September 20-21, 2007
Department of Health and Human Services
Public Health Service
National Institutes of Health
National Institute of Mental Health
The National Advisory Mental Health Council (NAMHC) convened its 216th meeting in closed session to review grant applications at 11:15 a.m. on September 20, 2007, at the Neuroscience Center in Rockville, Maryland, and adjourned at approximately 3:15 p.m. (see Appendix A: Review of Applications). The NAMHC reconvened for an open session at the same location at approximately 3:30 p.m., adjourning for the day at approximately 5:00 p.m. The open session continued on the following day, September 21, 2007, in Building 31C, National Institutes of Health (NIH), Bethesda, Maryland, from 8:30 a.m. until adjournment at approximately 12:30 p.m. In accordance with Public Law 92-463, the open policy session was open to the public. Thomas R. Insel, M.D., Director, National Institute of Mental Health (NIMH) chaired the meeting.
Council Members Present at the Grant Review and/or Open Policy Sessions
(See Appendix B: Council Roster)
Sergio A. Aguilar-Gaxiola, M.D., Ph.D.
Glorisa J. Canino, Ph.D.
Elizabeth Childs, M.D., P.C.
Jonathan D. Cohen, M.D., Ph.D.
Peter J. Hollenbeck, Ph.D.
Dilip V. Jeste, M.D.
Jeffrey A. Kelly, Ph.D.
Norwood Knight-Richardson, M.D., M.B.A.
Helena C. Kraemer, Ph.D.
John S. March, M.D., M.P.H.
Enola K. Proctor, Ph.D.
Peter Salovey, Ph.D.
Suzanne E. Vogel-Scibilia, M.D.
Stephen T. Warren, Ph.D.
Thomas R. Insel, M.D.
Jane A. Steinberg, Ph.D.
Liaison Representative Present at the Open Policy Session:
Anne Mathews Younes, Ed.D., representing A. Kathryn Power, Center for Mental Health Services (CMHS), Substance Abuse and Mental Health Services Administration (SAMHSA)
Margot Aronson, Clinical Social Work Association
Heather Benson, Research! America
Mark Blockus, National Council for Community Behavioral Healthcare
Alan Friedman, Transcriber
Reuven Ferziger, Johnson & Johnson
Marci Giang, Council on Social Work Education
Pat Kober, American Psychological Association
James McNulty, National Alliance for the Mental Ill
Noel Mazade, NASMHPD Research Institute, Inc.
Anne Michaels, National Foundation for Mental Health (Friends of NIMH)
Gerald Overman, College of Psychiatric & Neurologic Pharmacists
Amy Pollick, Association for Psychological Science
James Radack, Mental Health America
Jeanne Rivard, NASMHPD Research Institute, Inc.
Beth Roy, Social & Scientific Systems, Inc.
Bette Runck, Science Writer
Angela Sharpe, Consortium of Social Science Associations
Barbara Wanchisen, Federation of Behavioral, Psychological & Cognitive Sciences
Jun Yan, American Psychiatric Association
Joan Levy Zlotnik, Institute for the Advancement of Social Work Research
Open Policy Session: Call to Order and Opening Remarks
NIMH Director, Dr. Thomas R. Insel, called the open policy session to order, welcoming all in attendance. Dr. Insel acknowledged the contributions of Council members whose tenure ends on October 1: Drs. Sergio A. Aguilar-Gaxiola, Ned H. Kalin, Peter Salovey, Stephen T. Warren, and Ms. Martha E. Hellander.
Update on Council's Neurodevelopment Workgroup
Dr. Insel introduced John March, M.D., M.P.H., who co-chairs the Council Workgroup on Neurodevelopment with Pat Levitt, Ph.D. Dr. March updated Council on the workgroup's progress by reminding Council of the workgroup's focus on the gap in knowledge between behavioral development and brain development. The workgroup is charged with making recommendation to Council on the following topics:
- How to integrate developmental neuroscience, behavioral science, and psychopathology to forge a deeper understanding of the pathways leading to mental illness.
- How to stimulate discovery of the molecular, genetic, experiential, and environmental underpinnings of mental illness during development.
- How to foster translational neuroscience to improve indicators of risk, specify causal mechanisms, and develop effective preemptive interventions.
- How to use training, research resources, and infrastructure support to speed scientific discovery and produce a diverse workforce of translational researchers.
The workgroup has met several times via conference calls and met for a 2-day meeting in July. The workgroup will continue to meet via monthly conference calls and a 2-day meeting in December where it hopes to finalize its recommendations. A presentation of the workgroup's recommendations is expected at the next Council meeting in January 2008.
Dr. March highlighted the difficulties faced by investigators given the challenges of reasoning from genes to cells, cells to systems, systems to individuals, and individuals to context or social behavior. The workgroup's challenge is to take a truly developmental approach to reasoning across the various levels of analysis and experimental platforms.
To date, the workgroup has reviewed the current NIMH research portfolio related to neurodevelopment and is working to identify potential research opportunities to move the field forward. Dr. March said that the workgroup is struggling to overcome the "silos" in which research is conducted in an effort to unify the fields of developmental neuroscience, behavioral science, and psychopathology. Tool development is receiving special attention because the tools currently available are not optimally suited for developmental analysis. Emphasis is also being given to multivariable models that cross levels of analysis and experimental platforms. The workgroup plans to begin a discussion of training and infrastructure in the coming weeks.
NIH Office of Portfolio Analysis and Strategic Initiatives
Dr. Insel introduced Alan Krensky, M.D., Director of the NIH Office of Portfolio Analysis and Strategic Initiatives (OPASI). This office is responsible for several trans-NIH initiatives, including the Roadmap. Dr. Krensky came to NIH from Stanford University, where he had extensive experience in clinical research and policy issues. He noted that because OPASI is a new office, he and his colleagues are still defining its boundaries by consulting with each of the NIH Advisory Councils.
