Roundtable Discussion on Research Career Paths in Rheumatic Diseases

Tuesday, March 25, 2008 (historical)

NIH Campus, Building 31, Room 4C32
Bethesda, Maryland

Co-chairs
Stephen I. Katz, M.D., Ph.D.
Susana Serrate-Sztein, M.D.

Introduction

For nearly 30 years, there has been a recognition that the number of physician scientists has been in decline. This trend has been observed in the rheumatic diseases field both in terms of a reduction in the number of researchers, and also in a decline in the number of rheumatologists in general. According to the American College of Rheumatology (ACR), 180 rheumatology fellows are trained per year, but this is not enough to sustain current programs.

The path that medical students take to become rheumatology researchers involves several stages. First, medical students who become interested in rheumatology research typically enter an internship/residency program to obtain the practical training necessary for becoming a certified physician. After approximately three years of residency training, these individuals would then enter a 2-5 year fellowship program in rheumatology research. Depending on the program and the trainee's interests, his or her research may be entirely non-patient oriented or may consist of clinically-oriented, human subjects research. After completing specialty training, the fellow is then expected to obtain an independent position and establish his or her research program. It is clear that preventing trainees from dropping out of a research path and moving toward other career options requires addressing multiple challenges that arise at various points.

The purpose of this roundtable meeting was to discuss how to attract, train, and sustain a strong rheumatology research workforce. The participants represented major stakeholders in improving the academic rheumatology workforce: National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS), Arthritis Foundation (AF), ACR, and ACR-Research and Education Foundation (ACR-REF), several leading academic rheumatologists, and two current rheumatology fellows. Specifically, participants were asked to address the needs and opportunities for developing future generations of physician scientists who will play a critical role in rheumatic diseases research.

Background

The NIAMS, AF, and ACR-REF offer a portfolio of funding mechanisms to support rheumatology trainees at various stages of their career. The NIAMS offers National Research Service Award Institutional Training Grants (T32) and Postdoctoral Individual National Research Service Awards (F32) which provide support around the beginning of the specialty training program. Note that the T32 is awarded to an institution to support multiple trainees of their choosing. In contrast, all other NIAMS training mechanisms are awarded to individuals. The Mentored Scientist Development Award (K01) and the Mentored Clinical Scientist Research Career Development Award (K08) are typically obtained later in the specialty training period, as is the Mentored Patient-Oriented Research Career Development Award (K23) which is sometimes also obtained after an independent position has been secured. The relatively new K99/R00 Pathway to Independence Award is meant to act as a bridge of support to an independent research position. Additional information on NIAMS training programs can be viewed at http://www.niams.nih.gov/Funding/Funding_Opportunities/activity_codes.asp.

The AF offers a Postdoctoral Fellowship Award that typically supports fellows at the beginning of their specialty training, and a Career Development Award to provide support just after an independent position is obtained. The ACR-REF Resident Research Preceptorship Award provides three months of support to introduce medical residents to research. The ACR-REF Physician Scientist Development Award and Clinical Investigator Fellowship offer support during the specialty training period. More information on these AF and ACR-REF awards can be found, respectively, at the following links: http://www.arthritis.org/information-for-researchers.php and http://www.rheumatology.org/ref/awards/index.asp.

Discussion

Defining the problem - insights from evaluations and anecdotal evidence

In discussing the obstacles that prevent trainees from becoming independent investigators, it is useful to break up the pathway into two major stages: the T32 (fellow)-to-K award transition and the K award-to-R (independent investigator) award transition. Both of these transitions present major challenges to trainees, but the nature of these obstacles tends to be different.

Several evaluations that have been conducted recently have provided important insights into the factors that must be addressed in order to increase the number of academic rheumatologists. In September 2007, an outside working group released an evaluation report focused on the effectiveness of the T32, F32, K01, and K08 funding mechanisms offered by NIAMS. A major outcome measure used to determine success of these programs was the ability of recipients to later obtain R01 funding. One caveat to this approach is that some independent investigators are funded through other mechanisms (e.g., other R grants, program project grants, non-NIH awards, etc.). Nevertheless, the evaluation has provided many insights into the status of training programs at the NIAMS. The full report can be viewed on the NIAMS Web site at http://www.niams.nih.gov/News_and_Events/Meetings_and_Events/Reports/2007/training_grant_eval_final_report.asp.

