Schema of the academic incentive plan.
Clinical contribution as a function of academic productivity as assessed by the department chair. There was no inverse correlation between clinical and academic output.
Percentage impact of the academic incentive plan per surgeon. Percentage impact was calculated by dividing the incentive returned to each surgeon by the amount withheld for that surgeon. A baseline of 1 means that the amount returned equals the amount withheld. Those with values less than 1 received less than they contributed (n = 25 [42%]). Those with values greater than 1 received more than they contributed (n = 31 [56%]). To protect the identity of individual surgeons, the graph was capped at a value of 3 times the amount withheld.
Mitchell CC, Ashley SW, Orgill DP, Zinner MJ, Raut CP. Gauging Surgeons' Understanding and Perceptions of an Academic Incentive Plan. Arch Surg. 2009;144(5):421-426. doi:10.1001/archsurg.2009.47
To gauge the faculty's understanding and perceptions of a recently implemented academic incentive plan (AIP) to encourage productivity.
Design, Setting, and Participants
Surveys were administered to faculty at a university teaching hospital before and after allocation of the incentive during the inaugural implementation of an AIP.
Main Outcome Measure
Survey Likert scale.
Preallocation and postallocation survey response rates were 64% and 67%, respectively. Although 92% of respondents submitted the required self-reports of academic activities, only 25% met with their chiefs to assess productivity. Despite the small incentive, more than 50% believed that the AIP encouraged them to increase academic pursuits and rewarded activities important to the department that had not previously been reimbursed. Most did not believe that teaching, clinical research, or citizenship were adequately compensated before or after AIP implementation. However, more thought that citizenship (P=.02) and administration (P=.001) were adequately compensated in the postallocation vs preallocation survey; there were similar trends for clinical research (P=.17) and teaching (P = .06). Among 25 respondents who provided additional comments, 52% mentioned a lack of transparency in the method of incentive distribution as a concern.
Although only a few met with their chiefs for assessment, most faculty believed that even a minimal financial incentive might encourage them to increase their academic activities. We are encouraged to continue the AIP as a positive motivator for academic pursuits, and we plan additional measures to make incentive allocation more transparent.
Since 1976, the Department of Surgery at Brigham and Women's Hospital has used a compensation model that rewards clinical volume. Individual surgeons bill for patient services, pay for practice expenses and overhead costs out of their net patient collections, and retain 50% to 100% of their remaining practice surplus based on their years of service and academic rank in accordance with the Brigham and Women's Physician Organization's Professional Staff Compensation Policy and the Compensation and Practice Plan Guidelines of the Faculty of Medicine of Harvard University. In contrast, faculty members' teaching, research, and administrative activities are rarely, if ever, compensated. Despite the pressure to increase clinical productivity, nearly all faculty members have a major commitment to teaching students, residents, and fellows; 40% had at least 1 research grant; and 62% had at least 1 publication during the 2006-2007 academic year.
To recognize and encourage more academic contributions, the Executive Committee of our Department of Surgery convened a committee of faculty members to create an academic incentive plan (AIP). The goals of the AIP were to (1) establish a formal system for self-reporting and review of all academic pursuits critical to the mission of the department and (2) encourage increased productivity in these areas with a modest incentive. The committee reviewed physician compensation models and incentive plans from around the country. With no outside funding available, the committee proposed and implemented a modest plan whereby 1% of net patient collections would be withheld and reallocated based on separate assessments by the division chief and department chair of faculty self-reports at the end of each academic year (Figure 1). Faculty self-report forms classified academic activities into the following 6 categories: teaching, laboratory research, clinical research, citizenship/service, administration, and uncompensated clinical care. The AIP was approved by the Executive Committee and the division chiefs and was described to faculty members in letters and at quarterly and annual faculty meetings.
To gauge the faculty's understanding of the AIP, measure their perception of its impact, and solicit their feedback regarding changes that would increase the effectiveness of the plan, we conducted a departmentwide survey before and after allocation of the incentive.
