Background: Positive deviance (PD) can be a strategy for the improvement of hand hygiene (HH) compliance. Methods: This study was conducted in 8 intensive care units and 1 ward at 7 tertiary care, private, and public hospitals. Phase 1 was a 3-month baseline period (from August to October 2011) in which HH counts were performed by observers using iPods (iScrub program). From November 2011 to July 2012, phase 2, a PD intervention was performed in all the participating centers. We evaluated the consumption of HH products (alcohol gel and chlorhexidine) and the incidence density of health care-associated infections. Results: There was a total of 5,791 HH observations in the preintervention phase and 11,724 HH observations in the intervention phase (PD). A statistically significant difference was found in overall HH compliance with 46.5% in the preintervention phase and 62.0% in the PD phase (P < .001). There was a statistically significant reduction in the incidence of density of device-associated infections per 1,000 patient-days and also in the median of length of stay between the preintervention phase and the PD phase (13.2 vs 7.5 per 1,000 patient-days, respectively, P = .039; and 11.0 vs 6.8 days, respectively, P < .001, respectively). Conclusion: PD demonstrated great promise for improving HH in multiple inpatient settings and was associated with a decrease in the median length of stay and the incidence of device-associated HAIs.