Study design, materials and methods
Patients reported symptoms of AI on a self-reported StMWex questionnaire in the patients’ waiting area at their local anorectal outpatients’ clinic prior to clinical investigations. After clinical investigations, an interview-based StMWex questionnaire was completed by the consultant surgeon or physiotherapist who were blinded to the self-administered reports of AI. Assessment of correlation between the individual items of the interview-based and self-reported SMIS (iSMIS and sSMIS, respectively), and Wexner score (iWexner and sWexner, respectively) were assessed using Spearman’s rho, and agreement was assessed using weighted kappa statistics (κ). Data were treated anonymously, and no personal data except age and gender was recorded.
A total of 75 female and 30 male consecutive patient with a mean age of 55 years (SD:14.3; Range: 24-85) responded. The mean incontinence scores were 9.7 (SD:7.0) on the iSMIS, 9.5 (SD:6.9) on the sSMIS, 8.4 (SD:8.5) on the iWexner, and 8.3 (SD:5.8) on the sWexner scores. Spearman’s rho showed a strong relationship between the two total SMIS scores (r=.903, n=97, p<.001) and the two Wexner scores (r=.912, n=101, p<.001), and explained variance was 82% (r²=.815) and 83% (r²=.832) for the SMIS and the Wexner scores, respectively. The agreement of +/- 1 point and +/- 2 points between interview-based and self-administered were 54% and 60% , respectively for the SMIS and 60% and 75%, respectively for the Wexner score. The assessment of consistency between the individual items of the iSMIS and sSMIS showed substantial agreement (κ≥0.60) for all items except almost perfect agreement for the items regarding incontinence of gas and using pad or plug (κ≥0.81) (Table 1) (Landis & Koch, 1977).
Interpretation of results
The level of agreement in the two scores in this joint questionnaire is substantial and there is no trent of over- or underestimation of either the St. Mark's score or the Wexner score, though the Wexner score appears to have slightly better agreement.