Hand hygiene and the hawthorne effect in the Emergency Department

Abstract

Background: Hand hygiene (HH) is an effective way to reduce hospital-acquired infections.(1) The Hawthorne observer bias effect may overestimate compliance but is difficult to measure.(2) HH improvement in Emergency Department (ED) requires multimodal approaches. (3) Before implementing a “High Five for Hand Hygiene” program(4) we wanted to investigate observer bias. Objectives: Develop a tool to measure HH observer bias; and determine effect size in ED. Methods: Two investigators independently audited health care workers (HCW) on inpatient wards using standard technique. We recorded if HCW was; not aware; possibly not aware; unknown; possibly aware; or aware of being observed. Gold standard was determined by asking the HCW. One investigator audited HH in ED during three two-week periods; low visibility (no identification); standard technique; and high visibility (fluorsecent vest). Awareness was recorded. Results: Of 45 paired ward observations 6 HCWs were aware of observation (13%). Inter-rater agreement was “very high” (Kappa 0.81). Using “aware” and “not aware” alone, observer one sensitivity and specificity was respectively 75% (95%CI 21-98) ;100% (95%CI 87-100); and 80% (95%CI 30-99);100% (95%CI 88-100) for observer two. Of 817 ED HH observations awareness was recorded in 95%. Overall HH compliance was 71.8% (low visibility 63%, standard 73.5%, high visibility 76.8%; p=0.001). Using the above tool compliance was 79% when aware and 61% when unaware (p < 0.001). Conclusions: The Hawthorne effect may be measured with reasonable sensitivity and high specificity. This (Figure Presented) finding requires validation. Hand hygiene compliance of Emergency Department health care workers increased by 18% when staff were aware they were being observed.L6221672682018-05-22 <br />