BackgroundHand hygiene is the primary measure for reducing nosocomial infections based on 7 steps recommended by the WHO. The aim of this study was to assess the duration and the quality of hand hygiene before and after simulation-based training (SBT).MethodsThe study took place in a University Hospital Pediatric Department among its residents and nurses. In assessment A, 10 hand-rubbing procedures per participant during a work day were scored by observers using a validated, anatomically based assessment scale. Two weeks later, all participants received a didactic course and SBT, followed 1 month later by assessment B, observation of 10 hand-rubbing procedures. Assessments were performed by 2 independent observers. Before-and-after testing was used to evaluate the demonstration of theoretical knowledge.ResultsIn total, 22 participants were included, for whom 438 hand hygiene procedures were assessed: 218 for assessment A and 220 for assessment B. The duration of hand rubbing increased from 31.16 seconds in assessment A to 35.75 seconds in assessment B (P=.04). In assessment A, participants averaged 6.33 steps, and in assessment B, participants averaged 6.03 steps (difference not significant). Significant improvement in scores was observed between assessments A and B, except for the dorsal side of the right hand. The wrist and interdigital areas were the least-cleaned zones. A difference between assessments A and B was observed for nail varnish (P=.003) but not for long nails or jewelry. Theoretical scores increased from 2.83 to 4.29 (scale of 0–5; P<.001).ConclusionThis study revealed that an optimal number of steps were performed during hand-rubbing procedures and that SBT improved the duration and quality of hand hygiene, except for the dorsal right side. Emphasis should be placed on the specific hand areas that remained unclean after regular hand-rubbing procedures.

An anatomically based assessment scale of handwashing quality with alcohol-based hand rub was designed. Contents of the scale divided each hand into 40 zones. Psychometric measurements were studied in 30 participants (120 hand sides). The scale was both valid (Cronbach α: 0·83 dorsal side and 0·73 palmar side) and reproducible (linear regression R2, 0·91; intraclass correlation coefficient, 0·99).