The PSW-WG met by telephone in April and June of 2013 to begin planning its work, including data needs and organizational approaches that would most efficiently allow the geographically-dispersed members to address the Working Group’s charge.
Based on input from Working Group members, the Co-Chairs created three Subcommittees to consider issues specific to different segments of the physician-scientist workforce:
as well as a Data Subcommittee to oversee data collection and analysis to support the PSW-WG report. The Subcommittees invited ad hoc members to expand expertise in areas underrepresented by the original Working Group membership. These Subcommittees met on a regular basis by telephone between August 2013 and February 2014.
Appendix I contains a list of the Subcommittees and their membership.
Face-to-face meetings of the full PSW Working Group were held on July 19, 2013 and March 11, 2014, with monthly telephone meetings held between August 2013 and Feb 2014. The last group meeting by phone was held on April 23, 2014.
Contractor Thomson Reuters provided support to the Data Subcommittee in mining data from multiple sources to answer key questions identified by the various PSW subcommittees regarding historic trends and current patterns that impact the physician-scientist workforce. The two primary data sources for this Subcommittee were:
To best understand trends in the workforce, the Data Subcommittee sought to analyze individual physician-scientists, as opposed to applications, since many scientists submit more than one application in a given year. To do this, a record of an individual was created based on all available demographic records within the IMPACII and AAMC Faculty Roster data. As an initial data preparation step, similar applicant profile records were compared and duplicate records were collapsed using automated methods based on NIH’s own data management methods but with enhancements that consider additional demographic and application evidence to identify duplicate records. Next, each individual’s entire NIH application and appointment history was analyzed to generate a history of prior support at each year in the PSW analysis time frame during which the individual applied to the NIH for support. This history was used to establish an applicant’s prior training and fellowship support, as well as prior and subsequent research grant support. In addition, all available educational and demographic information was used to assign individuals to appropriate degree, gender, and race/ethnicity categories as described in Appendix III.
For the purpose of analysing trends over time, an individual applicant was counted only once per Institute/Center (IC) and mechanism in each Fiscal Year (FY). In the event an applicant applied more than once in a given FY to an IC and mechanism, the most recent awarded application was selected to designate him or her as an awardee. Even if an individual submitted applications to more than one IC, he or she was counted once within a given Fiscal Year. In select analyses, termed “5-year rolling windows,” an individual was counted once if he or she submitted one or more applications within a 5-year period.
The approach used by the PSW Data Subcommittee differs from standard data reporting at NIH, which normally focuses on reporting application and award total counts and breakdowns. These differences may lead to discrepancies when comparing trends presented here to those reported for applications and awards.
Unless otherwise noted, the source of data for all charts and tables included in this report are from NIH’s IMPACII data system, supplemented with AAMC Faculty Roster data, as provided under a data sharing agreement with AAMC. Select reports were generated using data from the NIH Medical Scientist Training Program (MSTP) and summary data from the AAMC’s Matriculating Student Questionnaire and Medical School Graduation Questionnaire. Aggregate data on faculty and physician-scientists were provided by the American Medical Association (AMA), the American Dental Education Association, the American Veterinary Medical Association and the Association of American Veterinary Medical Colleges.
In addition, data analyses were carried out with significant support form NIH’s Division of Statistical Analysis and Reporting (DSAR) within the Office of Extramural Research. Specifically, DSAR staff provided data on T32 appointees’ outcomes and other data review and analysis.
Appendix II contains information about the methodology used in the quantitative data analysis.
Appendix III is a description of data definitions used to extract the NIH IMPACII data.
Appendix IV contains the complete set of relevant graphs generated by the working group with links to underlying data available at NIH RePORT website.
The Lab-based PS Subcommittee members conducted focus groups with young faculty at their institutions to identify career retention and advancement concerns. Catalyst Research & Communications conducted telephone-based focus groups with medical, dental, and veterinary students. Questions focused on the factors that influenced their decision to pursue a research career. Telephone interviews were also carried out with deans at 15 US medical, dental, and veterinary schools to ascertain their perceptions of how students choose whether to pursue a career in research. Catalyst also conducted interviews with young faculty holding K08 and K23 grants that paralleled the demographics of the faculty members attending the Lab-based Physician-Scientist Subcommittee’s focus groups. Detailed findings of the qualitative research may be found in Appendix V. In January 2014, the National Institute of Nursing Research (NINR) conducted an open-ended survey of a purposive sample of nine deans of nursing schools with research-focused training programs. The purpose of the survey was to gather information regarding the experiences of schools of nursing in training successful nurse-scientists. A summary of the findings may be found in Chapter 4.