. She added, however, that HSR can converge with the interests of patients, for example, the interest of patients with a certain disease or condition in receiving a safe and effective treatment are not necessarily divergent from the interests of a researcher in finding a safe and effective treatment for the same disease or condition. While some would argue that there are differences between the two in terms of whether findings are viewed as generalizable vs. intended primarily for local application, the speaker felt that the findings of good QI activities are generalizable to some extent, and much depends on how “generalizability” is defined. Significant QI findings should be communicated and applied to a larger group of patients; in fact, there may be an ethical responsibility to disseminate information germane to improving care. Publication of findings should not be used to distinguish the 2 activities. The issue of whether the two activities differ in regard to risk is, she said, a “complicated subject we could talk about for a long time.” In both types of activity, there is the possibility of risk or burden undertaken for the benefit to society. Sometimes a patient participating in a QI effort in a local setting benefits at the moment; other times, it is the person in the same bed the next day who benefits. The speaker concluded that trying to find a line or clear conceptual divide between QI and HSR may not be possible. Because of this, both require oversight and attention to protecting the individuals involved. However, a key question is whether that oversight and the standards that define the oversight process must be the same for both. The speaker concluded that QI activities probably should not fall under the current regulatory framework for HSR. She noted that the process currently in place for HSR is often burdensome and cumbersome, even for research. Its focus is on compliance, and the process is often stringent and inflexible. Further, IRBs are often unfamiliar with QI goals and methods. Consequently, they may delay or prevent important initiatives. To avoid the delays associated with IRB review, sometimes people design projects so they can call them QI and avoid the process — an undesirable outcome that can result in lower levels of rigor in terms of methods and statistics. Other issues with the review process include unjustified variation in decisionmaking from institution to institution, including differences related to what constitutes expeditable research; the inability of the existing review process to accommodate the frequent changes that may be associated with QI initiatives; and what the speaker called an “over-emphasis” on informed consent, especially the use of long and complicated documents that are not accomplishing the main purpose of the consent process. As Cretin et al. observed in a 2000 article: “Quality improvement would be discouraged if thoughtful innovation and measurement cycles required regulatory review or informed consent.” What is needed? The speaker emphasized that both QI and HSR need oversight. However, a constructive oversight system (or systems) would be more efficient, less fragmented, less burdensome, and designed to help promote important progress (toward improved quality and useful knowledge) while providing appropriate protections. Clear organizational authority and criteria for determining what kind of review is needed. She called for:
She ended by noting that we may need to engage in quality improvement of the process of IRB review and regulatory compliance. Remarks by Nancy Neveloff Dubler: Process of Quality Improvement: Informed Participation and Institutional Process Reinforcing other speakers, Ms. Dubler contended that QI is morally mandatory for professions who treat patients and for the institutions in which they work. This obligation derives from individual professional ethical obligations and from organizational ethics analysis and the notion of institutional moral agency. She said institutions have an ethical obligation to implement QI findings and argued for a “nimble system” to oversee QI. Contrasting QI and HSR, she noted that research is not morally mandatory for institutions or for human subjects, who are morally free to consent to or to refuse participation. She contended, however, that patients are morally required to participate in QI – a responsibility that derives from possible immediate benefit to oneself and the responsibility to benefit others. The obligation of the physician and organization to give the best possible care would be impossible to effectuate without the participation of patients. From a moral perspective, in short, it is mandatory for institutions, physicians, and patients to participate in QI because it is part of the social contract of medicine that caregivers “do no harm” – an obligation that implies the need to improve care as the knowledge and tools that facilitate improvement are developed and made available. QI has been defined as “the systematic, data-guided activities designed to bring about immediate improvements in health care delivery in a particular setting” (Lynn et al., 2007). It must allow for a rapidly unfolding process in which an approach is piloted, revised, retested, and implemented. Ms. Dubler described QI as having the following elements:
QI is similar to research in that both endeavors involve human participants; generate a question through empirical or systematic inquiry that data collection is designed to answer; propose a set of outcome measures that will support a proposal; test solutions; and involve critical evaluation of data. Ensuring that any particular QI project meets applicable ethical standards requires attention to following:
The notion of informed consent does not apply to most QI projects. It would affect data and delay implementation of findings if informed consent were to be required. [Some projects that require patient cooperation would require informed consent]. Exceptions to the practice of informed consent are already in place in research. The present federal regulations permit an IRB to approve research without the provision for informed consent when the following are in place: [45 CFR §46:111 (d)]
“Informed participation” is an alternative concept that includes the following elements:
In conclusion, Ms. Dubler emphasized that QI should not be subject to review as if it were research, but should rather be open, transparent, and built into the culture of the medical center. Effective oversight requires a structure that is intellectually, practically, morally sympathetic to QI. DISCUSSION SACHRP members expressed uniform appreciation for the presentations, commenting or raising questions in a variety of areas. Informing and involving patients. Various SACHRP members observed:
Risk and harm. A SACHRP member wondered whether human beings are known to have been harmed by QI improvement activities, and if so, under what circumstances. The member suggested that answers could help see what kind of oversight is warranted. Dr. Powe queried whether harm could result from an inherent COI between healthcare managers who are concerned with cost containment as well as providing quality care. Ms. Dubler said she “worries a lot” about this issue. Her medical center, Montefiore, now has a “complex cases” committee that is tasked with examining issues related to patients’ length of stay and ensuring the system is transparent. That said, she added that she has never seen a QI intervention that caused harm. In general, she said, the notion of harm is overblown in research and QI and underacknowledged in clinical care. Dr. James pointed out that a modern view of efficiency and QI is that they are by no means opposites; both can be addressed at once. He gave the example of a new emergency room that opened and immediately reported an increase in average wait times from 14 minutes to 2 hours. The problem was solved through a monitoring system in which people were pulled together to try to understand why the problem was occurring and test different strategies until “we got it right.” Delivering good care, he said, requires a manager who is able to “steer the car.” Dr. James added that of the 1800 projects in the Intermountain Health Care data base, he can recall none that have resulted in harm. However, QI efforts typically target areas in which the performance has been poor and try for improvements. He says the organization has sponsored at least 120 projects for which the findings probably should be shared, but it has kept its focus on serving its patients and making sure the network is performing at peak level. Dr. Clancy said she knows of at least one study in which a QI intervention led to harm; she has also heard that bar coding on medications can have unintended consequences. Dr. Berwick added that “anytime people try to improve anything, there is a potential for trouble.” However, it is important to keep trying to achieve improvement in thoughtful and systematic ways. Oversight of QI activities. Dr. Genel was in agreement that QI and HSR have differences that can be used to determine the type of oversight appropriate to each activity. While QI does require oversight, he held that this should not be the same oversight used to supervise HSR. He added that “we are doing something wrong when the IRB system gets in the way of this type of work” and need to fix it. A SACHRP member said a “fundamental issue” was how to set up a structure capable of seeing what trajectory a given project should follow (i.e., whether it is QI or HSR). How can parameters be set to guide decisionmaking, given the fluid relationship between the two? Mr. Nelson saw a “huge amount of overlap” between the two endeavors and was concerned that the decision not fall to someone unprepared to make it. He saw this determination as a complex task that cannot easily be reduced to a practical algorithm. Dr. James responded that in fact the boundary between QI and HSR needed to “collapse”; day-to-day improvement should be normalized as a routine aspect of healthcare management. A key issue is how to give due ethical scrutiny to this function. However, Dr. Berwick held that not all QI requires such scrutiny. He gave the example of a management initiative to take down the wall between the doctors’ and nurses’ room to see if teamwork improved. Such an experiment, he argued, should be undertaken at the discretion of management without the need for oversight. Dr. James suggested that careful review is needed at the stage when the findings of a QI initiative are prepared for publication and recommended for larger-scale implementation. Another speaker hoped for a “sea change” in which organizations and healthcare providers “demand more of themselves,” adding that OHRP and SACHRP can play a critical role. The speaker called for experiments in oversight for projects that fall in the “grey area” between QI and HSR. Another SACHRP member, however, pointed out that certain provisions of the Health Insurance Portability and Accountability Act (HIPAA) would have a bearing on the creation of any oversight mechanism. Dr. Powe wondered whether professional societies have oversight responsibilities that encompass educating their membership on ethical principles that apply when QI becomes organized research. Another speaker suggested it would be helpful to articulate standards governing such activities. An ex officio added that professional societies also have a role in determining when QI findings should become recommended, established practice. An ex officio member referenced “hints” in the preceding discussion of a risk-based approach to grey zone activities. Would the inherent risk of an activity influence the type of oversight needed? Ms. Dubler responded that while American medicine generally is “very risk averse,” no one thinks that QI projects are likely to be harmful. On the other hand, some risk managers use IRB approval as a protective screen even when it is clearly inappropriate. Dr. Berwick suggested using the word “burden” rather than “risk,” adding that inaction has its own burden. An American should not, for example, experience harm from an avoidable central line infection. The risks of inaction must also be taken into account. Health disparities. A SACHRP member asked how QI could help ameliorate some of existing disparities in health care. Dr. Clancy responded that black-white differences in access to care have diminished in response to certain QI interventions, though results are less impressive for others. Better data are needed systemwide so that the application of useful findings can be accelerated. HSR vs. QI. Dr. Powe observed that the QI community has been pushed to make its activities more rigorous so they look more like research. For example, those concerned with improving health care in minorities demand the highest quality evidence available. Some journals that publish QI findings actually want to see IRB-approved studies. Dr. Grady thought it was a “shame” to have to worry about calling a QI project research. She suggested that guidance be created that would at least define what does not fall in the “grey area” in between the two types of activities, defining the features of QI projects that clearly do not require IRB review and those that clearly do. She added that Chief Quality Officers, who are competent to exercise oversight, are very hard to find and recruit. Ex officio Susan Gainer from the Department of Housing and Urban Development (HUD) commented that the agency has sponsored research and development projects for years that do not clearly fall under the research umbrella. For years, the agency has tried to move all such projects into the IRB arena. Listening to the discussion, she said she wonders if this was the wrong approach. HHS role. In response to a question from an ex officio, Dr. Berwick reminded SACHRP that OHRP has the authority to waive jurisdiction for certain types of activities. He added that simply by raising the question and holding this panel, it has helped acknowledge the issues. He said the Centers for Medicare Services (CMS) should be acknowledged as an “obvious actor” in related discussions since it has an abiding interest in assuring that hospitals are exercising appropriate oversight. Dr. Strauss suggested that SACHRP could help suggest a solution to the problem, which he characterized as a need for guidance to help the IRB system respond appropriately to a range of situations and avoid suppressing inquiry. The field should be educated to become aware of the harm that can result from unwarranted “mission creep.”
