This paper describes the development and evaluation of training intended to enhance students' performance on their first live-animal ovariohysterectomy (OVH). Cognitive task analysis informed a seven-page lab manual, 30-minute video, and 46-item OVH checklist (categorized into nine surgery components and three phases of surgery). We compared two spay simulator models (higher-fidelity silicone versus lower-fidelity cloth and foam). Third-year veterinary students were randomly assigned to a training intervention: lab manual and video only; lab manual, video, and $675 silicone-based model; lab manual, video, and $64 cloth and foam model. We then assessed transfer of training to a live-animal OVH. Chi-square analyses determined statistically significant differences between the interventions on four of nine surgery components, all three phases of surgery, and overall score. Odds ratio analyses indicated that training with a spay model improved the odds of attaining an excellent or good rating on 25 of 46 checklist items, six of nine surgery components, all three phases of surgery, and the overall score. Odds ratio analyses comparing the spay models indicated an advantage for the $675 silicon-based model on only 6 of 46 checklist items, three of nine surgery components, and one phase of surgery. Training with a spay model improved performance when compared to training with a manual and video only. Results suggested that training with a lower-fidelity/cost model might be as effective when compared to a higher-fidelity/cost model. Further research is required to investigate simulator fidelity and costs on transfer of training to the operational environment.
Introduction
Routine neutering of healthy small animals is a part of veterinary practice.1 The American Society for the Prevention of Cruelty to Animals reports that there are at least 70 million stray dogs and cats living in the US at present, and many of those animals require sterilization.2 It is estimated that only 10% of animals presenting to shelters have been neutered before presentation.2 Many surgical techniques have been described, but the most commonly practiced is the ventral midline celiotomy approach.1 A study of veterinary graduates entering the job market in California revealed that 17% of employers felt that the graduates were deficient in their ability to perform routine surgical procedures, including ovariohysterectomy (OVH) and castration.3 A survey of veterinary surgeons showed that over 60% expected Day One DVM graduates to have good surgical skills with minimal supervision required, especially in common surgical procedures like OVH.4 An Australian study found that over 95% of their new graduates reported having to perform an OVH without supervision soon after graduation, while as students only 41% had the opportunity to perform an OVH on a cat, and only 75% had the opportunity to perform an OVH on a dog.5 Yet employers consider OVH a critical procedure that new graduates should be able to perform proficiently upon entering the workforce.6 A recent study also commented on the high level of anxiety associated with learning surgical skills and showed a significant reduction in this anxiety when students were allowed to practice in a simulator lab before exposure to live animals.7
Practical training in veterinary surgery has traditionally involved the use of live animals or cadaver tissues.8 Live-animal use in veterinary surgical training brings a host of ethical and practical concerns, including acquisition of the animal subjects.8 Cadavers lack biologic responses and may undergo autolysis and rigor, which may make surgical training less realistic.9 Griffon and colleagues compared a reusable model, which mimicked canine anatomy and hemodynamics, to a cadaver for training veterinary students to perform a live-animal spay. The results indicated that the model was more effective than a cadaver for teaching basic surgery skills and OVH in canines.9
Models present a good alternative for teaching OVH because they allow the student to perform the procedure multiple times until an acceptable level of skill is reached.10 Having mastered the psychomotor skills and procedural steps, the learner can spay a live animal with less of the anxiety typically associated with the first performance of such a procedure.7 However, training on a model also has limitations, including the costs associated with development and use of the model and the ability to capture realistic features of the actual patient it is intended to simulate.10 Evaluation of teaching models is challenging, and, to date, studies in veterinary medicine have concentrated primarily on level 1 or 2 of Kirkpatrick's training program hierarchy, focusing primarily on student reactions and some observations regarding learning.11 Boet and colleagues reviewed the use of simulation for Crew Resource Management (team training) and noted that few studies in medicine have assessed Kirkpatrick levels 3 and 4.12 The authors are unaware of any simulation studies to date in veterinary medicine that have assessed these advanced levels.
Recently, some veterinary schools have developed surgical simulator training programs and created OVH models with varying levels of fidelity.8,13 Fidelity refers to the degree the simulator (or, in the current study, spay model) represents its real world counterpart.14 Fidelity analysis is a function of the nature of the task, the learning stage of the performer, and the identified learning objectives.14 Determining the amount of fidelity required for a simulator should be guided by an explicit task analysis to “determine the necessary physical fidelity and functional characteristics of the situation in order to provide the most cost effective training.”15(p.70) For example, Flexman and colleagues compared training in a Link Trainer (lower-fidelity ground trainer) to training in an actual plane (higher fidelity) on the number of errors, amount of time, and number of training trials required to achieve proficiency on a particular task being trained.16 Based on the calculation of transfer effectiveness ratios (TER) and percent transfer for the various tasks required to take off, fly, and land a plane (from the two training environments to the operational environment), they were able to identify where each training environment provided the most cost-effective training.16 The results indicated that training in the Link Trainer led to high, positive transfer on many of the flight tasks. Furthermore, they identified where training in the actual plane provided the most cost-effective training. The authors concluded there could be significant savings in flight training time if the ground-training environment was integrated appropriately into a training curriculum.16
To date, the authors are unaware of any published reports that compare teaching tools (simulator models) and their impact on the acquisition of surgical skills needed to perform the procedure on a live animal, especially with respect to the level of fidelity. Likewise, the authors are not aware of any published articles in veterinary medicine that have reported undertaking a fidelity analysis to determine whether a simulator model is suitable for the teaching and transferring of surgical skills into the live-animal surgical environment or whether there has been research that has investigated the return on investment in time and money spent to create surgical models.
Two veterinary schools have developed similar surgical skills training courses using part-task trainers, instructor-facilitated laboratories, and assessments that begin in the student's first year and culminate with the student performing a live-animal OVH in his or her third year. To facilitate practice of the OVH procedure before live-animal surgery, each school independently developed a spay model that allows students to practice all of the skills they have learned on individual task trainers in a single procedure. Although the models are similar in that they allow the student to perform most skills needed to spay a small animal, there is a significant difference in the level of fidelity and cost of each model. The medical education literature provides some evidence that low- to medium-fidelity models can be just as effective as higher-fidelity models when training students in basic surgical skills.17,18
The purpose of this study was to compare three teaching interventions designed to help students prepare for their first live-animal OVH and to determine which one gave the greatest return on investment based on performance outcomes and simulator cost. Institutional human ethics review board approval was obtained from the Conjoint Health Research Ethics Board at the University of Calgary.
Materials and Methods
Development of the OVH Models
A canine OVH model called the ROSSie (Ross Ovariohysterectomy Surgical Simulator) was developed at the Ross University School of Veterinary Medicine (RUSVM) and is in part manufactured by a local manufacturer, Sun Island Cloths.a The ROSSie model was adapted from the commercially available three-layer surgical approach and closure model known as the DASIE (Dog Abdominal Surrogate for Instructional Exercise).19 The exterior sleeve of the ROSSie model is made of various fabrics that differ from the original DASIE. The model contains a supportive PVC frame, foam bowel loop, uterus and ovaries made from latex tubing, and suspensory ligaments made from a plastic tablecloth material. The ROSSie model allows students to practice aseptic technique, making a midline celiotomy surgical approach, hemostasis of small vessels, exteriorization of the uterus and ovaries, rupture of the suspensory ligament, pedicle ligation, and three-layer closure of the abdominal wall. The initial iteration of the model was developed over a span of 3 months. Modifications to both design and use of the model in the surgical skills curriculum have continued over the 7 years of its use in the teaching program. The cost of raw materials and labor to make the ROSSie model and stand is approximately $64 Canadian, and the cost of the replacement parts to reset the model for its next use is $20 Canadian for the fabric body wall pad plus $5 Canadian for each latex reproductive tract. The fabric body wall can be used repeatedly (up to eight incisions can be made) before needing to be replaced by simply rotating the fabric cuff to a new section for each new incision. Only the latex reproductive tract needs to be reset for each use. The approximate cost to reset the ROSSie model for each subsequent student is $7.50 ($2.50 per incision on the fabric body wall pad and $5 for the latex reproductive tract).
The University of Calgary Faculty of Veterinary Medicine (UCVM) developed a silicone OVH model in collaboration with Veterinary Simulator Industries (VSI).b The model has a plastic base that houses silicone parts, including a spleen, kidneys, gastrointestinal tract, bladder, and female reproductive tract. This model allows students to practice aseptic technique, performance of a midline celiotomy surgical approach, location of the uterus and ovaries within an organ-filled abdominal cavity, exteriorization of the uterus and ovaries, rupture of the suspensory ligament, pedicle ligation, and three-layer closure of the abdominal wall. This model also allows students to practice manipulating organs to help assess hemostasis, and it allows them to complete an intradermal suture pattern. The model took several months to develop and was a collaborative effort between VSI and the UCVM clinical faculty members, supported by a $20,000 DATIVE grant (Development and Assessment of Technology in Veterinary Education) to develop the prototype model.c It had been used in continuing education for practitioners to date but had not been previously used in an undergraduate teaching environment. The cost of the UCVM model to purchase is approximately $675 Canadian, and the cost of the replacement parts to allow resetting for next use is $40 Canadian for the three-layer body wall pad plus $45 Canadian for each female reproductive tract. These replaceable parts (three-layer body wall and reproductive tract) need to be reset for each subsequent student, at an approximate cost of $85 Canadian.
Both models were developed with a primary focus on teaching and practicing the psychomotor skills and procedural steps needed to perform an OVH procedure. Neither was designed to replicate the feel of the tissue in an actual spay due to limitations in materials, cost, and technology. Figures 1 and 2 show the respective models.
