|
|
||||||||
1 Robert Wood Johnson Clinical Scholars Program, University of Pennsylvania, Philadelphia
2 Department of Family Medicine and Community Health, University of Pennsylvania, Philadelphia
3 Leonard Davis Institute of Health Economics, University of Pennsylvania, Philadelphia
4 Robert Graham Center, Policy Studies in Family Medicine and Primary Care, Washington, DC
CORRESPONDING AUTHOR: Sean C. Lucan, MD, MPH, Robert Wood Johnson Clinical Scholars Program, University of Pennsylvania, 423 Guardian Dr Blockley Hall, 13th Floor, Philadelphia, PA 19104-6021, sean.lucan{at}uphs.upenn.edu
| ABSTRACT |
|---|
|
|
|---|
METHODS We descriptively analyzed NIH grants to family medicine from 2002 through 2006 and the current NIH advisory committee memberships.
RESULTS Grants (and dollars) awarded to departments of family medicine increased from 89 ($25.6 million) in 2002, to 154 ($44.6 million) in 2006. These values represented only 0.20% (0.15% for dollars) and 0.33% (0.22% for dollars), respectively, of total NIH awards. Nearly 75% of family medicine grants came from just 6 of NIHs grant-funding 24 institutes and centers. Although having disproportionately fewer grant continuations (62% vs 72%) and R awards (68% vs 74%)—particularly R01 awards (53% vs 84%)—relative to NIH grantees overall, family medicine earned proportionately more new (28% vs 21%) and K awards (25% vs 9%) and had more physician principal investigators (52% vs 15%). Ten of the nations 132 departments of family medicine (7.6%) earned almost 50% of all family medicine awards. Representatives from family medicine were on 6.4% of NIH advisory committees (0.38% of all members); family physicians were on 2.7% (0.16% of members).
CONCLUSIONS Departments of family medicine, and family physicians in particular, receive a miniscule proportion of NIH grant funding and have correspondingly minimal representation on standing NIH advisory committees. Family medicines engagement at the NIH remains near well-documented historic lows, undermining family medicines potential for translating medical knowledge into community practice, and advancing knowledge to improve health care and health for the US population as a whole.
Key Words: National Institutes of Health family practice family physicians research advisory committees
It is not down in any map; true places never are.Herman Melville
| INTRODUCTION |
|---|
|
|
|---|
In contrast, family medicine has a well-developed clinical infrastructure with high productivity, boasting the second greatest number of clinicians14 and providing the highest number of health care encounters in the United States.15 Because of substantial clinical presence, family physicians may be instrumental in helping to bridge the chasm between medical knowledge and actual clinical care toward improved population health.16 Such bridging is a stated focus of the National Institutes of Health (NIH), highlighted for instance in NIHs Roadmap for Medical Research.17 In fact, Roadmap has been cited as "an immediate opportunity for family medicine to contribute to the mission of the NIH."6 Family medicines clinicians, for example, might participate in Roadmaps Clinical and Translational Science Awards,18 partnering with veteran investigators, to perform practice-based research in community practices.16 By contrast, family medicines researchers could potentially lead some of these efforts.7, 13
Unfortunately, though, neither family medicines clinicians nor its researchers have strong foundations in these areas: family medicine has had a historically poor track record with the NIH,10 perhaps in part because of the NIH structure. NIH, the nations leading source of support for biomedical research,19 comprises 20 institutes and 7 centers that focus primarily on specific organ systems, disease categories, and scientific specialties.20 Although new initiatives, such as Roadmap and Clinical and Translational Science Awards, could foster greater engagement with primary care, the research interests of family medicine may still not be well-aligned with the agendas of NIH institutes and centers—potentially to the disadvantage of family medicine research, the NIH mission, and overall population health in this country.
Because the agendas of the institutes and centers are largely set by NIH advisory committees, increasing family medicine representation on such committees provides an opportunity for greater engagement and is a goal of family medicines leaders.1,6 Through membership, family medicine can participate in research review and help set priorities for national research spending. As a baseline, we were unable to identify any reports quantifying family medicines membership on NIH advisory committees in the published literature.
