Systematic Review of Studies on Compliance with Hand Hygiene Guidelines in Hospital Care.
Erasmus et al. Infect Control Hosp Epidemiol 2010; 31: 283 - 294
We've all heard that hand hygiene is important to prevent healthcare associated infections. And we've also heard that in most hospitals, hand hygiene compliance is poor. In 2004 an editorial commented that in the United States " hand hygiene rates average 40 - 60% on a good day." (Weinstein 2004) We now have an article available that tells us exactly how poor hand hygiene compliance is in hospitals, based on published studies. And Weinstein's range seems to be fairly accurate.
Performing a comprehensive search for published articles that described compliance with hand hygiene in hospitals from countries with 'established market economies', these authors identified a final set of 96 articles. In these articles compliance with hand hygiene guidelines was assessed either by directly observing healthcare workers (90% of articles), or by self-reporting by healthcare workers.
Compliance was defined as the percentage of time healthcare workers cleaned their hands when they were supposed to (a 50% compliance means that healthcare workers cleaned their hands during half of all the times they were supposed to clean them). In two-thirds of these articles hand hygiene compliance was studied in intensive care units. Most of these articles were from the US, Canada, Europe and Australia.
The reported compliance across studies ranged from a low of 4% to 100%. Ranking the results from all the studies and for all types of healthcare workers (meta-analysis could not be done), the median hand hygiene compliance rate was only 40%. This means that in half the studies, healthcare workers cleaned their hands less than 40% of the time. In three-fourth of the studies, hand hygiene compliance was 50% or less. In other words, only a quarter of the studies found compliance rates higher than 50%. Even for the subset of ICU studies, the results were pretty similar: the median was 40 - 50%, and 72% of studies reported a compliance of 50% or less.
What about differences between doctors and nurses? Although the authors of this review did not report direct comparisons of physicians with nurses within each hospital, it seems that hand hygiene compliance is worse among doctors than among nurses. There were more published studies reporting that doctors were poorly compliant than were studies reporting that nurses were poorly compliant . Specifically, if we define 'poor hand hygiene compliance' as a compliance rate less than 50%, almost 80% of studies in this review found that doctors were poorly compliant, whereas almost 60% of studies found that nurses were poorly compliant.
This review and the included studies had some limitations. The included studies were heterogenous in their methods (hence a meta-analysis could not be done) and often did not report the specifics of how observation of hand hygiene was performed. Many of them used direct observation and self reporting to measure hand hygiene compliance which are not the ideal methods of measurement. Healthcare workers may change their behavior if they are being observed by hand hygiene auditors. But if this happened, one would expect the measured hand hygiene compliance to be falsely high. The fact that the reported compliance rates were so low in spite of directly observed measurement is alarming, because unobserved rates then are presumably even lower.
Reference
Weinstein RA. Hand hygiene - of reason and ritual. Ann Intern Med 2004; 141: 65-66
Comments