Doctor, do you have a moment? National Hand Hygiene Initiative compliance in Australian hospitals

In reply: Doctors' low rate of compliance with hand hygiene practice at the hospital level will continue to be concealed as long as rates for health care professional groups are aggregated. Neither the Hand Hygiene Australia (HHA) website nor the MyHospitals website provides the individual hospital rates for nurses and doctors separately and by moment for independent public scrutiny.

We are delighted that Grayson and colleagues are now cautious about linking Staphylococcus aureus bloodstream (SAB) infection and hand hygiene. When HHA plotted SAB infection over the first 2 years of the program,1 we could not reconcile their findings and counselled against this unwise causal linkage.2 Our generalised linear regression model specific for Poisson distribution offset by patient-days highlighted that, even with this superior analysis, linking this statistically rare event with hand hygiene performance will remain unproductive, owing to complex causal reasons.

Our cost estimates included the fourth quarter, as hospitals continue auditing to collect sufficient numbers of observations. Even with this additional period, our estimate was conservative, using 2.2 minutes per observation and a mid-level hourly rate. We excluded 50% of the estimated training costs and indirect costs of auditing.

Moment 1 compliance gauges clinicians' commitment to patient safety, and it is the most common indication for hand hygiene in all hospitals. Moments 2 and 3, before and after a procedure, are infrequently indicated, and we did not review these because of the additional focus these two indications receive from the mandatory aseptic technique policy.3 Moment 3, after exposure to body fluid, is universally well performed.4 Moments 4 and 5, after contact with a patient and their environment, have poor but higher compliance than Moment 1, as clinicians interpret these indications as an opportunity for self-protection.5

Drilling down through the data showed that actual levels of compliance over 5 years had been obscured.6 Any concerns that our finding may undermine the HHA program should be redirected to addressing both the uncomfortable knowledge that doctors are underperforming and the very real potential for an audit-associated Hawthorne effect inflating compliance. Using evidence-based findings is vital to the next phase of the quality improvement cycle. The goodwill generated by HHA should now be used to resolve poor performance with regard to Moment 1.