Results of the NIAMS training evaluation indicated that 17% of NIAMS T32 awardees go on to receive R01 funding, while much higher percentages of F32, K01, and K08 recipients (34%, 83%, and 55%, respectively) obtain R01 funding. The majority of NIAMS F32 recipients are Ph.D. fellows, while the majority of T32 recipients are M.D. fellows. While the T32 and F32 mechanisms are meant to target roughly the same career stage, anecdotal evidence suggests that for M.D. fellows, the T32 mechanism is used at an earlier stage than the F32. Therefore, because F32 and K awardees having M.D. degrees tend to be more senior than trainees appointed to T32s, it is perhaps not surprising that F32 and K recipients are more likely to reach successful research careers. It is difficult to predict early in the pathway which trainees will become successful. Thus, it may be necessary to support a relatively large number of fellows at this stage, even if only a small percentage go on to become independent researchers. Finally, the high rate of success of the K01 and K08 mechanisms, especially if one looks at success in terms of the percentage of recipients still involved in research, indicates that there is a robust rate of return for funds invested in these programs.

It is also important to consider that the T32 program has other indirect benefits to academic programs besides helping to produce investigators who obtain R01s. For example, clinical rheumatologists who don't have significant research funding, but are still involved in research projects, are often trained via the T32 mechanism. Also, when an institution has a T32 program, it is more likely to provide fellows with training in performing clinical trials that is much more extensive than what is offered by involvement in industry-sponsored clinical trials.

The quality of mentoring has received attention as a potentially important factor determining whether a rheumatology trainee progresses to an independent research position. In recent years, several surveys of rheumatologists have illuminated the obstacles that trainees face when considering a career in research. In particular, an insightful study of rheumatology fellows is currently being prepared for publication by the ACR Committee on Research, Subcommittee of Young Investigators. Based on the survey results, fellows generally thought that the training programs they were enrolled in, and the quality of mentoring that they received, were very good. However, a potential caveat is that that the fellows may have only considered mentoring related to research and funding when answering the survey, rather than mentoring related to career issues more broadly such as how to successfully manage a research program and navigate personnel issues. Thus, the quality of mentoring that relates to career issues appears to still be an open question.

The effect of academic salaries on the decision to enter, or stay in, a career in rheumatology research is complex. The relatively low salaries earned during the fellowship years, particularly during the K-to-R transition, are thought to be an obstacle to continuing down the career path to independence. This is due to the fact that many fellows are burdened by large educational debt and are often beginning a stage in their lives where there is a perceived need for higher income (e.g., due to a growing family). The NIH Loan Repayment Program (LRP; see below) is targeted to this need. Similarly, the salary level later in the career path for independent investigators tends to be lower than that of private practice or industry. The extent to which this affects the decision to choose a career in academic rheumatology appears to differ according to the gender of the respondents. The salary level for more senior positions tended to be an obstacle for many male respondents, but was much less of an obstacle for females. Generally, those that were strongly interested in research were willing to tolerate the lower salaries.

Based on the ACR survey, the biggest obstacle for rheumatology fellows is the anxiety over obtaining research funding at the NIH. This is a major impediment at the T32-to-K transition. As fellows look ahead to the possibility of independent positions in rheumatology research, their views are affected by observing their mentors struggling to get NIH research funding. The effect that the funding environment has on their decision to pursue a rheumatology research career has been exacerbated by a number of factors that have recently come into play. These include the reduced state and/or institutional support for research, the fact that fellows are older now and in more of a hurry to start independent careers, and the institutional pressure on principal investigators to have two or more R01 grants to support their research. Another important issue is that industry and private practice are exerting a greater pull on the available pool of trainees than ever before. This is in part due to the higher salaries offered by industry and private practice, and the perception that careers in these fields offer more stability than in academia.

Possible solutions to career path obstacles

Although several mechanisms are available for rheumatology fellows from the NIAMS and other organizations, making the transition from an early training fellowship (e.g., T32) to a K award is challenging. While some solutions already exist, others could be implemented. The NIH LRP plays a very important role in lowering the opportunity cost of staying in academia. This 2-3 year award pays trainees up to $35,000 per year toward the relief of their qualified student loan debt. The success rate for the LRP is quite favorable (around 50%). However, this program needs to be better publicized as many residents and fellows are still not aware of it. The LRP is currently directed towards clinical researchers; however, the meeting participants suggested that the NIH may want to consider expanding it into basic research as well. For more information on the NIH LRP, please visit http://www.lrp.nih.gov/index.aspx.

Another promising approach to reduce attrition during the T32-to-K phase is to enhance mentoring of trainees on career issues (e.g., understanding the NIH grants process, grant writing, project management, and enhancing the mentor-mentee relationship). There are some significant efforts in this arena already. For example, some NIH Institutes and Centers offer workshops that explain to junior scientists how the funding process works, how to improve the mentor-mentee relationship, and how to manage a research program. NIAMS staff members periodically give presentations at scientific conferences about career issues, focusing primarily on grant mechanisms and preparation. The ACR is organizing a new conference for junior rheumatology researchers to begin in 2008. Half of this conference will focus on presenting current research findings and the other half will be directed towards career issues and planning.

