Materials and Methods

Questionnaire development—Each of the authors (LNH, DDS, and LKL) independently generated a list of skills necessary to complete the most common surgical procedures performed in small animal,³ bovine,⁴ and equine⁵ practice. Independently generated lists were compared and consolidated to create a draft version of the survey. The draft version of the survey was completed and reviewed by small animal, large animal, and equine private practitioners and board-certified veterinary surgeons working in academic and private practice. Five experienced practitioners (2 large animal or equine and 3 small animal) working in the central Ohio region who were graduates of The Ohio State University College of Veterinary Medicine reviewed the draft version of the survey. Draft surveys were also reviewed by 4 American College of Veterinary Surgeons diplomates (3 small animal and 1 large animal and equine) who were active surgery instructors in the teaching program at the Colorado State University College of Veterinary Medicine. Input from these veterinarians was used to improve clarity of survey instructions and determine whether additional core surgical skills categories should be included in the survey.

Likert-type items comprised the main portion of the questionnaire. Participants were asked to select the proficiency (1 = no proficiency expected, 2 = some proficiency expected but may require supervision, 3 = good proficiency expected with minimal supervision required, or 4 = complete proficiency expected with no supervision required) and frequency of use (1 = seldom [0% to 25% of time], 2 = occasional [> 25% to 50% of time], 3 = frequent [> 50% to 75% of time], and 4 = routine [> 75% to 100% of time]) expected of entry-level veterinarians in regard to 26 core surgical skills. An open-ended question was also included to identify core surgical skills categories not included in the survey that respondents considered important.

Items were grouped into subcategories of skills that participants were asked to rate in regard to expected proficiency and frequency of use by entry-level veterinarians. These subcategories included skills pertaining to the surgical suite, surgeon, and preparation of patients for surgery; use of basic surgical instruments and suture and handling of soft tissues; orthopedics; and use of electrocautery and laser instruments. Additionally, demographic questions were included to obtain information regarding respondent gender, year of veterinary school graduation (categorized as time since veterinary school graduation [≤ 10, 11 to 20, or > 20 years]), practice type (companion animal, food animal, equine, mixed animal, or other), practice location (western United States, midwestern United States, northeastern United States, southern United States, or other), practice setting (urban, suburban, or rural), number of veterinarians employed in the practice (part-time veterinarians were counted as 0.5 veterinarian), number of surgical procedures performed by the respondent per week (< 10 or ≥ 10), and number of new graduates the respondent had mentored during the past 5 years (≤ 5, 6 to 20, or > 20). The present study was exempted from review and approved by The Ohio State University Institutional Review Board.

Sampling frame—The sampling frame used in the present study was a simple random sample of veterinarians selected from the membership database of the AVMA. The only information available from the database was the address of each member; no information regarding practice type was available. Selection of the sample was performed by AVMA staff, and the names and addresses of selected veterinarians were provided to the authors. By use of an estimate that 50% of respondents would choose a particular response (the most conservative estimate for calculation of sample size) with a 95% confidence level and a margin of error ≤ 3.6%,¹² a sample size of 750 participants was calculated for inclusion in the study reported here.

Study design—A standard survey method¹³ was used. The survey and all correspondence with selected veterinarians were distributed via regular mail. A letter explaining the purpose of the survey was sent to all selected veterinarians on June 1, 2009, before the survey was distributed. The survey was mailed to selected veterinarians on June 5, 2009. Survey participants were assured of anonymity, and participation in the survey was voluntary. Selected veterinarians were asked to contact the authors by phone or e-mail if they did not understand or had specific questions regarding the survey. Responses were recorded by manually entering survey data into a standard database program.^a A postcard reminder was sent to nonresponders 2 weeks after the initial mailing, and a second survey was sent to them 10 days after the postcard reminder. Survey responses were accepted through November 1, 2009.

Statistical analysis—Proportions were calculated for responses that consisted of categorical data. For Likert-type items, proportions, means, and 95% confidence intervals were calculated for each response. For expected proficiency questions for which ≥ 40% of respondents chose 1 of the 2 lowest expected proficiency ratings (expected proficiency rating, 1 or 2), Likert-type items were dichotomized into high expected proficiency (expected proficiency rating, 3 or 4) and low expected proficiency (expected proficiency rating, 1 or 2) categories. For expected frequency of use questions for which ≥ 40% of respondents chose 1 of the 2 lowest expected frequency of use ratings (expected frequency of use rating, 1 or 2), Likert-type items were dichotomized into high expected frequency of use (expected frequency of use rating, 3 or 4) and low expected frequency of use (expected frequency of use rating, 1 or 2) categories.

Multivariate logistic regression was used to identify demographic variables associated with the odds of items being assigned low expected frequency of use or proficiency ratings. Variables included in the logistic regression included respondent gender, time since graduation from veterinary school, practice type (restricted to companion animal vs food animal, equine, or mixed-animal practice), geographic practice location, practice setting, number of veterinarians employed in the practice, number of surgical procedures performed per week, and number of new graduates mentored during the past 5 years. Variables with values of P ≤ 0.25 in univariate analyses were included in multivariate logistic regression analysis. Variables were removed from the full multivariate model on the basis of results of the Wald test. Biologically meaningful interactions among variables were tested for inclusion in the model in a similar manner. For all analyses, values of P ≤ 0.05 were considered significant. Standard statistical software^b was used for all analyses.

Results

Response rate—Of the 750 AVMA members surveyed, 387 completed and returned the survey (52% response rate). Demographic characteristics of survey respondents were summarized (Table 1).

Expected proficiency ratings—For each of the 26 core surgical skill items, ratings for expected proficiency of entry-level veterinarians were summarized (Table 2). Five items (creation of secure square knots by use of hand ties [1-handed tie technique], use of basic orthopedic instruments, manual reduction of bone fragments and use of fixation techniques to stabilize fragments, use of electrosurgical [electrocoagulation or radiosurgery] instruments to incise tissue or assist in achieving hemostasis, and use of a laser to incise tissue or assist in achieving hemostasis) were assigned a low (1 or 2) expected proficiency rating by ≥ 60% of respondents. Results of multivariate logistic regression identifying demographic variables associated with assignment of low expected proficiency ratings were summarized (Table 3). In regard to the creation of secure square knots by use of hand ties [1-handed tie technique], male respondents were 1.7 times as likely as female respondents to assign a low expected proficiency rating, veterinarians who had mentored ≤ 5 new graduates in the past 5 years were 2.2 times as likely as veterinarians who had mentored ≥ 6 new graduates in the past 5 years to assign a low rating, and veterinarians who had graduated from veterinary school 11 to 20 years prior to completing the survey were 2.0 times as likely as veterinarians who had graduated ≤ 10 years before completing the survey to assign a low rating. In regard to use of basic orthopedic instruments, females were 2.8 times as likely as males to assign a low expected proficiency rating, veterinarians performing < 10 surgical procedures/wk were 1.7 times as likely as those performing ≥ 10 procedures/wk to assign a low rating, and veterinarians in the northeastern United States were 4.7 times as likely as those in the western United States to assign a low rating. In regard to manual reduction of bone fragments and use of fixation techniques to stabilize fragments, female respondents were 2.4 times as likely as male respondents to assign a low expected proficiency rating, veterinarians performing < 10 procedures/wk were 1.8 times as likely as those performing ≥ 10 procedures/wk to assign a low rating, and companion animal veterinarians were 2.4 times as likely as food animal, equine, or mixed-animal practitioners to assign a low rating. In regard to the use of electrosurgical (electrocoagulation or radiosurgery) instruments to incise tissue or assist in achieving hemostasis, female respondents were 2.2 times as likely as male respondents to assign a low expected proficiency rating. In regard to the use of a laser to incise tissue or assist in achieving hemostasis, no differences were detected among demographic groups.

Expected frequency of use ratings—For each of the 26 core surgical skill items, ratings for expected frequency of use by entry-level veterinarians were summarized (Table 4). Seven items (creation of secure square knots by use of the 2-handed tie technique, creation of secure square knots by use of the 1-handed tie technique, placing and securing indwelling drains and tubing, application of bandage materials to support limbs or joints, use of basic orthopedic instruments, manual reduction of bone fragments and use of fixation techniques to stabilize fragments, use of electrosurgical [electrocoagulation or radiosurgery] instruments to incise tissue or assist in achieving hemostasis, and use of a laser to incise tissue or assist in achieving hemostasis) were assigned a low (1 or 2) expected frequency of use rating by ≥ 60% of respondents. Results of multivariate logistic regression identifying demographic variables associated with assignment of low expected frequency of use ratings were summarized (Table 5). In regard to the creation of secure square knots by use of hand ties (2-handed tie technique), veterinarians who had mentored ≤ 5 new graduates in the past 5 years were 2.1 times as likely as those who had mentored ≥ 6 new graduates in the past 5 years to assign a low expected frequency of use rating. Similarly, in regard to the creation of secure square knots by use of hand ties (1-handed tie technique), veterinarians who had mentored ≤ 5 new graduates in the past 5 years were 2.3 times as likely as veterinarians who had mentored ≥ 6 new graduates in the past 5 years to assign a low expected frequency of use rating. In regard to placing and securing indwelling drains and tubing, food animal, equine, or mixed-animal veterinarians were 2.0 times as likely as companion animal veterinarians to assign a low expected frequency of use rating. In regard to the use of basic orthopedic instruments, female respondents were 4.1 times as likely as male respondents to assign a low expected frequency of use rating, and veterinarians in the midwestern United States were 2.5 times as veterinarians in the western United States to assign a low rating. In regard to manual reduction of bone fragments and use of fixation techniques to stabilize fragments, female respondents were 3.9 times as likely as male respondents to assign a low expected frequency of use rating. In regard to the use of electrosurgical (electrocoagulation or radiosurgery) instruments to incise tissue or assist in achieving hemostasis, female respondents were 1.7 times as likely as male respondents to assign a low expected frequency of use rating, and food animal, equine, or mixed-animal practitioners were 2.4 times as likely as companion animal practitioners to assign a low rating. In regard to application of bandage materials to support limbs or joints and the use of a laser to incise tissue or assist in achieving hemostasis, no differences were detected among demographic groups.

Open-ended questions—Fifty-four write-in comments were received for “Other skills not included in the list of 26,” and 77 write-in comments were received for “Please use box to provide any information you feel is relevant to the study.” No specific additional core surgical skills were identified by respondents who provided these comments; however, common comments from respondents included requests for veterinary schools to improve efficiency and abilities of new graduates in regard to performance of common surgical procedures, such as ovariohysterectomy, castration, declawing, cystotomy, mass removal, laceration repair, and basic wound management; management of intraoperative hemorrhage; management of complications after surgery; use of catheterization techniques; appropriate selection of suture; use of hand skills for the performance of dental procedures; performance of dental extraction techniques; staff and client communication skills; and general case management skills (such as preparing cost estimates and working within a budget).

Discussion

The primary purpose of the present study was to develop a ranked list of core surgical skills on the basis of results of a survey of general veterinary practitioners listed in the AVMA membership database. Evaluation of survey data revealed a broad consensus among general practitioners in regard to expected frequency of use of core surgical skills and expected proficiency of entry-level veterinarians in regard to those skills. Consensus was independent of practice type, setting, or geographic location; time since graduation; or gender of respondents.

Another objective of the present study was to obtain survey data from general veterinary practitioners that could be compared with survey data obtained from American College of Veterinary Surgeons diplomates. Results of the present study were similar to those of another study¹¹ in which American College of Veterinary Surgeons diplomates were surveyed to rate expected proficiency and frequency of use of surgical skills in an entry-level veterinary practice; these similarities were independent of most respondent demographic variables.

We believe results of the present survey and those of the other survey¹¹ of American College of Veterinary Surgeons diplomates may facilitate identification of which core skills are necessary and important for inclusion in surgical skills curricula for veterinary students and may help veterinary schools prioritize teaching of those skills. Because there were few differences detected in what constitutes core surgical skills among general practitioners (present study) and veterinary surgical specialists,¹¹ regardless of the animal species with which survey respondents worked, perhaps veterinary faculty can lower the emphasis on species differences during instruction of surgical skills. Even during later stages of a curriculum (when species-specific skills are more likely to be taught), veterinary schools should be able to provide instruction that is relevant to students with an interest in any species practice emphasis. Furthermore, results of the present study and the other study¹¹ could serve as justification for veterinary faculty to continue teaching the core skills that are currently included in surgical curricula. Results of the present study may help inform surgical specialists working in academia about the skills general practitioners perceive as important for veterinary students to have when they graduate. Likewise, results of the other study¹¹ may help inform veterinarians in general practice about the skills American College of Veterinary Surgeons diplomates perceive as important for surgical practice. As veterinary educators begin to assess student proficiency and ensure graduates have competency in core skills, practitioners may develop an improved understanding of what skills entry-level veterinarians have and how much supervision they require or expect during performance of surgery. Veterinary students may beneft by reviewing the list of core skills identified in the present study and by monitoring their own progress in proficiency with these skills. Results of the present study may also help students better understand expectations and outcome objectives of their surgical training programs.

Broad agreement was found in the present study among survey respondents regarding expected proficiency and frequency of use ratings for most of the core surgical skills. However, some differences were detected in ratings among respondents of various demographic groups. For example, male respondents were 1.7 times as likely as female respondents to assign a low expected proficiency rating for 1-handed knot tying skills by entry-level veterinarians, and respondents who had mentored ≤ 5 new graduates in the past 5 years were 2.2 times as likely to assign a low expected proficiency rating for that item, compared with respondents who had mentored ≥ 6 new graduates in the past 5 years. Similarly, the greater the time since graduation, the more likely it was that creation of secure square knots with the 1-handed technique received a low expected proficiency rating. In addition, the fewer new graduates who respondents had mentored in the past 5 years, the more likely it was that the creation of secure square knots with the 2- and 1-handed techniques received a low expected frequency of use rating. There seemed to be a distinct difference in expected proficiency ratings for the performance of 2-handed versus 1-handed ties. Greater than 60% of respondents rated the creation of secure square knots with the 2-handed tie technique as having high expected proficiency (entry-level veterinarians expected to have complete or good skills). In contrast, > 60% of respondents rated the creation of secure square knots by use of 1-handed ties as having low expected proficiency (entry-level veterinarians expected to have some or no skills). Not all surgical procedures require the creation of secure square knots by use of 1-handed ties for successful completion, which may be why most respondents indicated that entry-level veterinarians should be able to consistently create secure square knots with little supervision using 2-handed ties but not using 1-handed ties. The majority (98.4%) of respondents assigned a high rating for expected proficiency in the creation of secure square knots with instruments; there were no differences in rating of this item with regard to demographics, which could explain the low rating for expected proficiency in the creation of secure square knots by use of hand ties (2- and 1-handed ties). Experienced veterinarians may mentor new graduates more frequently than do less experienced veterinarians; those more experienced veterinarians often value economy and efficiency in performance of surgery. This could account for the low expected proficiency rating for the creation of secure square knots by use of 1-handed ties by respondents for whom a long time had passed since graduation. It would be interesting to determine the combined impacts of gender, amount of experience, and number of veterinary students mentored on ratings received for this item. It might be that highly experienced male veterinarians who mentor high numbers of new graduates discourage the creation of secure square knots by use of 1-handed ties in favor of the creation of square knots by use of instruments and 2-handed ties.

Most respondents assigned low expected proficiency and frequency of use ratings for orthopedic skills (use of basic orthopedic instruments and fracture fixation skills [manual reduction of bone fragments and use of fixation techniques to stabilize fragments]). The low ratings for these items probably reflect an expectation that entry-level veterinarians will use these skills seldom or only occasionally in practice. Female respondents were more likely to assign low ratings for orthopedic instrument handling and fracture fixation skills (2.8 and 2.4 times as likely as males, respectively, for expected proficiency ratings and 4.1 and 3.9 times as likely as males, respectively, for expected frequency of use ratings). This comparison may be worth reevaluating at a later time, given that most veterinary graduates entering practice are female and there are an increasing number of veterinary employers who are female.⁸ For both of the orthopedic skills items, those veterinarians performing < 10 surgical procedures/wk were almost twice as likely to assign a low expected proficiency rating as those performing ≥ 10 surgical procedures/wk, and companion animal veterinarians were 2.4 times as likely to assign a low expected proficiency rating as were food animal, equine, or mixed-animal practitioners.

Ostensibly, companion animal veterinarians are more likely to expect entry-level veterinarians to refer animals with orthopedic problems to specialists than are veterinarians in a rural practice. Large companion animal practices with multiple veterinarians are typically located in metropolitan areas and tend to have more experienced veterinarians available for the performance of specialized surgical procedures (eg, orthopedic procedures), compared with practices in rural areas. Therefore, companion animal veterinarians may not expect that entry-level veterinarians would perform these types of surgery as frequently as rural-practice veterinarians expect they would.¹⁴ In contrast, food animal, equine, or mixed-animal practice veterinarians may expect new graduates to perform some orthopedic surgeries because many veterinarians working in practices outside urban centers may not have as many opportunities or requests from owners to refer patients with orthopedic problems.

In regard to the use of electrosurgical instruments, female respondents were 2.2 times as likely to assign a low expected proficiency rating and 1.7 times as likely to assign a low expected frequency of use rating as male respondents. Food animal, equine, or mixed-animal veterinarians were 2.4 times as likely as companion animal veterinarians to assign a low expected frequency of use rating for this item. However, there was no difference in how respondents who graduated in various years or who mentored various numbers of new graduates rated expected proficiency or frequency of use for this item. This finding may be attributable to low availability of these types of instruments in general veterinary practices, yet it seems likely that recent graduates would be more familiar with the use of electrosurgical instruments than would veterinarians who graduated longer ago and therefore would be more willing to use these instruments in practice. However, it was difficult to formulate a reasonable explanation for the discrepancy in results between genders of respondents because there were no differences detected in expected proficiency ratings for this item on the basis of practice type. It is possible that because surgical specialty practices are increasingly available in urban areas and there are many small animal–exclusive practitioners,¹⁰ having electrosurgical, electrocoagulation, and electroincision devices (including recently developed tissue fusion devices) available may not be considered a priority in general veterinary practices.

It was not surprising that food animal and equine practitioners were twice as likely as companion animal practitioners to assign a low expected frequency of use rating for placing and securing indwelling drains and tubing. Large animal veterinarians may be less likely than small animal practitioners to treat patients having wounds with large dead space that require drainage. Many wounds in large animals are located on the distal aspects of limbs and are amenable to treatment with bandaging techniques.¹⁵ Many large animal veterinarians seem to think that if an appropriate bandage can be applied over a wound, there is a limited need for establishment of drainage; wounds on the main portion of the body (ie, wounds that are not on limbs) may be treated with placement of drains, but these wounds do not seem to occur as often as do wounds on the distal portions of the limbs.^c

Results of the present study in which AVMA member general practitioners were surveyed and results of another study¹¹ in which the American College of Veterinary Surgeons diplomates were surveyed were remarkably similar. No substantial differences were found in expected proficiency or frequency of use ratings for most core surgical skills included in these surveys. Findings of these studies were important because they provided a clear understanding of expected proficiencies and frequencies of use of core surgical skills for entry-level veterinarians. These findings also provided a basis for development of surgical curricula that meet expectations of employers and new graduates working in general and specialty practices. Although all core surgical skills included in the surveys of the present study and the other study¹¹ might be taught in veterinary curricula, we believe educators should design training programs that ensure graduating students have competency in the core surgical skills that were rated by ≥ 60% of respondents as having high expected proficiency (complete or good proficiency; 21/26 items). Skills assigned a low rating could be taught in elective courses to students with an interest in surgery or to those expecting to enter a postgraduate surgical training program.

Proficiency in many areas of knowledge and nontechnical and procedural skills, such as communication and interpersonal, physical examination, and problem-solving skills, are essential to becoming a competent entry-level surgeon.¹ However, the scope of the present study was limited to identification of core technical skills directly associated with performance of surgery. On the basis of responses to open-ended questions (which asked participants to list other skills not included in the survey), respondents seemed to have been confused regarding the definition of a core surgical skill in the survey. For example, some practitioners listed management of intraoperative hemorrhage, management of complications after surgery, use of catheterization techniques, appropriate selection of suture, use of hand skills for the performance of dental procedures, performance of dental extraction techniques, staff and client communication skills, and general case management skills (such as preparing cost estimates and working within a budget) as other skills not included 26 items in the survey. These are certainly important skills and areas knowledge, but were not within the scope of the survey. In retrospect, the authors could have more clearly defined core surgical skills by instructing participants to place an emphasis on the skills directly associated with performance of surgical procedures. This may have provided better focus in responses to open-ended questions, which may have allowed identification of additional skills not included in the list of 26 core surgical skill items in the survey.

An additional limitation of the present study was the inability to compare demographic characteristics (eg, practice type, age, or gender) of respondents with those of nonrespondents. This information was not available in the AVMA-provided list of veterinarians. However, the strong response rate to the survey gave us reasonable confidence that the results were representative of the general population of AVMA members.

Similar to the way core skills were allocated into discrete categories in the present study, distinct teaching modules could be developed for instruction of skills in each of these categories. This would allow for coordinated teaching of skills at various stages during veterinary education programs, the timing of which would depend on the curricular structure of each school. For example, modules in which the most basic core skills (eg, instrument handling and tissue dissection) were taught could be effectively introduced during first-year anatomy laboratory courses. If students could learn those skills as they started these laboratory courses, skills would be reinforced as students practiced dissection for identification of anatomic structures. Instead of having to work to correct inappropriate habits learned by trial and error at these laboratories, students could improve proficiency in core surgical skills in a coordinated and sequential manner. Introduction of surgical skills training early in a curriculum would allow students to repeatedly practice those skills during their veterinary education, which could improve retention and accelerate mastery of skills and ultimately improve competency of graduates.^16,17

Development of a list of core surgical skills is just one of the steps needed for identification of the curricular changes necessary to improve surgical competency of entry-level veterinarians. Educators should be compelled to use proven and effective methods to teach, assess proficiency with, and motivate students to learn these core skills. It remains to be determined how frequently performance of these skills should be repeated during a veterinary curriculum to ensure proficiency of students. A future survey of academic surgeons working at US veterinary institutions may provide useful information regarding the best time to introduce training of core surgical skills in a curriculum and the order in which instruction of skills should be introduced.