Diagnostic Performance of Digital and Clinical Approaches for Detection of Podoconiosis by Community Health Workers in Rwanda: A Comparison with Dermatologist Diagnosis

Introduction:

Detection of podoconiosis is essential to prevent disability, but diagnostic challenges mean it often goes unnoticed in Rwanda, leading to irreversible harm and social stigma.

Methods:

This study evaluates the diagnostic performance of the WHO Skin-NTD mobile application and a clinical algorithm implemented by trained community health workers (CHWs) in detecting podoconiosis in endemic communities comparing them with dermatological reports as the clinical reference standard method.

Results:

Both the WHO Skin-NTD mobile application and the clinical algorithm demonstrated high diagnostic performance for podoconiosis, with different proportion classified positive estimates (66.3% and 71.4% respectively). The mobile application showed a sensitivity of 89.8% (83.4%– 94.3%) and specificity of 88.1% (77.1%–95.1%), while the clinical algorithm demonstrated a sensitivity of 96.4% (91.7%–98.8%) and specificity of 86.4% (75.0%–94.0%). Overall diagnostic accuracy was high for both approaches, with Sensitivity + Specificity/2of 0.89 (0.84–0.94) for the mobile application and 0.91 (0.87–0.96) for the clinical algorithm.

Conclusion:

Both the WHO Skin-NTD mobile application and the clinical algorithm demonstrated high diagnostic performance for detection of podoconiosis by CHWs in Rwanda. However, their use should be guided by context-specific resource availability rather than simultaneous deployment. The clinical algorithm is particularly suitable for low-resource and offline settings, while the mobile application offers added value in settings with digital infrastructure through standardized decision support and enhanced data management. A flexible, context-dependent approach to implementation may optimize case detection and support podoconiosis control efforts in endemic settings.