Increasing funding from NIH is also a goal of family medicine leaders.1,6,13 Levels of funding for family medicine were reported most recently by Rabinowitz et al,21 but only for a single year (2003) and with no estimates of change through time. Trends in funding were assessed by Campos-Outcalt et al for the years 1984 to 1997.10 The conclusions of that study may no longer apply, because from 1998 to 2003 Congress doubled the NIH budget.22 Since 2003 no studies have evaluated trends in NIH grants to family medicine.
Our study had 2 specific aims: (1) to quantify current family medicine membership on NIH advisory committees, and (2) to assess the level and distribution of NIH grants to departments of family medicine during the past 5 years (2002 through 2006).
| METHODS |
|---|
|
|
|---|
NIH Grants to Family Medicine
From the NIH Office of Extramural Research, we received data on all grants awarded to departments of family medicine from 2002 through 2006. The Office of Extramural Research (OER) provided these data—including grant number, award amount, fiscal year, principal investigators degrees, and institution—in response to our Freedom of Information Act request. Although OER assured us of the accurateness and completeness of provided data, they would provide no information on unfunded applications "due to the privacy protection offered." From grant numbers, we determined application type, activity code, and the NIH institute or center.23 We found values for NIH grant totals on the NIH Web site.24
Principal investigators were categorized as 1 of 4 types: physician (medical degree [MD or DO] with or without other degrees but no other doctoral degree), scientist (doctor of philosophy [PhD], doctor of science [ScD or DSc], or doctor of public health [DrPH] without a medical degree), physician-scientist (medical degree with additional doctoral degree), or other (no medical or other doctoral degree).
Family Medicine on NIH Committees
We established a universe and typology (Table 1
) of NIH advisory committees from the NIH Web site.25 A high-ranking administrator in the NIH Office of Federal Advisory Committee Policy reviewed a preliminary committee list and ultimately endorsed our final spreadsheet as comprehensive and accurate.
|
| RESULTS |
|---|
|
|
|---|
|
More than two-thirds of family medicine awards were for R grants (Table 2
), of which one-half were for R01 awards (discrete projects by principal investigators). K awards (research career programs grants) accounted for 25% of family medicine grants. Over the 5 years, there was an increase in the proportion of K awards relative to R awards for family medicine. In contrast, grant proportions for the NIH in general remained stable over the 5 years, with R awards accounting for about 75% (nearly 85% R01) and K awards accounting for 9% of the total.28
Physicians in family medicine were awarded more than one-half of all grants (Table 2
). Two-thirds of these physician grantees had additional masters level training, and another 10% had doctoral degrees. In general, the great majority of NIH grants (as much as 85%) go to grantees with doctoral degrees.29
One-half to two-thirds of family medicine grants in any given year were noncompeting continuations (Table 2
). New applications accounted for as little as one-fifth or as much as one-third of awarded family medicine grants. For NIH totals, noncompeting continuations accounted for two-thirds to three-quarters of the total (average 72%), and new requests made up one-fifth to one-quarter (average 21%).30
Of the 132 departments of family medicine in US medical schools, 63 (48%) earned NIH grants in the last 5 years. Ten departments (16% of those funded, 7.6% of the US total) accounted for almost 50% of all family medicine awards; 4 departments cumulatively earned 25% of awards, and 1 department alone earned 11% (Table 2
). From 2002 through 2006, 25 of the funded departments (40%) received less than 1 grant per year; and 8 (13%) received only 1 grant total. Three of the top-10 departments of family medicine were from the top-10 medical schools in terms of total NIH grants; 6 were from the top-20 schools in terms of total NIH grants.31
Distributions by awarded dollars (displayed in the Supplemental Appendix, available at http://www.annfammed.org/cgi/content/full/6/6/534/DC1) were essentially similar to distributions by numbers of grants. Notably though, 4 of the top 10 departments of family medicine by number of grants did not appear in the top 10 departments for grant dollars—mostly because of relative proportions of R and K awards, and the greater dollar amounts (generally about 250% greater) awarded for R grants.
