FY2024 Individually Measured Phenotypes to Advance Computational Translation in Mental Health (IMPACT-MH) U01 and U24 NOFOs
Questions & Answers from the Technical Assistance Teleconference, May 2, 2023
General information
This document contains general information and Q&A (both prepared and live) from the teleconference presentation. The recording of this meeting is available online.
Notices of Funding Opportunity
- RFA-MH-23-105: Individually Measured Phenotypes to Advance Computational Translation in Mental Health (IMPACT-MH; U01 Clinical Trial Optional)
- RFA-MH-23-106: Individually Measured Phenotypes to Advance Computational Translation in Mental Health (IMPACT-MH): Data Coordinating Center (U24 Clinical Trial Not Allowed)
Note that these grants are cooperative agreements and there will be substantial involvement from NIMH staff and a steering committee.
Please carefully read:
- the applicant instructions in section IV.2
- the review information in section V, and
- the terms and conditions in section VI.2
Purpose of the webinar
The purpose of this webinar is to provide information about the RFAs and answer general questions. We cannot provide feedback about specific planned applications. Please contact a NIMH Program Officer to discuss specific applications. If you aren’t sure which program officer to contact, please contact Jenni Pacheco.
Purpose of the funding opportunities
This pair of Notices of Funding Opportunity (NOFOs) is intended to stimulate and support research that will use behavioral measures and computational methods to define novel clinical signatures that can be used for individual-level prediction and clinical decision making in mental disorders.
Timeline overview
Timeline (same for both RFAs):
- Letter of intent due: May 14, 2023 (Note: letter of intent is optional, but requested)
- Applications due: June 14, 2023
- Scientific merit review: Fall 2023 (locus of review is NIMH)
- Advisory Council review: January 2024
- Earliest start date: April 2024
Q & A: General Questions
Q. Will the RFAs be re-issued or submission dates added?
A. There is no plan to re-issue the RFAs or add submission dates.
Q. What are the budget caps?
A. Maximum annual direct costs for U01 applications are $2.5M.
Maximum annual direct costs for U24s are $1M.
Maximum grant duration is 5 years for both types of awards.
Q. Can prior approval be obtained to exceed the budget caps?
A. No.
Q. Are the set-aside amounts that are provided in the RFA for the full duration of the grant or only the first year?
A. The set-asides are for the first year of the grants. Specifically, NIMH intends to commit $30 million total in FY 2024 for this pair of NOFOs to fund up to 8 U01 awards and 1-2 U24 awards.
Q. Are there special page limits for applications under these NOFOs?
A. No. Standard page limits apply. The research strategy section is limited to 12 pages for both the U01 and U24 applications. See NIH page limits tables for more information.
Q. Are sub-contractors allowed to issue sub-contracts (i.e., are third-tier sub-contracts allowed)?
A. Third-tier sub-contracts are highly discouraged and would only be allowed under unusual circumstances. Applicants may contact NIMH Grants Management staff to discuss best strategies for their specific situations.
Q. Are applications from non-U.S. institutions allowable?
A. Foreign institutions are eligible to apply for the U01 awards.
Foreign institutions are not eligible to apply for the U24 award(s).
See Section III. 1. Eligible Applicants: Foreign Institutions.
Q. Are appendices allowed?
A. NIH policy regarding appendices (NOT-OD-17-098) is applicable.
Specifically, only the following documents are permitted:
- Blank data collection forms, blank survey forms, and blank questionnaire forms;
- Simple lists of interview questions;
- Blank informed consent/assent forms.
Letters of support are not considered an appendix and can be submitted as part of the application.
Q. What is a Plan for Enhancing Diverse Perspectives (PEDP) and how do I submit it?
A. All applicants must include a PEDP consisting of a summary of strategies to advance the scientific and technical merit of the proposed project through expanded inclusivity. The plan should provide a holistic and integrated view of how enhancing diverse perspectives is viewed and supported throughout the application and can incorporate elements with relevance to any review criteria (significance, investigator(s), innovation, approach, and environment) as appropriate.
The PEDP may be no more than 1 page in length and should include a timeline and milestones for relevant components that will be considered as part of the review.
