Managing physical activity through voice: a usability study of the PCHA app with blind and low-vision individuals

While mobile health (mHealth) technologies can support individuals' physical activity management, they often present accessibility barriers for blind and low-vision (BLV) users. To address this gap, we developed the Personalised and Conversational Health Agent (PCHA) app for BLV users.

We conducted a 7-day usability study with six BLV adults. Data were collected on physical activity indicators (daily step counts and walking distance), System Usability Scale (SUS) scores, and qualitative feedback from semi-structured exit interviews.

The PCHA app achieved a mean SUS score of 77.1, with ratings ranging from "Okay" to "Excellent." Participants valued the app's streamlined interface and reminder system but expressed mixed preferences for voice versus text-based data input. The AI-based chat feature demonstrated limited engagement, with participants reporting generic responses that reduced its perceived utility. The physical activity data revealed highly variable patterns, influenced by personal and contextual factors, and technical issues such as incomplete data synchronisation.

Voice interfaces offer important accessibility benefits but should complement screen reader-based interactions. Further improvements should include multi-modal interaction options, improved technical reliability, and adaptive personalisation features that account for individual circumstances and preferences. These findings provide design insights for advancing inclusive mHealth interventions to better support physical activity management among BLV individuals.