The Collaborative Scholarship Intensive: A Research-Intensive Course to Improve Faculty Scholarship

Balancing clinical, educational, and scholarly elements of an academic career can be challenging. Barriers to scholarship production are especially numerous (eg, time, competing priorities, inexperience),1,2 yet this is often a primary metric in faculty performance evaluation and promotion.3

Institutions have grappled with ways to support successful scholarship for clinical faculty. Workshops focusing on faculty publication productivity, peer support, and collaborative writing are common,2,4-12 although the structure of these varies greatly. Time frames range from 90 minutes to daylong events, held once or over the course of several months.8 Objectives vary from the writing process5,6,12 to training on specific writing skills.7 While beneficial, these models often don’t encompass all phases of research, such as data collection or analysis. Furthermore, as the majority of studies report pilot data or single cohorts, it’s unclear whether their impact is sustainable over more diverse groups or longer time frames.

Building on the strengths of previous models, we created the Collaborative Scholarship Intensive (CSI) with a goal of increasing scholarship output amongst faculty with minimal to no publication experience. We aimed to determine whether intensive instruction in research methodology, coupled with structured writing support and protected work time, increases faculty scholarship skills and output.

The University of Minnesota Medical School Institutional Review Board granted this study exemption from formal review for human subjects’ protection. We invited all 96 Department of Family Medicine and Community Health (DFMCH) faculty via email to register for CSI by completing an online form.

Participants included 19 physician (MD/DO) and four behavioral health (PhD/PsyD) faculty who participated in CSI over the first 3 years of the course. Nine faculty completed CSI in 2015, seven in 2016, and seven in 2017. All 23 participants were at the level of assistant professor with years on faculty ranging from 2 to 27.

Table 1 describes the six CSI sessions. In addition to 2.5 to 4.5 hours of focused instruction, each session included 1.5 to 3.5 hours of one-to-one work and writing time. Our faculty development team, career research faculty, and biomedical librarian provided the focused instruction and led the writing time. CSI participants were also supported by research staff experienced in practice-based research who provided assistance and instruction around specific aspects of research methodology.

One goal of CSI was for participants to add at least three scholarly products to their curriculum vitae including a 500 word peer-reviewed evidence summary, a presentation at a conference, and a manuscript submitted to a peer-reviewed journal.

Participant Feedback. At the conclusion of each session participants were asked to complete a brief evaluation of instructors and content. Although not used in course evaluation, this information was helpful in modifying course content and provided feedback on participant engagement.

Self-rated Competencies. Before and after the course participants used an 11-point Likert scale to rate their confidence in successfully performing 12 research competencies (Table 2).13 We calculated an average overall competency score for each participant as the mean of all 12 ratings (Chronbach α for pre=0.98, and post=0.93).

Scholarly Work. For each participant, we compared the quantity of scholarly work during the 1-year period preceding their start in the course to the quantity of scholarly work during the 1-year period following their start in the course. We defined scholarly work as the number of publications in any author position in a peer-reviewed journal, and oral/poster presentations at professional conferences.

We calculated means and standard errors for the 12 research competencies before and after the course. We used linear mixed models, each with a random intercept for participant ID, to test for differences in competencies before and after the course for all years combined. We used a paired t test to compare scholarly work from before to after the course.

All 12 competencies showed significant improvement pre- to postcourse (Table 2). The average overall competency score of pre- and postcourse ratings significantly increased from 3.8 (SE=0.3) to 6.7 (SE=0.3; 76% change, P<.001; Table 2).

Scholarly work significantly improved from before to after the course, with many participants exceeding productivity goals by partnering with classmates or other colleagues on multiple projects. Table 3 illustrates the increases in the number of publications (t₂₂=6.86, P<.001), number of conference presentations (t₂₂=3.38, P=.003), and total scholarly products (t₂₂=6.41, P<.001) for all participants combined pre- and postcourse. The number of total scholarly products per participant was more than two times higher after the course (percent change=132%).

This program evaluation study examined the effectiveness of an innovative faculty development program. CSI participants had (1) significantly improved self-ratings of 12 research competencies, and (2) significantly increased scholarly output in pre- to postcourse comparisons.

An acclaimed feature of CSI was structured writing support and work time led by faculty with research and publishing experience. Participants consistently rated this as the most impactful aspect of the program with one commenting,

I really just loved having the work time with so many great resources and people to help us out. Even just talking about roadblocks with others in the course was incredibly helpful.

CSI may have contributed to a culture of collaborative scholarship that is netting additional benefit. Since CSI class 1 commenced, there has been more than an 80% increase in faculty publications across the department. Clearly we cannot attribute this increase to CSI alone. But as one participant commented,

I really loved the course and learned a ton. So glad to learn that this is not just about writing in a cubicle by yourself.

The direct costs for running CSI were largely contained to food for participants and instructors. The indirect costs were more substantial. Time for course planning and coordination by the administrator and director, instructors to present content, and participants to attend were the primary expense. Most participants used allotted continuing education hours to cover their time in the course.

Our results have limitations. The study sample size was small and limited to a single site and the three classes that completed CSI at least 1 year ago. Additionally, self-ratings on the 12 research competencies are prone to individual bias toward improvement, although this may be offset by the other primary measure of program success, that being an objective count of scholarship output. Next steps include broadening CSI to enroll faculty from other specialties at our university.

Our program successfully engaged clinical faculty with limited publication histories in collaborative research instruction. Initial analyses and participant feedback were positive, with an increase in both participant and departmental scholarship production.