PUBLIC COMMENT Alan Trachtenberg of the Indian Health Service (HIS) said he had found the guidelines from the Association of State and Territorial Epidemiologists helpful in distinguishing between public health practice and research. He gave the example of a $50 million grant for competitive diabetes projects that adapted one basic approach in multiple ways at different sites. Although the central IRB thought the project was quality improvement, many local IRBs believed it was research. The project was one in which the protocol changed over time to reflect findings, which caused tension. The story, he said, is ongoing. Quoting the familiar saying that “if the only tool you have is a hammer, everything looks like a nail,” Dr. Trachetenberg observed that the Common Rule is “an awfully big hammer.” He said tools were needed that would indicate that something is “not a nail.” Chris Goeschel, Director of Quality and Patient Safety at the of Johns Hopkins Medical Center, commented that the Keystone collaborative project that helped create this opportunity for dialogue was intended to result in improved quality among participating hospitals. The project “pushed the envelope” between evidence and practice. He added that some of the 72 participating hospitals did not know how to design a good QI study at the outset; however, with support, all proved capable of understanding and participating in solid research. Participants are now eager to continue the work of closing the gap between “where evidence exists and practice lives.” He stressed that a means of providing oversight for quality improvement research is definitely needed. Dr. Goeschel urged SACHRP to take on the “larger issue” of collaborative quality improvement projects at a future meeting. He noted that following recent publicity, he received inquiries from many states and countries where clinicians are interested in participating in the work of the Keystone collaborative but are concerned about violating Federal regulations. He hoped that more clarity might be provided in the short term. David Vulcano, vice chair of the Association of Clinical Research Professionals (ACRP) and vice president for clinical research at the Hospital Corporation of America (HCA), saw research and QA/QI as very separate processes that can be part of each other. He suggested that the situation could be represented graphically as two circles with a smaller circle inside each one. The goal of treatment is to help a patient get better, while the goal of research is to create generalizable knowledge. When a physician prescribes a drug to help a patient get better, the physician has an obligation to measure the patient’s improvement. Similarly, when a QI activity seeks to provide improved care to an entire class of patients, it is essential to measure improvement. Clarification of the borders between QI and research is essential, though the speaker admitted he “shudders” at the thought of new regulations. He noted that changes in the reimbursement structure for the Centers for Medicare and Medicaid Services (CMS) have been a major driver of QI efforts; as pay-for-performance policies beginning to impact hospitals, timing is critical. Julie Handman of the Infectious Diseases Society of America (IDSA) saw efforts to reduce infections as part of QI. She emphasized the need to distinguish between research and public health practice such as routine surveillance for pathogens. While this surveillance process has become more rigorous and specialized, like other areas of medicine, she believed that neither publication nor the increasing specialization of methods used makes this critical effort research. She recommended a SACHRP panel focusing on CDC’s existing guidance for this activity. Further, she suggested expanded efforts to educate local IRB members. Denise Daugherty of the Agency for Healthcare Research and Quality (AHRQ) queried about the language needed to appropriately describe what is happening when QI activities are the subject of research and are intended to lead to generalizable knowledge. Several panelists responded. Ms. Dubler said there was a need to “bring light” to areas that might lead IRBs to identify a QI activity as research. For example, she held that randomization or the presence of data sets should not automatically classify an activity as research. Dr. James commented that evaluation of a QI project for the purpose of sharing results in a publication should be distinguished from operations. Dr. Berwick stressed that the regulations give OHRP the ability to classify an activity as research but still say that that activity is not under its jurisdiction. He added that collaboration among organizations also does not make the activity research. John Mather of Unicorn emphasized the need to “tease out” and “clarify” what key words mean (e.g., public health practice and research). Because the scientific method is used in both, this alone provides no distinction between QI and research. He agreed with others that oversight is needed for both functions. In support of Dr. Berwick, Dr. Mather also reminded SACHRP that OHRP has previously declared certain areas to be outside the OHRP process, including oral history. Report of Subcommittee on Issues Impacting Those with Impaired Decisionmaking Capacity (SIIIDR) Summary of public comment on OHRP’s Request for Information (RFI). Julia Gorey reviewed comments received from the public on adults with impaired decisionmaking as research subjects. The RFI was issued jointly by FDA and OHRP. It may be found at: www.hhs.gov/ohrp/documents/20070905.htm. The comment period was between September 5, 2007 and January 16, 2008. The RFI asked six broad questions:
The 53 respondents included 7 professional/nonprofit organizations, 2 stakeholder or consumer groups, 21 private citizens, 16 universities or affiliated medical centers, 2 pharmaceutical companies, and 5 Federal or state agencies, including substantive and extensive comments from NIH. New York State’s response was particularly strong. Reminding SACHRP that the RFI’s intention was to inform the deliberations of SIIIDR rather than preclude its work, she observed:
Other specific areas in which guidance was requested include
All comments received may be requested by writing: Julia.gorey@HHS.gov Other input received. Ms. Flynn thanked staff, committee members, and ex officio members for their support. She explained that in addition to the welcome input received through the RFI, Dr. Strauss facilitated a Town Hall meeting on the topic at the December, 2007 Public Responsibility in Medicine and Research (PRIMR) conference. The Town Hall was attended by 200 people. Comments uniformly emphasized the confusion in the current situation, which leads to variance in practices, variance in the way IRBs understand their role and execute it, and concerns in research community, as well as within the broader health community. People who attended the session were not eager to see a regulatory response but did look for guidance, including clarity in key definitions and assistance in improving practice. SIIIDR continues to seek stakeholder input to inform its process. Introductory remarks. After citing pertinent Federal regulations, Dr. Strauss stressed that while the Common Rule points to “mentally disabled” persons as an example of a group of subjects “likely to be subject to coercion or undue influence” and therefore in need of “additional safeguards,” it provides little direction either on how to identify this group of subjects nor on what additional safeguards are needed. The committee conceives of its efforts to address this gap in terms of a triangle framed by three fundamental issues:
Dr. Strauss explained that the remarks presented during this meeting will largely focus on issues related to the identification of subjects. In addressing this issue, SIIIDR is aware of the need for balance between protecting individuals and creating unnecessary burdens related to IRB review. The subcommittee formally requested guidance:
He stressed that the subcommittee is not writing guidance, but rather writing recommendations for guidance. It hoped and envisioned that all the Common Rule agencies would be included in a collaborative process leading to the acceptance of clarifications. The subcommittee proposed the following:
DISCUSSION Dr. Bankert asked whether the term was new or intended to replace an existing term. Dr. Strauss explained that the new term is intended to capture the group of people who need additional safeguards. It is more specific than the term “impaired decisionmaking.” Dr. Shore commented that NIH selected the term “consent capacity” because it focused on the specific abilities that are directly relevant to informed consent. Additional discussion points included:
ACTION The recommendation was approved unanimously with the following changes:
Recommendation 2. Dr. Strauss explained that Recommendation 2 is intended to describe what is meant by decision-making capacity, alerting IRBs to the circumstances in which they may choose to consider special safeguards. Noting that “final guidance may benefit from the use of case examples,” the subcommittee proposed the following:
DISCUSSION Preamble. Dr. Lux suggested the word “consent capacity” would be the best term for this context. Dr. Strauss rejoined that the subcommittee wished to acknowledge the fact that investigators will face people with a whole range of impairments in decision-making and cognitive abilities, some of whom have sufficient capacity to make a decision about consent. He agreed with Dr. Powe that the intent was to encompass not only a consent decision, but the entire chain of events that follow as the protocol is implemented. However, Dr. Shore argued that the term “consent capacity” was more appropriate in this instance, since OHRP would not be expected to provide guidance on the clinical issue of decision-making capacity. a. SACHRP members suggested that “a continuum of ability and impairment” implied two axes and was therefore confusing. b. Dr. Bankert questioned the need for the reference to impairment and suggested referring simply to “consent capacity.” Dr. Strauss indicated that wording was intended to apply to the group in need of additional protections. Dr. Marshall found the words, “address the phenomenon of impaired capacity itself,” unclear in the context of the sentence. Dr. Strauss explained that the intent is to ensure that people who have impaired capacity but do not have psychiatric disorders (for example, people with neurological disorders or those in intensive care units) also should be considered as potentially having impaired consent capacity. Members also questioned the reference to “policies” as opposed to guidance. Dr. Strauss said the subcommittee intended to suggest that OHRP’s guidance materials encourage and guide institutions, research sponsors, IRBs, or others to develop policies that show respect for subjects’ rights and welfare by addressing this issue. c. Dr. Lux wondered if the recommendation ought to specifically say consent capacity is “situation-specific” as well as task-specific. However, Dr. Tilden noted that the subcommittee had decided not to bring up the issue of voluntariness by including a reference to the situation; they wanted to focus on intrinsic factors that contribute to or impair the ability to consent rather than extrinsic factors such as the setting (for example, an emergency room or an in-patient service). Dr. Lux withdrew his suggestion. Dr. Strauss added that the subcommittee might want to reference the fact that it decided not to address the issue of voluntariness, an area of which practical ethics concepts are less well developed. Dr. Genel wanted to add wording that points to the possibility of change over time, but Dr. Strauss assured him this was brought out in a later recommendation. d. Wording related to policies was changed to be consistent with section b of this recommendation. ACTION
Recommendation 2 passed unanimously with the following changes:
Recommendation 2. Guidance should provide information for institutions, IRBs and investigators on the nature of consent capacity and its impairment as it relates to research participation.
Specifically:
a. An individual’s consent capacity is not simply present or absent; capacity is best understood as occurring along a continuum.
b. Impaired consent capacity occurs in a wide range of conditions and disease states. To respect the rights and welfare of all research participants, guidance should encourage the development of policies that acknowledge the many manifestations of impaired consent capacity and are not limited to consideration to specific disorders.
c. Consent capacity is task-specific and depends on the nature and complexity of the relevant decision-making process. Therefore, a judgment regarding an individual’s capacity to consent may not be the same for all research studies.
d. In many individuals, impairment in capacity to consent is not a static phenomenon. During the course of a research study, a research participant’s consent capacity may improve, fluctuate over time, or worsen with changes in the individual’s underlying condition. Guidance should encourage policies on consent, the assessment of capacity, and the use of surrogate-based consent procedures to reflect this fact.
Recommendation 3. In presenting the following recommendation regarding the need for assessment, Dr. Strauss conveyed the subcommittee’s concern that IRBs not overreact to guidance by mandating formal instruments regardless of their appropriateness to the situation. The subcommittee proposed the following:
DISCUSSION Preamble. Ms. Gilby wanted to soften the prescriptive start: “research…should.” Accordingly, the preamble was reworded in a manner similar to other recommendations, stressing the need for guidance. a. Dr. Strauss explained that the subcommittee wanted to be clear that the only way to know if a particular individual has impaired consent capacity is by making an assessment (whether formal or informal). This requirement is nowhere in the regulation, but it is pervasive in many international ethical guidelines. There should be some de minimus recognition of each prospective participant’s capacity to consent. In studies in which subjects are unlikely to have such an impairment, that judgment can normally be made as part of the “normal give and take” of the consent process. Dr. Shore questioned the phrase, “should be cognizant of” on the grounds that it implied there had already been an assessment. In the regulations, Dr. Pritchard pointed out, the term “cognizant” is used in reference to a wider category of problems than “capacity to consent.” The sentence was reworded. Dr. Powe was troubled by the phrase “unlikely to recruit,” which seemed vague. The thought was reworded to stress the idea that individuals with impaired consent capacity were not “anticipated” in the subject population. While one SACHRP member thought the import should not be limited to the recruitment process, Dr. Shore countered that these issues do come up before recruitment begins and investigators need to build in appropriate assessments for the study population. b. Dr. Tilden asked how the level of risk affected the assessment of capacity or the importance. Dr. Strauss responded that his institution studies individuals with schizophrenia, for example, the kind of assessment needed is determined based on the acuity of the patient’s illness, whether they are inpatients or outpatients, and the level of risk of the study. Language in (i) was revised using the word “anticipated” to remove the word “likely” and maintain parallelism with revisions made to (a). Dr. Strauss clarified that the subcommittee included the reference to documentation to remind the reader that it should be considered; however, it is not necessary in all instances. He added that the baseline requirements for documentation will vary from IRB to IRB. In regard to the reference to “clinical assessment” in (ii), a SACHRP member asked, “If an anthropologist were doing study on a population with impaired decisionmaking, would the word ‘clinical’ be appropriate?” The word was removed when Dr. Shore pointed out that many newer assessment instruments were designed for administration by lay interviewers. Dr. Strauss also provided the example of a minimal risk community-based study of people with mild cognitive impairment in which research assistants are trained to determine who is able to answer the questionnaire and consent. The relevance of “ordinary clinical treatment” was similarly questioned in (iv) and broadened to “ordinary practice or clinical care.” A reference to researchers in (v) was broadened to include staff, since Mr. Nelson pointed out that qualified and trained staff may be the ones to obtain consent. Dr. Strauss emphasized that the intent is for the reference to include all those responsible for the consent process. c. Dr. Strauss elaborated on the type of “enhancements” to the consent process that might be appropriate. Examples could include a cover page for the consent form, an outline, a slide show, or a multimedia presentation. He noted, however, that empirical research continues to shed light on what might be appropriate in a given situation. d. Dr. Tilden suggested using the word “robust” to describe approaches rather than the word “intensive.” Dr. Strauss demurred, suggesting that the two words had different meanings. e. Dr Powell raised the question of the frequency with which disclosure of relevant information, reconsent, and reassessment of consent capacity should be reconsidered. Dr. Strauss responded that this would depend on the nature of the study and the subjects, as well as local practices. Ideally, he explained, when deterioration in consent capacity is considered, ongoing participation would be enabled through an advance directive or appointment of a surrogate decisionmaker. The committee decided to add a sentence that spoke to the issue of frequency. ACTION The committee approved Recommendation 3 with the following revisions. On the second day of the meeting, it returned to this recommendation to make minor editorial changes and reapprove the recommendation as it stands below. Changes made on the second day also included the use of the word “requirements” in (e), strengthening the importance of the ethical standard implied.