Dr. Krensky began by explaining the provisions of the 2006 NIH Reform Act, the first omnibus reauthorization of the NIH in 14 years. The Act established the Division of Program Coordination, Planning, and Strategic Initiatives (DPCPSI) and the use of a common fund to support trans-NIH research. The Act also created a Council of Councils to guide trans-NIH priorities; established a Scientific Management Review Board (SMRB) to oversee structures and authorities that may be used for improvements at the NIH; and initiated a public process to review potential organizational changes.
OPASI's mission is to provide the NIH Institutes and Centers (ICs) with the methods, tools, and information necessary to improve the management of their large and complex scientific portfolios. With input from key stakeholders, it also identifies emerging scientific opportunities or rising public health challenges and facilitates the acceleration of NIH investments in these areas, focusing on those involving multiple ICs. OPASI is responsible for coordinating and increasing the effectiveness of NIH-wide evaluation processes. Dr. Krensky highlighted the Roadmap as an example of this process. Roadmap initiatives must demonstrate a high potential to transform how biomedical or behavioral research will be conducted as well as demonstrate the potential for synergistic promotion and advancement of the individual missions of the ICs to benefit health; the applicability to issues beyond the scope of any one or a small number of ICs; the likelihood that no other entity is able or likely to perform the work; and the potential for producing a public health benefit. The Roadmap mechanism provides an opportunity to cross disciplinary and organizational boundaries to carry out activities relevant to the mission of several ICs. The emphasis is on innovation. Portfolio analysis tools, which are being developed, will help promote a better understanding of the complete NIH portfolio, allow the various ICs to come together, and help identify gaps and opportunities.
During the discussion following Dr. Krensky's presentation, Dr. Knight-Richardson asked where OPASI fit in the NIH structure. Dr. Krensky said that OPASI does not have overarching responsibility for NIH programs, but rather serves a coordinating, supporting, facilitating, and communicating role.
All trans-NIH projects undertaken by OPASI are funded through the common fund, which is about 1.7 percent of the total NIH budget. Dr. Krensky pointed out that in the past, the ICs provided money for the common fund, but in fiscal year 2007, Congress appropriated a lump sum for the fund. Dr. Insel underscored the importance of this change. The contribution NIMH made to the Roadmap is now used to fund research project grants (RPG): In addition, many NIMH investigators are being funded through the Roadmap mechanism, further increasing the pool of funds.
Dr. March asked if OPASI will be addressing collaboration among intramural and extramural scientists. Dr. Krensky said that the only boundary that needs to be respected is that between peer review and the program administering grants. More interaction among intramural and extramural scientists involves complex issues but is being explored.
Dr. March also asked whether Roadmap funding would continue to support collaborative science–e.g., Clinical and Translational Science Awards (CTSA). Dr. Krensky reminded Council that the CTSAs, which grew out of the Roadmap process, were too expensive to be funded through the Roadmap; thus, they were taken over by the NIH National Center for Research Resources, with about $70 million coming from Roadmap funds. The CTSA concept has been widely praised and has resulted in greater communication across disciplines, but implementation of the idea has raised many issues in addition to those of financing and regulation.
Dr. Jeste asked whether OPASI is involved in career development for young scientists. He suggested an arrangement between NIH and universities that would, for example, require matching funds to support young scientists. Dr. Krensky said the IC Directors meet regularly to discuss the issue and OPASI contributes when there are issues of portfolio analysis or evaluation.
Dr. Canino asked if OPASI is involved in attracting senior investigators for peer review. Dr. Krensky pointed out that the peer review process is undergoing both internal and external review, the results of which will be announced in the coming months. Pilot programs using different approaches to peer review could be done through the Roadmap or individual ICs. He noted that the Pioneer Awards employed a different approach to peer review.
Dr. Cohen asked about informatics, which is plagued with duplication–for example, in statistical methods for analyzing brain images and microarray techniques. Dr. Krensky agreed that there is a need for consolidation as there are currently 17 different trans-NIH groups working on informatics. He noted that Dr. Zerhouni is interested in unifying this effort via a Deputy Director position within the Office of the Director.
Dr. Proctor asked about OPASI involvement in programs for managing chronic diseases and the science of disseminating and implementing research findings and changing provider behavior. Dr. Krensky pointed out that chronic illnesses are within the purview of several ICs, but that coordinating their activities is a long-term process.
Dr. Insel thanked Dr. Krensky for his informative talk and his attention to the questions from Council members.
Council Discussion of Program and Policy Issues
Dr. Kraemer observed that the discussion earlier in the day, during the grant review session, appeared to suggest that Council members do not agree on the definition of different types of studies–pilot, exploratory, and hypothesis-testing. Requirements of each are distinct and are sometimes confused. She offered the following definitions:
- A hypothesis-testing study requires one or two well-articulated hypotheses for which the researcher has a rationale and justification. Also needed are an analytic plan and a design that can generate enough power to test those hypotheses. It is only in the context of a hypothesis-testing study that concepts such as "significance level" and "power" have any meaning. To provide the rationale and justification, exploratory studies are usually needed. To ensure that the proposed design and analytic procedure are feasible, a pilot study is needed.
- An exploratory study is not designed to answer questions; rather, it generates questions for future studies to test. It usually involves multiple testing, with many measures and a great deal of data crunching. Exploratory studies, such as many epidemiological studies, can be large.
- A pilot study does not aim to answer any questions. It is a feasibility study to find out whether the desired population can be sampled in sufficient numbers, whether the proposed procedures can be implemented, whether there are unexpected side effects, or whether proposed measurements will work. Mistakes can be made in the pilot study to avoid making them in the hypothesis-testing study. Because pilot studies do not provide answers, asking whether they have sufficient power does not make sense.
