Printed from:

The Impact of Administrative Academic Units (AAU) Grants on the Family Medicine Research Enterprise in the United States

Christopher P. Morley, PhD, MA; Brianna J. Cameron, MPH; Andrew W. Bazemore, MD, MPH

Background and Objectives: The Health Resources and Services Administration (HRSA) awards funding to primary care departments—or “Academic Administrative Units” (AAUs) at US medical schools—to strengthen or grow these departments and ultimately increase the output of primary care physicians into the US workforce. One aspect of these AAU grants that is often overlooked is the fact that they support research infrastructure for these departments.

Methods: This study used multiple methods, including content analysis of current AAU grant abstracts (n=23), publications resulting from AAU funding (n=79), and survey responses from AAU project directors (n=19) to explore and describe the impact of current AAU grants on family medicine research in the United States.

Results: Federal support for family medicine departments remains very low compared to other disciplines. Several AAU grants have provided direct support for research activities as stipulated in the grant abstracts (6/23). However, most grants appear to have facilitated scholarly activity of some sort, including evaluation and quality improvement activities. Two practice-based research networks are supported with AAU funds, and at least 79 publications over the past 10 years, representing a wide variety of methodological approaches and topics, have been produced and indexed in PubMed with explicit acknowledgment of AAU funding.

Conclusions: In the absence of substantial NIH support for family medicine departments, the AAU funding mechanism remains a crucial, but often overlooked, factor in facilitating scholarly activity in departments of family medicine.

(Fam Med 2016;48(6):452-8.)

The absence of the specialty of family medicine from the funding pipeline provided to researchers by the National Institutes of Health (NIH) in the United States has been noted repeatedly.1-3 This issue has likely had a deleterious effect on the ability of faculty located within departments of family medicine (DFMs) and programs to conduct and disseminate research. For example, only half of all US family medicine residencies reported that a faculty member had produced a poster presentation or paper in the past 3 years.4

Given the scarcity of NIH funding awarded to DFMs, family medicine researchers often seek out funding from other sources, such as the Patient Centered Outcomes Research Institute (PCORI), the Agency for Healthcare Research and Quality (AHRQ), and private foundations. Another source of funding that potentially allows DFMs to pursue research, train future researchers, and participate in other scholarly activities is the Primary Care Training and Enhancement Program (PCTE), which is administered by the Health Resources and Services Administration (HRSA). Authorized by Title VII, Section 747 of the Public Health Service Act and the American Recovery and Reinvestment Act of 2009 (ARRA), PCTE provides grants that support capacity building in primary care.5,6 In 2013, $36.5 million was awarded to hospitals, medical schools, physician assistant training programs, and other public and private nonprofit entities under Title VII, Section 747 programs.5

PCTE provides funding in six program areas, one of which is the Academic Administrative Units in Primary Care Program (AAU). The AAU supports the establishment, maintenance, and improvement of academic units or other programs that improve clinical teaching and research or enhance interdisciplinary recruitment, training, and faculty development.7 AAU funding is available to departments of family medicine, general internal medicine, general pediatrics, or inter-departmental collaborations within accredited public or nonprofit private schools of allopathic or osteopathic medicine.7 Of all programmatic areas funded through the PCTE, the AAU mechanism is the only one that explicitly mentions research infrastructure within its mandate.

AAU grants allowed funding for a wide range of activities, including the development of learning collectives across regions, the development of new curricula in data analytics training for clinicians and for care transitions, clinical process improvements in academic medical practices, and a host of other projects. With $16.8 million awarded from 2010–2014, AAU grants also provide a substantial opportunity for DFMs to engage in departmental improvement and enhancement projects, including the maintenance of research infrastructure within and across primary care departments at medical schools.8 In the current 2011–2016 cycle, AAU grants provide up to $160,000 per year in funding for up to 5 years.8 By utilizing AAU grants to build research capacity and boost research productivity, the field of family medicine can increase knowledge and influence practice in areas closely tied to population health.

Although HRSA evaluates projects semi-annually to assess program progress, output, and impact, these evaluations are not publicly available. It is also not clear that the outcomes of research infrastructure support have ever been fully evaluated. Through performing a multi-method descriptive study examining current and past AAU grants, this study demonstrates and characterizes the substantial contribution that the AAU program makes to the research output of family medicine.