PUBLIC COMMENT John Mather expressed concern that sections (a) and (e) of Recommendation 3 may place a burden on IRBs when studies come up for continuing and annual reviews. He commented that the Subpart A subcommittee might wish to consider the implications of a possible change in mentation for continuing review. He also suggested that while OHRP has traditionally encouraged institutions to run their own OHRP-supported educational programs, it may wish to consider using these programs to “get the message out” in certain important areas. FRIDAY, MARCH 28 Opening Remarks The Chairman welcomed attendees and provided a brief overview of the agenda. He noted that Dr. Felix Gyi, the former Co-Chair of Subpart A subcommittee, will remain as a member of the Subpart A subcommittee but Dr. Liz Bankert, a SACHRP members, will assume the role of Co-Chair. Report of the Ad Hoc Committee: Proposal for Recommendations Based Upon SACHRP Panel “Issues Involving Research in Disaster Settings” The Chair reviewed four major points brought forward by members of panel that addressed SACHRP on the subject of issues involving research in disaster settings (October 30, 2007):
An ad hoc committee formed at the fall, 2007 SACHRP meeting developed the following recommendations:
DISCUSSION Consideration of ethical principles. Dr. Kirchner stressed the importance of ensuring that local and State planning be done in keeping with ethical principles. Dr. Tilden responded that the entire recommendation is aimed at encouraging proactive consideration of this issue. Speed and timeliness. Dr. Strauss said the recommendation should stress the issue of speed and timeliness. The problem lies not in the possibility that research may not be reviewed, but rather in how the review process is tailored to the specific demands of post-disaster research. Dr. Lux agreed. Based on the experience of the Environmental Protection Agency (EPA), he said there are organized, robust mechanisms to incorporate local considerations into the ethical review, but systems are unable to respond under the time constraints required by “acute” disaster research. A reference to the need for a “time-sensitive” process was added to reflect this concern. Coordination. Dr. Carr commented that the National Institute of Mental Health (NIMH) has developed initiatives to address these issues in advance of disasters. Dr. Carr expressed concern that the existing wording implied there was unethical research taking place, when the focus should be placed on how the ethical review process is coordinated. Dr. Tilden responded that “responsiveness” is an issue because of timing considerations, noting that the panel has heard that disaster victims are sometimes “surveyed to death” because of lack of coordination. The scientific aspects of research have been reviewed, but there are still issues related to how it will be conducted. Dr. McNeilly noted that in disaster response, “the shots are called by the States.” The role of the Federal government and the State-level perspective should be delineated. Dr. Lux said that EPA’s experience suggests that State and local health departments have resources and are familiar with research; however, there are difficulties coordinating in a timely way to facilitate research following a disaster. Dr. Genel summarized by saying the coordinated process ought, in itself, to be timely. Dr. Strauss called for language referring to the lack of “workable mechanisms for prior review and coordination.” Importance of research. Dr. Strauss proposed added language to stress that in the aftermath of disaster, research involving human subjects is of critical importance. Prioritization. Dr. Kirchner asked how priorities were identified when many proposals for research must be considered. Members agreed an understanding of ethical aspects of prioritization is encompassed in the recommendation, particularly in the call for criteria related to State and local coordination and oversight of research activities. Dr. Kirchner was especially concerned that research activities not interfere with salvage and rescue operations (a concern raised by the panel that addressed SACHRP on this topic). Accordingly, a reference was added to the need for sensitivity to the potential for research to adversely affect disaster management. Understanding the existing situation. Members agreed that it is important to confirm and supplement the panel’s observations with an assessment of the current situation: “The first step might be understanding what’s out there and how it works.” Language was refined to call for an “assessment of the current ethical review process and if indicated the feasibility of developing additional processes…” ACTION The proposed recommendations were approved unanimously with the following changes:
Diversity and Clinical Trials Dr. Tilden reported that the Subpart A Subcommittee (SAS) has made a preliminary review of issues presented by a panel that addressed SACHRP on the topic of diversity and clinical trials, concluding that aspects of this issue are relevant to the Common Rule and counterpart FDA regulations. A working group is developing recommendations for SAS consideration and possible forwarding to SACHRP. Recommendation for Modification of the Common Rule to Require Tribal Approval for Research The Chair also presented the following recommendation at the suggestion of Dr. Romera, a SACHRP member unable to attend the meeting. He suggested that it go to the Subpart A Subcommittee for consideration:
ACTION SACHRP approved the proposal that the concern and draft recommendation be forwarded to the Subpart A Subcommittee. No changes were incorporated other than changing (e) to (f) since there is already an (e) at that place in the regulations. Subpart A Subcommittee (SAS) Report Mr. Nelson announced that Elizabeth Bankert, a SACHRP member, has agreed to serve as SAS Co-Chair. Mr. Nelson gratefully acknowledged the many contributions of Dr. Felix Gyi, the former Co-Chair, to the committee. The Chair added his appreciation. SAS then presented recommendations related to specific exemption categories. He observed that there is considerable confusion within the research community regarding how the categories should be interpreted, adding that this leads to inconsistency in review and can create unnecessary delays in the review process for these low risk studies. Prior recommendations from SAS that apply across all exemption categories have already been approved by SACHRP. The recommendations presented at this meeting address specific categories of exempt research. All of them request OHRP to clarify the use of these categories through guidance and examples. Exemption Category 1 This category includes:
SAS proposed the following recommendations:
Recommendation 14, Mr. Nelson explained, addresses the specific concern that the Department of Education may interpret this exemption differently from OHRP. DISCUSSION Nontraditional settings. In response to a question from Dr. Strauss, Mr. Nelson clarified that SAS believes the exemption has not been used in areas where it should be applicable. He added that training and education occur in many varieties of settings, within walls or, as Dr. Powe suggested, in open spaces such as golf courses. Dr. Tilden suggested there was need for a flow chart that clarified the order in which questions should be asked (e.g., is it a normal educational practice? Is it a nontraditional setting?). Normal educational practice. Mr. Rodamar observed that some IRBs seem to believe anything that happens in a school must be exempt. He was surprised, however, that SACHRP members at a previous meeting seemed to believe a survey of students regarding substance abuse behavior would not be exempt or routine. He said there was much evidence of a large “gray zone” in this area. Mr. Rodamar asked whether a self-appointed educator who is actually using a Web site to collect information for financial gain might be considered exempt. Mr. Nelson emphasized that an IRB would need to assess the investigator’s intended use of information. Dr. Strauss observed that the nature of routine vs. nontraditional educational practices changes over time, which implies that OHRP would need not only to provide examples, but also to define the characteristics of routine and established practices. Dr. Tilden felt the language of Recommendation 12 conflated practices and settings: “The guidance also should provide the current OHRP interpretation of “normal educational practices” as not being restricted to only traditional settings.” He asked for new language that referred to “traditional approaches and traditional settings.” Cultural differences. Dr. McNeilly questioned whether all cultural differences could be considered acceptable practices on that basis. An example would be corporal punishment. Dr. Tilden said it would be helpful to understand how the funding agency and OHRP would view such a decision and whether they would consider it “normal educational practice.” Dr. Strauss noted that we are asking OHRP to develop examples to clarify its position. Community-based participatory research. Dr. Marshall was pleased that Recommendation 13 addressed this issue. Mr. Nelson observed that the issue was a broad issue that is not associated specifically with education. Since it may be exempt on grounds on than this specific exemption category, language was broadened to request clarification of when such research is “eligible for exemption.” Dr. Genel suggested the same recommendation refer to the “development” rather than the “design” of the study plan. Dr. Powe proposed language clarifying that community-based participatory research is an example of action research rather than vice versa. Dr. Marshall agreed, noting that either term implies a particular orientation to a problem that includes the involvement of and interaction with a population in a community setting, including working with a community member on the study design. She thought the term “action research” is more vague. Mr. Nelson added that definitions may vary for this relatively new genre. Members discussed eliminating the term “action research,” but Dr. Powe and Dr. Trachtenberg felt it had relevance in the educational context. ACTION Recommendations 12-14 were approved unanimously with the changes underlined below: Recommendation 12
Recommendation 13 Other Examples. Guidance should clarify whether and when”action research” (for example, community-based participatory research) is eligible for exemption. These types of activities usually are designed to contribute to generalizable knowledge; however, the development of the study plan (protocol) does not follow the “traditional” model. The study plan develops as part of the interaction with the community and/or subjects…. Guidance should use examples to illustrate when actions or community-based participating activities do not meet the definition of research, fit under one or more exemption categories, or are eligible for IRB review under an expedited review procedure. Exemption Category 2 This category includes:
SAS proposed the following recommendations:
DISCUSSION Public behavior. In regard to Recommendation 15, Dr. Powe wondered whether a classroom held in an open setting such as the mall or a park would be considered “public behavior.” Mr. Nelson suggested that something that does not require special permission to observe, such as class members looking at the Lincoln Monument, would come under this exemption in his opinion. Dr. Tilden expressed concern that the categories might be manipulated, since they are examined independently from each other. For example, students might be taken to an outdoor setting that is laden with drugs to observe their behavior. Dr. Strauss noted that there is a “reasonable expectation of privacy” (a concept documented in the guidance) even in some settings open to anyone, such as a meeting of a 12-step group or an online community of cancer survivors. A parenthetical reference to this topic was added. Audiotapes. Dr. Lux wondered whether there were methods available to investigators that would allow them to identify specific individuals on audiotapes. Mr. Nelson rejoined that this would depend on the availability of information other than the audiotape itself to match the voice to a particular individual. This is analogous to having a DNA sample without a way to match it to a particular person. Dr. Tilden added that analysis of whether audiotapes include identifiable data would differ in HIPAA and the Common Rule. Mr. Nelson contrasted the example of anonymous recordings of one hundred drug dealers, as opposed to a small sample of alleged drug dealers who were interviewed because their names were in the newspaper. Dr. Trachtenberg observed that even an anonymous recording might be seized and used by law enforcement in a small community where voices might be recognized. Dr. Strauss commented that this type of possibility is very real; he knows a researcher whose data was seized by a foreign government in turmoil in the belief her work had political implications. Dr. Tilden noted that the context other than the audiotape itself would clearly have to be considered, and examples should illustrate this. Dr. Strauss added that in some cases the investigator might be able to reduce the risk in order to allow the research to qualify for exemption. Risk determination. Mr. Nelson explained that there are instances in which research could pose risks even without identifiers, for example, through the harm that could ensue through a particular line of questioning – for example, interviewing a person with post-traumatic stress disorder (PTSD) about his or her experiences. Dr. Strauss gave the additional example of an intensive battery of questionnaires administered to people recently diagnosed with terminal illness. He was concerned, however, that some IRBs might inappropriately consider any interaction with a vulnerable population, such as psychiatric patients, to be above minimal risk. ACTION SACHRP approved Recommendations 15-18 with the modifications underlined below. Recommendation #15
Recommendation #18
Exemption Category 3 This category includes the following:
SAS made the following recommendations:
ACTION The proposed recommendations were accepted without discussion or modifications. Exemption Category 4 This category includes the following:
SAS made the following recommendations:
DISCUSSION Dr. Strauss noted that the exemption is understood to allow for access to private information so long as the research does not record or move data from the source. This raises the question of how people are protected who believe their information is private and would be unaware of such uses of their data. Dr. Tilden added that this issue commonly arises; many people do not want medical data accessed for any reason other than their own health care, for example. The question becomes, “what is a legitimate expectation of privacy?” Mr. Nelson noted that HIPAA may close the door on some such uses of data that would otherwise be eligible for exemption. ACTION SACHRP approved recommendations 22-24 without revisions. Exemption Category 5 This category includes the following:
SAS made the following recommendations:
Background for Recommendations on Exemption Category 101(b)(5) Background for Recommendations on Exemption Category 101(b)(5)
Recommendation 25: Definition of “Significant.” When developing new guidance, criterion (4) above should be deleted, because the foregoing recommendation stipulating “not greater than minimal risk” makes this a moot point.
Recommendation 26: Federal and state. Criterion (2) above is interpreted too narrowly and institutions should be able to apply this exemption to public programs supported by state departments and agencies as well as federally supported public programs. OHRP has explained that the broadening of the exemption is not possible within the constraints of the current regulations. When developing new guidance, OHRP should broaden its interpretation to include using the exemption for activities that involve federal flow-through monies to the state public benefit programs. Note that it would need to be made clear that state university research did not fit under an expanded exemption.
Recommendation 27: Federal versus state. If it is not possible to broaden the interpretation of this exemption category, the guidance should specifically indicate that state programs do not fit under this category and list the federal programs with the statutory authority to fit in the category.
DISCUSSION Dr. Kirchner raised the question of whether the reference to minimal risk in Recommendation 25 would provide sufficient protection from intrusions on participant privacy, such as an aggressive interview of a disaster victim whose spouse has just been killed. Mr. Nelson felt it would. However, Dr. Kirchner stressed that intrusions on privacy are increasingly common. SACHRP members voted to delete Recommendation 25. ACTION SACHRP disapproved Recommendation 25. Recommendations 24, 26, and 27 were approved without changes. Recommendations after 24 will be renumbered. Exemption Category 6 This category includes the following:
SAS made the following recommendation:
DISCUSSION Dr. Tilden asked for examples of how this exemption is being used. Dr. Marshall illustrated the category by describing a hypothetical obesity prevention program that researches issues of satisfaction associated with taste. Dr. Lux noted that “wholesome foods” are hard to define. Dr. McNeilly added that any food additives would have to have been shown to be safe in order for this type of study to be exempt. ACTION Recommendation 28 was approved without changes. Other Recommendations
DISCUSSION Mr. Rodamar, an ex officio member representing the Department of Education, commented that the agency frequently faces difficulties with the large number of statutes that address confidentiality. Also, both the Common Rule and the Family Educational Rights and Privacy Act (FERPA) specify elements that must be included in informed consent forms that differ slightly. Researchers can be blind-sided if they are unaware of these differences. Dr. Strauss proposed deleting the introductory words, “consistent application of exempt categories,” in order to ask separately for (1) a joint guidance document and (2) for an attempt to reconcile different exemption requirements. ACTION The recommendation was approved unanimously with the following change:
Need for Revised Regulations on Exemptions Mr. Nelson asked SACHRP to consider whether or not the work presented by SAS on exemptions “closes the book” on the topic or whether there was need for additional work. Specifically, he asked whether SACHRP saw a need for revisions to existing exemptions, new categories of exempt research, or a completely new approach to defining/applying “exemptions.” Dr. Bankert added that some exemptions “seem not even to apply to our world now.” She said SAS had considered whether categories of exempt research could be pulled from the regulations so they could be reworked, similar to categories of expeditable research. Even the word “exempt,” she pointed out, causes confusion. DISCUSSION Dr. Genel suggested requesting public comment on the issue. However, he was not sure that fine tuning the exemption category would address the “dysfunctional system” many believe exists at present. Dr. Strauss rejoined that SACHRP has an obligation to look at complaints and concerns about the functionality of the system, and making recommendations regarding the kind of activities that should and should not be covered under the Common Rule is one way to address these issues. However, he thought a broader discussion of the Common Rule and its application could be planned for another time. SACHRP considered the possibility of a panel to address larger issues related to the Common Rule. As Mr. Nelson put it, it would explore the question, “why is the system broken – if it is?” Dr. Marshall raised the question of how SACHRP would respond to information received from such a panel. Dr. Lux endorsed the need to look at this issue thoughtfully, but at the same time he stressed the fact that the Common Rule has provided enormous benefits and should not undergo change without careful thought. SACHRP agreed to request such a panel. In regard to exemptions, the Chair saw the key question as, “is research in 2008 substantially different form 30 years ago? Has the landscape changed enough to warrant change?” Mr. Nelson suggested that broader input is needed to answer that question. Dr. Pritchard noted that if the Request for Information (RFI) simply stated the agency was contemplating making changes to existing exemptions, it would receive a “fairly random array of recommendations.” More pointed questions may help probe the response to particular kinds of shifts in the way the regulations address exemptions. IRB Membership Rosters Dr. Bankert reviewed the requirement in the Common Rule related to the submission of membership rosters:
Noting that the regulation does not specify a time period for reporting changes in the membership roster, Dr. Bankert reported that SAS committee members found they all reported differently. OHRP guidance (FAQ 5) indicates that IRB registration is effective for three years and must be renewed at that time:
Mr. Nelson stressed that institutions are largely unaware of OHRP guidance and are confused as to when they should report roster changes, with resulting variability in practice. He added that there is an administrative burden for both institutions and OHRP related to filing roster changes, with little gain for human subjects protections. Some institutions update their rosters every time they change, and OHRP reviews every new roster. Nevertheless, because IRB memberships change regularly, the version filed with OHRP is frequently out of date. (However, institutions have reason to maintain current rosters for their own purposes, even in the absence of a requirement to submit to the Federal level.) He noted that the FDA does not require IRBs to be registered (i.e., rosters to be filed), thereby creating discord with OHRP. SAS therefore proposed the following:
DISCUSSION In response to a question from the Chair, Mr. Nelson explained that Recommendation 1 was considered feasible within the short term, while Recommendation 2 is less likely to occur. Dr. Powe wanted to understand the intent of the requirement to report the IRB roster. Ms. Bankert thought the purpose was to ensure the IRB is properly composed; however, OHRP has provided guidance to help assure this. Dr. Powe clarified that the proposal envisions a “snapshot” of IRB membership provided every three years. Dr. Pritchard shared that OHRP is waiting for an opinion from the Office of General Counsel (OGC) as to whether the 3-year reporting period is a legal interpretation. While it is possible that the institution will be more careful in ensuring the IRB is properly constituted if it must report membership to OHRP, he thought this possibility “sufficiently speculative” to be an inadequate basis for preserving the current system and its associated administrative burden. Dr. Tilden wondered whether it would be possible to institute a more user-friendly electronic reporting system in which only changes were reported. Dr. Pritchard responded that a dwindling number of institutions do not report electronically. Dr. Stith-Coleman clarified that it is already possible to submit only the membership change and not the entire roster when using the electronic report system. OHRP is currently enhancing the system to make it easier to identify the change. Dr. Carr asked FDA colleagues whether or not lack of a requirement to submit the roster seemed to result in problems. Dr. Less said the agency spot-checks the roster sometimes but not do so routinely, so he was uncertain. After discussion, wording of Recommendation 2 was clarified to make clear that OHRP is not proposing to eliminate IRB registration, only the submission of rosters. ACTION Both recommendations were approved with the following wording change to Recommendation 2:
PUBLIC COMMENT Dr. Trachtenberg (IHS) commented that he was walked several Tribal representatives through the IRB registration process and they found it user- friendly. Wrap-Up Discussion Dr. Genel expressed appreciation for the work of the Chair. Dr. Tilden reiterated his own appreciation for the efforts of OHRP staff and ex officios to make the meeting a success. The meeting was adjourned at 3:54 pm. References Agency for Healthcare Research and Quality (2007). National Healthcare Disparities Report. AHRQ Publication No. 08-0041. Retrieved April 5, 2008, from www.ahrq.gov/qual/nhdr07/nhdr07.pdf Berwick, D. M. (2008). The science of improvement. JAMA, 299 (10), 1182-1184. Committee on Quality of Health Care in America, Institute of Medicine (2001). Crossing the quality chasm: A new health system for the 21st century. Retrieved April 3, 2008 from http://books.nap.edu/openbook.php?record_id=10027&page=1 Casarett, D., Karlawish, J.H.T., Sugarman, J. (2000). Determining When Quality Improvement initiatives should be considered research: Proposed criteria and potential implications. JAMA, 283, 2275-2280. Cretin et al. (2000). Should patients in quality improvement activities have the same protections as participants in research studies? JAMA 284, 1786-1788. Gawande, A.A. A lifesaving checklist. New York Times. December 30, 2007, p. 8. Retrieved April 8, 2008 from: www.nytimes.com/2007/12/30/opinion/30gawande.html?pagewanted=print Kohn, L.T., Corrigan, J.M, & Donaldson, M.S., editors (2000). To err is human: Building a safer health system. Committee on Quality of Health Care in America, Institute of Medicine. Retrieved April 3, 2008, from http://www.nap.edu/catalog.php?record_id=9728#toc Lawrence, R.S., Mickalide, A.D. (1987) Preventive services in clinical practice: Designing the periodic health examination. JAMA 257, 2205-7. Lynn et al. (2007). The ethics of using quality improvement methods in health care. Annals, 146 (9), 666-674. Pronovost, P., Needham, D., Berenholtz, S., Sinopoli, D., Chu, H., Cosgrove, S. et al. (2006). An intervention to decrease catheter-related bloodstream infections in the ICU. New England Journal of Medicine, 355(26), 2725-32. Lawrence, R.S., Mickalide, A.D. (1987). Preventive services in clinical practice: Designing the periodic health examination. JAMA, 257, 2205-7. Lee, T.H. (2007) Eulogy for a quality measure. New England Journal of Medicine, 357, 1175-1177. Miller, F.G., Emanuel, E.J. (2008). Quality-improvement research and informed consent. New England Journal of Medicine, 358(8), 765-767. National Institute of Health (2008). Research involving individuals with impaired capacity to consent: NIH points to consider. Working draft document (unpublished). Wells, K., Sherbourne, C., Shchoenbaum, M., Ettner, S., Duan, N., Miranda, J. et al. (2004). Five-Year impact of quality improvement for depression: Results of a group-level randomized controlled trial. Archives of General Psychiatry, 61, 378-386. Secretary’s Advisory Committee on Human Research Protections March 27 and 28, 2008 Arlington, VA
Certification of the Summary of Minutes I hereby certify that, to the best of my knowledge, the foregoing summary of minutes is accurate and complete. __________________________________________________________________________________________________ Sam Tilden, M.D., J.D., L.L.M., Chair Date Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®. |
Last revised: July 9, 2008
|
