Reliability study of the Laparoscopic Skills Index, LSI, a new measure of gynecologic laparoscopic surgical skill

grantor: University of TorontoThis thesis deals with the construction and reliability testing of the LSI, a new multiitem measure of laparoscopic skills in gynecology to be used in the human model. 'Methods'. The construction of the LSI involved (1) item selection; (2) selecting a method to scale responses; (3) design; (4) choosing a scoring method. Reliability was evaluated in two phases (1) a 'pilot study' in which six raters reviewed six videotaped procedures (subjects), on two occasions, separated by one month; (2) a ' main study' in which four raters reviewed 20 videotaped procedures (subjects). The main study was conducted following efforts to improve the LSI reliability. 'Results of the main study'. Cronbach's alpha was 0.95 indicating a high level of internal consistency. A "best measure" reliability coefficient for interrater reliability was ICC = 0.77 (95 percent CI 0.56 to 0.90) which is considered to be very good. This result was attained after excluding one rater where a systematic bias was evident. The LSI appears to have the properties of a unidimensional index in which the item variables are true components of the overall attribute. Before entering into general usage, construct validity must be determined and the two methods of administering the LSI, in vivo and by videotape, must be compared.M.Sc