Dr. Kraemer suggested that all three types of study are important and require funding. Confusion over the requirements of each type of study has caused considerable controversy in psychology, where some researchers have advocated eradication of p values and elimination of significance testing altogether. Dr. Kraemer said she disagrees with that argument and instead would like to clarify the boundaries for each term.
Dr. Salovey thinks the corollary to misunderstanding these terms is disagreement on the part of reviewers about what a principal investigator is actually doing. Furthermore, investigators may prematurely submit applications for hypothesis-testing studies because they think that exploratory or pilot studies are less acceptable.
Dr. Nakamura said that several mechanisms have been specifically designed for exploratory or pilot studies. Still, review committees ask for the hypothesis that is being tested. Dr. Insel said that no matter how review processes are structured to accept pilot studies, innovative questions, and high-risk ventures, review committees appear to be resistant to scoring these types of studies positively. Dr. Cohen said the same is true of fellowship applications, where study sections ignore the critical element–the training environment–and instead focus on the specification of hypotheses and experimental design.
Dr. March noted that scientific and resource questions are raised by this issue. In child psychopharmacology, resources necessary to do the progression of studies from exploratory to pilot to hypothesis-testing studies are not available.
Dr. Salovey agreed that study sections tend to ignore the mechanisms for pilot and exploratory studies. When an investigator submits a 5-year research project grant that outlines pilot and feasibility studies in the first year or two, reviewers typically instruct the investigator to do those initial studies without funding. In many fields, that is impossible. He said the process is inefficient because it requires researchers to spend too much time writing grant applications.
Dr. Canino said she is acutely aware of the problem because she is in a field in which pilot testing is critical. She has begun studying preschoolers, an area in which few methodological psychometric studies have been done. The state of the art on ethnic minority children is even further behind. Funding for pilot studies is difficult to find.
Dr. Insel suggested that the Council Workgroup on Neurodevelopment address issues raised by Drs. March and Canino. He also said that the reviewers, particularly those on NIMH review panels, can be educated about the distinctions among different types of studies and the need for each. Review group training sessions–on experimental design and innovation, for example–have been well received. Dr. Steinberg noted that another training session for reviewers on design and analysis is planned for the summer.
Following this discussion, Dr. Insel adjourned the initial session of the Council meeting.
Call to Order and Opening Remarks
The Council reconvened the following morning on the main campus of the NIH in Bethesda, Maryland. Dr. Insel opened the meeting with a sad announcement. News of Paul Sirovatka death was received overnight. He had worked at NIH and NIMH for 25 years, retiring 3 years ago and joining the American Psychiatric Association. Dr. Insel said that Mr. Sirovatka's death represents a huge loss for the Institute and the entire field of mental health.
Approval of the Minutes of the Previous Council Meeting
Turning to the minutes of the May 2007 Council session, Dr. Insel asked if Council members had revisions or comments on the minutes. Hearing none, the minutes were unanimously approved.
NIMH Director's Report
Dr. Insel noted the end of fiscal year 2007 (FY07) is nearly upon us, and he reminded Council that FY07 budget saw only a 0.1 percent increase over FY06. FY06 budget reflected a 0.7 percent decrease from FY05. To cope with an essentially flat budget over the past few years, NIMH adopted the NIH funding strategy discussed at the May Council meeting.
Dr. Insel summarized the success of the new strategy. In 2007, 2,858 competing research project grant (RPG) applications were submitted, up from 2,757 in 2006. Between 2000 and 2007, an average of 95 new investigators were funded each year. That number dropped to 69 in 2006, but it rose to an estimated 89 in 2007. The average cost per grant rose sharply in 2007, despite the usual 10-percent cut. It is not possible to keep costs at the same level and expect researchers to do the same science in a flat-budget environment, he said.
Dr. Insel said that one measure of the Institute's budgetary health is the success rate of grant applicants. NIMH tries to maintain a success rate of 20 percent in the three submission cycles. Among principal investigators who had previously received funding, the 2007 success rate was 22.4 percent. Among new investigators, the rate was 17.3 percent.
Dr. Insel said it is clear that the culture of science is changing and that the next generation of scientific breakthroughs will require team players, technical innovators, and translators to do interdisciplinary research, team science, and discovery-driven science and to share data via public access databases. Translation involves reciprocal interchanges from bench to bedside to practice, as well as consideration of public health interventions, services research, implementation science, and real world relevance. "Team players" are being developed through the Roadmap's interdisciplinary research and training and the new multiple principal investigators mechanisms. Technical innovators are being trained through the Pioneer, Innovator, and Pathway to Independence (K99) Awards and by encouraging new investigators. Translation is being accomplished through the CTSAs and through institutional research training grants (T32s).
Dr. Insel highlighted the critical question of how many new investigators should be trained, given the need to balance the payline for grants and the pipeline of applicants. Training too many researchers may be setting researchers up for failure if grant money for new investigators is not available. In 2006, there were almost 1,700 investigators in the pipeline but only 69 new principal investigators. Dr. Insel asked if it is ethical to have that many individuals in training. The issue is particularly urgent for NIMH, since the Institute spends proportionately more on training than any other NIH Institute. The Council's 2004 7-year plan was to maintain pipeline support where outcomes were shown to be best; reduce support where growth cannot be sustained (e.g., K awards); and maintain support for a diverse workforce. The result of these efforts has brought the percentage of the total non-AIDS training budget down from almost 12 percent to less than 10 percent with a ultimate target of around 9 percent. There is a perception in the field that NIMH is withdrawing support for training, although even with the final reductions, the Institute will still be supporting 50 percent more training than the average NIH Institute.
In summary, Dr. Insel said that it is necessary to use an evidence based approach to decide which training mechanisms to use. He asked the Council to consider what kind of scientists will be needed in 2017 or 2020. What is the optimal ratio of grantees to trainees? How can investments be optimized to make sure that the Institute is training a diverse, innovative workforce?