Faculty in the divisions of cardiac surgery, general and gastrointestinal surgery, otorhinolaryngology, plastic surgery, surgical oncology, thoracic surgery, transplant surgery, trauma/burn/surgical critical care, urology, and vascular surgery were surveyed before and after allocation of the AIP in June and July of 2007 and January and February of 2008, respectively. Responses were collected using http://www.surveymonkey.com. Questions evaluated the faculty's understanding of the AIP, the percentage of time devoted to academic pursuits important to the departmental mission, perception of the adequacy of compensation for these activities, potential impact of the AIP, and opinions about the most effective size of the incentive pool (Box). The postallocation survey also included questions measuring the perceived impact and efficacy of the AIP. Most questions were scored on a 4-point Likert scale (“do not agree,” “somewhat disagree,” “somewhat agree,” or “completely agree”). For simplicity of presentation, the answers “do not agree” and “somewhat disagree” were scored as negative, and the answers “somewhat agree” and “completely agree” were scored as positive. Space for free-text comments was also included. A total of 87 faculty members received the preallocation survey and 75 faculty members received the postallocation survey. Statistical significance was calculated using the Mantel-Haenszel test for Likert scale questions and the Fisher exact test for yes/no questions. P ≤.05 was considered significant.
Questions in Both the Preallocation and Postallocation Surveys
The motivation for the new incentive plan.
Where the budget for the new academic incentive plan is coming from.
How the incentive pool will be allocated under the new plan.
Please indicate the percentage of time you spend on the following activitiesb:
The following activities are adequately compensated under the current systema:
The academic incentive plan willa
Encourage you to increase activities that are important to the missions of the department and the divisions but are not currently reimbursed.
Reward you for activities that are important to the missions of the department and the divisions but are not currently reimbursed.
The academic incentive plan would be most effective if the incentive pool were (check one):
Not increased (1%).
Increased to 2%.
Increased to 3%.
Increased to 5%.
Increased to >5%.
Demographic information (voluntary)
Additional Questions in the Postallocation Survey
For the first year of the academic incentive plan, 2005-2006d
Did you submit the self-reported achievements form?
Did you meet with your division chief about the academic incentive plan?
Did you consult with another mentor(s) about the academic incentive plan?
Are there any other specific changes you would recommend for the design or implementation of the academic incentive plan?c
a4-Point Likert scale (“do not agree,” “somewhat disagree,” “somewhat agree,” or “completely agree”).
bPercentage scale (0%-100%).
The response rates for the preallocation and postallocation surveys were 64% and 67%, respectively. The postallocation survey population was smaller because division chiefs, who were ineligible for the AIP by design, and faculty members who left during the fiscal year were excluded. Preallocation and postallocation survey respondents did not differ in their divisional representation (P = .99) or academic ranking (P = .75) (Table 1). Respondents to the 2 surveys did not differ in the percentage of time spent on teaching (P = .94), laboratory research (P = .29), clinical research (P = .88), citizenship/service (P = .66), administrative activities (P = .68), or clinical care (P = .43) (data not shown). Although 92% of respondents submitted a self-report form detailing their academic activities, only 25% met with their division chiefs during the evaluation process, and 8% consulted with another mentor for an assessment of their productivity.
In both surveys, nearly 20% of faculty failed to understand the source of the incentive pool budget (P = .95;Table 2). Even after allocation of the incentive, nearly 20% did not understand the motivation behind the AIP and nearly half did not understand how it was allocated.
Although most faculty did not consider teaching, clinical research, citizenship/service, or administrative activities to be adequately compensated before or after the incentive allocation, significantly more thought that citizenship/service (19% vs 31%; P = .02) and administration (25% vs 39%; P = .001) were adequately compensated in the postallocation survey (Table 2). There were similar but nonsignificant trends for clinical research (P = .17) and teaching (P = .06). No trend was observed for laboratory research (P = .68). Approximately 75% and 54% in the preallocation and postallocation surveys, respectively, believed that clinical care was adequately compensated under the current system (P = .08).
