734. Outcomes from Two Pilots of an Automated Observation System for Monitoring Hand Hygiene
Session: Oral Abstract Session: Preventing Hospital Transmission: Environment and Hands
Friday, October 9, 2015: 11:15 AM
Room: 7--AB
Background: Strategies to improve hand hygiene (HH) are a priority for infection prevention initiatives. Automated observations (AO) promote compliance through immediate  feedback and can facilitate monitoring of more hand hygiene opportunities (HHO) than feasible with direct observation (DO).  We describe our experience with AO and its effect on HH compliance.

Methods: An AO system was instituted on a medical floor (unit 1) and a surgical ICU (unit 2) for 90 and 60 days respectively. Baseline compliance was based on DO for 1 year. DO continued during pilots and for one year following completion of pilot 1. During pilots, healthcare workers (HCWs) wore alcohol-sensing badges that provided immediate feedback on HH compliance with a color change from red to green when alcohol was detected.  HHOs were monitored at room entry and exit, with compliance defined as alcohol detection within 30 seconds of entry and 60 seconds of exit.  Pearson’s coefficient calculations were performed to estimate correlation between HCW compliance at onset and completion of pilot.    

Results: Results are shown in shown in Table 1. The median number of HHO observed by AO per HCW during pilots was 242 (range 1-2097). In both units, overall compliance rapidly achieved 98% in week one, and remained stable at mean 97% (SD = 0.67%). HCW showed stable compliance during pilots. For unit 2, HCW compliance during the first week correlated with compliance during the final week (r=0.82, P<0.0001) and with overall compliance during the 60-day pilot (r=0.77, P<0.0001).


% Compliance


# of HCW badged

DO Baseline

(for 1 year pre-pilot)

AO during pilot

DO during pilot

DO Post-pilot

(for 1year)

Unit 1


54% (88 HHO)

98% (140,000 HHO)

93% (27 HHO)

86% (88 HHO)

Unit 2


52% (104 HHO)

97% (27,566 HHO)

99% (68 HHO)


Table 1. HH results obtained from AO system and DO for specified time periods

Conclusion: AO with immediate feedback demonstrated rapid and stable improvement in HH. As expected, the AO facilitated monitoring of a great degree more HHO than feasible with DO.  

The data provide insight into the trajectory of behavior change associated with AO systems. Improved HH was observed rapidly at pilot initiation. Post-pilot data from unit 1 suggests some durability in HH improvement despite withdrawal of the feedback. This finding has implications for the design of system-based HH approaches.

Heather Michael, MD, Infectious Disease, Cleveland Clinic Foundation, Cleveland, OH, Thomas G. Fraser, MD, FSHEA, Infectious Disease, Cleveland Clinic, Cleveland, OH, Colette Einloth, QPSI, Cleveland Clinic foundation, Cleveland, OH, Cynthia Fatica, RN, BSN, CIC, Quality and Patient Safety Institute, Cleveland Clinic, Cleveland, OH and Theresa Janszen, PSQI project manager, Cleveland Clinic Foundation, Cleveland, OH


H. Michael, None

T. G. Fraser, None

C. Einloth, None

C. Fatica, None

T. Janszen, None

Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 7th with the exception of research findings presented at the IDWeek press conferences.