Update on Peer Review Initiative
Dr. Insel introduced Lawrence Tabak, D.D.S., Ph.D., Director of the National Institute of Dental and Craniofacial Research, who co-chairs both the Advisory Council to the Director Working Group on Peer Review and the Steering Committee Working Group on Peer Review. As the NIH adapts to rapidly changing fields of science and ever-growing public health challenges, the principles of the peer review analysis are to ensure the processes used to support science are as efficient and effective as possible for applicants and reviewers alike and that NIH continues to draw the most talented reviewers. These two working groups are seeking broad input from both external and internal research communities.
The NIH issued a request for information from the scientific and academic communities that solicited opinions about issues related to the current system of research support, challenges and solutions specific to the peer review process, core values of peer review, review criteria and scoring, and whether the current peer review process is appropriate for investigators at different stages of their career. In a continued effort to seek input, the working group held two teleconferences with about 100 deans of colleges and universities, along with regional town meetings around the country, and a meeting is planned that will involve patient advocacy groups. The working groups have done an extensive analysis of the peer review literature and the process of peer review in other agencies, such as the National Science Foundation (NSF), and examined the results of prior experiments with peer review done within various ICs.
This information-gathering phase is expected to be complete by December. Then NIH leadership will devise a series of pilot experiments to test particular interventions. If all goes as planned, these pilots and their evaluations will begin by early spring. After that an implementation plan informed by those pilots will be developed. The implementation phase will include briefings for the NIH staff, scientific societies, NIH Councils, trade press, advocacy organizations, and the Congress, along with expansion of successful pilots and development of a new NIH peer review policy.
Dr. Tabak summarized some of the themes that have emerged during this information-gathering phase. Much attention has been given to review criteria. Should the project or the person be reviewed? Should reviews be retrospective or prospective? Should different types of applications with different review criteria be tried, such as some without preliminary data?
Several new mechanisms have been proposed to change the review process. These include a two-stage technical subject matter review, such as that used by journals; electronic review; virtual electronic dialogue between reviewers and applicants to answer questions, correct errors, and clarify ambiguous points in applications; and different types of review for different types of science.
Dr. Tabak acknowledged NIMH's interest in interdisciplinary research. Historically, he said, NIH reviewers have tended to view interdisciplinary applications as overly ambitious and unfocused. Dr. Tabak said that small business research grants may also require a difference type of review. Are academics the best people to review small business applications? Another suggestion is that investigators could designate one application as primary and subsequent submissions would be judged by different criteria. Should we give more useful feedback to applicants, particularly new investigators? Should applications that are triaged (unscored) be ranked? Should pre-applications be submitted to separate competitive from noncompetitive ideas and give meaningful advice to unamended applications?
Ideas have also been offered for maximizing review and reviewer quality. Do reviewers have sufficient context? Where should that context come? Should incentives be provided for reviewers? Should the NIH require funded investigators to serve as reviewers? Should service be more flexible? Should reviewers and scientific review administrators be rated? Should NIH, like NSF, not publish reviewer identities?
Ideas for scoring include using different kinds of voting scales, such as using psychometrics to group scores that are not statistically different. Should additional dimensions be added to extract more information from the review? For example, two scores would be given, one for the application as received and one for the best potential score. Should NIH require a minimum effort for a principal investigator or limit the percentage of effort that can be recovered?
After this brief overview, Dr. Tabak solicited input on the review process from the NAMHC members. Dr. Kelly noted that innovation is a review criterion, but the benchmarks and definitions of innovation are murky and subjective, whereas definitions of methodology are more objective. Reviewers appear to have few metrics for judging innovation but numerous metrics for judging methodology, and, as a result, innovation tends to get mere lip service. Dr. Hollenbeck noted a possible advantage of a multi dimensional rating scale but expressed concerns regarding reviewer burden with such a system. Dr. Aguilar-Gaxiola noted that with the growing numbers of applications and the variety within any given study section, reviewers may not be sensitive to the complex issues of particular types of research, such as the complexity of conducting community-based research. This review format appears to disadvantage these types of applications.
Dr. Salovey highlighted the lack of senior investigators who serve on study sections. Workload is certainly part of the reason for the reluctance; however, the question of incentives and obligation needs to be explored. There was a time when researchers who received funding felt a kind of moral or civic obligation to give back by serving on either Council or a study section, or both, or helping out in some other way. Dr. Canino endorsed the idea of offering incentive while Dr. Warren agreed with the suggestion of mandatory service for grantees as a means to recruit senior investigators to serve as reviewers. Dr. Childs agreed with the concept of required service for grantees and suggested adding this information to the grant application itself. Dr. Hollenbeck suggested another point of contact to serve as a reviewer might be at the award of a first renewal grant. Dr. Kraemer suggests that improving the review process might also improve the quality of the applications under review as one informs the other. Dr. Jeste also endorsed the inclusion of junior researchers into the review process as a learning experience.
Dr. Proctor reported on the problem of reviewers' occasional lack of familiarity with program announcements to which applications are responding. She suggested that reviewers may need more continued training on these and other issues.
Dr. Cohen highlighted the difficulties of review committees, which are now interdisciplinary and trans-Institute, to consider the Institute priorities. Dr. Cohen also questioned how fine grained a distinction reviewers should make. He proposed a three-tier review recommendation-- "fund now, come back, or go away"-- and beyond this distinction, funding decisions based on priority should be made by program in consultation with the IC Advisory Council.
Dr. Vogel-Scibilia proposed the use of two rating scales, one for the research idea and the other for the scientific design. This information might provide information to the applicant on how to proceed. She also suggested that a mechanism for raising the score for innovative ideas may be useful.