We utilized a multi-method descriptive approach to explore and characterize the impact of AAU funding upon the ability of recipient DFMs to conduct research and scholarship. Three convergent and complementary methods were used to assess AAU impact: a content analysis of the public abstracts of the 23 active AAU grants, a content analysis of publication abstracts identified through PubMed as acknowledging AAU funding, and a survey of current AAU project directors.

Content Analysis of Grant Abstracts for Current AAU Recipients

Active grant abstracts were obtained from the HRSA Data Warehouse8 (n=23) and imported into the Dedoose online qualitative analysis tool. An initial review of the 23 grant abstracts resulted in the development of a short code set. The codes were then applied to the text of the abstracts to assess the extent to which each grant abstract expressed a commitment to research, dissemination, and other activities. The overarching code “Research Support” was defined as the support of the conduct of original research (not including research training or translation of research). “Research mention” was applied when research was mentioned in a grant abstract but without direct reference to specific studies or activities.

Dissemination activities coded into three categories: “Publication,” “Conference,” and “Stakeholders,” which describe disseminating results through formal publication, academic conferences, and through stakeholders (eg, non-academic audiences or through non-specified means).

Codes for specific types of activities include “Evaluation,” which describes rigorous evaluation of educational programs, curricular components, or assessment methods, and “Quality Improvement” (QI), which describes support for rigorous QI activities. Additional emergent codes were developed for other cited topics, including support for Practice-Based Research Networks (PBRNs) and for patient-centered medical home (PCMH) activities.

Content Analysis of Abstracts Indexed on PubMed Citing Support From AAU

HRSA assigns AAU award numbers with a program prefix of “D54” to AAU grants. Additionally, during the 2009–2011 fiscal years, many awards were funded through the American Recovery and Reinvestment Act of 2009; AAU award numbers during these periods often had the prefix “D5A.” We therefore conducted a search of PubMed using the search string: D54*[Grant Number] OR D5A*[Grant Number]

This process revealed 79 articles as of May 2015. Supplemental searching of PubMed with relevant terms, such as “AAU” or “HRSA,” did not reveal unique results. The abstracts from the 79 articles were imported into ATLAS.ti version 7.5.4. Codes were developed within two broad themes: research methods and research topics. Within the research methods theme, codes were developed to describe the study type, data collection techniques, and data analysis methods. Within the research topics theme, emergent coding was utilized in order to capture the broad variety of research topics examined by family medicine research. This yielded two categories of codes within the research topics theme: disease-specific and other.

Survey of Program Directors

Each of the 23 project directors (PDs) are publically listed with contact information on the HRSA website. This information was collated and used to create an invitation list to participate in a brief, anonymous, web-based survey. In addition to four email invitations, which were sent between February and March 2015, a paper version of the survey was mailed with a pre-addressed, postage-paid return envelope. The corresponding author of this manuscript was included in the invitation list to respond on behalf of the funded AAU of which he serves as PD.

The survey instrument was constructed and deployed within and consisted of seven questions that asked about the PD’s primary department (family medicine, internal medicine, pediatrics, or other); any collaborative department; the year the current grant began (2010, 2011, or other); whether the respondent’s particular AAU funded research infrastructure in some way, offering several options and a comment field; whether AAU funding has lead to publication or similar scholarly work (peer-peviewed journal articles on PubMed, not on PubMed, scholarly books or chapters, conference abstracts and presentations, and an “other” comment field); whether the respondent always, usually, or never acknowledges AAU funding if they publish; and an open-ended field asking how the AAU affects formal research, scholarship, and evaluation activities in the funded department(s).

Human Subjects Protection

As all data analyzed for this study were either (1) publically available, (2) non-human in nature, or (3) collected anonymously, the Institutional Review Board of SUNY Upstate Medical University granted the study an exemption from review.




Content Analysis of Grant Abstracts for Current AAU Recipients

Out of the 23 active AAU grant abstracts, 10 (43.5%) explicitly referenced research in some way. Of those 10 abstracts, six abstracts expressed a commitment to the support of research activities or conduct of original research (21.7%), one of which also expressed a commitment to research training. Four abstracts mentioned research in a non-substantive way (17.4%), with references to incorporation of research training for faculty, residents or students, protecting faculty time for undefined research projects, or similarly non-specific goals.

