Comparing human to electronic observers to monitor hand hygiene compliance in an intensive care unit

OBJECTIVE: We sought to determine whether an electronic hand hygiene (HH) system could monitor HH compliance at similar rates to direct human observation. METHODS: This 4-year proof-of-concept study was conducted in an intensive care unit (ICU) of a private tertiary-care hospital in São Paulo, Brazil, where electronic HH systems were installed in 2 rooms. HH compliance was reported respectively using direct observation and electronic counter devices with an infrared system for detecting HH opportunities. RESULTS: In phase 1, HH compliance by human observers was 56.3% (564 of 1,001 opportunities), while HH compliance detected by the electronic observer was 51.0% (515 of 1,010 opportunities). In phase 2, human observers registered 484 HH opportunities with a HH compliance rate of 64.7% (313 of 484) versus 70.6% (346 of 490) simultaneously detected by the electronic system. In addition, an enhanced HH electronic system monitored activity 24 hours per day and HH compliance without the presence of a human observer was 40.3% (10,642 of 26,421 opportunities), providing evidence for the Hawthorne effect. CONCLUSIONS: The electronic HH monitoring system had good correlation with human HH observation, but compliance was remarkably lower when human observers were not present due to the Hawthorne effect (25%–30% absolute difference). Electronic monitoring systems can replace direct observation and can markedly reduce the Hawthorne effect.