Dr. Salovey suggested an invigoration of the esprit d'corps within a study section might improve the process. Reorienting reviewers to the overall task might create a more positive experience. Another idea (borrowed from NSF) is to build into review something other than grant review-for example, discussion of an interesting issue or a report from another part of the Institute. This might help to make the experience more rewarding. Dr. Insel said that NIMH study sections do seem to have members who are so invested that they readily sign up for a second term. Having an interest in the mission of the Institute is more likely to create that kind of enthusiasm than when generic reviews are done through a central body such as the Center for Scientific Review (CSR).
NIMH Strategic Planning
Dr. Insel reported on the strategic planning that has been underway for the last few months. This process is guided by the NIMH vision and mission statement. Stakeholders include the research community, the general public, patients and families, service providers and state mental health authorities, family and patient organizations, professional groups, the media, the NIH and DHHS, and Congress. After a series of brainstorming meetings, a plan has begun to emerge. A document will be ready for public comment in the fall and may be ready for formal approval by the January Council meeting.
Several fundamental observations have informed the planning. First, the landscape of science has changed tremendously and will change further in the next decade. How does the NIMH position itself with the right people, projects, and investments for the future? Second, the current approach to diagnosis and treatment of mental disorders needs to be transformed. This concern has been expressed by the public and mental health professionals. Third, rather than simply measuring success by the number of grants, dollars, or published papers, success should be measured by transformative scientific discovery and ultimately by public health impact.
With that background, Dr. Insel outlined the plan's four cross-cutting objectives:
- Build the pipeline of discovery through brain and behavioral science to fuel research on the pathophysiology of mental disorders.
- Chart the trajectories of mental disorders to determine when, where, and how to intervene.
- Develop new and better interventions that incorporate the diverse needs and circumstances of people with mental illness.
- Strengthen the pubic health impact of NIMH-supported research.
Dr. Proctor said that the day's presentations had underscored the need for more training. She endorsed the data-driven approach to determining the ingredients of successful training programs. Dr. Jeste reiterated his belief that NIH should consider partnering with universities and foundations. He also noted that trainees who are not successfully funded with research grants may have gone on to become very successful academics, clinicians, teachers, and administrators. Outcome need not be measured solely on whether recipients receive research grants. Dr. Jeste said he believes that training should begin at the earliest possible level of education. Dr. Aguilar-Gaxiola said he agrees with Dr. Jeste's points. He suggested that early training should be part of the strategic plan.
Dr. Insel suggested that the Council take a systematic approach to analyzing training needs by setting up a Council workgroup to address the issue. Several members volunteered to participate. Dr. Insel noted that the issue extends beyond NIMH. He asked that the workgroup consider both short- and long-term investment needs and the sacrifice of research grants that may be necessary to train future investigators. This workgroup might also consider the need for diversity in the workforce and in training.
Enhancing Data Sharing at NIMH
Dr. Insel said that NIMH has developed a data-sharing toolbox for program staff to enhance the utility of data and promote collaboration among investigators supported by both grants and contracts. He introduced Della Hann, Ph.D., Director of the Office of Science Policy, Planning, and Communication, who provided an update on how NIMH proposes to enhance data sharing and data deposition.
Dr. Hann highlighted that the atmosphere of data sharing across the scientific community is changing and that an enhanced system is necessary when large datasets with complex variables are being generated at taxpayer expense. She noted that there are barriers to data sharing including the need to safeguard participants' privacy and informed consent documents and the concern among the original investigators that others will not appropriately credit their work. When data are collected with grant support, investigators are expected to share the data, although NIMH cannot require data sharing since it is not a regulatory agency. With contracts, NIMH can require datasets as a deliverable.
Dr. Hann noted that the Genome Wide Association Studies (GWAS) policy released in August 2007 has broad implications for all forms of data sharing. One issue addressed by the policy is the setup of a database and data input. Another concerns the data output–the conditions under which individuals will have access to the data. She explained that NIH is the repository for the GWAS, and all of the data entered will be deidentified to ensure privacy. For the first 12 months or less, the original investigators will have exclusive access to the data. Each IC will form a committee to review requests for data access, following established standards and guidelines. The individuals requesting use of the data must specify to the committee what questions they plan to ask, why they want to look at the data, and their willingness to make the information public. All of these details will be posted on a web site.
Dr. Hann concluded her presentation by reiterating the changing culture of science and the increasing need for data sharing. Journal editors, among others, are considering requirements for data sharing, even for the review or publication of papers. Every policy varies somewhat because the data and the culture of science around that data are different. Expectations developed for genetic studies may not be applicable to clinical trial data or imaging data.
Philip Wang, M.D., Dr.P.H., Director of the Division of Services and Intervention Research (DSIR), continued the presentation by explaining the policy on the release of public use datasets from NIMH's large practical clinical trials. The goal is to maximize their usefulness so that mediators or moderators of effects can be identified and data can be pooled for meta analyses. To fulfill this commitment, the contracts for the trials required the preparation and release of a public use dataset after a limited proprietary period.
Dr. Wang emphasized, as had Dr. Hann, that NIMH is committed to ensuring that there are safeguards to protect the data and study participants. He noted that public use datasets will be deidentified before delivery to NIMH and established procedures will be used to address consent and privacy issues. A small internal working group is examining how to update the NIMH policy for the distribution of data. It is also considering whether requesters should be required to get approval from their institutional signing official or their institutional review board. In addition to the lessons of the GWAS, the group is drawing heavily on the experience of other ICs.