(A) Foam bowel loop, PVC support, and latex uterine body showing mounting of parts, wooden support for body sleeve pad, and body wall sleeve pad. (B) Side view showing body wall sleeve pad mounted in wooden support block (without PVC support inserted in sleeve). (C) Latex uterine body with attached plastic tablecloth suspensory ligaments. Knots are placed to represent the ovaries and cervix. Note: PVC support pipe fits into body wall sleeve and contains the latex uterus, which is attached to the PVC using binder clips.
Figure 1: ROSSie model components
(A) Complete simulator with body wall pad in place, simulator with lid removed from plastic base to demonstrate internal contents. (B) Gastrointestinal tract removed to show reproductive tract mounted in plastic base. (C) Reproductive tract is replaceable and is a one-time-use item that attaches to the base at three locations: both ovaries and cervix.
Figure 2: UCVM model components
Development of the OVH Teaching and Assessment Tools
A cognitive task analysis (CTA) of the OVH procedure was performed and used to create the teaching and assessment tools for measuring content validity.20,21 UCVM surgical teaching faculty members were recruited via email to perform the CTA of the canine OVH procedure.20 CTA is a technique used to break down a procedure into its component steps by videotaping experts, who are asked to describe every detail of what they are doing and why.20 Transcripts from each expert's recordings were then merged to create a gold standard checklist that was then used to build a consensus between experts on the critical steps in the procedure.20
A detailed standardized description was then produced, outlining the necessary steps to perform an OVH based on the CTA.20 A written learning tool was created from this description for student use (Appendix 1). It was seven pages in length and included 20 categories, with further detailed items. The categories included (1) surgical preparation of the patient (clip, prep, bladder expression), (2) draping, (3) laying out instrumentation, (4) ascertaining depth of anesthesia, (5) making the incision (through skin, subcutaneous, and muscle layers), (6) locating both of the uterine horns and the cervix, (7) rupturing the first ovarian suspensory ligament, (8) clamping the first ovarian pedicle, (9) ligating the ovarian pedicle, (10) transecting the ovarian pedicle, (11) locating the second ovarian pedicle, (12) manipulating the second ovarian pedicle: rupturing the suspensory ligament/clamping, ligating and transecting the second ovarian pedicle, (13) locating the cervix, (14) manipulating and clamping the uterine pedicle, (15) ligating the uterine body, (16) transecting the uterine body, (17) checking the ovarian pedicles and uterine stump for hemorrhage, (18) preparing to close the abdomen (gauze sponge count), (19) abdominal wall closure, and (20) subcutaneous closure and skin closure. The written learning tool was made available to all students in advance of the surgical laboratory. It was paired with a simulator model for those students in the appropriate groups.
A digital video recording was then made, showing a board-certified surgeon performing the surgical procedure on a canine patient, while demonstrating all of the steps listed in the standardized learning tool. Voice-over and captions highlighted important steps, relevant anatomic features, and decision-making points. The video was approximately 30 minutes in length. All students were provided with the video to review in advance of the surgical laboratory.
The OVH learning tool identified several objectives that needed to be taught and subsequently performed and assessed in the OVH procedure. These included gentle tissue handling; correct suture and instrument handling/selection; efficient exploration of the abdomen and location of the uterus/ovaries; ability to place secure ligatures, tie secure knots, and correctly perform suture patterns; and performance of the remainder of the technical steps involved in the OVH procedure.
Based on the CTA conducted with content experts, a 46-item standardized checklist was developed (by ER/RF) for assessment of student performance in the lab. Three major phases of surgery were identified: performing the midline celiotomy approach, performing the OVH, and performing a three-layer closure of the incision. The spay procedure was then further broken down into the following nine component parts: (1) making the body wall incision (7 items), (2) locating and manipulating pedicle #1 (9 items), (3) locating and manipulating pedicle #2 (9 items), (4) manipulating the broad ligament (1 item), (5) locating and manipulating the uterine pedicle (4 items), (6) checking hemostasis of the pedicles (4 items), and (7, 8, and 9) closure of the body wall, subcutaneous tissue, and skin (12 items). In summary, there were three major phases of the procedure, which were broken down into nine component parts identified by a clustering of several individual checklist items.
The checklist was designed to capture the quantitative aspects of the procedure by identifying which items were performed, while still allowing capture of the more qualitative aspects of performance using a 4-point Likert-type scale (unsatisfactory, borderline, good, excellent) to score performance on the component tasks according to how well they were performed. The checklist is provided in Appendix 2.
Evaluation of the Simulators: Expert Panel
To investigate the functionality and usefulness of the two spay models for training the OVH procedure, an expert panel of 15 small-animal or rural mixed practitioners, all having between 11 and 20 years of experience in private general veterinary practice, were recruited via email from the Distributed Learning Veterinary Learning Community (DVLC) associated with the UCVM. The practitioners were randomly assigned to two groups and asked to perform an OVH on either the UCVM model (n=8) or the ROSSie model (n=7). They were then asked to fill out a questionnaire designed to gain more knowledge about their current surgical experience, as well as collect comments on features of the models they were assigned to use. Specifically, they were asked about features of the models they found to be realistic, features they felt required improvement, and any challenges they encountered when using the model.
Evaluation of Training Interventions
Third-year students in the UCVM DVM program (class of 2015) were recruited by email. The students had prior anatomy teaching and basic surgical skills training using various part-task trainers in a vertically integrated clinical skills training program. Part-task surgical skills training took place in laboratories over the first three years of the DVM program, and their acquired skills were assessed using OSCE stations, including draping, gowning, and gloving; making a skin or body wall incision; performing a basic skin suture pattern; performing an inverting suture pattern; and three-layer abdominal wall closure. At the time of this OVH study, these third-year students had no prior whole-task training in the DVM program allowing them to put all the psychomotor skills together with the procedural and declarative knowledge necessary to successfully perform a spay.
Students were randomly assigned to one of three intervention groups (Group T: lab manual and video only, n=8; Group R: lab manual, video, and ROSSie model, n=9; Group U: lab manual, video, and UCVM model, n=10). Two weeks before the spay lab, all students in all groups were provided access to the digital video media and a detailed standardized learning tool that had been created following the CTA. Students in groups R and U were concurrently provided with their respective simulator models, suture material, and basic instruments. Students were encouraged to review the materials and practice with their simulator models as much as they felt they needed before the lab.
On the day of surgery, all student participants (n=27) completed a pre-operative survey designed to determine their prior surgical experience and confidence level going into surgery. A second survey, designed to gather data on the usability and acceptance of the models as a training device, was given to those students in the R (n=9) and U groups (n=10).
The use of animals at the UCVM was approved by VSACC (Veterinary Sciences Animal Care Committee of the University of Calgary) in conjunction with guidelines of the CCAC (Canadian Council on Animal Care). All patients were sourced from the DVLC following referral from veterinary practitioners. The students performed routine pre-operative physical examination and blood analysis, including a complete blood count and chemistry, the day before surgery. Only healthy animals had surgery performed as part of the lab experience.
Board-certified clinical skills instructors (surgeons, anesthesiologists, and small-animal specialists), who had taught and run the spay lab in previous years, supervised the pre-operative preparation. All instructors were blinded to the students' research intervention group assignment.
The students oversaw post-operative recovery, hospitalization, and care of the patients under the supervision of the clinical instructors and technicians. Animals were typically discharged on the day following the procedure.
Students were randomly assigned into pairs for the performance of their surgical procedures. One student acted as anesthetist, while the other acted as surgeon and performed the OVH. Following the completion of the first procedure during the morning session, the students traded roles for the afternoon session. This assignment strategy allowed each student to perform the complete spay as both anesthetist and surgeon. Student pairs included combinations of students randomly assigned to different groups. All surgeries were overseen by qualified veterinary instructors, who provided surgical or anesthetic interventions if necessary to safeguard animal welfare. Time for the students to perform the procedure was recorded on the anesthesia forms, which documented the time at which the surgeon draped the patient to the time he/she finished the skin closure.
Student performance was assessed in the surgery lab by expert blinded raters (n=2). Expert is defined here by the rater's experience as a board-certified surgeon, along with specific prior training on rating student performance using standardized checklists as part of UCVM's clinical skills OSCE program. To avoid rater fatigue, individual raters were randomly assigned to a maximum of four students to observe at one time; however, rating was not done independently for each student. This team assessment was performed to allow raters the opportunity to observe as many behaviors as they could as they circulated between the participants. The raters were there solely to observe student behaviors and score performance, and they did not provide instruction or assist students. In addition to the raters, there were two additional board-certified surgical specialists, who were present solely to instruct and assist the students. They did not provide input on the rating of the student performance.
Following surgery, all students (n=27) were asked to complete an exit survey to gather information about their experience during the surgery and their confidence level after the procedure was completed. The R and U groups were also asked to add any comments regarding their model's usability following surgery on a live animal.
Statistical Analyses
An ANOVA was conducted on four OSCE surgical skills station scores to establish whether there was homogeneity between the randomly assigned intervention groups. An ANOVA was conducted on the time required to complete the live-animal spay surgery between the three intervention groups, rank-ordered groups, and morning/afternoon session assignment.
The surgical checklist results were analyzed by tabulating rating frequencies using a contingency table. Due to the small sample size per intervention group, the decision was made to collapse the ratings into two categories: excellent/good and borderline/unsatisfactory. Furthermore, in consideration of participant skill level (partway through year three of training) and this being the first live surgical experience as the surgeon of record, the instructors were of the opinion there would be fewer excellent ratings and that excellent ratings could be used to further refine the rank ordering of participants.
