Fellowship Square-Mesa reduces monthly falls from 20 to 2 with Helpany AI radar monitoring

Fellowship Square-Mesa faced a persistent fall problem despite comprehensive prevention measures including quarterly staff training, hourly assurance checks, and wearable RCare devices. Staff recognized they were stuck in a reactive cycle — detecting falls after they happened rather than preventing them. Monthly fall counts averaged 20 incidents, with no meaningful downward trend despite sustained effort.

In July 2024, the community deployed Helpany's AI tool 'Paul' across its assisted living neighborhoods. Paul uses radar-based motion detection (no cameras or audio) to passively monitor residents' movements, analyzing sleep quality, gait strength, and activity patterns to identify residents at elevated fall risk. Real-time alerts are sent to staff via a mobile app, enabling proactive check-ins before incidents occur. Rollout included resident, family, and staff focus groups and care-level-tailored onboarding.

Monthly falls dropped from a baseline of 20 to 12 in Paul's first month (July 2024), then to 6 in August, 4 per month from September through November, and just 2 in December. Overnight falls were zero from August through December. Staff shifted from reactive response to data-driven intervention using daily and weekly behavioral reports.

• Radar-based passive monitoring can deliver dramatic fall reduction without compromising resident privacy through cameras or wearables.
• Structured onboarding with focus groups across care levels is critical to earning resident and family trust.
• Moving from fall detection to fall prevention requires continuous behavioral data, not just incident response protocols.