Background: Digital health history devices (DHHDs) have the potential to improve the quality and efficiency of medical consultations by providing standardized patient history summaries and aiding physicians in differential diagnosis (DD) formulation. These tools also engage patients in the diagnostic process, potentially enhancing clinical decision-making. This study evaluates the effectiveness of DIANNA, a DHHD, in improving DD accuracy in an outpatient setting.
Objective: This study aimed to determine whether DIANNA enhances the accuracy of physician-generated DDs compared to traditional history-taking methods.
Methods: A pseudo-randomized controlled trial was conducted at the University Hospital Geneva emergency outpatient department, including 101 patients presenting with limb, back, or chest conditions. The first 51 patients were assigned to the control group, where physicians formulated DDs based on standard history-taking and clinical examination. The subsequent 50 patients formed the intervention group, where physicians reviewed DIANNA-generated DD reports before consultation. A senior physician independently formulated DD lists in all cases, serving as the gold standard for comparison.
Results: DIANNA use was associated with a significant improvement in DD accuracy, with a mean accuracy of 79.3% (SD 24%) in the intervention group compared to 70.5% (SD 33%) in the control group (P = .014). Effectiveness varied by case complexity. In low-complexity cases, where one or two DDs were possible, the intervention group demonstrated an 8% improvement over the control group (P = .08). In intermediate-complexity cases involving three possible DDs, the intervention resulted in a 17% improvement (P = .03). In high-complexity cases, where four or five DDs were possible, there was a 15% improvement (P = .92), though this was not statistically significant. DIANNA determined appropriate DDs in 81.6% of cases, and physicians reported that it contributed to establishing the correct DD in 26% of cases.
Conclusions: DIANNA demonstrated the potential to improve DD accuracy, particularly in intermediate-complexity cases. However, its effectiveness was limited in low- and high-complexity cases, suggesting that further validation is needed to assess its broader clinical impact. These findings highlight the role that DHHDs such as DIANNA could play in enhancing diagnostic accuracy and supporting clinical decision-making in medical practice.
Clinicaltrial: This trial was registered on clinical trials (NCT03901495).