Chi-square analyses were conducted and the resulting likelihood ratio statistic used to establish whether there was a relationship and significance between the type of intervention and checklist ratings. The likelihood ratio statistic is preferred when the sample size is small.22 To account for the risk Type 1 error in the contingency analyses, a Bonferroni correction of 0.017 was used (p value/number of groups: .05/3=.017). Analyses were undertaken for each checklist item (1 to 46), phase of surgery, and individual surgery components.
One method for calculating an effect size with categorical data is the odds ratio (OR). A series of OR analyses were conducted to investigate the odds of scoring an excellent or good rating based on participant assignment. An OR analysis is best interpreted using a 2×2 contingency table, which is “the categorical data equivalent of a focused comparison.”22 Analyses were undertaken for each checklist item (1 to 46), surgery components, phase of surgery, and the total checklist score. Further analyses calculated the odds of scoring an excellent/good rating based on assignment to a simulator group (UCVM and ROSSie), spay model assignment (UCVM versus ROSSie), and morning versus afternoon sessions.
Participant performance was rank ordered based on the total number of excellent/good ratings. The rank ordering of participants with the same number of excellent/good scores was further refined based on the number of excellent ratings attained. The top and bottom 25% of participants were highlighted for checklist item and intervention group comparison.
ANOVA and Chi-square analyses were conducted using SPSS,22 and the OR was calculated based on instructions from Field (2009)22 and confirmed using a 2×2 calculator tool accessed at vassarstats.net/odds2x2.html.
Results
Expert Panel
The 15 practitioners recruited to evaluate the spay simulators all had between 11 and 20 years of experience in private general veterinary practice and completed spay procedures on a regular basis. The average number of reported spays per week based on simulator assignment groups did not differ significantly, p<.05 (Figure 3).
Figure 3: Mean OVH procedures performed per week by recruited DVLC small-animal practitioners (UCVM, n=8; ROSSie, n=7)
Student Participants
Thirty third-year students from the UCVM DVM program (class of 2015) volunteered and were randomly assigned to one of three intervention groups. Data from three students could not be collected after it was established their patient was either spayed or male during the pre-surgical physical examination.
Data from 27 students were analyzed (Group T: lab manual and video only, n=8; Group R: lab manual, video, and ROSSie model, n=9; Group U: lab manual, video, and UCVM model, n=10). Before their live-animal surgery lab, all students had access to the video media and standardized written learning tool outlining how to perform an OVH procedure. Only students in the R and U groups were assigned simulators to work with. All students had access to their teaching materials and models 2 weeks before their first live-animal OVH surgery.
There were no statistically significant differences between intervention groups relative to prior live-animal or OVH surgery experience (p<.05) or assessment scores on previous surgical skills OSCE stations (four OSCE station scores: instrument handling and making an incision, F[2, 24]=2.222, p<.130; simple suturing pattern, F[2, 24]=0.552, p<.583; inverted suturing pattern, F[2, 24]=0.015, p<.985; abdominal wall closure, F[2, 24]=1.224, p<.312). The results of the pre-operative survey are summarized by the rank order of the students' performance on their live-animal spay (Table 1). Students felt confident about performing the OVH, and there was no significant difference reported by rank ordering or group (p<.05).
Table 1: Summary of pre-surgical survey results pertaining to confidence level and prior surgical experience
Rank
Group
Confident to spay today?
Significant prior live- animal surgery experience?
Observed spays as part of a summer student job at a minimum of 7 months before the spay study
Usability of the Model
Findings from the DVLC practitioner questionnaire pertaining to the usability of the models were evaluated to identify useful aspects of the models and areas where improvements could be made. The results of the practitioner comments were tabulated, and the findings are summarized in Table 2.
Table 2: Summary of comments from general small-animal practitioners asked to perform a spay procedure on a simulator model (UCVM, n=8; ROSSie, n=7)
Feature of model
Summary of comments for UCVM model group
Summary of comments for ROSSie model group
Abdominal wall
Liked three-layer closure, but layers are a bit thin
Not fancy—skeptical at first of value Very clear for layers Great holding of sutures
Reproductive organs
Appearance good—obvious anatomy
An instruction sheet would be helpful
Maybe anchored too deep in abdomen
Use not intuitive by appearance
Penrose drain works well as uterine body and ovaries
Other internal organs
Perhaps unnecessary
No other organs may be drawback
Tissue handling
Great for breaking suspensory ligaments, clamping, ligating
No suspensory or broad ligaments to handle—may be limitation
Overall
Greasy (silicone oil eluted out of model organs during handling) Friable (silicone material can be delicate, especially for ligation) Very good for learning anatomy and mechanics of what to do in order
“Looking at model prior to the spay I was skeptical, but after the spay I am impressed. All students should have opportunity to practice with this” (direct quote from one participant summarizes it well)
The pre-operative survey results and post-operative comments concerning the usability and acceptance of the model as a training device are summarized by the rank order of students' performance on the live-animal spay (Tables 3 and 4).
Table 3: Pre-operative survey results for simulator groups by rank order of performance
Table 4: Post-operative student comments regarding spay model usability
Rank
Group
Comments
1
ROSSie
Model was confusing and not useful at all.
2
ROSSie
Difficult to close on this model. Like the ability to practice. Not realistic.
3
UCVM
The best thing was to develop muscle memory of the steps in the procedure. The tissue was very friable and tore when clamped or sutured. The video and description were great, but the model gave me more confidence because I could do it.
4
UCVM
Video was super helpful, and the model confirmed/tested my knowledge.
5
ROSSie
An instruction booklet provided with the model would be helpful—not intuitive. Did not feel real.
7
ROSSie
(No comment)
9
UCVM
Really liked the model. Realistic and definitely helped with landmarks and how to maneuver things. Wish the skin was thicker as I couldn't do a proper close.
10
UCVM
(No comment)
11
UCVM
(No comment)
12
UCVM
Model helpful for working through the steps, but did not like the texture of the silicone—not realistic and hard to manipulate (often broke when using instruments).
13
ROSSie
(No comment)
15
UCVM
(No comment)
16
UCVM
(No comment)
17
ROSSie
Model not helpful for learning anatomy and landmarks. Helpful to practice ligation though.
18
UCVM
Anatomy was good. Tissues were greasy, and sutures cut through easily. Can only use model once—seems wasteful. Practicing greatly enhanced my thinking about and confidence in the procedure.
19
ROSSie
The video and written instructions were more helpful than the simulator provided.
20
UCVM
Model increased my confidence in completing the procedure. I practiced the subcuticular pattern before surgery but afterwards realized that I should have done this a few more times.
22
ROSSie
Model not helpful for finding the linea alba, searching for the uterus, or suturing. I found the video and lab manual much more helpful.
25
ROSSie
I had no idea what the cervix was supposed to be. The subQ was frustrating to suture (foam) because it kept getting tangled in forceps and needle. I liked suturing the skin though.
Length of Surgical Procedure
The surgical procedure time was recorded for each participant, starting with the initial draping and concluding with the end of skin closure. There were no statistically significant differences in surgery time between the three intervention groups (F[2, 24]=1.749, p<.197), between the three rank-ordered groups (F[2, 24]=0.757, p<.480) or between the morning and afternoon sessions (F[1, 25] = 0.088, p<.770).
Expert raters used a standardized 46-item checklist to assess performance of each student during the procedure. Each item was rated using a 4-point Likert-type scale (excellent, good, borderline, unsatisfactory).
A Chi-square analysis comparing the three intervention groups revealed no statistically significant differences on any of the 46 individual checklist items. However, statistically significant differences between the three interventions were observed in the total score (combining the 46 checklist items). Statistically significant differences were noted when the checklist items were collapsed into the nine individual surgery components (making the incision, locating the second pedicle, closing the fascia of the body wall, and closing the subcutaneous tissue layer), and statistically significant differences were observed in the three main phases of the procedure (making the incision, performing the OVH, and closing the incision). The contingency table and likelihood ratio results for the comparisons of the three intervention groups are provided in Tables 5 and 6.
Table 5: Chi-square results presented by phase of surgery
Intervention
Phase of surgery
Rating categories
UCVM
ROSSie
Video
Likelihood ratio
Making the incision (7 items)
Excellent/Good
64
49
37
p≤.002
Borderline/Unsatisfactory
6
14
19
Performing the OVH (27 items)
Excellent/Good
244
218
170
p≤.001
Borderline/Unsatisfactory
26
25
45
Closing the incision (12 items)
Excellent/Good
106
92
65
p≤.001
Borderline/Unsatisfactory
14
16
30
Total
Excellent/Good
414
359
272
p≤.001
Borderline/Unsatisfactory
46
55
94
Table 6: Chi-square results presented by surgery component
Intervention
Surgery components
Rating categories
UCVM
ROSSie
Video
Likelihood ratio
Making the incision (7 items)
Excellent/Good
64
49
37
p≤.002
Borderline/Unsatisfactory
6
14
19
Pedicle 1—locating and manipulating (9 items)
Excellent/Good
79
74
62
p≤.593, NS
Borderline/Unsatisfactory
11
7
10
Pedicle 2—locating and manipulating (9 items)
Excellent/Good
87
77
61
p≤.017
Borderline/Unsatisfactory
3
4
11
Broad ligament—manipulating (1 item)
Excellent/Good
10
9
5
p≤.019, NS
Borderline/Unsatisfactory
0
0
3
Uterine body—locating and manipulating (4 items)
Excellent/Good
37
28
22
p≤.046, NS
Borderline/Unsatisfactory
3
8
9
Hemostasis—assessing (4 items)
Excellent/Good
31
30
20
p≤.132, NS
Borderline/Unsatisfactory
9
6
12
Close body wall (6 items)
Excellent/Good
58
47
37
p≤.013
Borderline/Unsatisfactory
2
7
10
Close subcutaneous (3 items)
Excellent/Good
27
25
14
p≤.017
Borderline/Unsatisfactory
3
2
10
Close skin (3 items)
Excellent/Good
21
20
14
p≤.481, NS
Borderline/Unsatisfactory
9
7
10
Total (46 items)
Excellent/Good
414
359
272
p≤.001
Borderline/Unsatisfactory
46
55
94
Odds Ratio
Given that statistically significant differences were observed in the Chi-square analyses, OR analyses were undertaken to identify specifically where these differences occurred during the OVH procedure. The decision was made to calculate the odds of a participant scoring an excellent or good rating based on intervention group assignment. An OR value equal to or larger than 2.0 was arbitrarily selected as a cutoff point, which would then be interpreted as the odds of attaining an excellent or good rating being two times higher when assigned to a particular intervention.