Nearly 60% of respondents in the postallocation survey agreed that the AIP encouraged them to increase their academic activities in the next academic year and rewarded them for activities important to the missions of the department and the divisions (Table 2). In the postallocation survey, significantly more faculty suggested that the funding pool remain at 1% of net patient collections (preallocation, 47%; postallocation, 67%; P = .03) (Table 3).
Of respondents who submitted free-text comments with the postallocation survey, 52% stressed a need for more transparency in the allocation of the incentive. A second concern raised in the comments was the potential for redistribution of income from clinically productive to academically productive faculty.
In the present health care climate, it is difficult for academic departments to encourage academic activities critical to their mission while balancing substantial growth in clinical volume. At our institution, clinical productivity is directly compensated via net patient collections. In contrast, academic activities are inconsistently compensated. Some such activities are funded. For instance, external sources provide grant funding for research, Harvard Medical School provides a stipend for teaching medical students during their preclinical training, and our hospital provides funds for covering trauma call. Until implementation of the AIP in 2006, there was no direct or consistent mechanism for compensating a range of academic activities, including teaching residents, citizenship and service, and administrative duties. In addition, there was no system of regular reporting and review of such activities. Given the difficulty of creating such a plan, we thought that a survey gauging the faculty's perceptions of the AIP and eliciting their feedback would provide useful information for improving the plan.
The survey results highlighted 4 issues: (1) the importance of communication in implementing an AIP; (2) the challenge of creating an evaluation method that is fair yet easy to administer; (3) the difficulty of building an AIP without outside funding; and (4) the power of a small incentive to change surgeons' perceptions. The survey results also pointed to potential areas for improvement.
First, nearly 1 in 5 faculty members did not understand the motivation behind the AIP and the funding source for allocations. This was particularly surprising because the chair, vice chair, division chiefs, and members of the AIP Committee addressed this topic in numerous meetings with and letters to the faculty. This percentage would undoubtedly have been smaller had more faculty met with their division chiefs. However, we suspect that some faculty understood the purpose of and funding source for the AIP but were instead questioning the legitimacy of an incentive plan funded by revenue withholds as well as the use of revenue withholding as the sole source of funding for the AIP.
A second issue, related to the aforementioned difficulty in communication, was the perceived need for more transparency in the allocation process. Nearly half of faculty members surveyed did not understand how the incentive pool would be allocated. To create a plan that is easy to explain and administer, the AIP Committee decided against adopting an incentive plan based solely on a quantitative assessment of faculty achievements. Some institutions have a relative value unit scale or a point system for measuring academic and administrative productivity.1- 4 However, such point systems fail to distinguish quality of work or publication (eg, prestige of the journal) or extent of participation in a committee or organization (leadership vs participatory role). Furthermore, most point systems do not adequately evaluate educational contributions. Our present system of regular self-reports by faculty and review by a division chief or chair accounts for qualitative differences in achievements.
Third, the source of funding for the incentive pool adversely affected survey responses. The AIP Committee originally planned to withhold only 1% of net patient collections for 2 reasons. Because this was a new plan, the committee believed that the withholding amount should be small. Furthermore, the goal was not to construct a system of monetary rewards for academic productivity but rather to reinforce the establishment of a system of reporting and review. More respondents in the postallocation vs the preallocation survey believed that the incentive pool should remain at 1% of net patient collections. Some faculty were concerned about potential redistribution of income from the clinically to the academically productive, which would go against the prevailing culture within the department. However, we found no evidence of such income redistribution. In a postsurvey analysis, we compared clinical productivity, measured by net collections, to the chair's assessment of faculty members' academic contributions and found no such inverse correlation.Figure 2 shows such data from 2 subspecialties with comparable service reimbursements. These results were not surprising. Academic institutions such as ours rarely attract physicians who aim to be extremely busy clinically without also being academically productive. Most physicians managed to do both with varying success. The few who had both low clinical and academic output were new faculty who were building their practice volume and research profiles.
