The abstract is read at the publisher; the summary is JClub's.
Lotus AI enrollees (3,223 users, Feb–Jul 2026) skew older (median age 51) and male (62%), with chronic conditions exceeding 2024 NHIS estimates in every age band by 4–7 points.
- Why it matters: Agentic, physician-supervised AI platforms are expanding, but their real-world user populations and care needs are poorly characterized against national benchmarks.
- What they did: The study analyzed Lotus AI enrollees who completed onboarding, messaged the agent, and had linked health records, then compared chronic-condition and utilization patterns to 2024 NHIS benchmarks.
- The result: Users with functional or symptom-defined disorders (18%) generated 1.8× physician escalations and 2.6× laboratory orders, and care among frequent visitors was highly fragmented (44% meeting a high-fragmentation threshold).
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.