Dissemination of results was a common theme emerging from content analysis of grant abstracts. One third of abstracts expressed intent to disseminate results through stakeholders or other non-academic means (34.8%). One quarter expressed a commitment to presenting results and grant outcomes at an academic conference (26.1%), and 17.4% declared an intention to publish results in an academic journal.

The vast majority of grant abstracts described a commitment to formal evaluation (82.6%), and 26.1% described engagement in quality improvement activities. Nine abstracts described PCMH practice transformation activities (39.1%), and two abstracts cited support or utilization of PBRNs (8.7%). In total, all but one abstract (n=22) described research activities, research-related activities, or knowledge dissemination of some sort (95.7%). See Table 1 for additional detail.




Content Analysis of Abstracts Indexed on PubMed Citing Support From AAU

Among PubMed-indexed publications citing support from the AAU program (n=79), the most prevalent study design was cross-sectional (36.7%). Cohort studies were also common; one fifth were prospective cohort studies (n=15, 19.0%), and four were retrospective cohort studies (5.1%). Eleven studies described high-level evaluation of education, training, or research programs (13.9%). Ten studies were review or summary articles (12.7%), five were interventional or randomized controlled trials (6.3%), and two were ecological or population-level (2.5%). Three studies were classified as “other” (3.8%), including two program evaluations not related to education, training, or research, as well as one article that tested an Internet search method for gathering information on chronic disease.

In terms of analytical methods, most studies used exclusively quantitative methods (64.6%) or mixed methods approaches (20.3%); only one used exclusively qualitative methods (1.3%). Eight studies utilized narrative literature review as the primary analytical method (10.1%), two performed a systematic review without a meta-analysis (2.5%), and one performed a systematic review with a meta-analysis (1.3%).

Data collection methods varied widely between studies, so coding categories are not comprehensive. One quarter of studies distributed surveys to collect original data (26.6%). Fourteen studies collected data through a PBRN (17.7%), and 11 studies collected data through literature review (13.9%). Seven studies used existing survey data (8.9%), six utilized chart review (7.6%), and five performed focus groups (6.3%).

Twenty-nine studies, or 36.7%, focused on a particular disease or type of disease, either being chronic (17.7%), acute or infectious (5.1%), disabilities (8.9%), or mental health (7.6%). Two studies focused on diseases that fell into more than one of the disease-specific categories. Just over a third of studies were categorized as health services research (34.2%), and just under a third examined health disparities (31.2%). Other common topics were medical training and curriculum development (15.2%), preventive medicine (11.4%), nutrition (7.6%), tobacco and alcohol use (6.3%), and technology (5.1%). Other topics were mentioned infrequently, including research capacity and funding (3.8%), quality improvement (2.5%), and pharmaceuticals (2.5%). See Table 2 for additional detail.




Survey of Project Directors

A total of 19 out of 23 current project directors, or 82.6%, responded to the survey invitation. Of those responses, 18/19 came from PDs within DFMs, with one respondent indicating affiliation with a department of pediatrics; this response indicated that the DFM at that institution, as well as internal medicine, were collaborating on the grant. Twelve of the 19 respondents indicated there was formal collaboration with other departments, including internal medicine, pediatrics, obstetrics and gynecology, psychiatry, geriatrics and palliative care, a physician assistant program, and a federally qualified health center. Of the grants given to each of the 19 respondents, 14 began in 2011, four began in 2010, and one began in a year other than 2010 or 2011.

Respondents indicated that the AAU program funded a variety of research types, including four PBRNs, baseline funding for faculty to conduct research and develop proposals, and formal training for both faculty and learners in research, evaluation, and QI methods (see Table 3). Ten of the 19 respondents indicated their grant funding had led to peer-reviewed publication, and 14 indicated that AAU funding had led to conference abstracts and presentations. Additionally, five of the respondents indicated the funding had supported publications not indexed on PubMed, and one indicated the funding had supported production of a book or book chapter. Fourteen respondents indicated they always acknowledge AAU funding when disseminating results, and two indicated they generally do (but a few papers or other items may be missing AAU acknowledgement).




