Basic study design.
Wang TS, Schwartz JL, Karimipour DJ, Orringer JS, Hamilton T, Johnson TM. An Education Theory–Based Method to Teach a Procedural Skill. Arch Dermatol. 2004;140(11):1357-1361. doi:10.1001/archderm.140.11.1357
Copyright 2004 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2004
To determine the effectiveness of an education theory–based method to teach students to place and tie a simple interrupted stitch.
A teaching intervention before-after trial.
Dermatology department, academic university.
Fourth-year medical students and dermatology residents.
Main Outcome Measures
Scores on a 12-criterion grading instrument before and after instruction.
The scores for medical students and residents in each class showed significant improvement. The mean score for all participants (N = 23) rose by 24% after instruction (P< .001). Scores in 9 of the 12 graded performance areas improved significantly after instruction, including scores in tissue damage/teeth marks (P<.001), needle dulled/bent (P< .001), needle loaded properly and knots square (P = .01), throws done correctly (P = .01), stitch tension and needle entry/exit angle (P = .02), amount of suture used (P = .03), and correct number of throws (P = .04). In addition, participants’ confidence increased significantly after instruction (P<.001). No difference was noted between men and women in preinstruction vs postinstruction score improvement.
This teaching method can be effectively used to teach students to place and tie a simple interrupted stitch. Once validated and expanded, it may prove useful in shortening and standardizing procedural skill training and in objectively documenting competency.
Traditional surgical training is largely an apprenticeship. Students observe and assist as cases present rather than in a systematic manner. The teaching of manual procedural skills is also rarely formally addressed; instead, teaching is usually done during surgery when mentoring is often random and situational. Learning is by observation, trial and error, and practice; and competency is difficult to measure. Differences in surgical experiences and aptitude can lead to variability among students in their manual skills.
Many articles address the teaching and learning of manual procedural skills. Although pigs’ feet and surgical tubing have been used for years,1- 3 technology has allowed the development of more advanced simulators.4- 7 Dexterity testing, teaching strategies, grading criteria, workshops, neuropsychologic predictors, and even force-measuring devices have also been described.8- 25 Few authors, however, have used education theory–based methods to teach manual procedural skills. Fewer still have used such methods to teach these skills in a workshop setting.
In 1998, Walker and Peyton26 outlined a 4-step method to apply education theory to the teaching of manual procedural skills. This method and current approaches to the teaching of manual procedural skills were reviewed by Hamdorf and Hall27 in 2000 (Table 1). If the 4-step method of Walker and Peyton26 could be proven effective in a workshop setting, the teaching of manual surgical skills might become better standardized and more efficient. Teaching in this way might also facilitate objective grading and documentation of competency.
The need for improvement is evident. Currently, learning even basic manual procedural skills via “bedside” apprenticeship can be a lengthy process. In medical school and residency, it is often difficult to assign trainees additional surgical rotations or allow them to simply “do” more. Recent Accreditation Council for Graduate Medical Education requirements mandate documentation of competency in procedural skills. Under the current system, objective measurement of these skills can be difficult.
The purpose of our pilot study was to investigate the applicability and utility of Walker and Peyton’s teaching method26 by attempting to teach participants to place and tie a simple interrupted stitch. The longer-term goals are to develop methods to enhance the teaching and learning of core procedural skills and to develop ways to objectively measure and document competency in those skills. Results may prove useful in the development of more advanced simulators.
Approval for this study was obtained from the University of Michigan Medical School institutional review board. The basic study design is outlined in the Figure. The participants were fourth-year medical students (M4s) rotating through dermatology and first-, second-, and third-year dermatology residents (R1s, R2s, and R3s) at the University of Michigan.
Applying Walker and Peyton’s method26 required that the process of placing and tying a simple interrupted stitch be broken down or deconstructed into its component steps (Table 2). Twelve steps were identified as objectively measurable and/or observable in the placement, and these formed the basis of our grading instrument (Table 3). For most criteria, 1 point was given if the task was performed correctly; 0 if it was done incorrectly. Some steps such as number of suture packets used or suture breaks were graded across the session rather than per stitch. A maximum of 60 points was available per session.
