To examine whether improved hand hygiene compliance in health care workers after a statewide hand hygiene campaign in New South Wales hospitals was associated with a fall in rates of infection with multiresistant organisms.

Data on rates of new methicillin-resistant Staphylococcus aureus (MRSA) infections (expressed as four clinical indicators) are reported by some Australian hospitals to the Australian Council on Healthcare Standards (ACHS) for accreditation purposes and are mandatorily reported by all NSW hospitals to the NSW Department of Health. Infections are classified according to whether they are acquired in the intensive care unit (ICU) or other wards and whether they are from sterile sites (blood cultures) or non-sterile sites. The clinical indicators reflect four different site categories (ICU sterile site, ICU non-sterile site, non-ICU sterile site and non-ICU non-sterile site) and are expressed as the number of new health care-associated infections per 10 000 acute care bed-days. Clinical indicator rates were examined for any decline between the pre-campaign period (July–December 2005) and post-campaign period (January–July 2007), and were compared with trends over a similar period in states without a hand hygiene campaign.

Pre-campaign and post-campaign rates for four MRSA clinical indicators.

Between the pre- and post-campaign periods, there was a 25% fall in MRSA non-ICU sterile site infections, from 0.60/10 000 bed-days to 0.45/10 000 bed-days (P = 0.027), and a 16% fall in ICU non-sterile site infections, from 36.36/10 000 bed-days to 30.43/10 000 bed-days (P = 0.037). The pre- and post-campaign rates of MRSA infection from ICU sterile sites (5.28/10 000 bed-days v 4.80/10 000 bed-days; P = 0.664) and non-ICU non-sterile sites (5.92/10 000 bed-days v 5.66/10 000 bed-days; P = 0.207) remained stable. Australia-wide MRSA data reported to the ACHS showed a 45% decline in infections from ICU non-sterile sites, from 25.89/10 000 bed-days to 14.30/10 000 bed-days (P < 0.001), and a 46% decline in infections from non-ICU non-sterile sites, from 3.70/10 000 bed-days to 1.99/10 000 bed-days (P < 0.001) over the period 2005–2006.

Two out of four clinical indicators of MRSA infection remained unchanged despite significant improvements in hand hygiene compliance in NSW hospitals. The reduction in MRSA infections from ICU non-sterile sites in NSW hospitals was mirrored in ACHS data for other Australian states and cannot be assumed to be the result of improved hand hygiene compliance. Concurrent clinical and infection control practices possibly influence MRSA infection rates and may modify the effects of hand hygiene compliance. More sensitive measurements of hand hygiene compliance are needed.