|Home | About | Journals | Submit | Contact Us | Français|
Successful strategies by which to effectively recruit and retain academic subspecialists in benign hematology have not been established.
To evaluate the effectiveness of a grant-funded, mentored fellowship with respect to retention and early career goals in hemostasis/thrombosis.
Via a nested case-control survey study, we sought to compare outcomes for graduates of a grant-funded, mentored fellowship training program in hemostasis/thrombosis (the National Hemophilia Foundation [NHF]-Baxter Clinical Fellowship Award) during conventional hematology/oncology fellowship training (cases), versus their training peers who were graduates of conventional hematology/oncology fellowship training alone (controls).
Survey response rate was 85% (11/13) for cases and 90% (9/10) for controls. All respondents had pursued careers in academic hematology/oncology. Median (range) percent time spent in benign hematology post-fellowship was 98% (70–100%) for cases versus 0% (0–20%) for controls. Time spent in research was significantly greater among cases than controls (median 80% [range: 42–90%] vs. 55% [10–80%], respectively; P=0.01). By years 3–4 post-fellowship, median annual number of peer-reviewed publications was higher for cases than controls (3.5 vs. 1.0; P=0.01). Cases were also more successful in grant funding (including K-awards).
These data suggest that a grant-funded, mentored fellowship training program in hemostasis/thrombosis may be superior to conventional hematology/oncology fellowship training alone with respect to outcomes of retention in clinical care/research, early-career grant funding, and publication productivity.
Successful strategies by which to effectively recruit, train, and retain academic subspecialists in benign hematology remain unclear. In the UK, a particular challenge has been identified for transfusion medicine subspecialists , and throughout Europe and the United States for hemophilia treatment experts . Innovative means of recruitment into training programs have been deemed critical to the future of these areas and the academic hematology subspecialty in particular.
The National Hemophilia Foundation (NHF)-Baxter Clinical Fellowship Program, funded through an educational grant by Baxter Biosciences, was developed in 2002 to educate and train new physicians in comprehensive care and research for individuals with bleeding and clotting disorders. “Centers of excellence” within the Centers for Disease Control and Prevention and Maternal and Child Health Bureau-established Hemophilia Treatment Center network were selected as training sites for the Program following institutional application, based upon the following criteria: 1) established hemophilia/thrombophilia treatment center with both clinical and research faculty; 2) affiliation with major universities and teaching hospitals; 3) strong track record and future plan for hemostasis/thrombosis trainee recruitment; 4) adequate patient base and volume in hemostasis/thrombosis; 5) institutional track record and resources for training; 6) depth of research training opportunities; 7) development of a curriculum for education in: hemostasis; thrombosis; treatment of hemophilia, von Willebrand disease and other bleeding disorders; and treatment of thrombophilia and thrombotic diseases. Via a peer review process involving representatives of the NHF's Medical and Scientific Advisory Council, seventeen institutions were approved for the Program, based upon fulfillment of the aforementioned qualifications.
Approved institutions were subsequently invited to nominate trainee candidates. Candidates were selected for the NHF-Baxter Clinical Fellowship Award based upon the following criteria: 1) medical degree; 2) eligibility for subspecialty board certification upon completion of the two-year fellowship; 3) United States citizenship or permanent resident status; 4) qualifications and previous experience; 5) long term goals, including desire to remain in the hemostasis/thrombosis field; 6) letters of recommendation provided by the hemophilia/thrombophilia treatment center Medical Director and from other faculty within the institution. The mentorship plan was not prescribed in detail by the Program, but was comprised of the following elements: 1) direct clinical and research mentorship by an expert in hemostasis/thrombosis who has a leadership role in a hemophilia/thrombophilia treatment center; 2) regular mentorship interactions in clinical care, research, and career development; and 3) generation and NHF panel review of semi-annual progress reports addressing progress in clinical training, research, and scientific presentations/publications.
