Surveillance of handwashing episodes in adult intensive-care units by measuring an index of soap and paper towel consumption.

Objective: To determine whether handwashing surveillance could be conducted by measurements of soap and towel consumption.

Design and participants: In the medical intensive-care unit (MICU) of the Omaha Veterans' Affairs Medical Center, 10 4-hour day-time observation periods encompassing 409 handwashing episodes were scheduled in a 51-day period. In the surgical intensive-care unit (SICU), 7 4-hour periods encompassing 350 episodes were scheduled in a 49-day period. An observer measured paper towel height, towel weight, and soap weight at each sink. The observer also counted handwashing episodes and bed occupancy. Using handwashing episodes as a dependent variable, stepwise linear regression was performed with changes in towel height, towel weight, and soap weight as independent variables.

Results: Mean handwashing episodes per hour per occupied bed were 2.39 +/- 0.80 (standard deviation) in the MICU and 2.83 +/- 0.72 in the SICU. Correlation r with handwashing episodes for MICU changes was 0.891 for towel height, 0.950 for towel weight, and 0.882 for soap weight. Corresponding correlations for the SICU were 0.881, 0.918, and 0.904. For both units, stepwise regression retained changes in the weight of towels and soap as independent variables (P < .0001), with R2 0.965 (MICU) and 0.981 (SICU).

Conclusion: Because soap and towel consumption measurements are closely related to handwashing frequency and because these measurements are easy to obtain, they offer a means of handwashing surveillance that can be sustained indefinitely. This can facilitate feedback-based interventions to improve handwashing frequency.