It should be uploaded as an "Other Attachment."
See Section IV.2 SF424(R&R) Other Project Information for more information.
Q. Can an investigator be PI (or MPI) on both a U01 and a U24 application under these RFAs?
Q. Can an investigator be PI (or MPI) on a U01 or a U24 application and be co-investigator or consultant on another U01 or U24 application under these RFAs?
Q. Can an investigator be PI/MPI (or other role) on more than one application of the same type (i.e., two U01s or two U24s) under these RFAs?
A. Yes. If an investigator is included on both a U01 and a U24 application, or on more than one U01 or U24 application, the applications should not mention/refer to any other applications.
Q. Should U01 or U24 applicants coordinate their applications with other U01 or U24 applicants?
A. It is not necessary for applications to be coordinated. Each application will be reviewed independently of other applications and must be able to stand on its own merits.
Applications should avoid coordinating budgets, to avoid possible budget shortfalls if coordinated applications are not funded.
Applications should not reference other submitted applications.
Once the grants are awarded, a process of coordination and harmonization of methods and measures will be organized by the DCC.
Q. What is the governance structure and decision-making process for the U01 projects, DCC, and Steering Committee?
A. The IMPACT-MH Steering Committee will oversee collaboration and harmonization efforts among the IMPACT-MH projects and help to set priorities for analyses of combined datasets. The Steering Committee will consist of the U01 PI(s), the U24 PI(s), NIMH Project Scientist(s) for the U01s and U24(s), and external members selected by the DCC team for their expertise in the overall goals of the IMPACT-MH initiative. NIMH Program Officer(s) for the U01s and U24(s) will be non-voting members of the Steering Committee.
External members of the Steering Committee will be identified after awards are issued. Potential external members of the Steering Committee should not be named in applications.
See Section VI.2: Award Administrative Information: Cooperative Agreement Terms and Conditions of Award.
Q. Will the U01 teams or the U24 team(s) conduct data analyses?
A. Data analyses specific to the aims of the U01 awards funded under this NOFO will be the responsibility of the U01 recipients. Analysis of combined datasets (in addition to other coordination and data sharing activities) will be the responsibility of the DCC (in collaboration with the IMPACT-MH Steering Committee).
See RFA-MH-23-105: Part 2. Section I. Notice of Funding Opportunity Description: Partnership with the IMPACT Data Coordinating Center
Each U01 project is responsible for validation and quality assurances on each project's data and for transferring data to the DCC. The DCC will verify the integrity of data, monitor the completeness of any combined data sets that are generated, and upload the data to the NIMH Data Archive (NDA).
See RFA-MH-23-106: Part 2. Section I. Research Objectives: 3. Develop and maintain a robust informatics infrastructure.
Q & A: U01 Questions
Q. Will applications be assessed for responsiveness before being reviewed? What are the responsiveness criteria?
A. Yes, applications will be assessed for responsiveness prior to peer review. The following will be considered non-responsive to this NOFO and will not be reviewed:
- Applications that do not incorporate both behavioral measure(s) and clinical records data to derive or validate a clinical phenotype
- Applications focused on developing a new intervention and clinical trials to test treatment efficacy and effectiveness.
- Applications aimed primarily at elucidating mechanism
- Applications proposing animal research
- Applications that do not include adherence to FAIR data principles
- Applications proposing the use of extant databases that do not provide evidence that the data structures are (or can be made) fully consistent with FAIR
- Applications that do not address the inclusion of individuals from groups that are underrepresented in biomedical research
- Applications that lack a PEDP
See RFA-MH-23-105: Part 2. Section I. Applications Not Responsive to this NOFO.
Q. What is the optimal sample size?
A. The number of participants will depend on the types of data proposed to be collected and the statistical power needed to achieve the scientific aims. Justification for the proposed sample size should be detailed in the Research Strategy section.
Q. Should applicants expect that the usual policy requiring a single IRB for multi-site studies applies?
A. Yes, multi-site U01s should plan to use a single IRB in accordance with NIH policy.
Q. Can a site be included in more than one U01 application?
A. Yes. If a site is included in more than one funded U01, however, the site would need to review and potentially revise its budget to avoid duplicative expenses and address any concerns about recruitment capacity.
