The majority of federal funding for health research and related activities is spent by the National Institutes of Health (NIH), within the Department of Health and Human Services (HHS), the Department of Defense (DOD), and the Department of Veterans Affairs (VA). In fiscal year 2010, NIH, DOD, and VA obligated about $40 billion, $1.3 billion, and $563 million, respectively, for activities related to health research. Applications for federal funding of health research are typically submitted by principal investigatorsthe lead researchers for research projectsthrough their institution, and in some cases they may submit applications to multiple agencies at the same time for funding consideration. It is common for agencies to fund health research on topics of common interest, such as breast cancer and post-traumatic stress disorder (PTSD). In some cases, funding similar research on the same topics is appropriate and necessary, for example, for purposes of replicating or corroborating prior research results. However, without effective sharing of information among federal agencies about their funding decisions, they may use available funds inefficiently due to duplication of effort.
Specifically, about 94 percent of federal funding for medical sciences research in fiscal year 2008 was obligated by these three federal agencies, according to data from the National Science Foundation.
With respect to DOD, we obtained data on obligations of funds made available for research, development, testing, and evaluation in the annual appropriation for the Defense Health Program. With respect to VA, we obtained data on obligations of its appropriation for Medical and Prosthetic Research.
Principal investigators are typically individuals designated by the applicant organization, such as a university receiving federal grants, to have the appropriate level of authority and responsibility to direct the project or program to be supported by the award.
Agency officials told us that multiple agencies cannot fund the same research application unless they work together to jointly fund it.
In some instances, research is initiated in response to congressional direction. For example, according to DOD, the Office of Congressionally Directed Medical Research Programs is funded through the annual Defense Appropriations Act and manages research in many areas, including breast cancer. According to DOD, funds identified during the appropriations process at the request of members of the House and Senate are used for congressionally directed research. http://cdmrp.army.mil/about/fundingprocess.shtml (last visited Dec. 2, 2011). Future GAO work is expected to examine the Office of Congressionally Directed Medical Research Programs.
GAO recognizes that, in some instances, it is appropriate for multiple agencies or entities to be involved in the same programmatic or policy area due to the nature or magnitude of the federal effort. For purposes of this report, the term unnecessary duplication refers to duplicative research funding that is not necessary to corroborate or replicate prior research results for scientific purposes.
NIH, DOD, and VA each lack comprehensive information on health research funded by the other agencies, which limits their ability to identify potential areas of duplication in the health research they fund. NIH, DOD, and VA program managersofficials who typically manage agency research portfolios and may provide input to senior agency officials responsible for making funding decisionstold GAO that, when reviewing health research applications, they typically search publicly available databases for potentially duplicative research projects funded by other federal agencies. These databases are used by various federal agencies, including NIH, DOD, and VA, to maintain information on funded health research applications. For example:
According to NIH, DOD, and VA officials, the information provided in the research databases they use to identify any potential duplication when making funding decisions is generally not sufficient. For example, NIHs public database provides basic application information such as the title, principal investigator name, abstract, and agency contact information for each application. However, program managers said they need more details on the aims and methodologies of funded applications in order to determine whether applications considered for funding are duplicative of funded research. Officials noted that even applications with identical titles may have different aims. In such cases, officials said they typically obtain information not contained in the databases by contacting colleagues at other federal agencies to obtain details on specific applications.
Officials at NIH, DOD, and VA added that they also communicate with officials at other agencies through participation on joint committees that have members from various federal agencies. For example, NIH officials stated that the Interagency Breast Cancer and Environmental Research Coordinating Committee, a committee established in 2010 by NIH, facilitates exchanges of information about breast cancer environment and research efforts across various agencies. While DODs database for applications funded through its Congressionally Directed Medical Research Programs provides information about applications aims and methodologies, DODs database does not provide contact information for the officials associated with specific applications. One program manager at NIH and several VA officials said that they had difficulty knowing who to contact at DOD to obtain further information on specific applications.
Another limitation of the databases is that they do not always allow for efficient, comprehensive searches to identify unnecessary duplication of research. As stated earlier, information on health research funded by NIH, DOD, and VA is in different databases with varying types and amounts of information. DOD and VA officials told GAO that, in general, when searching multiple databases for potential duplication, the large number of funded applications on related topics makes comprehensive checks difficult and time-consuming. Because of this, officials at NIH, DOD, and VA told GAO that they often limit searches to principal investigators other federally funded research projects, which they are generally required to list on their applications. To address this challenge, VA officials told GAO that they are working to make comprehensive searching of the various databases less time-consuming. VA awarded a contract for the development of an electronic tool to search multiple databases and check for potential duplication among health research applications funded by various agencies and other sources. According to VA officials, this tool, when implemented, will allow these officials to identify in a timely manner applications that are most likely to be duplicative.
Officials at NIH, DOD, and VA acknowledged that duplication may sometimes go undetected. GAO performed searches on funded applications for breast cancer and PTSD research in NIHs database and DODs Congressionally Directed Medical Research Programs website using various key words frequently found in related research. While most of the applications identified did not appear to be duplicative, GAO identified two applications, one funded by VA and the other by DOD, that a VA program manager confirmed were duplicative as described in the databases. However, the databases were not updated to reflect modifications that had been made to the applications aims. The VA official told GAO that these two applications were originally identical and submitted by the same principal investigator. VA funded one of the applications with the understanding that DOD would not fund the second, duplicative application. Subsequently, according to DOD officials, DOD funded the second application after the principal investigator made some modifications to its aims in order to make it no longer duplicative. However, VA officials did not have information on DODs funding of the application or on how it had been modified. This example illustrates how the databases used to check for duplication in health research do not always provide comprehensive information needed to evaluate research for potential duplication across federal agencies during the funding decision process.
Officials at NIH, DOD, and VA also stated that they consider the opinions of peer reviewers, who are typically scientists or professors who score proposals for scientific merit, to determine whether applications may be duplicative of other research. NIH and VA applications have a required section where principal investigators and other key personnel must list all current funding they receive and all other applications they have submitted at the time of their application. Peer reviewers generally have access to this information when scoring the proposals.
According to VA officials, NIHs database contains information on about one quarter of all VA-funded health research applications. VA officials told us that they are working to add information on most VA-funded applications to this database by August 2012. In addition, NIH officials stated that they search NIHs database for information on proposals funded by NIH.
NIH, DOD, and VA officials told us that they also may search other databases, such as clinicaltrials.gov, DeployMed ResearchLINK, and PubMed, which contain information on federally funded health research.
NIH officials said the system that provides information to NIHs database may contain additional information for VA applications, such as the actual application and supporting documentation; however, this information is only available to NIH and VA officials.
Officials told us that they check this information prior to funding to ensure that the application is not duplicative of other federally funded research conducted by the principal investigator.
This tool will be completed by June 28, 2012, according to VAs contractor. After its completion, VA plans to use it internally to analyze its research portfolio and to identify potential duplication across research funded by various entities. VA also plans to make some information resulting from its use of the tool available to the public.
The searches we performed were not comprehensive or generalizable.
Because multiple federal agencies fund research on topics of common interest, there is potential for unnecessary duplication. As long as research on similar topics continues to be funded by separate agencies, it is incumbent on the agencies to coordinate effectively with each other. While NIH, DOD, and VA take steps to check for duplication in the health research they fund, the agencies have opportunities to improve sharing of information needed to evaluate research for potential duplication when making funding decisions. In order to do so, the Director of NIH as well as the Secretaries of DOD and VA should
For example, NIH, DOD, and VA could collaborate to allow for more efficient, comprehensive searches to identify duplication, by, for example, increasing commonalities among their respective databases; providing additional information in their respective databases, such as more details on the aims and methodology of applications that may be useful to program managers evaluating applications for duplication; and ensuring contact information for agency officials associated with specific applications is made available in their respective databases, if possible. NIH, DOD, and VA could also provide program managers with information to help them identify when they receive similar applications and to monitor the funding status of these applications, such as which applications receive funding, and which are modified during the funding process.
Determining ways to improve access to comprehensive information and to improve officials ability to identify duplication could help agency officials in their efforts to avoid duplication when determining which health research applications to fund.
The information contained in this analysis is based on findings from the products listed in the related GAO reports section as well as additional work GAO conducted.GAO used breast cancer and PTSD research as examples of areas of research that are funded by these three agencies. Within NIH, GAO focused on the National Cancer Institute and the National Institute of Mental Health, because these entities fund the majority of breast cancer and PTSD research within NIH, respectively, according to NIH officials. Within DOD, GAO focused on the Defense Health Program and, within VA, the Office of Research and Development, because these entities fund the majority of health research within DOD and VA, according to officials with whom GAO spoke. GAO focused its work on coordination across federal agencies that impacts decisions to fund health research. GAO collected and analyzed documents provided by NIH, DOD, and VA officials. GAO did not focus its review on coordination within federal agencies. In addition, GAO searched the available databases containing information on applications funded by NIH, DOD, and VARePORTER and DODs Congressionally Directed Medical Research Programs websiteto identify examples of potentially duplicative research applications funded by these agencies. GAO searched for the terms breast cancer and PTSD and then searched for terms that were frequently cited in titles that appeared to indicate potential duplication. GAO also interviewed 23 officials at NIH, DOD, and VA whom it selected because of their involvement in coordination across federal agencies when determining which research applications to fund in the areas of breast cancer and PTSD.
GAO provided a draft of this report section to HHS, DOD, and VA for review and comment. HHS and DOD provided written comments. DOD generally agreed with GAOs findings, and HHS did not state whether it agreed or disagreed. In its comments, on behalf of NIH, HHS provided more detail on NIHs policies and procedures concerning monitoring and managing potential overlap in funding, particularly within NIH. HHS also described an internal NIH database that is also available to VA staff and that provides more detailed information on grants than is included in NIHs public RePORTER database, but is not generally available to staff at other agencies. For this work, GAO focused on RePORTER because it is the NIH database that officials at other agencies told GAO they use when checking for information on NIH- or VA-funded research and is available to officials at all agencies. HHS and VA also provided technical comments, which were incorporated as appropriate. All written comments are reprinted in appendix IV of the PDF version of this report. As part of its routine audit work, GAO will track the extent to which progress has been made to address the identified actions and report to Congress.
For additional information about this area, contact Linda T. Kohn at (202) 512-7114 or firstname.lastname@example.org.
Comments from Department of Defense
Comments from Department of Health and Human Services