Open-ended responses were recorded from 16 of 19 respondents. Table 3 includes the full text of these responses and general categorizations of those responses. Three responses indicated the AAU program supported the rigorous development of clinical innovations, particularly related to PCMH initiatives. Two respondents mentioned using AAU funds to augment educational programs and to evaluate those innovations. Five respondents described AAU funding as providing general protected time for faculty to do research or to partner with others, and two specifically mentioned proposal development stemming from AAU infrastructural support. Additionally, these open-ended respondents mentioned PBRN support, research training, and medical workforce research. Only two of the 16 open-
ended responses indicated no support for scholarly activity.




Out of the 23 active AAU grant abstracts, 10 (43.5%) explicitly mentioned research or cited support for research activities. However, all but one (n=22, 95.7%) referenced research, dissemination, or research-related activities. Since the grant abstracts of current AAU recipients were very general, these results likely underestimated the amount that grant recipients participated in each of these activities.

Although the majority of AAU grant abstracts did not specifically mention research, it is clear from the 79 PubMed-indexed articles that research output from AAU grants is strong and varied. Since 2005, the research output of AAU grants has been building each year, and diversity is growing in terms of study design, analytical tools, data collection methods, and research topics. It is also likely that the 79 papers that explicitly acknowledge AAU funding are a smaller part of a larger pool of published literature that had either less direct or unacknowledged connections to AAU grant support.

In addition to the content analysis of the PubMed-indexed abstracts, the survey of PDs helped to overcome the lack of detail included in the AAU grant abstracts by inquiring specifically about research activities conducted using grant funding. The survey of PDs, as well as the content analyses of grant abstracts and publication abstracts provided unique and corroborating evidence demonstrating the role of AAU grants in supporting family medicine research infrastructure.

The AAU mechanism provides about $4 million in funding to primary care departments each year, depending upon budgetary fluctuations. To place the AAU mechanism in perspective, Lucan and colleagues2 found that family medicine departments received about $57 million in funding from NIH per year, on average, between 2002–2006, which amounts to less than 0.2% of the total NIH budget. A preliminary analysis we have conducted indicates that this trend in NIH funding has not changed since Lucan’s findings were published. The AAU mechanism is therefore tiny in comparison to even the scant funding that comes to family medicine departments from NIH, but the relative scarcity of NIH funding for family medicine departments is precisely the reason that the AAU funds are important.


There are several limitations to this study that must be noted when considering the findings reported above. First, it is possible that the public abstracts of AAU grants do not completely reflect the nuance or full detail of the scholarly work performed by DFMs and other primary care departments. Similarly, it is possible that not all research papers have fully acknowledged the support of AAU grants, especially if the AAU mechanism primarily provided infrastructural background support for projects. Although we identified roughly 87 separate institutional affiliations on the 79 publications, we did not have the resources available to accurately parse the number of institutions or studies represented across this body of literature. The survey was limited by being designed to be quickly completed by PDs or their designees in order to facilitate a good response rate. An in-depth interview process or a lengthier survey may have extracted more detail from respondents. An additional shortcoming was that this project was conducted with a limited amount of time and resources available for the purpose. It is possible that with more time, a more thorough interview process might have been conducted with PDs. Also owing to limited resources, we did not explore the effects of other alternative (to NIH) funding sources, such as the Patient-Centered Outcomes Research Institute, the Agency for Healthcare Research and Quality, or other federal or nonprofit funders. We also did not extensively explore other PCTE mechanisms for research support, as the AAU mechanism is the only one that explicitly ever funded research infrastructure, and empirically, this can be seen in quick replications of our methods. For example, entering the grant prefix for the Title VII “Predoctoral Education” program—“D56 [Grant Number]”—into the PubMed search interface yields only five published articles, and there are no results found when using the “Faculty Development” prefix (D86). We could not detect any effect of Title VII impact upon family medicine research beyond the AAU mechanism. Given these preliminary results, it is possible, but not likely, that other Title VII PCTE grants also had measurable effects on family medicine research that our study was not designed to detect and describe. This study focused solely on one Title VII funding mechanism that explicitly allows support for research infrastructure.