Q. Is it allowable to propose a clinical trial?
A. Clinical trials are allowable but not required. Under this NOFO, NIMH will accept applications that propose studies that meet the definition of a clinical trial if the prospective assignment to intervention and assessment of outcomes serves the purpose of deriving or validating a novel clinical signature. Trials to test the efficacy or effectiveness of interventions are not considered responsive to this NOFO and will not be reviewed.
For information about NIH’s definition of a clinical trial and NIMH funding for clinical trials, see Support for Clinical Trials at NIMH.
Q. Should specific data types be proposed?
A. For the purpose of this NOFO, a clinical signature is a clinical phenotype comprised of a combination of features derived from behavioral data and the clinical record that can be assessed at the level of the individual and is prospectively associated with a specific clinical and/or functional trajectory, differential response to specific treatment types, and/or biological or psychological mechanisms.
Behavioral tasks and measures may include computerized tasks administered via web- or device-based platforms and/or passive collection of behavioral data. Behavioral measurements should be selected with scalability in mind, including their potential to be implemented in routine clinical practice.
Data types other than behavioral data and clinical records data, such as EEG, MRI, blood-based measures, or genomics, could be used in conjunction with behavioral data to further refine or disambiguate novel clinical phenotypes. However, the primary goal of this initiative is to develop highly accessible tools, so the use of costly, resource-intensive methods should be considered judiciously.
See RFA-MH-23-105Section I. Notice of Funding Opportunity Description.
Q. Will data collection measures and methods be harmonized across U01 projects?
A. When two or more funded U01 projects propose to measure the same information (e.g., the same cognitive construct, symptoms, or demographic information), awardees will be asked to use identical measurement tools and collection procedures whenever feasible and should otherwise establish a means to aggregate across similar measures.
The DCC will be responsible for facilitating the coordination and communication among the IMPACT-MH projects and leveraging opportunities for harmonization of methods and measures.
See RFA-MH-23-105 Section I. Notice of Funding Opportunity Description, Section IV.2. Content and Form of Application Submission: Approach, and Section V. Application Review Information.
Q. Given likely delays in initiating data collection due to post-award consultation with the DCC regarding harmonization of measures and methods and potential resulting protocol revisions, should U01 applicants indicate that the study will be Delayed Onset with regard to Human Subjects Protections or should U01 applications include the PHS Human Subjects and Clinical Trials Information form based on the network’s preliminary proposed research in the application?
A. The U01 studies should not be marked as delayed onset. The Human Subjects and Clinical Trials Information form should be completed based on the proposed research.
Q. Is there a preference for U01s to use existing datasets or to collect new data (or both)?
A. The scientific goals of the IMPACT-MH initiative and the availability of appropriate datasets should guide decisions about whether to use existing data or collect new data.
Both new and existing datasets may be appropriate for the purposes of this NOFO. All proposed datasets should be compliant with the FAIR data principles (ensuring that data are Findable, Accessible, Interoperable, and Reusable) and should include:
- data from individuals with mental disorders or at risk of developing mental disorders,
- behavioral data from either task-based and/or naturalistic/passive assessments,
- clinical records data,
- sufficiently large sample size to allow for detection of robust novel phenotypes that can be identified onan individual basis.
See RFA-MH-23-105 Section I. Notice of Funding Opportunity Description.
Q. If the use of an existing dataset(s) is proposed in a U01 application, should the specific dataset(s)s to be used in the analysis of existing data be identified in the application?
A. Yes. If the U01 team does not currently have access to the dataset(s), documentation that the investigator or entity that controls access to the dataset will provide access to the U01 team should be provided in a letter of support.
Q. Should U01 applications be organized like a center grant application (with cores and projects)?
A. No, U01 applications should be structured similarly to R01 applications.
A single Research Strategy section (page limit:12 pages) should be submitted with each U01 application.