Dr. Wang said that the working group continues to wrestle with a few issues. The group is also considering how to ensure the validity of the reanalysis of clinical trial data. The large clinical trials are comparative effectiveness trials with multiple active agents that are still on patent in most cases. Currently, scientific journals have the responsibility for reviewing analytic plans and results. It may not be feasible for NIMH to assume a more active role in ensuring validity of analyses, and doing so may not further the overall goal of promoting use of public use datasets. A second issue is what to do if the results of reanalysis are misleading and used in marketing by other industries such as pharmaceutical companies. To address this issue, the working group is consulting with the FDA, which has an office that regulates the use of research data for marketing purposes.
During the discussion period, Dr. Kraemer recommended that the workgroup attend to the unit of data to be archived. She recommended a staged approach to data release given that it often takes several years for the investigators to publish their planned papers. She asked the group to consider archiving data only after it has been used in a published paper suggesting that by the end of the study, all the data would be archived.
Dr. Wang said that, conceptually, staged release of data is possible. The plan may become problematic, however, for some types of studies. For example, epidemiologic studies often need available variables for control and adjustment. A limited dataset may allow reanalysis of the main hypotheses, but the full dataset is needed to explore additional factors. Dr. Insel pointed out that an unintended consequence of staged release of data might be that the original investigators would do many more papers or slow down publication so that they can hold onto the data longer.
Dr. Kelly reported great interest in the field in whatever policies NIMH develops for large trials, which may serve as models for single studies. Dr. Hollenbeck agreed and stated that basic research has faced a similar issue around the sharing of simple reagents. It is essential to systematically ensure that data are submitted. Whatever plan is devised, it should be followed because investigators are collecting far more data than they can analyze themselves.
Dr. March said he thinks a data sharing plan should be worked out with investigators for all research project grants, contracts, or projects done in response to solicitations. The problem is cost, particularly with large datasets. A great deal of quality assurance, programming, and other effort must go into generating a cleaned, locked, and intelligible dataset. Once the data are reanalyzed, a series of derived variables would need to be added to the dataset. The costs involved in this process are not, at present, reimbursable under the original grants; therefore supplemental funding is needed.
Dr. Kraemer agreed that costs associated with archiving data must be built into the grant application. She pointed out that the original study would probably be better designed as a result of archiving because investigators would have to define and justify every variable and clean the data as the study progressed.
Dr. Insel said that these ideas and others will be considered when developing plans for data sharing. He said it is likely that profound changes in grant and publication practices may emerge.
Cognitive Neuroscience Treatment Research to Improve Cognition in Schizophrenia
Dr. Insel introduced Jonathan Cohen, M.D., Ph.D., who discussed the Cognitive Neuroscience Treatment Research to Improve Cognition in Schizophrenia (CNTRICS) project, which was undertaken to fill a gap between basic cognitive neuroscience and practical clinical applications. He noted that developments in cognitive neuroscience over the past 30 years have been substantial, and those developments include sophisticated cognitive measures for functions relevant in psychopathology. It is now possible to measure memory, language, and executive function in the intact and the impaired brain. Despite these advances, there has been remarkably little penetration into either clinical research or clinical practice.
Dr. Cohen said that two reasons, one scientific and one sociological, largely account for this state of affairs. Scientifically, the problem is that cognitive neuroscience measures are not always suitable for off the shelf clinical application; the need for reliability often conflicts with the need for sensitivity. Among the sociological obstacles appears to be the lack of interest among basic scientists in developing their measures for clinical use. These obstacles suggest that translational research may need a translator. Feasibility studies are needed to ascertain which instruments can be adapted for clinical research and practice.
Dr. Cohen introduced Robert Heinssen, Ph.D., ABPP, Chief of the Schizophrenia Spectrum Disorders Research Program in the Division of Adult Translational Research and Treatment Development (DATR), to describe an initiative being developed to address the need for better clinical measures. Dr. Heinssen reminded Council that cognitive deficits are a core feature of schizophrenia and a major determinant of poor functional outcome. While a small empirical literature supports several promising rehabilitation strategies, no drug has yet been approved for cognitive enhancement purposes. In 2002, NIMH established the Measurement and Treatment Research to Improve Cognition in Schizophrenia (MATRICS) initiative to stimulate the development of drugs to treat cognitive deficits in schizophrenia.
As an outgrowth of MATRICS, the CNTRICS initiative includes an investigator initiated conference series that focuses on neuroscience based approaches for assessing and treating cognitive deficits in schizophrenia. The CNTRICS meetings bring together basic cognitive and affective neuroscientists with clinical schizophrenia researchers, industry representatives, and members of the NIMH cognition working group. The group explores the development of brain based measures of specific cognitive systems and the underlying neural circuitry that would support rational therapeutic development. The conference series was organized around three broad themes: identifying cognitive constructs; psychometric, translational, and practicality issues; and selecting specific measures for adaptation. Constructs that were selected were believed to have the greatest empirical evidence supporting a relationship between the cognitive process and the underlying neural system implicated in schizophrenia and include attention, executive control, long-term memory, working memory, perception, and social/emotional processing. For each construct, one or two mechanisms for translation were selected that had the highest degree of construct validity and greatest potential relevance to schizophrenia. They included, for example, long-term memory mechanisms of relational encoding and retrieval and reinforcement learning.
Dr. Heinssen proposed an initiative that would support the initial adaptation, testing, and refinement of experimental cognitive science tasks for measuring cognitive change over time in persons with schizophrenia. The initiative encourages translational approaches, and requires the coordinated efforts of systems neurobiologists, human cognitive neuroscientists, clinical schizophrenia researchers, psychometricians, and methodologists in clinical trial design–all working in interdisciplinary research teams. Specific goals of the initiative would be to standardize the design and administration of well validated experimental cognitive science tasks. Thus, the initiative starts with construct validity and moves toward practical utility. To be successful, investigators would have to base their selection of tasks for translation on initial construct validity and relevance to schizophrenia. The tasks must be related either empirically or theoretically to neural circuits or cognitive processes implicated in schizophrenia. Dr. Jeste said he hopes these assessments go beyond the usual paper-and-pencil test for cognition. Dr. Childs added that one of the challenges in getting clinicians to implement up to date assessment measures is their belief that the assessment will not change how they treat patients because other treatments are not available. This barrier can be overcome with education, but it is also a stimulus to improve treatment. She challenged the field to think about how clinicians can be encouraged to do assessments to obtain critical information while dealing with time constraints and patient and family needs.
