Evidence and methods from studies of symptom-assessment applications under real-world conditions: a scoping review
File(s)
Preprint version
Author(s)
Kopka, Marvin
Altalib, Sami
Hayhoe, Benedict
El-Osta, Austen
Type
preprint
Abstract
Background Symptom-assessment applications (SAAs) are widely used as patient-facing digital tools for triage and diagnostic support, yet most evaluations have relied on vignettebased designs that poorly reflect real-world use. Evidence generated under real-world conditions, where users experiencing symptoms enter their own information, remains fragmented and methodologically heterogeneous. Objective We conducted a PRISMA-ScR compliant scoping review to synthesise evidence from studies evaluating SAAs under real-world conditions and to characterise the methods used. Methods Searches of MEDLINE, Embase, Web of Science and medRxiv identified 20 eligible studies published between 2018 and 2025. The aims of the included studies can be clustered into three domains: (i) performance and safety, (ii) user characteristics and usability, and (iii) impact on decision-making and healthcare utilisation. Results Across studies, triage accuracy ranged from 34-85%, with a consistent tendency toward over-triage, that is, assigning a higher urgency than necessary. Reported safety concerns were uncommon but non-negligible, particularly at scale. Reported diagnostic accuracy varied widely depending on operationalisation and reference standards. Users were predominantly younger and female, and acceptability was generally high. Evidence on behavioural and system-level impact showed decreased healthcare utilization and a tendency towards lower perceived urgency, although the evidence remains limited. Methodologically, the studies varied substantially in sampling strategies, reference standards, matching procedures, and blinding, which limits comparability. Conclusions These findings indicate that while SAAs may provide comparatively safe triage advice
Date Issued
2026-03-07
Citation
Figshare, 2026
Journal / Book Title
Figshare
Copyright Statement
Copyright © 2026 The Authors. This work is licensed under a Creative Commons Attribution 4.0 International License.
License URL
Description
Preprint version
Identifier
https://figshare.com/articles/journal_contribution/Evidence_and_Methods_from_Studies_of_Symptom-Assessment_Applications_Under_Real-World_Conditions_A_Scoping_Review/31563451?file=62498515
Subjects
Symptom assessment
real-world evidence
symptom checkers
methods
review
implementation
accuracy
safety
impact