Q & A: U24 Questions
Q. Will applications be assessed for responsiveness before being reviewed? What are the responsiveness criteria?
A. Yes, applications will be assessed for responsiveness prior to peer review.
Applications proposing the following will be considered non-responsive to this NOFO and will not be reviewed:
- Applications that do not include a plan to ensure that datasets adhere to FAIR data principles
- Applications that do not provide a detailed description of experience and expertise in data infrastructure management and computational methods
- Applications that do not include strategies to monitor and address potential biases in aggregated data and analytic approaches
- Applications that lack a Plan for Enhancing Diverse Perspectives
See RFA-MH-23-106: Part 2. Section I. Applications Not Responsive to this NOFO.
Q. For the U24 applications, should specific data types be anticipated?
A. Behavioral data (including task-based and naturalistic/sensor-based data) and clinical records data should be anticipated. Other types of data, such as EEG, MRI, blood-based measures, and genomics may be proposed by U01 applicants. The specific data types to be used won’t be finalized until after the U01 grants are funded, so U24 applications should describe in general terms the processes and tools necessary to gather, combine, store, analyze and manipulate de-identified health information datasets as well as combined/multi-modal data from multiple U01 studies in a timely manner.
See RFA-MH-23-106 Section IV.2. Content and Form of Application Submission: Approach and Section V. Application Review Information: Approach.
Q. Are the U24 teams expected to have expertise in collection and storage of blood biomarkers such as metabolomics, proteomics, etc.?
A. U24 applicants should include descriptions of their capacity to handle and process data and biosamples that they anticipate will be relevant to the scientific goals of the RFAs. U24 grantees will not be expected to establish or serve as a repository for biosamples. Existing repositories, such as the NIMH Repository and Genomics Resource (NRGR), will be used for storage of biosamples. If expertise in any specific data type(s) to be collected by the funded U01 teams is not available on the DCC team, individuals with appropriate expertise can be invited to serve on the Steering Committee.
Q. If genetic variables are included in U01 projects selected for funding, will the U01 projects arrange for DNA extraction, genotyping, and sequencing independently or will this be coordinated centrally by the DCC? Will the U01s or the DCC be responsible for these costs?
A. If a sufficient number of funded U01 projects include genomic data, the DCC will coordinate the choice of technology, timing, and facility to conduct genetic analyses in order to minimize site-related variability and batch effects in genetic analyses. The costs for generating and analyzing genetics data will be the responsibility of the U01 teams, with coordination of data analysis by the U24.
Q. What type of expertise is optimal for the study team?
A. The PI (or Multi-PIs) of the DCC must be experienced in the coordination and management of multi-site clinical research studies, including success in meeting milestones and timelines.
Additional expertise on the DCC team should include proficiency with multimodal data, and state of the art computational and data analytic skills, including:
- Experience with large-scale, multi-site studies, including computational approaches to the analysis of multivariate data and ethical considerations associated with large-scale analyses,
- understanding of the characteristics and criteria needed for the application of computational algorithms to inform clinical decisions, with attention to mitigating potential biases, and
- data science best practices to ensure that data are amenable to future computational techniques.
See RFA-MH-23-106 Section IV.2. Content and Form of Application Submission and Section V. Application Review Information: Investigators.
Content from Live Q&A
General questions
Q: Confirming that the Plan to Enhance Diverse Perspectives (PEDP) is not part of the 12- page Research Strategy?
A: Correct, the PEDP is not part of the 12-page Research Strategy section. As described in the IMPACT-MH NOFOs, the PEDP is a one-page document that should be submitted as Other Project Information as an attachment (see Section IV ).
Q: Can you clarify the discouragement of sub-contracts mentioned at the beginning of the call?
A: Sub-contracts are acceptable, but sub-sub-contracts are discouraged due to the added complexity for managing the award. Applicants who expect to need sub-sub-contracts should contact RDoC so that we can coordinate with NIMH Grants Management to assist.
U01 questions
Q: If the DCC is responsible for NDA uploads, is a Data Management and Sharing Plan required in the U01 application? Should the costs associated with data management and sharing (including uploading data to the NIMH Data Archive) be included in U01s budgets?
Here we will share with the U24 center instead and they will handle NDA. Is that right?
A: U01 applications should include a Data Management and Sharing Plan. Their budgets should include the personnel time needed to prepare the data to be transferred to the DCC in a format that is amenable to upload to the NDA. The DCC will do the uploading to the NDA and work with the U01s on any data QC issues that are flagged by the NDA.