Odds Ratio: Simulator Versus No Simulator
Participants assigned to a spay model were two or more times more likely to achieve an excellent/good rating on all three phases of surgery (making an incision, 2.9; spay, 2.4; and suturing, 3.0) and on six of nine surgery components (making the incision, 2.9; locating and manipulating pedicle #2, 4.2; locating and manipulating the uterine body, 2.4; assessing hemostasis, 2.4; closing the body wall, 3.2; and closing the subcutaneous layer, 7.4). Based on all of the items scored (total score), participants in the simulator group (groups R and U) were 2.6 times more likely to score an excellent/good rating in comparison to those assigned to the group without a simulator (group T). An OR of 2.0 or higher was calculated for 25 of the 46 checklist items. The contingency table and OR results are listed in Table 7.
Table 7: Odds ratio results: model (UCVM or ROSSie) versus no model by individual checklist item
Section
Item
Intervention
B/U
E/G
OR
Making the incision
Adequate length
Model
1
18
10.8
No model
3
5
Adequate hemostasis
Model
1
18
2.5
No model
1
7
Proper sharp blunt dissection
Model
6
13
2.2
No model
4
4
Safe abdominal approach
Model
4
15
3.8
No model
4
4
Extends safely
Model
2
17
8.5
No model
4
4
Pedicle #1
“Window” safely broken
Model
1
18
6.0
No model
2
6
Gentle tissue handling
Model
1
18
6.0
No model
2
6
Pedicle #2
“Window” safely broken
Model
1
18
2.8
No model
1
7
Gentle tissue handling
Model
1
18
6.0
No model
2
6
Places secure ligation
Model
2
17
2.8
No model
2
6
Uterine body
Cranial to cervix
Model
3
16
2.1
No model
2
5
Places secure ligation
Model
2
17
5.1
No model
3
5
Places secure transfixation
Model
1
18
10.1
No model
3
5
Hemostasis
Check mesocolon
Model
2
17
5.1
No model
3
5
Check mesoduodenum
Model
4
15
2.3
No model
3
5
Check colon and bladder
Model
4
15
2.3
No model
3
5
Closing body wall
Appropriate suture size
Model
1
18
6.0
No model
2
6
Adequate bite size
Model
1
18
6.0
No model
2
6
Adequate bite spacing
Model
2
17
2.8
No model
2
6
Linea layer ventral
Model
1
18
6.0
No model
2
6
Knot at end secure
Model
1
18
3.0
No model
1
6
Closing subcutaneous
Correctly bury knot
Model
1
18
18.0
No model
4
4
Simple continuous pattern
Model
1
18
10.8
No model
3
5
Ends with proper buried knot
Model
3
16
3.2
No model
3
5
Closing skin
Ends with proper buried knot
Model
10
9
2.7
No model
6
2
There was no statistically significant difference found for component 4—manipulating the broad ligament.
Odds Ratio: UCVM Simulator Versus ROSSie Simulator
Participants assigned the UCVM simulator were two or more times more likely to achieve an excellent/good rating on one phase of surgery (making the incision, 3.0) and three of the nine surgery components (making the incision, 3.0; locating and manipulating the uterine body, 3.5; and closing the fascia of the body wall, 4.3).
A calculation of the individual checklist items indicated an advantage on six items using the UCVM model (proper sharp blunt dissection, safe abdominal approach, locating the first horn and ovary, locating the cervix, secure knot at end of closing the body wall, and ending with a proper buried knot). An advantage on one of the checklist items was found using the ROSSie model (suspensory ligament broken easily). The contingency table and OR results for these items are listed in Table 8.
Table 8: Odds ratio results equal to or larger than 2.0 for UCVM versus ROSSie model
Section
Item
Intervention
B/U
E/G
OR
Making an incision
Proper sharp blunt dissection
UCVM
1
9
11.3
ROSSie
5
4
Safe abdominal approach
UCVM
1
9
4.5
ROSSie
3
6
Pedicle #1
Locates first horn and ovary
UCVM
1
9
2.6
ROSSie
2
7
Suspensory ligament broken
UCVM
4
6
5.3
ROSSie
1
8
Uterine body
Locate cervix
UCVM
2
8
2.0
ROSSie
3
6
Closing body wall
Knot at end, no gap
UCVM
1
9
2.6
ROSSie
2
7
Closing subcutaneous
Ends with buried knot
UCVM
1
9
2.6
ROSSie
2
7
Odds Ratio: Morning Versus Afternoon Surgical Assignment
Participants assigned to the afternoon session were two or more times more likely to achieve an excellent/good rating on one phase of surgery (making the incision, 2.4) and two of the nine surgery components (making the incision, 2.4; locating and manipulating pedicle #2, 2.2).
A calculation of the individual checklist items indicated an advantage on three items for participants performing surgery in the morning session (locating the first horn and ovary, suspensory ligament broken, places secure circumferential ligation). An advantage on two checklist items was found in the afternoon session (proper sharp blunt dissection, safe abdominal approach). The contingency table and OR results for these items are listed in Table 9.
Table 9: Odds ratio results equal to or larger than 2.0 for morning or afternoon sessions
Section
Item
Intervention
B/U
E/G
OR
Making an incision
Proper sharp blunt dissection
AM
8
7
5.7
PM
2
10
Safe abdominal approach
AM
6
9
3.3
PM
2
10
Pedicle #1
Locates first horn and ovary
AM
3
12
2.8
PM
1
11
Suspensory ligament broken
AM
4
11
2.8
PM
4
8
Pedicle #2
Placing secure ligation
AM
3
12
2.8
PM
1
11
Rank Ordering of Participants
Participants were rank ordered based on the number of excellent/good ratings attained out of 46 checklist items. The top and bottom 25% (n=7 each) and middle 50% (n=13) were categorized. The top seven included four participants using ROSSie, two using the UCVM simulator, and one from the video intervention group. The bottom seven included two participants using ROSSie and five from the video intervention group.
The number and type of checklist ratings for the top seven (total=322) included 84% excellent (n=270), 15% good (n=47), and 1% borderline ratings (n=5). There were no unsatisfactory ratings in the top seven ranked students. The number and type of ratings for the bottom seven (total=320, 2 missing) included 24% excellent (n=77), 40% good (n=128), 31% borderline (n=99), and 5% unsatisfactory (n=16). The rank ordering of participants, rating on each checklist item, and amount of faculty assistance required are listed in Appendix 3.
There are no statistically significant differences on third-year OSCE performance between the rank-ordered groups (F[2, 24]=0.477, p<.626).
Post-Operative Surveys
All participants using one of the OVH models (U and R, n=19) were surveyed post-operatively to gather data on how they used the model to prepare and how confident or anxious they felt during the live-animal procedure (Table 10). Answers on the survey were categorized using a 5-point Likert-type scale with categories of strongly agree (5), agree (4), neutral (3), strongly disagree (2), and disagree (1). All participants reported that they used their model only once during their preparation and that they practiced alone.
Table 10: Results of post-operative survey of model groups by rank order
Worldwide, veterinary schools are exploring the use of various task trainers to provide students with surgical skills training.23–25 More specifically, several schools have developed OVH models for students to practice the procedure.8,13 In recent literature reviews of simulator evaluation and instructional design in simulation-based medical education, researchers generally concede that more well-designed research projects are needed that collect the appropriate data to support the use of specific training methods and simulators.26–28 This project served to develop the teaching and assessment tools required to critically evaluate two OVH models of varying fidelity.
Chi-square analyses determined statistically significant differences between the three intervention groups on four of nine surgery components (making the incision, locating the second pedicle, closing the fascia of the body wall, and closing the subcutaneous tissue layer), all three phases of surgery (making the incision, performing the OVH, and closing the incision), and the overall score (all 46 items).
To establish whether training with a spay model improved performance over training with video and lab manual only, we conducted OR analyses. The results indicated that training with a spay model improved the odds of attaining an excellent or good rating on almost half of the checklist items, six of nine surgery components (making the incision, locating and manipulating pedicle #2, locating and manipulating the uterine body, assessing hemostasis, closing the body wall, and closing the subcutaneous layer), all three phases of surgery (making the incision, performing the OVH, and closing the incision), and the overall score.
We also conducted OR analyses to compare the $675 ($85 to reset for the next student) UCVM silicon-based model with the $64 ($7.50 to reset for the next student) ROSSie cloth and foam model. The results indicated an advantage for the UCVM model on six checklist items (proper sharp blunt dissection, safe abdominal approach, locating the first horn and ovary, locating the cervix, secure knot at end of closing the body wall, and ending with a proper buried knot), three of nine surgery components (making the incision, locating and manipulating the uterine body, and closing the fascia of the body wall), and one phase of surgery (making the incision), with the ROSSie model demonstrating an advantage on one checklist item (suspensory ligament broken easily).
