TITLE: A POSTOPERATIVE SATISFACTION INSTRUMENT WITH PROVEN RELIABILITY AND VALIDITY IN DIVERSE PATIENT POPULATIONS

AUTHORS: Allan Klock, MD, Michael A. Paul, BS, Suanne M. Daves, MD, Jerome M. Klafta, MD, William R. Furman, MD, Michael F. O'Connor, MD
AFFILIATION: University of Chicago, Chicago, IL.

Despite calls for a valid and reliable tool to assess perioperative patient satisfaction, no instrument currently has demonstrated these necessary statistical properties.¹ We have previously reported the development of a perioperative satisfaction instrument and demonstrated its reliability and validity in patients undergoing prostatectomies.² We sought to determine if these results could be replicated in patients undergoing a variety of surgical procedures.
METHODS: After receiving IRB approval, 330 consecutive patients who had undergone surgery by 10 different surgeons and gynecologists were selected from our hospital billing database. The patients were mailed a 22-item questionnaire with a 0 to 3 response format. To assess the construct validity of the instrument, written patient comments were evaluated by a blinded observer who assigned a ranking of highly satisfied, neutral, or dissatisfied for each subscale (surgery, nursing, and anesthesia) and overall satisfaction. ANOVA was used to evaluate differences between groups. In order to assess test-retest reliability an identical second survey was mailed to the first 188 respondents 14 days after their initial response.
RESULTS/RELIABILITY: We received responses from 217 of 330 patients (66% response). Kappa values for individual items ranged from 0.47 to 0.69, indicating good reliability for all items. Of the 1954 test-retest pairs, 74% had perfect agreement, 24% of the pairs differed by a score of 1.
VALIDITY: The ratings of written comments showed significant differences in the numerical scores for overall satisfaction and each of the subscales (p<0.001 for each).

Table 1: Subscale scores (mean±SD) grouped according to rating of written comments (p<.0001 for each group)

DISCUSSION: In a diverse population of surgical patients, our instrument has proven reliability and validity. In a separate study, the instrument has also shown the ability to discriminate between surgical specialties and surgeons. This instrument can now be used to determine the important correlates of patient satisfaction.
REFERENCES:

¹Anesth Analg 87:1089-98, 1998.

²Anesth Analg 84(4):S193, 1997.