When AI Meets Accessibility: How Perceived Ease of Use Mediates and Disability Type Moderates Healthcare Website Usability

Abstract

Purpose:    This study examines the impact of AI-powered accessibility features on healthcare website usability for differently abled people in the Colombo District. It also investigates the mediating role of perceived ease of use and the moderating effect of disability type.

Design/methodology/approach:  A quantitative cross-sectional survey design was adopted. Data was collected from 106 respondents with visual, auditory, and motor impairments in the Colombo District using stratified sampling to ensure representation across disability categories. Data analysis was conducted using Partial Least Squares Structural Equation Modelling (PLS-SEM).

Findings:    The findings reveal that AI-powered accessibility features do not have a significant direct effect on healthcare website usability. However, perceived ease of use fully mediates this relationship. Disability type significantly moderates the relationship between AI-powered accessibility features and usability, with users with motor impairments showing a stronger effect compared to those with visual and auditory impairments. Additionally, AI-powered accessibility features have a direct positive impact on perceived ease of use.

Originality:   This study contributes to the existing literature on digital accessibility and healthcare usability by integrating AI-powered accessibility features with usability outcomes, while highlighting the mediating role of perceived ease of use and the moderating influence of disability type.

Implications:   The study emphasizes the importance of user-centered accessibility design in the healthcare industry. It provides practical insights for healthcare providers and developers to enhance digital platforms, ensuring inclusivity and improved usability for differently abled users.