In general, the results of this study provide preliminary evidence that using an OVH simulator in preparation for a student's first live-animal spay provides better training than video and written resources alone. These results are similar to those of Griffon and colleagues, who determined that training with a reusable spay model was more effective than cadavers in teaching basic surgical skills and the OVH procedure in canines.9
In addition, the results suggest that training with a lower-fidelity and less expensive spay model might be just as effective in teaching the necessary skills to perform an OVH when compared to training with a higher-fidelity and more expensive spay model.
Usability of the Simulators
Overall, the expert panel comments on the usability of the UCVM and ROSSie models suggested that neither model was perfect, particularly in regard to the feel of the tissue. Yet both models provided the experience of completing most procedural steps required in performing an OVH. As such, practitioners stated the models should be made available to students for practice.
This consensus matches the developers' design intentions: developing a model that provides students with repeated opportunities to practice the psychomotor skills and procedural steps associated with an OVH but without necessarily replicating the tactile feel of a live-animal surgery.
There were no statistically significant differences in student performance between the two simulator groups for the live-animal surgery. Despite this, students in the UCVM model group were more likely than those in the ROSSie model group to state that the model was helpful and that it increased their knowledge and confidence level pre-operatively. Despite their inexperience, students noted similar issues with the models as those identified by the expert panel; however, perhaps not surprisingly, they were not as open minded. Practitioners admitted skepticism regarding the ROSSie model based on appearance but went on to appreciate its value when they worked with it. Our results would seem to suggest that a student's perception of the model's training effectiveness alone may not be the most reliable measure and should not be the sole validation of a model.
Baum et al. acknowledged the dichotomy of learner perception of usefulness versus actual impact on learner performance when developing a research strategy to investigate simulator fidelity and training effectiveness in the military, stating that “a device may be unpopular or lack face validity, but nevertheless leads to excellent performance on the operational equipment.”29(p.25) Research in simulation-based medical education further supports that this dichotomy is likely due to a misalignment of measures with objectives. Students with limited experience may not be able to assess how much they have actually learned from a particular simulation.30
The more favorable student comments concerning the UCVM model could be a result of participant bias. The students were well aware of the model's existence and knew that it was developed by UCVM before the study was conducted. Alternatively, the UCVM students did not consider the ROSSie model intuitive and were quick to dismiss it. Providing a guide for use of the ROSSie model may have improved perceptions of its usability at the outset. This would have been similar to how instructors at RUSVM introduce it in their program. To determine the usability of each model on its own, the study design here did not include a guide for either model.
The dichotomy between what users think of a model and the actual usefulness of the model has led over time to face validity being considered less important when considering validity of a simulator.27 Recent work in the area of simulation has suggested that face validity should no longer be used when evaluating validity of assessments, and our work here would seem to support that.31,32
Regardless, satisfaction data are important to collect regarding use, because student perceptions of the model will affect their motivation to use it. If students do not feel the model is helpful, they will be less motivated to practice on it.31 Given that deliberate practice is reportedly important for good outcomes in simulation training, trainee motivation to practice on a model is imperative.26
In summary, the qualitative data collected from the experts regarding the utility of both models provides preliminary evidence to support integration of the UCVM or the ROSSie model in a surgical skills training program. Feedback also helped inform the model developers of areas for improvement.
Comparison of Performance on the Live-Animal OVH and Previous Surgical Skills OSCE Scores
Analyzing the prior surgical skills OSCE data, there were no statistically significant differences between the intervention groups before training. Nor were there statistically significant differences between the rank-ordered groups after live surgery. This is important in establishing that the student surgeons were essentially equivalent in skill level no matter which group they were assigned to. It also suggests that OSCE task performance is not necessarily predictive of live-animal procedural performance, where students are required to integrate multiple tasks, such as making an incision, suturing, and closing the body wall, into a single procedure on a real patient. One explanation for this is that OSCEs do not necessarily integrate important, non-technical components such as clinical reasoning, confidence, group work, and support, which may also lead to procedural competence and success.33
Surgery Time
The surgery time did not vary significantly based on intervention, rank ordering, or morning/afternoon assignment. One possible reason for similar times between model groups and the non-model group is the lack of repetitive practice coupled with feedback. The researchers gave the students the models but did not provide specific instruction on how many times they should practice the procedure or in what context, alone or paired with a partner. All of the students indicated that they practiced once on their own, providing neither repetition nor feedback from a peer or instructor. Given that repetitive practice with feedback is considered an important factor in the success of a simulation, the lack of repetitive practice could be a major contributing factor to the lack of impact on the surgery time seen in the model groups versus those without model training.26 The authors plan to repeat this study with a broader group in the future and plan on providing more explicit instructions regarding practice. In this pilot study, the authors were not prescriptive about the amount of practice because they noted that the UCVM students had a large amount of work on task trainers in their curriculum previously. The authors thought it would be of interest to see how much practice students felt they required in putting those tasks together into one single procedure.
Transfer of Skills to Live-Animal Procedure
The results indicate that participants in the simulator groups outperformed those in the non-simulator group. A comparison between the two spay simulators indicated there was no statistically significant difference in performance on the majority of items.
This is important when considering modifications to the models. If students can perform these aspects of the procedure without physically practicing them on the whole-task trainer, there may not be a need to improve upon these areas of the models. Providing the correct level of fidelity at the right time in the training process can potentially be a more cost-effective use of simulation.34
Transfer of Skills to Live-Animal Procedure: UCVM Simulator versus ROSSie Model
From the perspective of skill transfer to the live-animal surgery, students in the UCVM model group were two or more times more likely than students in the ROSSie model group to attain an excellent or good rating on their midline celiotomy approach, locating and manipulating the uterine body, and closing the fascia of the body wall.
An inspection of individual checklist items indicated participants trained using the UCVM model group performed better on six items (proper sharp blunt dissection, safe abdominal approach, locating first horn and ovary, locating cervix, secure knot at end of body wall closure, and ending with buried knot in skin) and the ROSSie model participants performed better on one (breaking the suspensory ligament).
These findings suggest that the fabric sleeve of the ROSSie model does not provide as good an experience to practice a midline celiotomy approach or closure of the fascia on the body wall as the UCVM model. Students also do not have the ability to practice an intradermal pattern on the ROSSie model as the dermis is too thin. This could have contributed to the diminished performance of the ROSSie model group when burying the knot because the students in this group did not have the opportunity to practice this skill when closing the dermis.
Interestingly, the veterinary surgeons in charge of the surgical skills training program at Ross University also identified similar weaknesses of the ROSSie model. A cadaver lab is now used in place of the ROSSie model to teach skin incisions and suturing, and a bucket covered with a thick felt layer for skin is used to teach the intradermal pattern and burying of the knot. The results of this study along with the increased commercial availability of higher-fidelity silicone multilayer suture pads at reasonable prices—and, more recently, do-it-yourself guides on pouring silicone suture pads—have motivated the developers of the ROSSie model to pursue a better three-layer closure model to use as a sleeve for the model.35
The one area where the ROSSie model appeared superior to the UCVM model was in the breaking of the suspensory ligament. The plastic tablecloth material used to re-create the suspensory ligament, while crude in appearance, was more realistic in feel than the more greasy silicone material of the UCVM model, and this likely accounted for a better practice experience for this part of the procedure. Some students commented on this positive aspect in their feedback on the ROSSie model.
Surgical Preparedness
In general, students felt that their preparation was good. As you get lower in the rank ordering, the students self-reported a lower level of confidence and a higher level of anxiety during surgery. Although some high-performing students in the top of the rank order reported anxiety during the procedure, they were more likely to simultaneously report that they felt strongly confident. This would seem to suggest that anxiety was manageable, which fits with the findings of others and may be one of the greatest advantages to practicing with a model.7
Cost
The authors believe that improvements to the ROSSie model in the areas identified could make it just as effective a training tool as the UCVM model. These findings could have huge cost implications regarding use of one of these models over another for a surgical skills program given the current trend toward larger cohort sizes in veterinary medicine.
Providing students with a ROSSie model to practice on has an initial cost of $64 Canadian per student, and students can practice the entire OVH procedure up to eight times for as little as $40 Canadian for replacement parts. This equates to a total cost of $104 Canadian.
Providing students with a UCVM silicone model to practice on is $675 Canadian per student, and students can only practice once on the model. Each subsequent practice session requires purchase and replacement of both the suture pad and the female reproductive tract, costing $85 Canadian. If a student were to practice the OVH eight times on this model, it would cost $595 Canadian for replacement parts plus the initial cost of $650. This is a total of $1,245 Canadian for the same amount of experience. The return on investment is obviously not justified when one considers the results obtained in this pilot study. To help offset the cost of use, the UCVM has recently started to create in-house replacement silicone parts and body wall pads, as mentioned in the previous section on skills transfer.
This study also highlights that a fidelity analysis is crucial in determining what aspects of training a model should address and what is needed to teach that skill or procedure most appropriately. The analysis would allow developers to create the most effective training model possible at the lowest cost, thereby maximizing return on investment.
This study would seem to suggest that the UCVM model is slightly superior when compared directly to the ROSSie model, but only in a few specific parts of the procedure. Perhaps it would be easier to justify use of the UCVM model with reusable suture pads and recyclable organs, allowing more uses per purchase, or alternatively with supplemental homemade parts. The UCVM model did have a clear advantage in looking more realistic, and this may be worth the investment to encourage students to practice on it; however, it is unclear whether this advantage justifies such large expense, especially with a large student cohort, as is typical in many veterinary schools today. With minor modifications, the ROSSie model is actually the more cost-effective solution and would provide a suitable training experience for this procedure.