The objectives of the present study were to compare outcomes of this grant-funded, mentored fellowship training program in hemostasis/thrombosis during conventional hematology/oncology fellowship training, versus conventional hematology/oncology fellowship training alone, via a nested case-control survey study. In particular, we sought to evaluate retention into a benign hematology career, time spent in research, publication trajectory, and grant funding success.
We conducted a nested case-control survey study involving NHF-Baxter program trainees (cases) and their contemporaneous institutional colleagues in hematology/oncology fellowship training (controls). A standardized survey was developed via a consensus process involving the authors, and then produced in web-based format. Prior to administration of the survey, cases were contacted with a request to provide (with permission) names and contact information for all controls for whom this information was available via their fellowship program, for the purposes of survey research. Trainees who had not yet completed fellowship were excluded. All cases and controls were then emailed a brief description of the study, along with a request to complete the survey via a link provided in the email, as well as to upload their current (i.e., up to date) curriculum vitae. To optimize response rates, personalized follow-up email reminders were sent to all subjects by one of the authors (N.A.G.). The study was approved by the Tulane University Institutional Review Board, with waiver of written informed consent.
Quantitative components of the survey used in the present analysis are shown in Figure 1. Additional data fields from the curriculum vitae included number of publications and all grant funding amounts (measured in direct costs, excluding NHF-Baxter Clinical Fellowship Award funding); these data were reported for the two-year period immediately preceding, the period during, the two-year period immediately after, and years 3–4 after fellowship training. Further data collection from the curriculum vitae consisted of gender, current age, and time (in years) post-fellowship entry to gaining appointment as assistant professor and to obtaining a National Institutes of Health K08 or K23 award.
Descriptive statistics were used to define distributions of continuous variables and frequencies (i.e., proportions) of categorical variables. Continuous data were compared between case and control groups by Mann-Whitney U test, and proportions were compared between groups via chi-squared test or Fisher's exact test, as appropriate based on cell frequencies in two-by-two tables. All hypothesis tests were two-tailed, with a P-value of less than 0.05 considered to be statistically significant. Statistical analyses employed SAS 9.1 software (SAS Institute, Cary, NC, USA).
Survey response rate was 85% (11/13) for cases and 90% (9/10) for controls. Ninety percent of respondents had trained in pediatric programs. Distribution of cases and controls (respectively, by institution) was as follows: The Children's Hospital, University of Colorado (n=2, n=3); Children's Hospital of Philadelphia, University of Pennsylvania (n=2, n=2); Department of Pediatrics, University of Michigan (n=2, n=2); Department of Medicine, Tulane University (n=1, n=1); Michigan Children's Hospital, Wayne State University (n=1, n=1); Department of Pediatrics, University of Iowa (n=1, n=0); Children's Medical Center, Emory University (n=1, n=0); and Cincinnati Children's Hospital, University of Cincinnati (n=1, n=0).
Demographic data and outcomes (excluding publication and grant funding trajectories) are given in Table 1. Median (range) age at survey response was 39 years (33–46 years), and did not differ between case and control groups. Fifty percent of respondents were female, with similar gender distribution between groups. All respondents had pursued careers in academic hematology/oncology. Median (range) percent time spent in benign hematology post-fellowship was 98% (70–100%) for cases versus 0% (0–20%) for controls, and time spent in research was significantly greater among cases than controls (median 80% [range: 42–90%] vs. 55% [10–80%], respectively; P=0.01).
Figure 2 displays peer-reviewed publication rates over time, comparing cases versus controls. By years 3–4 post-fellowship, median annual number of peer-reviewed publications was higher for cases than controls (3.5 vs. 1.0; P=0.01). Figure 3 shows grant funding trajectory for cases versus controls, in annualized direct costs, excluding NHF-Baxter Clinical Fellowship Award funding. Although not statistically significant, median annual grant dollars were appreciably higher among cases than controls ($80,000 vs. $23,000; P=0.20), as was the percentage of individuals who obtained K awards within 5 years of beginning fellowship (33% vs. 0%; P=0.21 [Table 1]).
