Wearable technology-derived digital biomarkers for non-invasive assessment of erectile dysfunction: a cross-sectional...

Erectile dysfunction (ED) represents a significant health concern that affects not only physical well-being but also psychological health and quality of life. This study aimed to investigate the potential of digital biomarkers derived from smartwatches for the risk assessment of ED.

This prospective, comparative cross-sectional study investigated digital biomarkers derived from smartwatch technology for ED assessment in 74 male participants (38 ED patients, 36 controls) between May and September 2025. Participants wore iOS-based smartwatches continuously for 14 days to monitor physiological parameters including heart rate variability (HRV), sleep architecture, and physical activity levels. ED diagnosis was established using the International Index of Erectile Function 5-item form (IIEF-5) with scores ≤ 21 indicating ED. Statistical analyses included univariate/multivariate logistic regression and ROC curve analysis.

Compared to controls, the ED group demonstrated significantly lower HRV (32 vs. 38.5 ms, p < 0.001), higher resting heart rate (69 vs. 58 bpm, p < 0.001), increased wake time percentage (p < 0.001), decreased deep sleep duration (20% vs. 25%, p < 0.001), and reduced daily step count (4196 vs. 5142 steps, p = 0.001). Multivariate analysis identified HRV (OR: 0.884, p = 0.019), wake time percentage (OR: 1.832, p = 0.005), and daily step count (OR: 0.872, p = 0.002) as independent associated variables. ROC analysis revealed wake time percentage demonstrated highest screening performance (AUC: 0.752, sensitivity: 68.4%, specificity: 69.4%).

Preliminary findings from this pilot study suggested that digital biomarkers from smartwatch technology could offer potential for non-invasive ED risk assessment. These results provided a feasibility foundation for integrating wearable technologies into clinical practice for personalized sexual health management.