Another major obstacle to getting past the T32-to-K transition is the difficulty in obtaining the K award itself. The success rates for the NIAMS K08 and K23 awards over the past three years have averaged 34% and 40%, respectively. The ACR-REF plans to soon offer bridge funding, available for 1-2 years, for trainees who received a good, but not fundable, score on their applications for K08 and K23 awards. The purpose of this bridge funding is to ensure that qualified trainees have the highest likelihood of achieving success in obtaining future NIH awards. Thus, it allows them to continue to fund their fellowship while they prepare a resubmission of their K application.

To address challenges in the K-to-R01 transition, there are a number of possible approaches. There is a general feeling that the AF and ACR-REF can come together to find ways to move more fellows to independent positions. The AF is currently designing a program to supplement K awards. A major goal of this program is to address the relatively low salaries earned by trainees during their fellowship. However, in some cases, it may be more productive for some or all of the supplement to go to purchase of research supplies and equipment and/or to support research staff salaries. In this case, the supplement could enable the fellow to move his or her research further along and thus produce more of the preliminary data needed for future R01 applications. Potentially, the use of the supplemental funds could be determined on a case-by-case basis in order to have the greatest positive impact.

NIAMS Program Directors sometimes observe extremely qualified rheumatology fellows applying for K grants, when they would probably be competitive for an R01. A number of possible approaches to encourage these individuals to apply for R01 grants more quickly were discussed. Some participants in the discussion felt that academic institutions should try to be more flexible with the timing of tenure and promotion for fellows who have family responsibilities. More effort also needs to be put into getting the word out to junior investigators regarding the flexibility that already exists in this regard. In addition, all stakeholders should continue to consider whether it is more productive to place additional resources at the earlier part of the training pathway or at the late stages. Perhaps the earlier part of the pathway would be most productive due to the fact that institutions are generally more amenable to providing support to more senior trainees given that much has been invested in them by that stage already. Additional training surveys and evaluations could potentially shed light on this issue.

Summary of next steps

In the near future, the NIAMS, AF, ACR, and ACR-REF will be involved in various activities with the goal of increasing the number of rheumatology trainees that become successful independent investigators. The ACR is currently carrying out a large rheumatology workforce study and will introduce an action plan based on the results by the end of 2008. In addition, the ACR survey of fellows will soon be published. The ACR-REF will implement the bridge award that targets the T32-to-K transition. Likewise, the AF will implement their plan to offer a supplement to K awards, pending decisions on how the funds are to be used. The NIAMS will continue to assess policies that help facilitate the retention of junior investigators. To better inform these policy decisions, NIAMS plans to soon implement a mechanism for prospective data collection for recipients of training awards. Finally, NIAMS will continue to explore potential partnerships with organizations such as the AF and ACR, as appropriate.

Related Links

Description of NIAMS Research and Training Mechanisms
http://www.niams.nih.gov/Funding/Funding_Opportunities/activity_codes.asp

Evaluation of NIAMS T32, F32, K01, and K08 Mechanisms
http://www.niams.nih.gov/News_and_Events/Meetings_and_Events/Reports/2007/training_grant_eval_final_report.asp

Description of the NIH Loan Repayment Program
http://www.lrp.nih.gov/index.aspx

Description of grants and awards offered by the AF
http://www.arthritis.org/information-for-researchers.php

Description of the ACR-REF grants and awards program
http://www.rheumatology.org/ref/awards/index.asp

Outside Participants

CROFFORD, Leslie, M.D.
Gloria W. Singletary Professor
Chief, Division of Rheumatology
Director, Center for the Advancement of Women's Health
University of Kentucky

DAIKH, David, M.D., Ph.D.
Assistant Professor, Medicine
Chief, Division of Rheumatology
University of California at San Francisco

DAVIDSON, Anne, M.D.
Associate Professor
The Feinstein Institute for Medical Research
Northshore University Hospital

FITZGERALD, John, M.D., M.P.H., M.B.A.
Assistant Professor, Medicine
Department of Medicine/Rheumatology
University of California at Los Angeles

FOX, David, M.D.
Professor, Department of Internal Medicine
Chief, Division of Rheumatology
Director, Rheumatic Disease Core Center
University of Michigan

HARDIN, John, M.D.
Department of Medicine
Albert Einstein College of Medicine

HERNANDEZ, Milton, Ph.D.
Director, Office of Special Populations and Research Training
National Institute of Allergy and Infectious Diseases
National Institutes of Health

KARP, David R., M.D., Ph.D.
Associate Professor
Rheumatic Diseases Division
University of Texas Southwestern Medical Center

LANE, Nancy, M.D.
Director, Center for Healthy Aging
Vice Chair of Research, Department of Medicine
Professor of Medicine and Rheumatology
University of California at Davis

PARK, Christy, M.D.
Department of Medicine
University of Tennessee

ROSEN, Antony, M.D.
Mary Betty Stevens Professor of Medicine
Director, Division of Rheumatology
Johns Hopkins University