Despite these limitations, we believe the multi-method approach triangulated on an accurate picture of the role of the AAU grant program in supporting scholarly activity within DFMs. The survey had a strong response rate with excellent coverage of family medicine PDs, and the results provide unprecedented insight into the impact of the AAU program on research and scholarship.




The primary initiative currently exploring the future direction of family medicine as a specialty in the US, Family Medicine for America’s Health (FMAHealth) has declared research as one of the six core areas that need to be addressed over the next 5 years.9 Additionally, the Research Core team of FMAHealth has specifically declared a need to link the family medicine research enterprise to the Triple Aim.

The AAU grant program provides an important source of research infrastructure support for family medicine in the United States. All three methods of examining the impact of AAU funding upon family medicine research are convergent—the AAU supports protected faculty time and the development of proposals, enables collaboration, and provides a backbone for several PBRNs.

If family medicine as a specialty is to advance research and scholarship, funding sources such as the AAU mechanism are vital. This is especially true, given the nature of what the AAU has funded over the past decade—projects that are core to Triple Aim values, as well as infrastructural support for PBRNs and research training. Unfortunately, PCTE, or “Title VII” funding, has waned over the years and is constantly threatened.10 The AAU program is also only a small part of PCTE. However, the AAU grant mechanism provides an important developmental step in furthering family medicine research and scholarship and should be continued in future funding appropriations.

Acknowledgments: This project was partially supported by the Health Resources and Services Administration (HRSA) of the United States Department of Health and Human Services (HHS) under grant number D54HP23297, “Academic Administrative Units,” for roughly $2,000 (total award amount $154,765; 90% financed with nongovernmental sources), corresponding to the amount of effort dedicated to this project by the lead author. This information or content and conclusions are those of the author and should not be construed as the official position or policy of, nor should any endorsements be inferred by HRSA, HHS, or the US Government.

Prior presentations: An abstract describing this work was accepted for inclusion at the North American Primary Care Research Group 2015 Annual Meeting in Cancun, Mexico.

Conflict of Interest Statement: The corresponding author of this report is also the project director of a current AAU grant, and the second author dedicated AAU-funded effort to this project while previously employed as a graduate student.

Corresponding Author: Address correspondence to Dr Morley, SUNY Upstate Medical University, Department of Family Medicine, 750 East Adams Street, MIMC 200, Syracuse, NY 13066. 315-464-7010. Fax: 315-464-6982.



  1. Lam CLK. The 21st century: the age of family medicine research? Ann Fam Med 2004;2 Suppl 2:S50-4. doi:10.1370/afm.191.
  2. Lucan SC, Phillips Jr RL, Bazemore AW, Phillips RL, Bazemore AW. Off the roadmap? Family medicine’s grant funding and committee representation at NIH. Ann Fam Med 2008;6(6):534-42. doi:10.1370/afm.911.
  3. Lucan SC, Barg FK, Bazemore AW, Phillips RLJ. Family medicine, the NIH, and the medical-research roadmap: perspectives from inside the NIH. Fam Med 2009;41(3):188-96.
  4. Young RA, Dehaven MJ, Passmore C, Baumer JG. Research participation, protected time, and research output by family physicians in family medicine residencies. Fam Med 2006;38(5):341-8.
  5. American College of Physicians. Changes to primary care training programs—Title VII, Section 747. 2013.
  6. Health Resources and Services Administration. American Recovery and Reinvestment Act (ARRA) Primary Care Training and Enhancement Program (PCTE). 2010.
  7. Health Resources and Services Administration. Academic Administrative Units in Primary Care (AAU). 2014;(December 5).
  8. Health Resources and Services Administration. Active Grants for HRSA Program(s): Academic Administrative Units in Primary Care (D54). 2014;(December 5).
  9. About FMAHealth. Accessed May 18, 2015.
  10. Harrison B, Bazemore AW, Dodoo MS, Teevan B, Wittenburg H, Phillips RL. Title VII’s decline: shrinking investment in the primary care training pipeline. Am Fam Physician 2009;80(8):872.

From the Department of Family Medicine, SUNY Upstate Medical University (Dr Morley and Ms Cameron); and Robert Graham Center for Policy Studies, Washington, DC (Dr Bazemore).

Copyright 2017 by Society of Teachers of Family Medicine