Eunwoo Yoo is an Infection Preventionist with extensive clinical experiences more than 12 years in the teaching hospital. She holds BS degree in Nursing from Yonsei University in South Korea and licensed registered nurse of 3 countries (South Korea, USA, and UAE) with the passion of nursing. Currently, she is working on Infection Prevention and Control Team for the government hospital in UAE runs by Seoul National University Hospital. As one of a key member of JCI accreditation in 2017, she has established comprehensive risk assessment and hospital-wide strategy through gap analysis. Her professional interests focus on healthcareassociated infection and achieve safe and clean environment by basic practices with high quality such as hand hygiene which is essential of infection prevention.


Background: Covert observation (CO) is a useful method as an effective and direct observation of hand hygiene (HH) compliance monitoring to overcome the Hawthorne effect. However, it is not clear whether the CO is effective in improving HH compliance. Accordingly, we conducted a study to determine the impact of covert observation on HH compliance. Method: We designed a prospective interventional study. Trained mystery shoppers monitored the entry and exit HH compliances in 3 different phases; the first CO (f-CO) in March and May 2017, CO combined with just-in-time coach (JITC) in July and October 2017, and second CO phases without JITC (s-CO) in March 2018. Overt observations (OO) were done simultaneously with above three phases by the infection control link personnel. We used Hand Hygiene Targeted Solutions Tool® for data collection. Results: Total of 10857 opportunities for HH were observed (3473 for f-CO, 2800 for JITC, and 1807 for s-OC) including 2777 opportunities for OO. The HH compliance rate of OO was 91.0% (2577/2777). HH compliance rate of f-CO was 34.3% (1192/3473). The rate was increased to 62.7% (1756/2800) by JITC (p=.00). HH compliance rate of s-CO was decreased to 57.5% (1039/1807) again but was still higher than the rate of f-CO (p=.00). HH rates on the exit opportunities were higher than the entry in all observations (89.7% (1257/1401) versus 92.3% (1270/1376) (p=.02), 31.5% versus 37.6%, 57.0% versus 69.0%, and 51.9% versus 62.9% for OO, f-CO, JITC, and s-CO respectively). Conclusions: CO method was useful for minimizing Hawthorne effect. However, CO was not effective for improving the HH compliance rate without the real-time feedback (JITC). Repeated JITC along with monitoring is needed to improve and maintain the good HH compliance.