Prior to each session, a 3.5 × 0.5-cm ellipse was cut onto a pig's foot, which the participant would be required to repair. Adson forceps (4.75 in, 1×2 teeth, delicate), a 6-in Mayo-Hegar needle holder, suture scissors, an unlimited quantity of 4-0 polypropylene suture (Prolene; Ethicon Inc, Somerville, NJ), and a PS-2 needle were used. A single instructor (T.S.W.) performed the teaching session and graded the entire study.
When the participant and instructor entered the treatment room, a clock was started to mark commencement of overall time, and then the basic study design was explained. Participants were asked to self-assess their previous surgical exposure and confidence. They were then instructed to place 5 simple interrupted sutures to close the defect. Task time commenced as the participant picked up the instruments. As participants performed each stitch, the instructor graded their performance using the grading instrument. At no time did the participant know the grading criteria. On completion of the fifth stitch, task timing ended, and the teaching segment of the study began.
During the teaching segment, Walker and Peyton’s 4 steps26 were performed. First, the instructor placed 5 simple interrupted stitches at normal speed (demonstration). Then, using the steps described in Table 2, the instructor performed the task while verbally “breaking it down” (deconstruction). Next, the instructor performed the task while being talked through it by the participant (formulation). Finally, the participant performed the task while talking herself or himself through it (performance). During formulation and performance, the participant verbally repeated the steps and was corrected if a mistake was made. Each step was repeated until the participant felt ready to proceed.
After the teaching segment, the participant was given a new identical defect and asked to again place 5 simple interrupted stitches to close the wound. The steps were graded, and the task time was recorded as before. When finished, the students were asked to provide another self-assessment. As they exited the room, overall timing ended.
Study end points before and after the structured skills workshop were compared via the paired t test. Correlations between various end points were made through Pearson product-moment correlation analysis. All P values are 2-tailed. Descriptive statistics are presented as mean ± SEM. The analyses were performed with SAS statistical software (SAS Institute Inc, Cary, NC).
A total of 23 participants completed the study: 8 M4s and 15 dermatology residents (R1s, R2s, and R3s). All M4 participants had completed their required surgical rotations prior to participation in the study. None of the residents had performed an internship in a surgical specialty. Seven participants were men and 16 were women.
Results are summarized in Table 4 and Table 5. The mean score for all participants combined (N = 23) rose from a preinstruction 47.6 to a postinstruction 58.9, an increase of 24% (P< .001). Preinstruction and postinstruction scores by class (M4-R3) all showed a significant increase. Medical students had the lowest mean preinstruction score (37.9), and R3s had the highest (56.1). Medical students showed the greatest increase from mean preinstruction to postinstruction scores (37.9 to 57.6, P = .01), and R3s showed the smallest increase (56.1 to 59.6, P = .04). A small but significant difference was noted between the mean postinstruction scores of M4s and residents (57.6 vs 59.6, P = .02). While both men and women showed significant increases in their mean postinstruction scores, no significant difference was noted between the mean preinstruction and postinstruction scores of men vs women (data not shown).
The graded item with the lowest mean preinstruction score was tissue damage/teeth marks. In order of increasing score, this was followed by needle dulled/bent, throws done correctly, knots square, and needle loaded properly. Statistically significant increases in the postinstruction scores of all participants were found in 9 of the 12 graded criteria.
A significant difference in the overall time was noted between M4s and residents (53.5± 3.1 vs 39.8 ± 2.8 minutes, P = .01). No significant change was noted in preinstruction vs postinstruction task time (data not shown).
Participants’ confidence in their ability to perform the task correctly increased significantly after instruction (3.1 ± 0.2 vs 4.3 ± 0.1, P<.001). In addition, a significant positive correlation was noted between participants’ preinstruction confidence in their ability to perform the task correctly and both their preinstruction and postinstruction scores (r = 0.73, P = .001 and r = 0.61, P = .002, respectively).
The current apprenticeship method of surgical training provides the breadth of clinical experience necessary to develop good surgical judgment; however, it can be a lengthy process and, in regard to manual surgical skills, can have inconsistent results. We studied a structured, education theory–based method to teach a manual surgical skill, and our results show it to be effective.
Medical students had the lowest preinstruction scores, which may imply that this basic task was not learned to proficiency through core surgical rotations during medical school. Residents, on the other hand, had higher preinstruction scores that increased with year of training. After instruction, all participants increased their scores to nearly the maximum of 60 points. Accordingly, M4s showed the greatest increase in score, while residents had less room for improvement and thus showed smaller but still significant improvement. It is interesting to note that the mean postinstruction score for the M4s was slightly but significantly lower than that of the residents. This may reflect the small sample size of our pilot study. However, we believe that this does not detract from the fact that near-parity was achieved between the 2 groups after instruction.
