UNIFORM DISTRIBUTIONAL BENEFIT OF COMMUNITY-DELIVERED SOCIAL PRESCRIBING ON HEARING-RELATED INTRINSIC CAPACITY IN OLDER ADULTS: A QUANTILE REGRESSION REANALYSIS OF THE HEARS TRIAL

Background: The HEARS randomized trial demonstrated that a low-cost hearing care intervention delivered by community health workers improved hearing - related communication function in older adults. However, mean - based analyses do not characterize how treatment effects vary across the outcome distribution. We examined the distributional pattern of the intervention effect using an aggregate- data- inform ed simulation approach. Methods: Synthetic baseline and 3-month Hearing Handicap Inventory for the Elderly-Screening Version (HHIE-S) scores were generated for 68 observations per group using published means and standard deviations.Scores were sampled from group-specific truncated bivariate normal distributions bounded between 0 and 40, assuming a baseline – to -follow-up correlation of 0.50. Baseline-adjusted conditional quantile regression coefficients were estimated from τ=0.05 to τ=0.95. Clinical response was assessed using improvements of at least 8, 10, and 12 points. Stability was evaluated across 200 independently generated synthetic datasets.