Influence of role models and hospital design on hand hygiene of health care workers - Research

Emerging Infectious Diseases, Feb, 2003 by Mary G. Lankford, Teresa R. Zembower, William E. Trick, Donna M. Hacek, Gary A. Noskin, Lance R. Peterson

Hand-hygiene compliance on room entry was significantly greater in the old hospital at 12% (36/304) compared to the new hospital at 6% (26/424) (p=0.006). After all hand-hygiene opportunities were assessed, we found that hand-hygiene compliance was significantly better in the old hospital compared to the new hospital (161/304 [53%] vs. 97/417 [23%]; p<0.001). Hand-hygiene compliance was significantly better after a hand-hygiene opportunity (258/721; 35.7%) compared to before a hand-hygiene opportunity (62/727; 8.5%; p<0.001). By univariate analysis, characteristics significantly associated with hand-hygiene compliance after a hand-hygiene opportunity included working at the old hospital, having patient contact, performing an invasive procedure, using gloves, and performing hand hygiene on room entry. A key finding was that when a higher ranking person in the room did not perform hand hygiene, other health-care workers were significantly less likely to wash their hands (Table 2).

During multivariate analysis, we identified the following independent predictors of hand-hygiene compliance: using gloves, performing an invasive procedure, working at the old hospital, performing hand hygiene on room entry, and having patient contact. Again, health-care workers present in the room with a higher ranking person or peer who did not perform hand hygiene were significantly less likely to wash their hands (Table 3).

When we further evaluated group behavior, we found that compared to single person room entry, health-care workers in a room with a higher ranking person who did not wash were significantly less likely to wash their own hands. In each of these episodes, the higher ranking person was a physician or nurse. Surprisingly, if either a higher ranking person or peer was in the room and performed hand hygiene, then the frequency of hand hygiene for others in the group was no better than that of a room which only one person entered (Table 4). This observation suggests that the effect of a role model is highly significant but most potent in negatively influencing hand-hygiene behavior.

Discussion

Despite construction of a new hospital with an increased number of sinks, we found that hand-hygiene compliance in the new facility decreased substantially. We demonstrated that health-care workers were significantly less likely to wash their hands if they were in a room with a peer or higher ranking person who did not perform hand hygiene. Not unexpectedly, hand-hygiene compliance was better after patient contact, performing an invasive procedure, and removing gloves.

Health-care workers were much less likely to perform hand hygiene if a peer or a higher ranking person in the room did not perform hand hygiene. Compared to health-care workers who entered a room alone, group behavior did not seem to improve if the higher ranking person or peer did wash their hands. Although these findings suggest that hand-hygiene behaviors can be affected by role model or peer hand-hygiene compliance, learned behaviors or time constraints may negatively influence group compliance with hand-hygiene procedures.