Fourth, nearly 60% of respondents in the postallocation survey believed that the AIP would encourage them to increase these academic activities. Under such an incentive plan, nearly half the faculty received an incentive amount that was less than the actual dollar amount of net patient collections they put into the pool. Thus, we expected that only half would view the AIP favorably.Figure 3 illustrates the relative amount of incentive allocated to each faculty member as a function of his or her own withheld funding.
The AIP Committee reconvened to review the survey results and the AIP process. Several changes were recommended to improve transparency and address other concerns. First, because physicians' understanding and “buy-in” are keys to the plan's effectiveness, the leadership proposed a communication strategy that includes more one-on-one meetings with chiefs or mentors and more division meetings in addition to departmental meetings. Second, the department will include a detailed list of examples of academic contributions in each of the 6 categories with the next round of faculty self-report requests. The list will encourage more comprehensive reporting and define what activities are recognized. Third, the chiefs and chair will individually score their faculty on a scale from 0 to 5 based on their academic contributions and will meet to create a unified and comprehensive evaluation. The new allocation schema will take into account not only the academic score but also the physician's contribution to the incentive pool because the intent is not to penalize those who are both clinically and academically productive. Fourth, each faculty member who submitted a self-report will receive a letter showing their scores and evaluations from the chief and their chair, which will complement the more rigorous one-on-one meetings with the chief.
Some issues remain unresolved. First, although most faculty members surveyed want to keep the AIP funding at 1% of revenue, is the size of the AIP pool adequate to change behavior? We believe that the allocation amount for and the net impact on individual physicians were not immaterial to the participants. Given the source of the AIP funding, the AIP Committee decided that it was appropriate to start small during the initial implementation. Second, some division chiefs questioned whether faculty with similar academic activities should receive similar incentive amounts, despite disparate net patient collections. Our goal was to acknowledge academic productivity but not punish clinical productivity. We hope that by accounting for clinical productivity to a degree, the new AIP addresses this problem. Third, the distinction between “compensating” and “recognizing” academic efforts is an important one. In the surveys, we asked whether academic activities were adequately compensated. In truth, it is difficult to determine what constitutes adequate compensation for such activities. At best, we are recognizing academic achievement with a modest incentive. Last, because the AIP is perceived to have a significant impact in some areas but not others, we are interested in identifying modifications that would make the AIP more effective in recognizing and encouraging all academic pursuits.
The AIP helps recognize and encourage mission-critical work within our Department of Surgery. However, there is clearly room for improvement. We are encouraged to continue the AIP to promote academic pursuits, and we will continue to measure its impact as we implement appropriate modifications.
Correspondence: Chandrajit P. Raut, MD, MSc, Department of Surgery, Brigham and Women's Hospital, 75 Francis St, Boston, MA 02115 (email@example.com).
Accepted for Publication: December 8, 2008.
AuthorContributions: All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Mitchell, Ashley, Orgill, and Raut. Acquisition of data: Mitchell and Raut. Analysis and interpretation of data: Mitchell, Ashley, Orgill, Zinner, and Raut. Drafting of the manuscript: Mitchell and Raut. Critical revision of the manuscript for important intellectual content: Mitchell, Ashley, Orgill, Zinner, and Raut. Statistical analysis: Mitchell and Raut. Administrative, technical, and material support: Mitchell and Ashley. Study supervision: Ashley, Orgill, Zinner, and Raut.
Financial Disclosure: None reported.
Previous Presentations: This paper was presented at the 89th Annual Meeting of the New England Surgical Society; September 28, 2008; Boston, Massachusetts; and is published after peer review and revision.
Additional Contributions: Jessica Bowen, BA, and Lizabeth Edmondson, BA (Department of Surgery, Brigham and Women's Hospital), and Mary Pat Johnston, BA (formerly in the Department of Surgery, Brigham and Women's Hospital; presently at Boston College Graduate School of Social Work), administered the surveys and tabulated the data. We also thank Stuart Lipsitz, ScD (Department of Medicine, Brigham and Women's Hospital), for statistical analysis support and Michael Jackson, MBA (Department of Surgery, Brigham and Women's Hospital), for details regarding the current compensation plan.