Presentation (CTI)

OBJECTIVES: Community-recruited, patient- and caregiver-reported studies often rely on self-reported diagnoses to determine eligibility. Validating diagnoses can improve confidence in participant authenticity and data accuracy, particularly in digital studies using broad online recruitment methods (e.g., social media). Despite rapid growth in this area, the extent and consistency of diagnostic validation remain unclear.
This study synthesizes recent literature to assess whether and how diagnoses are validated in community-recruited digital real-world studies.
METHODS: A targeted PubMed review identified digital real-world studies published within the past 5 years that recruited participants from the community and relied on self-reported diagnoses. Studies were screened against predefined criteria, and data were extracted on whether diagnosis validation was undertaken, the methods used, and how limitations were reported.
RESULTS: Sixty studies met inclusion criteria. In 86.0% of studies, diagnoses were based solely on self-report without supporting evidence of clinical confirmation; among these, 54.0% acknowledged this as a limitation. Indications varied widely, with endometriosis (11.7%) and mental health conditions (10.0%) most frequently represented.
Diagnosis validation was uncommon and heterogeneous. Methods included use of validated instruments (n=3), submission of medical records (n=2), and individual approaches such as additional surveys, recruitment from clinical datasets, or clinical testing (n=1 each). Requirements for validation ranged from mandatory to optional, with some studies conducting subgroup analyses restricted to clinically confirmed populations.
CONCLUSIONS: Diagnosis validation is infrequently implemented and lacks standardization in community-recruited digital real-world studies. Establishing best practices could improve the credibility and interpretability of findings in this growing research area. Future research should assess validation approaches across indications, refine screening strategies to reduce misclassification, and determine when self-reported diagnoses are an appropriate proxy for clinician-confirmed diagnoses.