ETC 46 - Engaging Physician Learners in Covert Hand Hygiene Observations: A Pilot Educational and Infection Control Initiative

Background: Hand hygiene (HH) is a widely recognized key standard practice to preventing healthcare-associated infection. Traditional direct observation methods are not reliable due to observer bias. Automated measurement systems generally do not capture specific moments of HH. Covert HH monitoring is a promising way of overcoming these limitations yet is difficult to sustain longitudinally.

Methods: We engaged physician learners in measuring HH adherence during their clinical rotations. The program was introduced during orientation and volunteer learners attended 1-hour training that included didactic learning and simulations. Data was entered into an application on personal devices, linked with a central server. Learners were not directed on which units or situations to observe, but rather to observe HH adherence within their usual workflow from 1 September to 30 November, 2023. Infection Control (IPC) auditors not known to the unit completed parallel audits. Learner and IPC audits were compared using the Chi-square for quality assurance purposes.

Results: Aggregate HH adherence collected by covert learners (n=4) of 51%(368/721) was not significantly different from that collected by IPC auditors (n=8) 51%(211/418) (p=0.8). Including only observations before and after patient/environment contact, HH adherence collected by learners was similar to that collected by IPC auditors; 49%(326/670) compared to 51%(205/404) respectively(p=0.51). HH adherence observed by learners by opportunity were: 45% before patient/environment contact; 88% before aseptic procedure; 77% after body fluid exposure; 53% after patient/environment contact. HH adherence observed by learners by healthcare worker group were: 48% physicians, 58% nursing, 76% allied health, 38% support staff, and 27% other.

Conclusions: Engaging physician learners in covert HH observation provides similar results to IPC auditors who are not known to the unit, suggesting learner data accurately reflects HH practices. This method better enables bedside HH observations, generally missed by traditional observation methods. Further evaluation is needed to confirm that covertness of observations can be maintained.