The findings of this nested case-control survey study suggest that a grant-funded, mentored fellowship training program in hemostasis/thrombosis may be superior to conventional hematology/oncology fellowship training with respect to retention of physicians in benign hematology clinical care and research, and academic outcome measures such as publication and grant funding trajectories. The study has important potential implications for future training in hemostasis/thrombosis and perhaps other specialties in benign hematology wherein recruitment and retention of physician-investigators has been deemed a critical challenge. Key enhancements of conventional fellowship training—specifically, grant-funding and mentorship in a particular area of subspecialty focus—likely serve as strong contributors to the comparative success of the grant-funded, mentored fellowship training program. For this reason, the authors believe that future efforts at recruitment and retention of hemostasis/thrombosis physician-investigators should emphasize the development of grant-funded, mentored fellowship training programs to augment conventional hematology/oncology training.
A national decline in academic subspecialists and trainees in benign hematology has been emphasized by the American Society of Hematology . This trend is exacerbated by the economic disparity between academic and private practice physician salaries in hematology/oncology, the latter of which emphasize competence in the management of malignant diseases. A survey study of adult hematology/oncology fellowship program directors revealed that only 24% of graduating trainees pursued an academic career . Furthermore, a survey study of adult and pediatric hematology/oncology fellowship program directors demonstrated that benign hematology serves as the clinical focus for only 5–6% of adult training graduates in private practice or academia, and for less than 1% of pediatric graduates in private practice; by contrast, 13% of pediatric hematology/oncology training program graduates who remained in academia pursued benign hematology as a clinical focus .
Various curricula for training and/or competence in benign hematology have been reported in the past several years. On behalf of the American Society of Pediatric Hematology/Oncology (ASPHO), Hastings and colleagues on the ASPHO Training Committee outlined goals and core components for pediatric hematology/oncology training. While both clinical and research training were emphasized, mechanisms for recruitment and retention in academic hematology/oncology were outside the scope of the Committee's report . Astermark and group, for the European Association of Haemophilia and Allied Disorders, described detailed competencies for clinical practice in hemostasis/thrombosis, but strategies for recruitment and retention into the field were not among the aims of the consensus criteria . Through a curriculum development effort funded by the American Society of Hematology, Abshire proposed criteria for clinical and research training in benign hematology within a two-year adult Hematology fellowship program, which includes three months of inpatient benign hematology and 4–5 months of outpatient benign hematology in the first year, and 75% time in benign hematology research in the second year . The research component emphasizes mentorship as well as a faculty oversight committee.
The present findings for successful retention and early career development in hemostasis/thrombosis build upon prior literature in the field, with particular regard to mentorship and training grant support in benign hematology. In a survey study among U.S. adult hematology/oncology training programs, Gitlin and coworkers investigated prognostic factors for successful fellowship training toward a career as a physician-investigator in hematology/oncology. This study revealed that completion of a clinical research project, presence of a clinical research track/program, availability of a formal research curriculum, mentorship with a faculty oversight committee, and acquisition of independent career development grant funding were all independently associated with trainee pursuit of a clinical research career in hematology/oncology .
Consistent with our observations in the present work, a systematic review of 39 studies on the impact of mentorship revealed positive associations with mentee career choice, retention in academia, and research productivity, including publication and grant funding . “Extensive mentorship” has also been emphasized as a key ingredient in curriculum guidelines for training in Hematology put forth by a subcommittee of the American Society of Hematology Committee on Training Programs, to supplement the basic structure for subspecialty training provided by the Accreditation Council on Graduate Medical Education . Yet, adequate mentorship is not easy to achieve. As Kaushanski and Shattil noted in 2007 (and which remains equally if not more relevant in 2011), “Now, more than ever, we are in need of outstanding mentors, but… all too often institutions do not reward mentoring” . By targeting funds toward both trainees and their mentors, perhaps future grant-funded, mentored fellowship training programs hemostasis/thrombosis can overcome a key institutional barrier to realizing the benefits of mentored training in a subspecialty at risk of attrition.