Limitations of this Study
The authors recognize that our participant numbers were low, and we plan to undertake a subsequent study with another veterinary program and larger student numbers. Due to the small cohort studied here, the authors used maximum likelihood estimates and a Bonferroni adjustment to reduce the risk of Type 1 error.
There was no assessment of inter-rater reliability, and this is a weakness of the present study. In future, student performance should be digitally recorded and independently assessed by a larger pool of raters over each item. In this current study, two raters assessed the students, but both raters did not assess all items. While they were overlapping in their observation of the student performance, they were not exhaustive.
The arbitrary selection of 2.0 for the OR cut-off was intended to inform the model developers and study authors of differences between models (UCVM versus ROSSie) or differences between interventions (training with a model versus without). The concern was that having a value too stringent (high) might result in failure to identify specifically where a model or intervention might have an advantage over the other. Future research might be undertaken to investigate cut-off values and their efficacy in informing simulator updates.
The reliability of the assessment tool and the internal consistency of the checklist were also not evaluated. However, it should be re-emphasized that the checklist was developed following CTA of four surgical experts to create a teaching tool. The CTA represents a consensus among the experts as to the simplest method for a novice to perform an OVH in a small-animal patient. The checklist was then further validated by comments made by the expert panel regarding the steps of the procedure most needed to be practiced by the learner.
There was potential bias for the UCVM model in the UCVM students who participated in this study. UCVM students recognized the model developed at their own school and were suspicious of the model developed elsewhere. As has been shown, the UCVM model ultimately did not appear significantly superior to the ROSSie, so student perception of a simulator should only be one part of the evaluation. Surveys given to the learners should include questions geared at determining user acceptance and motivation to use the model. A repeat study with a larger cohort of students from another institution that did not develop either model might provide more conclusive evidence, especially where this study relates to the return on investment. Providing an instruction guide might also remove this confounding factor. Finally, asking students to practice on each model a certain number of times as opposed to giving them a choice as to how many times they used the model would also strengthen this study.
Conclusions
The evidence gathered supports the use of both the UCVM and ROSSie OVH models combined with video and written instruction in preparation for a student's first live-animal spay.
Item-specific outcome measures suggest that slight modification to both models should be considered. Changes to the ROSSie model might include replacing the fabric abdominal wall with material that more realistically mimics tissue and providing a guide to its use detailing the anatomy of the model. Changes to the UCVM model might include increasing the strength and improving the feel of the tissues, as well as finding a way of making the suture pad more reusable. The authors have considered that a hybrid model may actually be most useful for teaching and learning the OVH procedure, and may provide the best return on investment for the program. This pilot study shows the importance of including fidelity analysis in model design and highlights that the validation of models in veterinary medicine is a complex process that should consider return on investment.
ACKNOWLEDGMENTS
The authors acknowledge Aylin Atilla, James Dundas, Priti Karnik, Graham Keys, Eric Pope, Matt Read, Terri Schiller, Claire Spackman, and Julie Williamson.
Footnotes
a
Sun Island Cloths, Bay Road, Basseterre, St. Kitts, West Indies
DATIVE (Development and Assessment of Technology in Veterinary Education) grant, University of Calgary Faculty of Veterinary Medicine, Calgary, Alberta, Canada
REFERENCES
1.
Pukacz M, Kienzle B, Braun J. Simple, minimally invasive technique for ovariohysterectomy in the dog. Vet Rec. 2009;165(23):688–90. Medline:19966332
Walsh DA, Osburn BI, Schumacher RL. Defining the attributes expected of graduating veterinary medical students, part 2: external evaluation and outcomes assessment. J Vet Med Educ. 2002;29(1):36–42. https://doi.org/10.3138/jvme.29.1.36. Medline:11932839
Smeak DD, Hill LN, Lord LK, et al. Expected frequency of use and proficiency of core surgical skills in entry-level veterinary practice: 2009 ACVS core surgical skills diplomate survey results. Vet Surg. 2012;41(7):853–61. https://doi.org/10.1111/j.1532-950X.2012.00978.x. Medline:22381004
Bowlt KL, Murray JK, Herbert GL, et al. Evaluation of the expectations, learning and competencies of surgical skills by undergraduate veterinary students performing canine ovariohysterectomies. J Small Anim Pract. 2011;52(11):587–94. https://doi.org/10.1111/j.1748-5827.2011.01120.x. Medline:22026742
Gopinath D, McGreevy PD, Zuber RM, et al. Developments in undergraduate teaching of small-animal soft-tissue surgical skills at the University of Sydney. J Vet Med Educ. 2012;39(1):21–9. https://doi.org/10.3138/jvme.0411.044R. Medline:22430078
Langebæk R, Berendt M, Pedersen LT, et al. Features that contribute to the usefulness of low-fidelity models for surgical skills training. Vet Rec. 2012;170(14):361–8. https://doi.org/10.1136/vr.100181. Medline:22331503
Boet S, Bould MD, Fung L, et al. Transfer of learning and patient outcome in simulated crisis resource management: a systematic review. Can J Anaesth. 2014;61(6):571–82. https://doi.org/10.1007/s12630-014-0143-8. Medline:24664414
Langebæk R, Toft N, Eriksen T. The SimSpay—student perceptions of a low-cost build-it-yourself model for novice training of surgical skills in canine ovariohysterectomy. J Vet Med Educ. 2015;42(2):166–71. https://doi.org/10.3138/jvme.1014-105. Medline:25862402
Andrews D, Bell H. Simulation-based training. In: Tobias S, Fletcher J, editors. Training and retraining: a handbook for business, industry, government, and the military. New York: MacMillan Reference; 2000. p. 357–84
Hays RT, Singer MJ. Simulation Fidelity as an Organizing Concept. In: Hays RT, Singer MJ, editors. Simulation fidelity in training system design: bridging the gap between reality and training. New York: Springer-Verlag; 1989. p. 47–75. https://doi.org/10.1007/978-1-4612-3564-4
Grober ED, Hamstra SJ, Wanzel KR, et al. The educational impact of bench model fidelity on the acquisition of technical skill: the use of clinically relevant outcome measures. Ann Surg. 2004;240(2):374–81. https://doi.org/10.1097/01.sla.0000133346.07434.30. Medline:15273564
DASIE Surgical Training Tools. DASIE™: Dog Abdominal Surrogate for Instructional Exercise [Internet]. [cited 2016 Apr 4]. Available from: http://www.dasiesurgery.ca/DASIE/DASIE.html
Cook DA, Zendejas B, Hamstra SJ, et al. What counts as validity evidence? Examples and prevalence in a systematic review of simulation-based assessment. Adv Health Sci Educ Theory Pract. 2014;19(2):233–50. https://doi.org/10.1007/s10459-013-9458-4. Medline:23636643
Issenberg SB, McGaghie WC, Petrusa ER, et al. Features and uses of high-fidelity medical simulations that lead to effective learning: a BEME systematic review. Med Teach. 2005;27(1):10–28. https://doi.org/10.1080/01421590500046924. Medline:16147767
Baum D, Smith DA, Hirshfeld SF, et al. Specification of training simulator fidelity: a research plan [Internet]. Alexandria, VA: Army Research Institute for Behavioral and Social Sciences; 1982 [cited 2015 May 24]. Available from: http://files.eric.ed.gov/fulltext/ED241793.pdf
Appendix 1: OVH Learning Tool Developed by CTA Small-Animal Ovariohysterectomy Surgical Procedure
Please note: This written protocol represents what the small animal surgery faculty at the University of Calgary have agreed is the preferred method to be taught to our DVM students for performance of an ovariohysterectomy in a dog or a cat. This protocol was derived using a cognitive task analysis. Where there are slight differences of opinion between the experts, then this will be noted, and all of the acceptable options will be noted using initials to identify the surgeon's individual preferences.
1. Surgery Prep
While the anesthetized patient is on the table (after being clipped and before being sterilely prepped), the bladder should be expressed into a stainless steel bowl.
2. Draping
After sterile prep is completed, then the patient is draped using four square drapes.
Drape 1—
Cranial, approximately 4 cm cranial to the umbilicus;
Drape 2—
Side, 5 cm from midline on the side or along the nipple line, whichever is closer to the midline (cats);
Drape 3—
Caudal, approximately level with pubis; and
Drape 4—
Side, 5 cm from midline on the side or along the nipple line, whichever is closer to the midline (cats).
Note: Size of draped area is variable depending on patient size, age, and condition (obese, pregnant, or in heat) but these are guidelines commonly used in practice.
Place towel clamps with handles facing away from incision area, one jaw in each drape at the corners, then push down and catch skin. You may (CTH, AA) or may not (TS) tuck handles to avoid trauma to bowel or organs placed on surface of the drape. If you do not tuck handles, then you are relying on the split sheet alone to protect viscera.
Apply a split sheet over top. Can be disposable or cotton. If using disposable, then place the drape over top, and cut out surgical site “window” using operating (suture) scissors. May (CTH, AA) or may not (TS) towel clamp. If towel clamps used, then use one cranial and one caudal (placed between the clamps used to fix the corners of the four square drapes).
3. Lay Out Instrumentation
Lay out your instruments on the back (prep) table. Drape a Mayo stand if using one (using two drapes, one for each half of the stand). Gauze count—Some surgeons count the number before and after surgery (CTH, AA), and some do not but take great care to avoid use of gauze sponges in or near open abdomen and only allow lap sponges with radio-opaque handle once the abdomen is open (TS, EKR). You should also get in the habit of scanning the abdomen before you close to make sure that you have not left anything in the abdomen. Counting is not always reliable.