Dr. Cohen suggested that more sensitive instruments may help refine treatments. For example, they might help predict who will respond to different classes of antidepressants without having to wait 3 weeks. Imaging techniques have now provided some evidence that it might be possible to predict response to pharmacological or psychosocial treatments.
Dr. Aguilar-Gaxiola said that he believes the project will add significantly to the phenotype of schizophrenia. His own epidemiological research using standardized clinical interviews has been frustrated by the inability to identify people with schizophrenia. The outcome of this initiative holds promise of identifying patients with more precision and fine tuning the measurement of their cognitive changes.
Dr. Insel called for a vote of concurrence for the concept, and the Council concurred. Dr. Insel asked Dr. Heinssen to give a progress report on the initiative in a year.
Programs of Excellence in Scientifically Validated Psychosocial Treatments
Michael Kozak, Ph.D., Program Chief of the Affective Processes and Anxiety Disorders Program, DATR, noted that not enough clinician-scientists are available to develop, evaluate, and disseminate psychosocial treatments despite their proven value in treating mental and substance abuse disorders. To speed the translation of psychotherapy findings to clinical care, NIMH, the National Institute on Drug Abuse, and the NIH Office of Behavioral and Social Sciences Research propose to establish programs of excellence to train the needed clinician scientists.
This initiative would support the development of curricula in science based psychosocial interventions that would integrate clinical and research components. The goals of the programs are to provide a model for clinician education nationwide; more integration of science and practice than in the pre award training program; immediate improvements in the competence of program graduates; and public availability of the awardee curricula. Award criteria include a strong preexisting training program in clinical behavioral science and research projects that have extramural grant support, as well as a curriculum that is well integrated with training in behavioral science that yields competency in one or more psychosocial interventions of scientifically established efficacy.
Dr. Jeste asked if an abbreviated version of the program would be available for psychiatric residents. Dr. Kozak said that any type of program can apply for the support. In answer to questions from Dr. March, Dr. Kozak said that the program is aimed at traditional academic institutions training new clinician-scientists and that interventions to be developed include those for children. He said that a curriculum for retraining established clinicians, as Dr. March suggested, might be adapted from the programs at some later date. Dr. March encouraged Dr. Kozak to build in dissemination and other processes by which evidence is translated from the research clinic into the community and, correspondingly, into the training program and into the curricula.
Dr. Vogel-Scibilia said she would approach the problem from the opposite side. Her own experience has taught her that new clinicians come out of training with the knowledge of the most effective treatments. After a year and a half or two in a clinic, they have adopted the culture of that clinic, not what they learned in training. Given this reality, it might increase dissemination into practice to train senior clinicians working in settings that do not have a research focus. If the senior clinicians are trained, they can train the staff in their clinics.
NIMH Career Opportunities in Research (COR) Honors Undergraduate Research Training Program
Michael Sesma, Ph.D., Chief of the Research Scientist Development Program within the Office for Special Populations, presented an overview of the COR program as a training program for undergraduate honors students that is designed to increase diversity among investigators. This program began 3 decades ago and is now being revised and renamed.
Dr. Sesma reiterated the concern over the inadequate number of research scientists from some ethnic minorities. This program aims to support undergraduate juniors and seniors who take on a mental health research curriculum including a summer mentored research experience with an NIMH grantee or NIMH intramural scientist. The goals are to prepare students to enter training for a Ph.D., M.D. /Ph.D., or other research doctoral degree program in biomedical or behavioral sciences or relevant health services. Outcome data from the first thirty years of the program show that this program has been very successful with 86 percent of program participants advancing to graduate programs within a year of completing their training.
The revised program will prepare students interested in doing research to continue their education in mental health disciplines and to become the translators discussed earlier. The programs must provide a contemporary curriculum to introduce multidisciplinary and translational science and quantitative and cross disciplinary research methodologies. Courses in research design and statistical methodology are mandatory, as are a range of courses in psychology and behavioral science, neurobiology or neuroscience, developmental psychology and neuroscience, genetics of behavior, epidemiology of mental health disparities, and the biological basis of mental disorders.Unlike the existing program, the revised version requires the COR training program to provide a mentored research experience with an NIMH supported researcher to develop the skills necessary for a career as an independent biomedical or behavioral investigator. Under the revised program, eligibility is expanded to include all colleges and universities that have a substantial enrollment of students from underrepresented backgrounds.
Dr. Salovey said that the big challenge at the college level is what is called "melt," that is, students melting out of scientific research. High school students with strong math and science backgrounds often turn to other subjects when they get to college. The best predictor of retention in sciences is hands on experience in laboratories. Therefore, mentored laboratory experience is an important component of this program.
Dr. Aguilar-Gaxiola said that he and his colleagues decided to recruit students who were late bloomers, those coming from disadvantaged situations, primarily poor families from the rural areas who were unaware of their own potential or the opportunities open to them. He recommended that the curriculum should allow students to decide how their exposure to the rich curriculum makes sense for their own lives. For example, the focus on health disparities raises issues of culture and language.
Dr. Sesma said that once an organized curriculum is introduced, it generates interest in younger students who compete for the training slots when they are juniors. A program such as COR can change the culture for science education in mental health disciplines.