A few limitations of the present work are noteworthy. First, the possibility exists for selection bias, in that the award of an NHF-Baxter traineeship may serve as an a priori marker of academic success. However, this potential bias was largely overcome by the selection of controls who were contemporaneous peers matched by institutional training program. Furthermore, this potential bias is also mitigated by the fact that no significant difference in grant funding and publication amounts were apparent between cases and controls in the immediate pre-fellowship period. Secondly, the study is challenged by a rather small sample size, rendering statistical estimates imprecise. For this reason, the findings are best described as preliminary, and warrant further investigation. Third, while we observed that time from beginning of fellowship training to acquisition of an Assistant Professor position was shorter among cases than controls, this potential marker of favorable outcome of the grant-funded, mentored training program could alternatively be explained by greater funding availability and/or clinical demand for junior faculty positions in hemostasis/thrombosis than in other areas of hematology/oncology, upon completion of fellowship training (i.e., in lieu of Instructor positions). Fourth, our study and the Program both assume that grant-funded, mentored training in hemostasis/thrombosis during fellowship—as opposed to during a post-fellowship Instructor period--is the appropriate alternative/comparator to hemostasis/thrombosis training during traditional hematology/oncology fellowship. An alternative training strategy for hemostasis/thrombosis is a post-fellowship Instructorship during which mentored training could take place in this subspecialized area. However, given the challenges of recruitment and retention in hemostasis/thrombosis discussed previously, we favor the approach of earlier immersion in grant-funded, mentored training. Lastly, despite the success of the NHF-Baxter Program in trainee recruitment and retention, adult (as opposed to pediatric) trainees were few. Given the broader salary gap between oncology and hematology in Internal Medicine than Pediatrics, and the delay in focused benign hematology training in some adult programs until the second year (at which time decisions regarding research focus are being made), larger measures will be necessary to boost recruitment and retention of trainees into clinical and research careers in adult benign hematology, including hemostasis/thrombosis. Notwithstanding these potential limitations, the results of this nested case-control survey study suggest that future efforts at recruitment and retention of physician-investigators in hemostasis/thrombosis should emphasize the development of grant-funded, mentored fellowship training programs to augment conventional hematology/oncology training.
The authors thank the survey respondents for their generous participation in the research, and Ms. Kelli Soare (a communication consultant contracted by Baxter Biosciences, and directly overseen by theh authors) for administrative support in production and execution of the web-based survey.
AUTHORSHIP N.A. Goldenberg designed the research, analyzed data, drafted the manuscript, revised the manuscript, and approved its submission for publication. R. Kruse-Jarres designed the research, analyzed data, drafted the manuscript, revised the manuscript, and approved its submission for publication. N. Frick revised the manuscript and approved its submission for publication. S.W. Pipe designed the research, revised the manuscript, and approved its submission for publication. C.A. Leissinger designed the research, revised the manuscript, and approved its submission for publication. C.M. Kessler designed the research, revised the manuscript, and approved its submission for publication.
Disclosures: N.A.G. is supported in part by a career development award from the National Institutes of Health, National Heart Lung and Blood Institute (1K23HL084055). N.A.G. and R.K.-J. are former recipients of an NHF-Baxter Clinical Fellowship Award. All authors except N.F. have received honoraria from Baxter Biosciences for their participation in the National Hemophilia Foundation-Baxter Clinical Fellowship Program Advisory Board and other advisory boards. R.K.J, C.A.L. and C.M.K. have received other grant funding and C.A.L. has received speaker fees from Baxter. N.F. is an employee of NHF. The authors have not received any honoraria or remuneration from Baxter or the NHF for their participation in the design, performance, assessment of the data, or generation of this manuscript. The authors have no other relevant conflicts of interest to disclose. Administrative support in web-based production and conduct of the survey was provided by an independent communication consultant contracted by Baxter, with direct oversight from the authors