Scanner-Measured Hand Hygiene Performance After Automated Visual Feedback-Supported Training: A Multicenter Quality Improvement Project in Italian Hospitals

Background: Healthcare-associated infections (HAIs) remain a major patient safety issue, and hand hygiene (HH) is a key preventive measure. Technologies providing immediate visual feedback may support HH training by allowing objective assessment of hand-rubbing technique. Methods: A multicenter quality improvement project was conducted in three hospitals in northern Italy to assess scanner-measured HH performance over time after a standardized HH training program supported by an automated visual feedback scanner, the Semmelweis Hand Hygiene System®. Nurses and nurse aides performed scanner assessments before training (T0) and up to 6–9 months after training (T1–T4). Data included demographic variables, baseline HH knowledge, ring detection, and scanner pass/fail outcomes. Longitudinal changes were analyzed using adjusted generalized estimating equation (GEE) logistic regression models. Results: A total of 292 healthcare professionals participated in the project (69.2% nurses; mean age 42 years; 81.1% female). At baseline, 44.2% failed the scanner assessment, with dorsum areas being the most frequently insufficiently covered, and 36.0% wore rings. Across 2422 scanner assessments, pass rates improved over time. In adjusted GEE models, each additional scheduled month from baseline was associated with higher odds of passing the scanner assessment (OR = 1.18, 95% CI 1.13–1.24, p < 0.001). This association remained significant in the first-scan-only sensitivity analysis (OR = 1.16, 95% CI 1.11–1.21, p < 0.001). The association was also confirmed in a completers-only sensitivity analysis. Ring detection was associated with lower odds of passing, whereas baseline HH knowledge was not independently associated with scanner performance. Conclusions: Scanner-based immediate feedback was associated with improved scanner-measured HH technique over time. However, the quality improvement design, absence of a control group, use of the scanner as both feedback tool and outcome measure, and declining retention across follow-up assessments limit causal interpretation. Scanner-supported feedback may represent a useful adjunct to traditional HH education, but further controlled studies are needed to assess sustainability and real-world HH compliance.