Artificial intelligence (#AI )–powered digital front door tools are increasingly being used to guide patients to appropriate care and alleviate health care system pressure. However, most evaluations offer limited insight into stated care intent, real-world behavior, or care appropriateness.
Methods
The E-Health Self–Symptom Assessment as a Front Door and Facilitator of Care (ESSENCE) study was a prospective real-world quality improvement evaluation embedded in Portugal’s largest private health care network (CUF). Adults using the Ada Health diagnostic decision support system through the myCUF app reported their care intent before and after symptom assessment. We tracked actual behavior through electronic health records and surveys. Physician panels retrospectively assessed the appropriateness of intended and observed care.
Results
A total of 1470 adults (≥18 years of age; mean age, 38.5 years; 57.7% female) were enrolled. Of the 1338 participants with pre- and postassessment intentions, 33.0% revised their planned care level immediately after assessment. Uncertainty decreased from 12.6% to 5.0% (P<0.001). Among 721 participants with observed behavior, 59.1% changed their care pathway: 28.9% de-escalated, 17.2% escalated, and 13.0% resolved prior uncertainty. Primary care consultations increased from 16.3% to 42.1% (P<0.001), whereas specialist visits decreased from 49.7% to 29.8% (P<0.001). Among nonemergency participants with preassessment intentions and sufficient clinical documentation (n=382), appropriate care increased from 29.8% preassessment to 64.4% postbehavior (95% confidence interval, 27.8 to 41.4; P<0.001). Of the 96 participants who planned an emergency department (ED) visit, 38.5% selected lower-acuity care after assessment. In the subset with clinician-rated follow-up, 93% (27 of 29; 95% CI, 78.0 to 98.1%) were judged to have appropriately avoided an unnecessary ED visit.
Conclusions
Integrating an AI-supported symptom assessment and follow-up service options within a digital front door was associated with shifts in patient intentions and behaviors, reducing uncertainty and promoting appropriate health care use. These findings suggest that diagnostic decision support systems shape real-world decision-making in addition to generating accurate recommendations, warranting further evaluation across diverse health care settings. (Funded by the Federal Ministry of Research, Technology and Space [Bundesministerium für Forschung, Technologie und Raumfahrt].)
Skip to content
Every Day, 2 Medical News from the most Trusted Journals
2 Medical News 