POSTER - #KU0038

Using Positive Deviance in Medical Education to Empower Learning

Badgett RG, MD; Brumfield B, DO; Wissman K, PharmD
Internal Medicine, KU School of Medicine - Wichita

Physician comment on the conventional mode of lifelong learning from mistakes:
“It is a crime. I'm not sure why or how it gets translated that way, but it is. Medicine has always had this very high ethical standard and to fail that standard is to be guilty. There's some anonymous court that's been set up someplace--I mean Osler or God somewhere at Massachusetts General Hospital-- and you‘ve been convicted and tried at the same time.”
- Christensen. Heart of Darkness.
JGIM 1992 PMID 1506949

Observations on from using positive deviance for lifelong learning:
“Interviews revealed enthusiastic support for surgeon-specific outcome reports (SSORs) as a means to improve patient care through awareness of personal outcomes with blinded divisional comparison for similar operations and diseases, and apply the learning objectives to continuous professional development and maintenance of certification."
- Ivanovic. Surgeon-Specific Outcome Reports and Positive Deviance.
Ann Thorac Surg. 2015 PMID 26188970

Background
Positive deviance (PD) has been used successfully in clinical care but not applied to medical education. We introduced PD to help residents’ tactics to speed the handling outpatient test results, in order to: 1) improve the clinical process, and 2) expose trainees to PD.

Methods
Each resident attended one PD seminars of 7 to 8 residents. During the seminar, residents’ opinions were collected with a RedCap survey.
Prior to the seminar, forest plots determined our overall proportion of results handled within 2 days (timely rate, TR), the TR of each resident, and the heterogeneity (I2) of the group. The forest plots were anonymized and emailed before the seminar to each resident with their own result identified.

Results
Our baseline TR was 52% (range 0% to 97%; I2 = 85%) with a mean of 7.4 days. In each seminar, after collecting and showing each member’s tactics without attribution, 72% and 58% of residents voted to encourage the deviants and all participants, respectively, to identify uniquely successful tactics.
After the seminar, 83% of residents assessed the session positively (p = 0.96 for comparison with the same assessment of other components in our last year-end curriculum survey).
Over the subsequent 10 weeks, a secular increase in performance occurred.

Conclusion
In this initial use of PD to empower learning on a mundane task, the residents received the experience similarly to established components of our curriculum. The finding that 72% of residents encouraged identifying positive deviants suggests that about a quarter of residents were initially uncomfortable with this introduction to PD.

Figure 1. example email to residents the day before the session.
(View poster)

Figure 2. Forest plot of test results turnaround times for all residents (anonymized). This slide was shown to residents when they voted whether to ask the positive deviants to identify themselves and their tactics from Figure 2.
(View poster)

Figure 3. Grid of test result handling tactics submitted at one session. Note this is useless without knowing which tactics belong to the positive deviants.
(View poster)

Figure 4. Test result handling for academic year 2019-2020 to present
(View poster)

Future plans
In 08/2020 we introduced peer coaching in which senior residents were asked to coach struggling residents. All residents were trained on peer coaching and delivering feedback.

References
1. Christensen et al. Heart of Darkness. JGIM 1992 PMID 1506949
2. Hazy JK, Uhl-Bien M. Towards operationalizing complexity leadership: How generative, administrative and community-building leadership practices enact organizational outcomes. Leadership. 2015 Feb 1;11(1):79–104.
3. Ivanovic et al. Surgeon-Specific Outcome Reports and Positive Deviance. Ann Thorac Surg. 2015 PMID 26188970