In 2006, family medicine earned no awards from one-third of all grant-funding NIH institutes and centers (Table 3
).
|
|
|
| DISCUSSION |
|---|
|
|
|---|
Second, family medicine has minimal representation on NIH advisory committees. Stakeholders from departments of family medicine are, thus, largely disenfranchised, having few seats at the table and little say in setting the national research agenda. This lack of representation is particularly true and perhaps most ironic for family physicians, whose extensive clinical practices32,33 and community ties make them ideally suited for advising on dissemination, implementation, and adoption of research.34
As the NIH struggles to translate the fruits of biomedical research into practice, family medicine struggles to fortify and invigorate its research capacity. A less-understood outcome of this mutual frustration is the failure to make the more than 200 million annual patient visits to family physician practices15 a source of medical discovery, new knowledge, and greater engagement between family medicine and NIH. Results from our study and previous work suggest potential strategies to facilitate greater connection and mutual benefit.
One strategy focuses on committee membership, both for practicing family medicine clinicians and career family medicine investigators. Practicing family physicians could serve in public seats on NIH advisory committees, offering community experience, clinical insights, and practice perspective. The Federal Advisory Committee Act requires that membership on NIH advisory committees be fairly balanced in terms of points of view represented, and national advisory councils and boards and program advisory committees specifically must have one-third public members by law.35 The other two-thirds are for scientific members—such as career family medicine investigators—with established records of research success.35 Although it is unclear that greater participation on NIH advisory committees will result in greater funding for family medicine research, it will certainly give family physicians greater voice in setting national research priorities and in shaping NIH efforts to translate basic science discovery to front-line clinical practice. Self-nomination to advisory committees is one mechanism that makes increased membership possible.35
A second strategy is greater research training. In our data, most family physician grantees and family medicine committee members had additional advanced degrees. NIH and family medicine could strengthen partnerships for advanced degrees and research training through National Research Service Award expansion, greater involvement in the Howard Hughes Medical Institute, NIH fellowships, or novel research training programs.
A third strategy is to draw on past successes by family medicine. As shown in our study and as noted by others,36 only a select group of departments of family medicine have any substantial engagement with the NIH. Through hub-and-spoke networking,13 these sites might supply role models and teachers to bring lessons about building successful research enterprise to the larger family medicine community (perhaps even through novel programs and financial backing from the NIH). Lessons from leaders could be helpful, not only for community family medicine clinicians,37 by fostering participation in practice-based investigations and quality-improvement initiatives, but also for aspiring career family medicine investigators,13 by providing guidance on grantsmanship and counsel on doing national-level research.
A final, related, strategy involves partnerships with existing research infrastructure, both within and outside family medicine.2,13 Family physicians in community practices might develop collaborative relationships with practice-based research networks38 and academic health centers that are part of the Clinical and Translational Science Awards consortium.6 They could thereby contribute to generating new knowledge by informing research questions,13 carrying out trials, and developing ideas for new kinds of research.16 For established family medicine investigators, expanding roles as collaborators, consultants, coinvestigators, and mentors will be important. Mentoring in particular will be key as family medicine strives to increase numbers in its research vanguard, developing new talent capable of independent investigation.1,7,13 Burgeoning junior investigators could look to veteran grantees—preferably in family medicine but also in other disciplines—for mentoring through career development programs (ie, K awards).1,6,13,21 Encouragingly, with disproportionate percentages of new awards and K awards going to family medicine during the last 5 years, partnering with mentors and existing research infrastructure may already be occurring.
Our study has notable strengths. Analyses of awards over the last 5 years allowed us to appreciate that there has been little substantive change in funding to family medicine since previous analyses, specifically in relation to the doubling of the NIH budget. We also considered application type, which had not been reported previously. Data on advisory committee memberships provided additional perspective and corroboration not found in published literature.