Developing Innovative Approaches to Personalized Care in Mental Health
Benedetto Vitiello, M.D., Chief of the Child and Adolescent Treatment and Prevention Intervention Program, DSIR, described an initiative to develop innovative approaches to personalized care in mental health. With personalized care, treatment and prevention decisions are based on individual patient characteristics, which can vary from genetic polymorphism to demographic, economic, and contextual factors. Past research has examined databases of naturalistically treated patients or clinical trials retrospectively to identify predictors and moderators of response. Although useful for generating hypotheses, the results cannot necessarily be incorporated into treatment guidelines, which tend to be very general– a one size-fits-all approach. Such generality is not suited to individuals with mental illness because their responses are heterogeneous.
This initiative is aimed at formulating new strategies, proposing new approaches, and identifying new ways of personalizing care with prospective studies. It accommodates a variety of approaches–from hypothesis generating studies to formulating and testing strategies, at least for their feasibility. It will not be primarily aimed at supporting infrastructure for data mining. The research can involve several NIMH units.
Developing State Mental Health Services Research "Laboratories" for Policy Relevant Studies
David Chambers, Ph.D., Chief of the Dissemination and Implementation Research Program, DSIR, cited "The Road Ahead" Council Workgroup Report, which included several recommendations on services research. The first recommendation is that NIMH should create a means of identifying policy changes and other trends likely to have the most significant effect on mental health services and then monitor their impact. The proposed initiative aims to follow up ongoing work with state mental health agencies. Previous work focused on enhancing the implementation of effective practices connecting science and service. A major challenge was a lack of coordinated data, and that lack made it difficult to understand the broader impact of system-wide policies. Although data systems exist in most states across multiple agencies, they are rarely integrated making it difficult to study the implications of policy changes. This initiative will help States develop studies to identify the impact of different policies on cost, quality of care, and patient outcomes. The long-term goal is to create a "natural laboratory" to facilitate comparative research across States.
Dr. Mathews-Younes said that the Substance Abuse and Mental Health Services Administration (SAMSHA) fully supports the initiative. Dr. Chambers had noted that SAMHSA's Center for Mental Health Services is making efforts to connect research and practice. Dr. Insel said this is an ideal opportunity for NIMH and SAMHSA to interact at the State level.
Dr. Childs said it would be helpful to the States to incorporate suggestions for overcoming barriers about data sharing. The experience of pooling utilization and claims data in Massachusetts was very complex and took more than a year and a half. Dr. Chambers agreed with the idea. Dr. Aguilar-Gaxiola said the initiative is particularly timely for some States, such as California, that have begun this activity.
Following the discussion periods, the Council voted to approve all of the concepts.
Dr. Joan Zlotnik, Executive Director of the Institute for the Advancement of Social Work Research, said she was pleased that the concept clearances have been sought and that a Council workgroup to focus on training is being formed. She suggested that the workgroup might consider a strategy to begin training early. It is also important to consider diversity in terms of the kinds of science and to think about traineeships in terms of mentorship and cross-site training programs in nonacademic settings. She said her organization and others would like to help with the workgroup's endeavors.
Dr. Insel welcomed the suggestion and said he would like to see the Council workgroup bring in former grantees and others to aid in its deliberations.
Dr. Insel adjourned the 216th meeting of the NAMHC at approximately 12:30 p.m. on September 21, 2007.
I hereby certify that, to the best of my knowledge, the foregoing minutes are accurate and complete.
Thomas R. Insel, M.D., Chairperson
Appendix A: Review of Applications
Review of Applications with Primary Assignment to NIMH
September 2007 Council
Date/Time Report Run: 09/20/2007 09:09:21 AM
|Scored #||Scored Direct Cost $||Not Scored (NRFC)#||Not Scored (NRFC) Direct Cost $||Other #||Other Direct Cost $||Total #||Total Direct Cost $|
Appendix B: Council Roster
(Terms end 9/30 of designated year)
|Thomas R. Insel, M.D.|
National Institute of Mental Health
|Jane A. Steinberg, Ph.D.|
Division of Extramural Activities
National Institute of Mental Health
|Sergio A. Aguilar-Gaxiola, M.D., Ph.D. (07)|
Professor of Clinical Internal Medicine
Director, Center for Reducing Health Disparities
University of California, Davis School of Medicine
Department of Internal Medicine
Glorisa J. Canino, Ph.D. (09)
Director, Behavioral Sciences Research Institute
University of Puerto Rico
Medical Sciences Campus
San Juan, PR
Elizabeth Childs, M.D., P.C. (10)
Jonathan D. Cohen, M.D., Ph.D. (08)
Eugene Higgins Professor of Psychology
Director, Center for the Study of Brain, Mind and Behavior
Director, Program in Neuroscience
Raquel E. Gur, M.D., Ph.D. (08)
Director, Neuropsychiatry Section
University of Pennsylvania Medical Center
Martha E. Hellander, J.D. (07)
Founder, Child and Adolescent Bipolar Foundation
Peter J. Hollenbeck, Ph.D. (08)
Professor of Biological Sciences
Department of Biological Sciences
West Lafayette, IN
Dilip V. Jeste, M.D. (10)
Ester and Estelle Levi Chair in Aging
Distinguished Professor of Psychiatry and Neurosciences
University of California, San Diego
VA San Diego Healthcare System (116A-1) (10)
La Jolla, CA
Ned H. Kalin, M.D. (07)
Hedberg Professor and Chairman
Department of Psychiatry
University of Wisconsin Medical School
Jeffrey A. Kelly, Ph.D. (08)
|Ex Officio Members||Liaison Representative|
|Office of the Secretary, DHHS|
Michael O. Leavitt
Department of Health and Human Services
National Institutes of Health
Elias A. Zerhouni, M.D.
National Institutes of Health
Department of Veterans Affairs
Ira Katz, M.D., Ph.D.
Department of Veterans Affairs
Office of Mental Health Services
|A. Kathryn Power, M.Ed.|
Director, Center for Mental Health Services