836

In response to a widening disparity between the num-ber of endoscopic gynecologic procedures that are per-formed and the extent of residency training, the Councilon Resident Education in Obstetrics and Gynecology de-veloped a set of training modules to help training pro-grams incorporate an effective laparoscopic educationalcurriculum.1 Cundiff2 reported the effectiveness of theseguidelines in a prospective study of residents who wereundergoing didactic and practical instruction. Despitethese global programs, however, many residents still re-port varied laparoscopic experiences and have varied lev-els of comfort with their own laparoscopic skills on

completion of their residency. In the experience re-ported by Cundiff, only 53% of residents reported com-petency by self-assessment (15% competency by facultyassessment) after incorporation of the guidelines, andsubjective improvement in laparoscopic performance byfaculty observations showed only upward, nonsignificanttrends. Still, many authors advocate animate and inani-mate laboratories as methods to improve skill precisionand to learn new techniques.3-5 The disparity betweenskills testing and competency partly results from a lack ofreproducible criteria by which to assess surgical skill.

Objective measures of surgical performance are difficultto quantify. Good operative skills require not only motor-skill finesse, but also knowledge, judgment, and experi-ence. Schueneman et al6 demonstrated that surgical skill(by faculty assessment) correlated, not with speed and pre-cision, but with the residents’ abilities to (1) use landmarksto create a mental 3-dimensional space, (2) interpret sen-sory cues based on previous experience, and (3) distin-guish essential details from nonessential details. Goff et al7

reported their initial experience in validating an objective

From the Department of Obstetrics and Gynecology, Division of Gyneco-logic Oncology, and The Southwestern Center for Minimally InvasiveSurgery, University of Texas, Southwestern Medical Center.Supported by the Association of Professors of Gynecology and Obstetrics(APGO)/Ortho-McNeil Faculty Development Award.Received for publication August 24, 2001; accepted October 26, 2001.Reprints not available from the authors.Copyright 2002, Mosby, Inc. All rights reserved.0002-9378/2002 $35.00 + 0 6/1/121254doi:10.1067/mob.2002.121254

EducationEffects of a laboratory-based skills curriculum on laparoscopic proficiency: A randomized trial

Robert L. Coleman, MD, and Carolyn Y. Muller, MD

Dallas, Tex

OBJECTIVE: The purpose of this study was to determine the effect and validity of an intensive laboratory-based laparoscopic skills training curriculum on operative proficiency in obstetrics and gynecology residents.STUDY DESIGN: This was a prospective, randomized, block-design trial of postgraduate year 3 and 4 resi-dents. The following study schedule was used: week 1 (prerandomization) included an orientation to studyobjectives, the administration of a laparoscopic experience questionnaire, timed video-laparoscopic drills (5 total), and the performance of a video-recorded laparoscopic partial salpingectomy; during weeks 2 and 3,the skills group residents repeated the laparoscopic drills 30 minutes daily for 10 days, and the control groupresidents had no formal practice sessions; during week 4, the week 1 evaluation was repeated. Operativeproficiency was quantified by the Global Skills Assessment Tool through blinded, independent scoring ofvideotapes.RESULTS: Twenty-six residents (skills group, 12; control group, 14) consented to the trial. Patient-related issues excluded 8 residents (30%). At week 1, no significant differences existed in previous laparoscopic experience, timed video skills, or resident operative proficiency (Global Skills Assessment Tool score) be-tween cohorts. At week 4, both groups significantly improved their timed drill test scores. The percent reduc-tion in time from baseline was of greater magnitude in the skills group versus control group (51% vs 18%, P < .0001). Laparoscopic performance also improved in both cohorts (P = .002). However, only the skillsgroup demonstrated significant intracohort improvement from baseline (mean, 4.9 points; P = .015; 95% CI, 1-7.5).CONCLUSION: A core curriculum of intensive video laparoscopic skills training improves not only technicalbut also operative performance among postgraduate year 3 and 4 residents. (Am J Obstet Gynecol2002;186:836-42.)

Key words: Video endoscopic, trainer, laparoscopic, proficiency, global surgical assessment tool

Volume 186, Number 4 Coleman and Muller 837Am J Obstet Gynecol

measuring tool for surgical technical skills. In this animal-based assessment model, they were able to demonstrateconstruct validity among junior and senior residents by ob-serving their performance of laparoscopic and open ab-dominal procedures. A follow-up study from theseinvestigators who used inanimate tools suggests this type ofassessment can be adequately accomplished in a morecost-efficient manner.8 Incorporation of concepts andtools (such as these) into a laparoscopic surgical curricu-lum would be critical to an expectation of improved surgi-cal skill. A recent randomized study by Scott et al9 suggeststhat intensive laboratory-based endoscopic video trainingcan enhance the skill performance of general surgical res-idents who perform laparoscopic cholecystectomy.

The purpose of this study was to evaluate whether a lab-oratory-based video laparoscopic skills curriculum couldtranslate into improved surgical skills in the operatingroom among our senior residents.

Material and methods

All postgraduate year (PGY) 3 and 4 obstetrics and gy-necology residents at the University of Texas, Southwest-ern Medical Center, who rotated through the generalgynecology rotations at Parkland Health and HospitalSystems were asked to participate in this prospective ran-domized trial. These residents were recruited during thefirst gynecology rotation of their respective year. At eachrotation block, there were potentially 6 residents (2 resi-dents from each of 3 gynecology services) who were eligi-ble to participate in the trial. The study was approved andmonitored by the University of Texas, Southwestern Med-ical Center Institutional Review Board, and confidential-ity of results was ensured.

Each study rotation was 26 days in duration and wasstructured in the following manner:

Week 1. A description of the program and its goals andobjectives was presented. Informed consent was ob-

Fig. Randomization schema and number of participants in each phase of the trial.

838 Coleman and Muller April 2002Am J Obstet Gynecol

tained, and participating residents were given a question-naire, the answers for which detailed their previous la-paroscopic experiences. They were asked to grade theirlaparoscopic skills in 5 areas (instrument handling,knowledge of instruments, use of assistants, knowledge ofspecific procedures, video hand-eye coordination) on a 5-point scale, with “1” representing uncomfortable, “3”representing average, and “5” representing proficiency. Abaseline, timed assessment of their performance in thepelvic trainer stations was recorded. We based our cur-riculum on 5 established laparoscopic drills (Table I):running string, block move, checkerboard, bean drop,and suture foam.10-13 The average time to complete 3 at-tempts in each skill station was used as the participant’sscore. To maintain objectivity, no other assessment of skillwas made for baseline testing.

Residents were then oriented to a step-by-step “task-to-be-performed” bilateral laparoscopic partial salpingec-tomy. Identified patients were previously scheduled forelective tubal ligation. The specific skills monitored byvideotape in each case were (1) visualized trocar place-ment, (2) the inspection and restoration of normalanatomy (if altered), (3) the identification of the fallop-ian tube to the fimbria, (4) the proper use of bipolar elec-trocautery to create a V across the tube joining in themesosalpinx, (5) the proper use of endoscopic scissors/forceps to excise and retrieve a portion of fallopian tube,(6) Endostitch (US Surgical Co, Norwalk, Conn) place-ment of the distal tube or mesosalpinx for hemostasis,and, (7) the removal of laparoscopic instruments undervisualization. Skills 4, 5, and 6 were specifically designedto assess the translation of skills mimicked in the videolaboratories. Skills 1, 2, 3, and 7 were skills that are taughtcustomarily to residents as a part of the surgical trainingduring the gynecology rotation. Qualified surgical assis-

tance was provided by the participating faculty (informedof the study objectives) or a PGY-4 resident. Assistantswere to respond only to the operating surgeon’s requestsand were not to volunteer input unless it was requested.Techniques for port placement, the ability to troubleshoot equipment mishaps, and closure techniques werenot assessed. After baseline data collection, residentswere randomized to either intensive skills training or noadditional skills training (control group). A block designwas used to ensure that equal numbers of PGY-3 and -4 residents were in each of the study arms.

Weeks 2 and 3. Residents who were randomized to the ex-perimental group underwent skills training in the video la-paroscopic training modules, as described, for at least 30minutes daily for 10 days. A journal was kept of their dailyprogress. No particular drill rotation was required, but allskills stations were to be practiced until the participant wascomfortable performing the task. Both groups underwentstandard didactic and operative instruction, within their gy-necologic rotation. Residents in the control arm had theopportunity to practice in the skills stations during theirsubsequent gynecologic rotations.

Week 4. The last week of the rotation repeated theevents of week 1 (questionnaire, a timed assessment ofthe 5 skills stations, and a videotape of a partial salp-ingectomy). The same faculty Attending Physician waspresent for both pretest and posttest surgical procedures.Residents were asked to report their laparoscopic experi-ence during the rotation and their comfort level with en-doscopic procedures.

All videos were randomly numbered and blinded toPGY level, randomization cohort, and sequence (first orsecond procedure). Two faculty members independentlyreviewed and quantified resident operative proficiencyusing the global surgical assessment tool (GSAT).14-17

Table I. Video-endoscopic skills stations used for evaluation

Drill maneuver Task Test

Running string Participant uses 2 laparoscopic graspers to run a Dominant and nondominant 140-cm string from 1 end to the other by grasping only hand-eye coordinationthe colored sections that are arranged at 12-cm intervals.

Block move Participant uses needle driver and a curved needle to Dominant hand pronation-supinationhook lift and individually move 4 blocks from 1 designated and 3-dimensional orientationarea to another 15 cm away. The nondominant hand controls the endoscope.

Checkerboard drill Participant picks up 16 metallic numbers/letters (1-8; A-H) Dominant and nondominant fine-motor individually with fine-tipped laparoscopic graspers from a flat grasping and orientationsurface and transports them to a flat arranged grid. One halfof the items were to be moved with the right hand, and one half of the items were to be moved with the left hand.

Bean drop Participant picks up 5 beans individually from a flat dish and Dominant hand fine-motor grasping,drops each of them through a 1-cm hole in an elevated cup that video orientation, and 3-dimensionalis 15 cm away. The dominant hand moves the beans, with the trainingnondominant hand holding the endoscope.

Suture foam Participant places a single intracorporeal knot that binds 2 Intracorporeal knot technique, pieces of foam using and Endostitch (US Surgical Co, dominant and nondominant handNorwalk, Conn) and fine graspers. coordination, and video orientation

Volume 186, Number 4 Coleman and Muller 839Am J Obstet Gynecol

Participants were graded on a 5-point scale for each ofthe 7 categories (respect for tissue, time and motion, in-strument handling, knowledge of instruments, flow of op-eration, use of assistants, and knowledge of specificprocedure). For example, in the time and motion cate-gory, a resident who demonstrated many unnecessarymoves was graded a “1”; efficient time and motion butwith some unnecessary moves was graded a “3”; and aclear economy of movement and maximal efficiency wasgraded a “5.” A total score of the possible 35 points wasrecorded from each reviewer. All scoring was performedafter the study’s conclusion to reduce bias that may havebeen related to individuals or problematic surgical pro-cedures, which may have been made known indirectly tothe reviewers. In our efforts to reduce bias, a subjective as-sessment from the attending surgeon who was involved inthe surgical procedures was not collected.

Statistical considerations. All comparisons were per-formed with the StatView 5.0 statistical program (SASInstitute, Cary, NC). Binomial data from the preques-tionnaire and postquestionnaire were analyzed with theFisher exact comparison. Ordinal categorical variables(such as comfort levels for individual tasks and global as-sessment scoring) were evaluated by Friedman’s 2-wayanalysis of variance. To test the hypothesis that no dif-ference existed in video task performance between thecohorts, a Wilcoxon rank-sum test was calculated. Re-peated-measures analysis of variance was used to com-pare timed skills tasks between cohorts at pretesting andposttesting. Paired intracohort continuous variables

that were not normally distributed were assessed withthe Mann-Whitney U test. A paired t test was used to as-sess the effect of individual normally distributed vari-ables within cohorts. Interrater reliability was assessedwith the κ statistic.

Results

The University of Texas, Southwestern Medical CenterObstetrics and Gynecology Residency program consists of14 residents at each PGY level. Overall, 26 of 28 potentialresident candidates agreed to participate in the trial. TheFig shows the randomization schema and the number ofresidents that completed the trial. At the trial’s conclu-sion, 18 (skills cohort, 11; control cohort, 7) of 26 resi-dents (69%) completed all phases of the protocol.Reasons for noncompletion of the trial were a lack of sur-gical case availability during the weeks intended (n = 6residents) and a cancellation of a planned surgical case(n = 2 residents). Complete pretest and posttest laparo-scopic drill data were collected on 20 residents (77%), 11residents in the skills cohort and 9 residents in the con-trol group. There was no crossover in this trial.

Questionnaire. On the baseline questionnaire, both co-horts reported similar basic laparoscopic experience as awhole, and no significant difference was observed in anycategory of previous laparoscopic experience or in com-fort with laparoscopic procedures. As expected, PGY-3residents reported significantly fewer total laparoscopicprocedures as compared with PGY-4 residents (median,13 vs 45; P < .0001) and were more likely to grade them-

Table II. Mean drill completion times (sec) between week 1 (pre) and week 4 (post) testing within and between cohorts

Checkerboard (s) Bean drop (s) String (s) Block (s) Suture foam (s) Totals (s)

Pre Post P value* Pre Post P value* Pre Post P value* Pre Post P value* Pre Post P value* Pre Post P value*

Skills 178 95 .0002 77 34 .001 68 41 .002 62 24 .002 66 28 .001 451 222 < .0001Control 171 141 .0008 75 53 .05 70 56 .02 52 33 .04 52 46 NS 420 329 .0003P value† NS .006 NS .001 NS .03 NS .02 NS .04 NS .0008

NS, Not significant.†Mann-Whitney U test, adjustment made for baseline variation.*Paired t test.

Table III. Percent improvement by task between randomization arms

Method used

Task Skills (Ơ) Traditional (Ơ) P value*

Checkerboard 45.4 19.6 .0007Bean drop 52.4 21.3 .013Running string 38.9 19.5 .053Block move 60.4 23.6 .016Suture foam 56.2 6.8 .001Mean 51.0 18.0 <.0001

*Mann-Whitney U test.†Pretest-posttest(s)/pretest(s).

840 Coleman and Muller April 2002Am J Obstet Gynecol

selves as having “low comfort” (on a 5-point scale, with “3”being “average”) with their endoscopic skills (P = .013).However, this effect was balanced by the block random-ization design. Previous or concurrent video gaming wasreported by 6 of 20 respondents. Two of these residentswere in the skills cohort and 4 of these residents were inthe control group (P = .2). Interestingly, although gamersreported similar comfort with laparoscopy at pretest asdid nongamers, they quantified their skill levels signifi-cantly higher than nongamers (P = .043). Other demo-graphic features were balanced between cohorts.

Video skills testing. Baseline and posttest drill timesfrom each of the video trainers are given by cohort inTable II. No difference was observed between the cohortsat baseline testing. However, posttest drill times were sig-nificantly different in all tasks between the cohorts. In ad-dition, although all residents showed improvement intask completion times at the end of their gynecology ro-tation, those residents who were randomized to the ex-perimental group showed the greatest improvement(construct validity; Table II).

The magnitude of this effect was quantified by an ad-justment of the within-group pretest to posttest varianceas a measure of the individual baseline completion times.Overall, drill performance was enhanced in the skillsgroup compared with the control group. Task comple-tion times were reduced an average of 51% (range, 39%-60%) versus 18% (range, 7%-24%; P < .0001; Table III).As expected, less marked improvement is seen among themost experienced surgeons, and critical minimum valueswere discernible on each skill. For example, if a residentperformed the checkerboard task in ≤150 seconds, theywere unlikely to be able to improve further from morepractice (data not shown). This performance at baselinewas within the 95% CI for the best posttraining times andessentially represented mastery of this skill. There is alsomuch less variance among the performance of individualtask trials at the conclusion of the rotation in the experi-mental group only.

Intraoperative skills assessment. Assessment of crite-rion-related validity was determined by grading intraop-erative resident surgical performance by the GSAT scores

both between and within the randomization strata. Col-lectively, mean GSAT overall scores were similar betweenthe cohorts at baseline and at study conclusion (TableIV). As expected, both cohorts demonstrated significantimprovement in laparoscopic skill with the completion ofthe gynecology rotation. However, only within the skillscohort was there a significant improvement from baseline(mean, 4.3; P = .0151; paired t test). In this manner, PGY-3 and PGY-4 residents who underwent brief, but inten-sive, video endoscopic skills training appeared toenhanced their laparoscopic skills above what might havebeen obtained through regular rotation-related practice.Although this study was not sufficiently powered to strat-ify by PGY level, it was observed that the improved intra-cohort skill assessment was most striking among PGY-3residents as compared with PGY-4 residents (data notshown). Interrater agreement was good between our in-dependent reviewers on the GSAT (κ, 0.72).

Resident skills perception. Both cohorts reported alower than average pretest comfort in these areas (me-dian, 12 [skills group] and 13.5 [control group] of a pos-sible 25 total points). As expected, self-assessmentcompetency improved in both cohorts after their rotation(median, 16 [skills group] and 17 [control group]).Pretest to posttest self-assessment was improved signifi-cantly (P = .001), yet no difference was observed betweenthe skills group and the control group after their rota-tion. Further, only a weak association (r = .11) was ob-served between self-perception and objective operativeevaluation by the GSAT, even among the skills group.

Comment

Effective instruction of surgical skill is central to all res-idency training programs. The manner in which theseskills are taught and the assessment of an individual’s skilllevel are heterogeneous and often ambiguous.7,18 Tech-niques include lecture, book instruction, animal modeldissection, and dry laboratory instruction. However, themost common theater in which to provide surgical in-struction is the operating room. A recent survey of resi-dency directors reported that 99% of responders taughtsurgical skills in the operating room.19 However, only29% had a formal surgical skills program, and only 17%used testing or standardized assessment. Still, there ismounting evidence that objective measuring tools can bevalidated and reproduced for both open and laparoen-doscopic techniques.7,20-22

Various authors have suggested that educational objec-tives that reach through laboratory-based skills practicecan lead to improved endoscopic proficiency by fine-tun-ing motor skills such as dexterity and hand-eye coordina-tion. Derossis et al23 demonstrated among generalsurgical residents that dexterity and visual-spacial skillscould be improved through repetition in video drillboxes. Similarly, Cundiff et al2 showed significant im-

Table IV. GSAT scoring of intraoperative performanceby randomization cohort

P† Pretest* Posttest* (within cohort)

Skills group 17.4 21.7 0.0151Control group 16.4 20.3 0.0923P value‡ (between cohorts) NS NS

NS, Not significant.*Mean of 35 total possible points.†Paired t test.‡Mann-Whitney U test.

Volume 186, Number 4 Coleman and Muller 841Am J Obstet Gynecol

provement in endoscopic competency with repetitivepractice among a cohort of obstetrics and gynecology res-idents at various levels of training. Their principal out-come measure was task completion time. Improvementin laboratory-based skills correlated with improvement inendoscopic self-assessment among the residents. No invivo confirmation was performed. However, faculty as-sessment of resident proficiency did follow a nonsignifi-cant upward trend.

Using laboratory-based proficiency as a surrogate forintraoperative skill is a harder bridge to cross becausemost measures of success (such as short time to task com-pletion) do not adequately evaluate precision or judg-ment, which are important intraoperative necessities.Martin et al,15 who used open surgical bench models,compared their results with similar tasks in live anesthe-sized animals. Highly significant correlation was observedbetween the 2 models, which validated their bench test-ing facility. Fried et al20 demonstrated similar findings ina randomized trial of 12 general surgery residents whounderwent laparoscopic skills evaluation. These investi-gators, after baseline assessment in both in vitro (videoboxes) and in vivo (porcine model) models, randomizedparticipants to either 5 days of skills training or notraining. Repeat testing in both models demonstrated “value-added” improvement of in vivo skills after labora-tory-based practice. Scott et al9, in one of the few humanin vivo trials, randomized 27 junior general surgery resi-dents to either 10 days of endoscopic video training or noadditional training after baseline proficiency evaluationby inanimate task-completion testing and intraoperativeGSAT scoring of laparoscopic cholecystectomy. Theydemonstrated improvement in video-eye-hand skills aftertraining that translated into improved intraoperative skillassessment. “Value-added” improvement was evident in 4of 8 individual parameters in the GSAT, when comparedwith control subjects.

In the current trial, we suggest a similar extrapolationamong PGY3 and PGY4 obstetrics and gynecology resi-dents in the performance of laparoscopic partial salp-ingectomy. Previous reports that used the same videoendoscopic tasks as used in this trial have demonstratedconstruct validity among varying levels of residents atbaseline.2,7,9,12,15 This effect was balanced in the currentreport by the block randomization technique. However,construct validity was demonstrated after training be-cause significant “value-added” improvement was ob-served in our skills group compared with control subjectsin task completion efficiency. Certain tasks were more dis-criminating than others, such as intracorporeal knottying and the checkerboard drill. Modification of thevideo drills to more closely mimic intraoperative skillsmay more efficiently teach these desired principles.

Intraoperative skill between cohorts as measured bythe GSAT was similar at baseline and significantly im-

proved at posttest but not statistically between the randomized cohorts. However, operative intracohort im-provement from baseline was observed significantly onlyin the skills cohort. This finding suggests that the endo-scopic instruction and experience during a customary gy-necologic rotation affects surgical endoscopic skillpositively. However, the performance of laboratory-basedskills training appears to provide an added benefitthrough repetitive video-eye-hand practice. Criterion-related validity is demonstrated.

In the current study, a specified task-to-be-completedpartial salpingectomy was chosen as the evaluating proce-dure. This was chosen because candidate patients wereabundant and could be enrolled electively at specifiedtimes that were relevant to resident training. Althoughconstruct validity has not been demonstrated with respectto this procedure, a much greater effect of practice and in-tracohort improvement was seen in our PGY3 residents.This suggests that the level of complexity for the assess-ment operation may have discerned critical skills that weregained from laboratory practice among less-experiencedsurgeons. Formal evaluation should be conducted.

There are several important considerations about theconduct of this trial, which require discussion. First, thenumber of residents who completed all phases of the protocol was approximately two thirds of the initialrandomization. Reasons for this (patient availability andcase cancellation) were not controllable; however, weare not able to assess whether this excluded resident co-hort was representative of the measured cohort. Thedisproportionate number of residents who completedall tasks (n = 11 [skills cohort] vs 7 [control cohort])might suggest less motivation in finding a suitable surgi-cal candidate. However, comparison of posttest videoscoring that was available for 2 of these residents whodid not perform a second surgical case was similar to thecontrol group as a whole. This suggests that the attritionbias may be minimal.

Second, adherence to the planned surgical procedurewas variable but random. Despite standardization tools, in-trinsic patient anatomy, comorbidities and difficulties re-lated to patient factors, entry complications, and trocarand instrument malfunctions could not be controlled.Likewise, the degree to which these deviations were scoredby the independent blinded judges was not standardized.However, good overall interobserver agreement was foundbetween our independent reviewers (κ = 0.72).

Third, resident proficiency was measured with theGSAT. This objective assessment of surgical skills was pio-neered and validated at the University of Toronto in gen-eral surgery residents in both animate and inanimatemodels.14-17 The tool has been validated recently by ob-stetrics and gynecology residents. Investigators at the Uni-versity of Washington have used the tool with bothanimate and inanimate models to assess resident surgical

842 Coleman and Muller April 2002Am J Obstet Gynecol

skill effectively and are suggesting the tool be used toidentify residents in whom additional surgical trainingmay be beneficial.8 Our study similarly supports its objec-tive use of surgical performance.

Last, long-term skill retention was not addressed. Be-cause testing (both laboratory and in vivo) was in closeproximity to skills training, the maximal effect of this in-tervention is measured. Derossis et al24 demonstrated thatcomposite precision/accuracy scores could be followed ingeneral surgical residents over a 2-year period with simula-tor models. However, whether improved skills by trainingpersist without specific skills training is not known. We in-tend to incorporate “booster” skills sessions throughout allgynecology rotations in our residency program.

In this trial, we demonstrate that a short-term intensivelaboratory-based video laparoscopic skills curriculum cantranslate into better individual operative proficiency.These findings have significant implications not only inthe efficiency with which new procedures can be learnedbut also ultimately in improved patient safety and low-ered operative cost. We suggest that formal skills trainingshould be incorporated into obstetric and gynecology res-ident surgical curriculum to enhance surgical skill.

REFERENCES

1. Sanfilippo JS, Laube DW, Sammarco JJ, Thompson RJ, Visscher HC,Youngblood JP. CREOG training guidelines: advanced surgicaltechniques for residency training programs in obstetrics and gy-necology. Washington (DC); 1994.

2. Cundiff GW. Analysis of the effectiveness of an endoscopy edu-cation program in improving residents’ laparoscopic skills. Ob-stet Gynecol 1997;90:854-9.

3. Melvin WS, Johnson JA, Ellison EC. Laparoscopic skills en-hancement. Am J Surg 1996;172:377-9.

4. Sammarco MJ, Youngblood JP. A resident teaching program inoperative endoscopy. Obstet Gynecol 1993;81:463-6.

5. Wolfe WM, Levine RL, Sanfilippo JS, Eggler S. A teaching modelfor endoscopic surgery: hysteroscopy and pelviscopic surgery.Fertil Steril 1988;50:662-4.

6. Schueneman AL, Pickleman J, Hesslein R, Freeark RJ. Neu-ropsychologic predictors of operative skill among generalsurgery residents. Surgery 1984;96:288-95.

7. Goff BA, Lentz GM, Lee D, Houmard B, Mandel LS. Developmentof an objective structured assessment of technical skills for obstet-ric and gynecology residents. Obstet Gynecol 2000;96:146-50.

8. Lentz GM, Mandel LS, Lee D, Gardella C, Melville J, Goff BA.Testing surgical skills of obstetric and gynecologic residents in abench laboratory setting: validity and reliability. Am J Obstet Gy-necol 2001;184:1462-70.

9. Scott DJ, Bergen PC, Rege RV, et al. Laparoscopic training onbench models: Better and more cost effective than operatingroom experience? J Am Coll Surg 2000;191:272-83.

10. Jones DB, Brewer JD, Soper NJ. The influence of three-dimen-sional video systems on laparoscopic task performance. Surg La-parosc Endosc 1996;6:191-7.

11. Nguyen NT, Mayer KL, Bold RJ, et al. Laparoscopic suturingevaluation among surgical residents. J Surg Res 2000;93:133-6.

12. Rosser JC, Rosser LE, Savalgi RS. Skill acquisition and assess-ment for laparoscopic surgery. Arch Surg 1997;132:200-4.

13. Rosser JC Jr, Rosser LE, Savalgi RS. Objective evaluation of a la-paroscopic surgical skill program for residents and senior sur-geons. Arch Surg 1998;133:657-61.

14. Faulkner H, Regehr G, Martin J, Reznick R. Validation of an ob-jective structured assessment of technical skill for surgical resi-dents. Acad Med 1996;71:1363-5.

15. Martin JA, Regehr G, Reznick R, et al. Objective structured as-sessment of technical skill (OSATS) for surgical residents. Br JSurg 1997;84:273-8.

16. Reznick R, Regehr G, MacRae H, Martin J, McCulloch W. Test-ing technical skill via an innovative “bench station” examination.Am J Surg 1997;173:226-30.

17. Winckel CP, Reznick RK, Cohen R, Taylor B. Reliability and con-struct validity of a structured technical skills assessment form.Am J Surg 1994;167:423-7.

18. Reznick RK. Teaching and testing technical skills. Am J Surg1993;165:358-61.

19. Mandel LP, Lentz GM, Goff BA. Teaching and evaluating surgi-cal skills. Obstet Gynecol 2000;95:783-5.

20. Fried GM, Derossis AM, Bothwell J, Sigman HH. Comparison oflaparoscopic performance in vivo with performance measuredin a laparoscopic simulator. Surg Endosc 1999;13:1077-82.

21. Goff BA, Lentz GM, Lee DM, Mandel LS. Formal teaching of sur-gical skills in an obstetric-gynecologic residency. Obstet Gynecol1999;93:785-90.

22. Shapiro SJ, Paz-Partlow M, Daykhovsky L, Gordon LA. The useof a modular skills center for the maintenance of laparoscopicskills. Surg Endosc 1996;10:816-9.

23. Derossis AM, Fried GM, Abrahamowicz M, Sigman HH, Barkun JS,Meakins JL. Development of a model for training and evaluationof laparoscopic skills. Am J Surg 1998;175:482-7.

24. Derossis AM, Antoniuk M, Fried GM. Evaluation of laparoscopicskills: a 2-year follow-up during residency training [see com-ments]. Can J Surg 1999;42:293-6.