In spite of these strengths, we must consider limitations in interpreting our results. First, we can make no comment about family medicine investigators who were collaborators or co–principal investigators, as such data were not available from NIH. This omission is important because research is increasingly interdisciplinary and networked.39 Although the ability to perform autonomous investigation has traditionally been a measure of the status and vitality of a discipline,29 research as a culture is shifting to encompass more collaborative and team-based models.39,40 NIH just started collecting reliable data on co–principal investigators in 2006, and future studies should evaluate family medicines role in collaboration. Indeed, interdisciplinary cooperation may be essential for family medicine to develop a critical mass of competent investigators.1,13
Second, there was some risk of misclassifying researchers in our data owing to family physicians working outside departments of family medicine or principal investigators having multiple departmental affiliations, for example. Rabinowitz et al reported 22 out of 149 cases (15%) in 2003 in which family medicine researchers were actually listed under departments of public health and preventive medicine.21 Although we received assurance from Office of Extramural Research that the data provided for our study was complete and accurate, if the same level of misclassification were at play, family medicine would have actually received 0.32% of NIH awards from 2002 through 2006 rather than the 0.28% we report. The potential for misclassification or missed identification for advisory committee members is also a concern. Only a single affiliation for each committee member was listed; thus, any family medicine researchers using a primary, non–family medicine affiliation would have been uncounted. Conversely, any principally non–family medicine researcher using family medicine as a primary affiliation would have been captured inappropriately. Given the tiny membership percentages we report, however, even large errors would not have changed our overall findings meaningfully.
Third, we had to limit our analysis of advisory committees to those with standing memberships—that is, we had to exclude special emphasis panels. To be generous, even if we were to make the unlikely assumption that every special emphasis panel had at least 1 family medicine member, then the percentage of advisory committees with family medicine members would increase from the 6.4% we report to 13.5%, and family medicine membership would rise from 0.38% of all members to 0.82%. In our view, this change would not represent a meaningful difference.
Finally, we do not provide a comparison group—another medical specialty, for instance—for our analyses. Listed affiliations on committee rosters were not sufficient to distinguish generalists from subspecialists in other specialties. Based on prior work, however, family medicine spends substantially less time on research than other primary care fields.1,7,41 In fact, according to a 2001 report, departments of family medicine receive roughly 6% as much NIH funding as do departments of pediatrics, and about 1% as much as departments of internal medicine.42 Negotiations with NIH for data comparing grant applications and success rates among specialties are ongoing by our group.
Despite limitations, we can confidently conclude from our study that family medicine is not well-engaged with the NIH and is essentially off the Roadmap. The NIH director has written, "NIH needs to have an explicit and dynamic process for supporting critical scientific programs that cut across scientific areas and that none of the individual institutes could support on its own."26 This kind of sentiment drives the Roadmap and Clinical and Translational Science Award programs and speaks directly to the integrative role of family medicine.36 At the same time, the Future of Family Medicine project and family medicine leadership have called for increasing emphasis on research1,13,43,44 and engagement with the NIH.1,13 Family medicine and the NIH have what each other wants, but the potential for greater partnership has not been realized. Lack of family medicine involvement in the planning and performance of federally funded research has implications for the direction of biomedical research in this country, its relevance to actual patient care, and its ultimate impact on public health. Lack of NIH support for family medicine investigational initiatives has implications for family medicines research capacity and the status and vitality of the specialty. Through greater research training, collaborations, and self-advocacy, family medicine might foster a better relationship with the NIH, to the benefit of both.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Disclaimer: The information and opinions contained in research from the Robert Graham Center do not necessarily reflect the views or policy of the American Academy of Family Physicians.
Received for publication December 22, 2007. Revision received June 25, 2008. Accepted for publication June 30, 2008.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. James, A. Davis, J. Borkan, and the Association of Departments of Family Medicine THE CHALLENGE TO BUILD RESEARCH CAPACITY IN FAMILY MEDICINE: IS OUR DISCIPLINE READY? Ann. Fam. Med, July 1, 2010; 8(4): 371 - 373. [Full Text] [PDF] |
||||
![]() |
K. C. Stange In This Issue: Cohort Studies Ann. Fam. Med, November 1, 2008; 6(6): 482 - 482. [Full Text] [PDF] |
||||
Read all TRACK Comments
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |