Gains in Quality of Life in Chronic Otitis Media Patients After Surgery, Adjusted for Placebo Bias.

Abstract

Using a short-form specific questionnaire, we sought to quantify specific and generic benefits of surgery for chronic otitis media (COM) while adjusting for expectancy (placebo) bias. A prospective observational pretreatment/posttreatment study. A national tertiary referral center. Consecutive adult COM patients were enrolled for 1 year. Of 246 patients, 205 were recommended for surgery, and 167 were operated on the (more) affected ear. All patients filled out two questionnaires, one specific (Chronic Otitis Media Questionaire-12) and one generic (Short Form-36) at baseline, and then again 6 and 12 months after surgery. Factor-based scores of questionnaires, standardized response means (SRMs) for treatment effects, with multiple linear regression for implementing bias adjustment. All but one (generic Short Form-36 "mental" scale) of the seven considered measures gave very highly significant ( p < 0.001) improvements. Unadjusted SRMs were large (1.0-2.0 standard deviation units) for audiometry, symptoms of ear discharge, reported hearing, and aggregate specific quality of life, but only moderate for the less specific activity/healthcare. The proposed bias adjustment reduced SRM magnitudes for most measures by about a third, for activity/healthcare and ear discharge by only a tenth to a fifth, and for audiometry not at all. Most scores of the specific questionnaire displayed definite placebo-like biases; this demands caution in interpreting improvement after COM surgery. With bias adjustments, credible and worthwhile magnitudes of improvements remained for Chronic Otitis Media Questionnaire-12 total and subscores (0.5-1.09 standard deviation SRM), but not for generic quality of life.