Use your Mayo stand for your immediate needs instruments and keep the rest on the back table to avoid clutter and to keep things clean.
4. Check with Anesthesia
Ask anesthesia if you can start surgery. They will indicate if it is appropriate to do so.
5. Making the Abdominal Incision
Load a #15 blade on a #3 scalpel handle. Make your incision starting at the umbilicus in a dog (and slightly caudal, 1–2 cm, to the umbilicus in a cat) (a cat's reproductive tract is slightly more caudal than a dog's and more difficult to exteriorize, so a caudal incision can be helpful).
Incise skin for distance of 2–10 cm depending on condition of the patient (age, reproductive status) and operator experience (larger incision in less experienced operator).
Use a combination of sharp and blunt dissection to clear and expose the linea alba. Avoid excessive (lateral) dissection because you will create a large amount of dead space that will lead to seroma formation post-operatively. The goal is to only create enough space that you can easily identify the linea alba to enter the abdomen and differentiate your fascial plane from your subcutaneous tissues during closure. The amount of lateral dissection performed should decrease with experience.
Blunt dissection using scissors or forceps is completed with tips down and by spreading the jaws. Gentle sharp dissection along midline with a scalpel blade is also acceptable. Care should be taken not to go through too many layers at once.
Tissue forceps are used to grasp the linea alba—use a #11 or #15 blade on a #3 handle, with the tip inserted on midline 1–2 cm caudal from the forceps. Make sure the scalpel blade is turned upside down to cut with the cutting surface pointing at the ceiling. Once a small incision has been made, then a pinky finger can be introduced to palpate the underside of the linea alba to check for adhesions. If no linea alba or body wall adhesions are identified, then the incision can be safely extended.
The linea incision can be extended by placing a thumb, forceps and scalpel into the incision to push cut with the blade, while using the forceps to lift up the body wall to protect the viscera (TS, AA). Alternatively, the linea can be transected with scissors used in a push cut (none of the surgeons at UCVM do this, but it is considered acceptable technique by many surgeons. None of the surgeons here like to do this because scissors shear tissue, while a sharp blade transects in a less traumatic manner).
Once the linea incision is completely open, then have a cursory look at the organs in the abdomen. Is there anything amiss before you start?
6. Locating the Uterus and Ovaries
Locate the first uterine horn. If you are left-handed, then stand on the patient's left, and if you are right-handed, then stand on the patient's right. You can locate either uterine horn first.
To locate the horn:
1.
The uterus will be between the bladder and the colon. Find the bladder first, and then locate the uterus underneath it. Follow the uterus along either horn to the ovary (TS).
OR
2.
Use right index finger at the caudal aspect of the incision, and put tip of finger under the body wall. Follow the wall down and progress dorsal and cranial. Palpate the caudal pole of the kidney and try to feel the ovary. If not located, then rotate the finger medially and try to hook the ovary or uterine horn and bring it medially. Gently lift the tissue out of the abdominal incision, and check to see if you have the uterus elevated (CTH, AA).
No spay hook required (all agreed). If using a spay hook, then hold up the midpoint of the abdominal wall opening, turn the hook against the abdominal wall, and slide it down into the abdomen. Rotate the hook 180 degrees and sweep medially. Pull up gently. Evaluate the tissues that you bring to the surface on the hook to check if the uterus is present.
To ensure that you have the uterus, follow the ovary and uterine horn to the bifurcation and then up to the other ovary. Look at the uterine arteries—they will be parallel to the uterine horn. The mesenteric arteries will be branching and perpendicular. Do not confuse the two.
Once you have the first uterine horn and ovary, place a hemostat on the proper ligament of the ovary. This is a tough ligament between the caudal aspect of the ovary and the tip of the uterine horn. This is to avoid losing it into the abdomen again.
7. Rupturing the Suspensory Ligament
With your right hand on the ovary and pedicle (and hemostat on the proper ligament), holding the ovary in your palm and the pedicle (vascular supply to the ovary) behind it, apply caudal medial traction to the suspensory ligament with your left hand to break it (AA, CTH). The suspensory ligament anchors the cranial aspect of the ovary to the body wall in the region of the diaphragm attachment near the last rib.
Alternatively, place two curved mosquitoes across the suspensory ligament just proximal to its attachment to the ovary. The tips of the hemostat are placed perpendicular to the suspensory ligament, the handles are angled away from each other, and the curved parts of the hemostats are placed back to back. Bringing the hemostat handles together will tear the suspensory tissue between the tips of the hemostats apart (TS).
Alternative methods for breaking the suspensory include strumming the ligament with a thumb and index finger, using a gauze 4×4 to gently pull it, or, finally, visualizing it and transecting it (especially if in heat, pregnant, or overweight and it is tough to break down).
Once the suspensory ligament is ruptured, then the ovary and the uterine horn can be retracted further out of the incision. Partial tearing of the suspensory ligament may still allow for exteriorization of the ovary and pedicle, but the remaining ligament will interfere with ligation efforts.
8. Clamping the Ovarian Pedicle
Using metzenbaum scissors or a mosquito forceps, blunt dissect a hole in the broad ligament (mesovarium) parallel to the ovarian vasculature and caudal to the ovarian pedicle (blood supply to the ovary). You can also use a gauze square to gently pull at ligament to create hole (but not for animals in heat or Collie/Sheltie breeds—may tear a vessel accidentally). Note this can be associated with more risk of hemorrhage and is not recommended to start with.
1.
Place three hemostats at 90-degree angle to the pedicle/uterine horn and perpendicular to your incision (this will help keep the pedicle lifted out of the abdomen as you work). Orient with handles in opposite orientation as you stack on top of one another. Triple Clamp Technique (TS/AA/CTH). Ideally, there is a few mm of tissue between each pair of clamps.
OR
2.
For smaller patients, where it is difficult to place three clamps below the ovary on the ovarian pedicle, place two clamps on ovarian pedicle perpendicular to the incision, as you normally would (with the handles in opposite orientation as you stack them). The third clamp can be placed across the vasculature and the proper ligament (between the ovary and the uterine horn). This is called the Modified Triple Clamp Technique (AA, TS).
9. Ligating the Ovarian Pedicle
Remove the most proximal clamp (the one closest to the body wall and furthest from the ovary), and place a circumferential ligature of appropriate size in the groove/crush site. Then place a transfixation suture more distal (closer to the ovary) to this suture to avoid ligature slippage. (You may choose to do multiple ligations if the patient is obese).
Tie the first throw as a square knot, preferably with three throws on top. Ensure each throw is firmly tightened (TS). When transfixing, tie a square knot after first partial circumferential, and then go around entire pedicle before tying another square plus three on top (five throws total). May use a surgeon's throw as first throw if the patient is obese and especially if the first throw of the square knot is slipping. You should still place an additional four throws on top of that initial throw (AA).
Always double ligate—but you may not always double transfix. You may only do one circumferential ligation followed by a transfixation. (Transfixation implies taking a bit of tissue and tying a knot around half of the diameter of the tissue before wrapping around the entire pedicle and tying. A circumferential ligation is encircling the whole pedicle at once and tying a knot without passing a needle through the actual tissue mass itself).
10. Transecting the Ovarian Pedicle
Take a #15 blade or your metzenbaum scissors and transect between the remaining two clamps. Cut along the edge of the clamp closest to the body wall. Palpate your ovary before you start cutting to make sure you will not be transecting part of it—especially if using the modified three-clamp technique (you don't want to leave ovarian tissue in the patient). Use tissue forceps to grasp the pedicle lightly (not across any vessels). While holding the pedicle with the forceps, carefully remove the clamp that is closest to the body wall. Check for hemorrhage. If none is noted, then return the pedicle to the abdomen and watch it for any signs of hemorrhage. If hemorrhage is noted, then gently grasp the pedicle or vessel with a hemostat or forceps and apply another transfixation. Be careful not to grab a lot of tissue at this point—this is often when ureters get accidentally ligated!
If the modified triple clamp technique is used, take a #15 blade and cut along the edge of the remaining clamp (i.e., between the clamp and the ovary). To be certain that the ovary has been removed in its entirety, open the ovarian bursa and examine the ovary. Use tissue forceps to grasp the pedicle lightly (not across any vessels). While grasping the ovarian pedicle with forceps, carefully remove the clamp that is closest to the body wall. Check for hemorrhage. If none is noted, then return the pedicle to the abdomen and watch it for any signs of hemorrhage. If hemorrhage is noted, then apply another transfixation suture.
11. Locating the Second Ovary
Keeping a hemostat across the distal end of the ovarian pedicle to prevent backflow and bleeding through the vasculature, follow the uterine horn caudally while manually (and gently) tearing the broad ligament as you go. If the animal is pregnant or in heat, then clamp and ligate larger broad ligament vessels as needed. Follow the uterine horn to the bifurcation and find the other uterine horn. Follow it cranially to the other ovary.
12. Transecting the Second Ovarian Pedicle
This is performed in a manner similar to that which was used for the first ovarian pedicle. You can also manually tear through the broad ligament on the uterus as described previously.
13. Locating the Cervix
Take both uterine horns (with attached ovaries) and reflect caudally.
The cervix is identified as a firm, thickened region that can be palpated just caudal to the uterine body (AA).
OR
Apply tension cranially on the uterus to bring the cervix cranial to the bladder (TS).
14. Clamping the Uterine Body
Cranial to the cervix but caudal to the uterine bifurcation, place three clamps at 90 degrees to the uterine body and the abdominal incision. Sometimes you only have room for two. (With cats or dogs that are in heat and may have a friable uterus, we often don't clamp at all as this might sever the uterus accidentally).
