Digital Self Screening for Arthritis Diagnosis Save

There are numerous impediments to diagnosis or referral; and a new study evaluates patient use of a mobile artificial intelligence (AI)–based symptom checker for diagnoses, but demonstrated only modest accuracy when applied to a cohort of patients with joint symptoms.

Researchers presumed that the use of digital diagnostic decision support systems (DDSSs) may expedite diagnosis and referral to health care providers. They sought to assess the diagnostic accuracy of a mobile artificial intelligence (AI)–based symptom checker (Ada) and a web-based self-referral tool (Rheport) to facilitate the diagnosis of inflammatory rheumatic diseases (IRDs).

A prospective, multicenter, open-label, crossover randomized controlled trial included 600 new patients presenting to 3 rheumatology centers. Participants were randomly assigned to complete a symptom assessment using either Ada or Rheport. The primary outcome was the correct identification of IRDs by the DDSSs, defined as the presence of any IRD in the list of suggested diagnoses by Ada or achieving a prespecified threshold score with Rheport. This was compared to diagnoses made by rheumatologists.

From the cohort of 600 patients, 36% were diagnosed with an IRD. Rheumatoid arthritis was most frequent, seen in 69 (11.5%) patients. Both the web (Rheport) and mobile (Ada) top disease suggestions demonstrated overall diagnostic accuracies of 52% to 63%, for IRDs.  The sensitivity of these tools were 62% to 52%, with specificities 47% to 66% for IRDs. 

Both tools were heterogenous, and somewhat inconsistent. For instance, Ada was better diagnosing RA, than diagnoses.  Overall, there was poor agreement between the 2 methods in diagnosing rheumatic disease (Cohen κ statistic was 0.08 to 0.15). 

The diagnostic accuracies of both digital diagnostic tools for were not promising in this selected or enriched patient population. Regulation and improvements are needed to improve the efficacy of digital diagnostic decision support systems.