Q: Do we need to include a budget for transfer of any biospecimens to the NRGR?
A: Yes, the costs of transferring samples should be included in U01 budgets.
Q: Should data analysis be included in the aims or should the aims focus on data collection, with analysis only in the research strategy section?
A: The initiative has the goal of identifying clinical phenotypes, so it is expected that the aims will include the integration of analyses for phenotype identification.
Q: Is there a preference for behavioral data collected using tasks, which can have well-controlled parameters, or is there equal interest in measuring naturalistic behaviors?
A: There is no preference for task-based measures over passive/naturalistic measures. The main consideration is that, given the interest in clinical utility, the measures should not be prohibitively challenging to implement across different clinical settings; priority should go to measures that can realistically be used as part of routine clinical care.
Q: Biological measures seem to be not discouraged though secondary, but are there any specific concerns with using them?
A: The potential concern with biological measures is that they can be resource-intensive, making it challenging to include in routine practice outside of research settings. Nevertheless, there are instances when more expensive or resource-intensive biological measures may be justifiable, for example as a means of validating a clinical signature. A biological measure could be used in addition to clinical and behavioral data to derive a clinical signature, but the added cost and complexity in the context of clinical settings should be addressed.
Q: By clinical record, are you referring to data from the electronic health record (EHR), traditional clinical research assessments (structured/self-report), or both?
A: Ideally, clinical data would be taken directly from the EHR to exemplify the feasibility of eventual clinical use of the novel clinical signatures identified in IMPACT-MH research. We recognize, however, that obtaining EHR data is not always feasible and that some applicants may have access to non-EHR-derived data that are similar to what is available in the EHR. Accordingly, the IMPACT-MH NOFOs allow use of data that are similar to information typically found in the EHR. These data can include items such as diagnoses, hospitalization history, and medication prescription history. Given the emphasis on practical implementation, data should not include specialized information that requires highly expensive and/or invasive data collection method, such as lengthy clinical interviews, however, clinical records data could include self-report measures or brief clinical interviews that are feasible to use across a variety of settings, even if the data are not pulled directly from EHR.
Q: To what extent should applications adhere to the RDoC matrix?
A: While RDoC principles are woven into this initiative, the RDoC matrix serves to provide exemplars of domains, constructs, units of analysis and data elements. Applications need not to be limited to items that appears in the RDoC matrix. Applicants should pursue topics and constructs consistent with appropriate evidence and scientific justification, regardless of explicit reference in the matrix.
Q: There are many computational models in addition to AI/ML and Bayesian methods that were specifically mentioned in the funding announcement. Is it permissible to use other approaches?
A: There are no restrictions on the types of computational models to be used. Those examples were noted as common techniques but applicants may propose any computational methods that are best suited to their scientific question and proposed data.
Q: Can a U01 application include optimization of computational phenotyping in non-clinical cohorts?
A: Yes, such research in non-help-seeking individuals would be appropriate to include in an IMPACT-MH U01 application, particularly in light of the RDoC’s emphasis on dimensional approaches and understanding a range of functioning. That said, because the focus is on clinical decision making, applications should be sure to demonstrate the relevance and utility of measures for people at risk of or experiencing mental disorders.
Q: A clinical trial could be about a diagnosis (e.g., major depression) or it could be more consistent with RDoC (e.g., patients with high levels of mood and/or anxiety). Would either design be responsive to this NOFO?
A: In keeping with RDoC, there should not be assumptions about the validity or homogeneity of any diagnostic classification, although diagnostic information is considered to be a signal of interest as part of understanding clinical phenotypes.
Note: Clinical trials focused on testing the efficacy of novel clinical interventions are not considered responsive to the IMPACT-MH U01 NOFO and should be directed to the NIMH Clinical Trials NOFOs. Under the IMPACT-MH U01 NOFO, studies using clinical trial methodology would be acceptable, for example, if using an intervention with established efficacy to test to the validity of a novel clinical phenotype.
If you have a question about these NOFOs that was not addressed in these materials, please email RDoCAdmin@mail.nih.gov.