Of the 12 performance areas graded in this study, the scores in 9 significantly improved after the teaching segment. Those 3 criteria for which the scores did not significantly improve had the highest preinstruction scores, which made significance more difficult to achieve given our small sample size. The criterion with the lowest preinstruction score was tissue damage/teeth marks. This was graded by judging whether excessive pressure was applied to the tissue by the pickups that would leave teeth marks. Anyone who has attempted to teach a student to suture knows that this is a very common error indeed. Many wounds heal poorly owing to trauma to the wound edges caused by excessive pinching. We were encouraged that the score for this, as well as for other common mistakes, significantly increased after the teaching segment.
It took significantly longer for the M4s to complete the entire study (overall time) than it took the residents. This finding seems to make sense in that those with lower preinstruction scores might be expected to require more teaching time. Surprising to us, however, was the finding that no change in task time was noted after the teaching segment. Before the study, we hypothesized that task time would increase after the teaching segment because participants would have more to remember. However, this was not borne out.
As we proceeded through the study, it was interesting to note how often during formulation and performance the participant missed steps that were discussed minutes earlier during deconstruction. Sometimes, the participant was performing the task incorrectly while concurrently describing it correctly. This highlights the large difference between verbalizing a task correctly and performing it correctly. While being taught at the bedside, how often does the student nod his or her head in apparent understanding only to perform the task incorrectly minutes later?
Finally, participants reported significantly more confidence in their ability to perform this task after the teaching segment. In addition, their preinstruction confidence showed a significant positive correlation with both their preinstruction and postinstruction scores. In other words, those participants who were more confident before the teaching segment tended to have higher preinstruction and postinstruction scores. This may imply that participants were good judges of their ability and were not overconfident.
Some may propose that other methods if used in the teaching segment would be equally effective; however, we chose to study Walker and Peyton's method.26 Others may feel that it was merely the 1-on-1 time spent with an instructor that improved the participants’ ability to perform the task. We believe it is the structure of Walker and Peyton’s method that enhances teaching and learning rather than simply the 1-on-1 time. Teaching students 1-on-1 is time consuming, but if Walker and Peyton’s method can be used to teach multiple students at once, efficiency will improve. We would point out that apprenticeship too is lengthy and sometimes inefficient.
Some may also find flaws in the way we deconstructed the task. The steps we identified were based on our experience only; our goal was simply to study a teaching method. Less important were the specifics of what was taught.
No control group was used in our study. However, it seems unlikely that repeating the task without the teaching segment would improve scores so dramatically.
Finally, we acknowledge the potential bias introduced by using a single grader who was also involved in the study design. Although we believe that the grading instrument accurately evaluates objective, observable criteria, future studies will be aimed at better validating it.
Although we taught only a single, simple manual skill, it is reasonable to believe that other more complex skills could also be effectively taught using this method. If validated and expanded, this method could be used to develop a curriculum of manual procedural skills and to teach such skills in a structured workshop setting, away from the patient. This would hopefully result in quicker learning and decrease the number of mistakes performed on patients. Workshops using advances in teaching methods and technology could be tailored to meet the needs of the target audience and incorporated into both medical school and residency education.
Our project studied an education theory–based method to teach a manual procedural skill. Results suggest that basic suturing skills can be effectively taught and objectively graded in this way and also point to improvements that might be made in current teaching methods (eg, apprenticeship, medical school core surgical rotations, and textbooks). If a more effective or efficient method to teach and measure competency in these skills is found, improvements in procedural training may occur.
Correspondence: Timothy S. Wang, MD, A. Alfred Taubman Health Care Center, University of Michigan, 1500 E Medical Center Dr, Room 1910A, Ann Arbor, MI 48109-0314 (email@example.com).
Accepted for Publication: July 23, 2004.
Funding/Support: Supported in part by a grant from the Gilbert Whitaker Fund for the Improvement of Teaching, University of Michigan, Ann Arbor.
Previous Presentation: This study was presented at the American Society for Dermatologic Surgery–American College of Mohs Micrographic Surgery and Cutaneous Oncology combined annual meeting; October 9, 2003; New Orleans, La.
Financial Disclosure: None.