Outcomes and patient-based hearing status in conductive hearing loss.

Abstract

To study hearing status and quality of life outcomes in patients with conductive hearing loss (CHL), and to develop a prognostic disease severity staging system for use in future clinical outcomes studies in CHL. Clinical outcomes research study in three stages: 1) Prospective development and psychometric validation of a disease-specific hearing status outcomes instrument for CHL. 2) Prospective observational outcomes study of patient-based hearing status and quality of life (QOL) outcomes after treatment of CHL. 3) Development of a prognostic disease-severity staging system to stratify patients into groups of similar treatment outcome. Psychometric instrument validation: assessment of test-retest reliability, internal consistency reliability, content, construct and criterion validity, and responsiveness to change. Statistical assessment of the impact of CHL on patient-based hearing status and QOL, and the change in hearing status and QOL after treatment. Using data from the prospective study and multivariate regression, identification of pretreatment independent variables, which predict good outcome after treatment, and development of a severity staging system. Test-retest reliability (r >0.70) and internal consistency reliability (alpha >0.70) were adequate for both subscales of the hearing status instrument. Construct validity was demonstrated by adequate convergent/divergent validity with other health status instruments on multi-item multitrait correlation matrixes. Criterion validity was demonstrated using audiometric threshold data. Responsiveness to change was demonstrated using the standardized response mean after treatment (SRM = 0.40 and 0.61 for the two subscales). Emotional hearing status and social/situational hearing status improved significantly after treatment of CHL: emotional score, 31.1 to 56.6 (P < .001); social/situational score, 56.7 to 63.7 (P = .01). Mean QOL subscale scores for patients with CHL were not significantly different from healthy populations, and global QOL scores did not change significantly after treatment of CHL. Using proportional improvement in emotional hearing status as an outcome, we identified several pretreatment factors that predicted good outcome (i.e., larger proportional improvement): level of pretreatment emotional hearing status, presence of sensorineural component of HL, and etiology of CHL (otosclerosis vs. all other causes). Some hypothesized predictive independent factors, such as bilateral HL, audiometric threshold level, pretreatment functional status, employment status, age, and duration of HL, were not identified as significant predictors of outcome in the multivariate analysis. We constructed a prognostic disease severity staging system in which the three predictor variables were each given a score of 0 or 1, and these three scores were added to obtain the disease stage (stage 0, 1, 2, or 3). Higher disease stage indicated more severe disease or a lower proportional improvement in emotional hearing status after treatment. This staging system appropriately stratified patients into groups of similar outcome after treatment (P = .02) and has significant potential use for future outcomes research on CHL. We have developed and validated a patient-based hearing status outcomes instrument for use in CHL. We have measured hearing status and quality of life outcomes after treatment of CHL, and we have developed a prognostic disease severity staging system that predicts the improvement in emotional hearing status after treatment of CHL.