15. Ligating the Uterine Body
Remove the clamp closest to the cervix—this is called “flashing” the tissue. Place a circumferential ligation in this site where the forceps have crushed the tissue. Use a square knot and then add three throws (preferably not a surgeon's knot) (TS). May decide to use a surgeon's if the animal is obese or if first throw security is questionable (AA, TS).
Place a transfixation suture proximal to the circumferential (closer to the bifurcation). In dogs with large vessels, ligate the uterine arteries separately and transfix to the uterine body—this ensures hemostasis.
16. Transecting the Uterine Body
Transect between the two remaining clamps. Cut along the caudal aspect of the clamp that is closest to the cervix.
Using tissue forceps, grasp the uterine pedicle/cervix and carefully remove the remaining clamp. Assess for hemorrhage. If no hemorrhage, then release into the abdomen. If there is hemorrhage, then place a second transfixation ligature to stop the bleeding.
17. Checking the Ovarian Pedicles and Uterine Stump for Hemorrhage
To check the left ovarian pedicle, use the mesocolon as an abdominal retractor to pull all the abdominal contents to the animal's right and look at the pedicle. Observe for hemorrhage.
To check the right ovarian pedicle, use the mesoduodenum as an abdominal retractor to pull all the abdominal contents to the animal's left and look at the pedicle. Observe for hemorrhage.
If hemorrhage is noted, then retrieve the pedicle with tissue forceps. Add another ligature to stop the bleeding. As noted above, take care not to grab excessive amounts of tissue because this is when ureters get accidentally ligated!
Check the uterine stump between the colon and the bladder. Observe for hemorrhage.
18. Prepare to Close the Abdomen
Count gauze (if you counted before commencing the procedure). If all are present, then you can close the patient. If not, then count again and search the abdomen (AA, CTH). If you haven't counted gauze, a thorough evaluation of the abdomen to ensure no gauze has been left behind is conducted (TS).
19. Abdominal Wall Closure
In general, use:
2–0 or 3–0 PDS if <10 kg
2–0 PDS if 10–20 kg (or cat)
0 PDS if 20–30 kg
1 PDS if >30 kg
SQ closure will be down one size from the linea suture size (TS). SQ closure is done with 3–0 or 4–0 Monocryl (AA).
Start at the distal end of the incision. If right-handed, then suture right to left; if left-handed, then suture left to right.
Take bites 5–10 mm from edge of linea and incorporate only the ventral rectus fascia. Take bites 5–10 mm apart.
Use a simple continuous in the linea alba. Start with six throws and end with eight (AA) OR start with a square knot, plus one more throw for security, plus one for PDS, and one more for starting a continuous—five throws (TS). Finish your continuous pattern with one more throw than you started with—therefore, six throws (TS). Cut suture tails 2–3 mm length.
Apply a subcutaneous layer next. Bury the knot. (Remember: deep to superficial and superficial back to deep). Use 3–0 or 4–0 in a cat, 3–0 in a dog.
Tack down to the ventral rectus fascia every few bites to close the dead space—especially if a lot of dead space was created (AA, TS).
Skin closure next. Intradermal is preferred, especially for animals going back to rescue societies that may not be monitored as closely (AA, CTH). Use a buried knot with 3–0 or 4–0 Monocryl.
Appendix 2: Checklist for Assessment of OVH Procedure Developed by CTA
Unsatisfactory
Borderline
Good
Excellent
The candidate performs the following (mark box with an X):
Fails to perform skill or task on own
Able to perform skill or task but requires significant coaching
Able to perform skill with minimal guidance
Able to perform skill competently first time without guidance
Making the incision:
1. Incision is in correct location (from umbilicus in dog, 1–2 cm caudal to umbilicus in cat)
□
□
□
□
2. Incision is of adequate length for the procedure (2–10 cm)
□
□
□
□
3. Makes smooth incision through skin (using proper fingertip grip)
□
□
□
□
4. Provides adequate hemostasis
□
□
□
□
5. Proper sharp and blunt dissection to locate linea alba; not too lateral; avoids creating dead space
□
□
□
□
6. Makes safe abdominal approach; uses reverse stab with scalpel and tents sufficiently to avoid injury to patient
□
□
□
□
7. Extends the incision safely; stays on midline, keeps straight incision, does not go too deep
□
□
□
□
Performing the OVH—locating the first ovarian pedicle
8. Safely and efficiently locates the first uterine horn and ovary
□
□
□
□
9. Identifies ovarian pedicle
□
□
□
□
10. Proper ligament is identified and grasped with hemostat
□
□
□
□
11. Suspensory ligament is broken down easily (to allow proper exteriorization of the first ovarian pedicle)
□
□
□
□
12. Window is safely made in the broad ligament (mesovarium) at correct location
□
□
□
□
13. Demonstrates gentle tissue handling
□
□
□
□
14. Applies proper three-clamp or modified three-clamp technique
□
□
□
□
15. Places secure circumferential ligation in the groove created by the most proximal clamp (closest to body wall and furthest from ovary)
□
□
□
□
16. Places secure transfixation suture above the circumferential ligation (closer to ovary) using the correct number of throws
□
□
□
□
Performing the OVH—locating the second ovarian pedicle
17. Locates the second ovary and uterine horn using safe, efficient technique
□
□
□
□
18. Identifies the second ovarian pedicle
□
□
□
□
19. Proper ligament is identified (adjacent to second ovary) and grasped with hemostat
□
□
□
□
20. Suspensory ligament is broken down easily (to allow proper exteriorization of the second ovarian pedicle)
□
□
□
□
21. Window is safely made in the broad ligament (mesovarium) at correct location near second ovary
□
□
□
□
22. Demonstrates gentle tissue handling of second ovarian pedicle
□
□
□
□
23. Applies proper three-clamp or modified three-clamp technique
□
□
□
□
24. Places secure circumferential ligation in the groove created by the most proximal clamp (closest to body wall and furthest from ovary)
□
□
□
□
25. Places secure transfixation suture above the circumferential ligation (closer to the ovary) using correct number of throws
□
□
□
□
Performing the OVH—broad ligament
26. Breaks down the broad ligament on both sides using careful hemostasis and tissue handling
□
□
□
□
Performing the OVH—uterine body
27. Efficiently and correctly locates the cervix
□
□
□
□
28. Cranial to the cervix but caudal to the uterine bifurcation, places three clamps using proper technique
□
□
□
□
29. Places secure circumferential ligation at location where hemostat closest to cervix has flashed the tissue
□
□
□
□
30. Places secure transfixation suture (may also individually ligate the uterine arteries and transfix to the uterine body in a large dog)
□
□
□
□
Performing the OVH—checking for hemostasis
31. Checks left ovarian pedicle by using mesocolon as abdominal retractor to pull all contents to right
□
□
□
□
32. Checks right ovarian pedicle by using mesoduodenum as abdominal retractor to pull all the contents left
□
□
□
□
33. Checks uterine pedicle between colon and bladder
□
□
□
□
34. Handles pedicle hemorrhage with ease; finds pedicle, retrieves with forceps, and ligates
□
□
□
□
Closing the body wall
35. Uses appropriate suture size for size of animal (2–0 or 3–0 for <10 kg, 2–0 if 10–20 kg, 0 if 20–30 kg, 1 if >30)
□
□
□
□
36. Adequate bite sizes (5–10 mm)
□
□
□
□
37. Adequate bite spacing (5–10 mm)
□
□
□
□
38. Knot started at end of incision (no gap)
□
□
□
□
39. Linea layer incorporates bites in ventral rectus fascia only
□
□
□
□
40. Knot ended at end of incision (no gap)
□
□
□
□
Closing the subcutaneous
41. Correctly buries knot at start
□
□
□
□
42. Simple continuous pattern placed correctly
□
□
□
□
43. Correctly ends with buried knot
□
□
□
□
Closing the skin
44. Correctly buries knot at start
□
□
□
□
45. Proper tension upon completion of closure, apposed but not too tight
□
□
□
□
46. Correctly buries knot at end (no need for tissue glue)
□
□
□
□
Further comments about student performance
Global score for overall performance
1
2
3
4
5
6
Inferior
Poor
Borderline unsatisfactory
Borderline satisfactory
Good
Excellent
Time of surgery—total (draping to skin closure):
Signature of examiner:
Appendix 3: Rank Ordering of Participants, Ratings on Each Checklist Item, and Amount of Faculty Assistance Required
Biography: Emma K. Read, DVM, MVSc, DACVS, is Associate Dean, Academic (DVM Program), University of Calgary Faculty of Veterinary Medicine, Teaching Research and Wellness (TRW) Building, 2nd Floor, 3280 Hospital Drive NW, Calgary, AB T2N 4Z6 Canada. Email: [email protected]. Her research interests include clinical skills, assessment, clinical reasoning, and simulation.
Biography: Andrea Vallevand, MSc, PhD, is Project Coordinator for Educational Technology, University of Calgary Faculty of Veterinary Medicine, 3280 Hospital Drive NW, Calgary, AB T2N 4Z6 Canada. Her research interests include simulation and assessment.
Biography: Robin M. Farrell, DVM, is Assistant Professor, Ross University School of Veterinary Medicine, PO Box 334, Basseterre, St. Kitts, West Indies. Her research interests include teaching, learning, and assessment using simulation.
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
Purchase this article, unlimited access to all formats
Checkout
Restore your content access
Note: This functionality works only for purchases done as a guest. If you already have an account, log in to access the content to which you are entitled.
