This report looks at the rate of bloodstream infection caused by one type of bacteria, called Staphylococcus aureus (S. aureus), that patients sometimes develop during the course of medical care or treatment provided by hospitals.
Although commonly found on the skin or in the nose or throats of healthy people, S. aureus can cause serious health complications for some patients and significant extra costs to the health system. In the most severe cases, S. aureus bloodstream infection can prove fatal; between one-fifth to one-third of people who experience this type of infection die as a result.
The annual number of healthcare-associated S. aureus bloodstream infections is small compared to the total number of patients that pass through Australia’s hospitals each year. However, every case of S. aureus bloodstream infection is considered potentially preventable, so there has been an increasing focus on monitoring its impact and introducing strategies to combat it.
The National Healthcare Agreement signed in 2011 sets a target for no more than 2.0 cases of healthcare-associated S. aureus bloodstream infection per 10,000 patient bed days for each state and territory. The rate of S. aureus bloodstream infection has been set as one of the 48 indicators agreed by the Council of Australian Governments to guide the Authority’s work.
This report uses the best available data to show how many healthcare-associated S. aureus bloodstream infections were reported by each of 352 public hospitals in 2011–12. The intent is to inform consumers, help hospitals and health service managers to identify opportunities to reduce the rate of preventable infection, and to stimulate improvements in the collection and consistency of national statistics.
Note: Since publication in May 2013, some figures have been revised following updates to methods and revised information from states and territories. The most up-to-date results for each hospital are available on the hospital results pages.
There were 1,725 cases of healthcare-associated S. aureus bloodstream infection in 2011–12 reported by the 352 public hospitals covered by this report. This represents over 99% of all cases of this type of infection reported by public hospitals nationally.
The rates of infection varied markedly between different public hospitals, even within groups of similar hospitals (peer groups). This suggests there may be opportunities at a number of hospitals to reduce infection rates, improve detection and infection reporting systems, or a combination of both.
The report has found two main factors affected the rate of S. aureus bloodstream infection within each hospital, namely:
- Hospital size
- The proportion of vulnerable patients within each hospital.
Vulnerable patients are people who have one or more of the risk factors shown to increase the chance of acquiring a healthcare-associated S. aureus bloodstream infection. For more information see Vulnerable patients.
To permit fairer comparisons, the Authority allocated each hospital to one of eight peer groups, according to its size and the proportion of its patients deemed at higher risk of developing a healthcare-associated S. aureus infection (Appendix table 1).
Bigger hospitals have higher rates
The two peer groups covering the largest hospitals, referred to as major hospitals, accounted for a disproportionately large share (82%) of all cases reported nationally.
The two peer groups covering the next category, large hospitals, accounted for 6% of all reported cases, while medium hospitals accounted for 5% of all reported cases, and small hospitals accounted for 2% (Figure 1a and Figure 1b).
Hospitals with more vulnerable patients have higher rates
Hospitals with higher percentages of vulnerable patients also reported more cases of S. aureus bloodstream infection in 2011–12 than hospitals with fewer vulnerable patients.
In major hospitals with more vulnerable patients, the average rate of healthcare-associated S. aureus bloodstream infection was 1.38 cases per 10,000 patient bed days.
In major hospitals with fewer vulnerable patients, the average rate of infection was 0.90 cases per 10,000 patient bed days.
Large, medium and small hospitals with more vulnerable patients had higher average rates of healthcare-associated S. aureus bloodstream infection than hospitals of similar size but fewer vulnerable patients (Key findings).
Big differences in infection rates between similar hospitals
Wide differences were seen in the rate of healthcare-associated S. aureus bloodstream infection reported by hospitals within the same peer group.
This finding is important because it suggests a component of the variation in infection rates between hospitals may relate to the performance of the hospital, and not to differences in the types of patients seen or treatments provided.
Among major hospitals with a higher proportion of vulnerable patients, the rate of reported healthcare-associated S. aureus bloodstream infection was over four times higher in some hospitals compared to others in the same group. The range reported was from 0.47 cases per 10,000 patient bed days to 2.15 cases per 10,000 patient bed days (Figure 1a).
In major hospitals with fewer vulnerable patients, the rate of reported healthcare-associated S. aureus bloodstream infection was over 15 times higher in some hospitals compared to others, ranging from 0.17 cases per 10,000 patient bed days to 2.59 cases per 10,000 patient bed days (Figure 1a).
Each of the remaining six peer groups (two peer groups for each of the size categories covering large, medium and small hospitals) had one or more hospitals that reported no cases of S. aureus bloodstream infection.
In Figure 1a and Figure 1b, the Authority has named the hospitals that have reported the highest and the lowest rates of healthcare-associated S. aureus bloodstream infection in the two major hospital peer groups.
Hospitals in other peer groups have not been named for a number of reasons. In smaller facilities, a single extra case can be enough to cause a dramatic increase in the annual rate for a hospital, while zero cases reported by some hospitals could reflect more immature data collection systems rather than zero infections.
Many hospitals and states have been collecting data on healthcare-associated infections for a number of years. However, submission of healthcare-associated S. aureus bloodstream infection figures to national data collections is relatively recent, and interpretations of new definitions may vary between hospitals.
Data is typically most consistent in the major hospital peer groups, which see large numbers of patients and have sophisticated healthcare-associated infection monitoring and reporting systems.
The figures for these and the other peer groups only reflect confirmed cases of S. aureus bloodstream infection. Other cases that were not confirmed or reported - for example, because tests were not done, or because test results were not properly recorded - could comprise extra cases not currently included in the results.
The Authority is aware of these limitations of the data and is undertaking work to improve future reporting on the comparable performance of healthcare organisations.
Information on healthcare-associated S. aureus bloodstream infection for more than 600 public and private hospitals is available on the National Health Performance Authority website at www.myhospitals.gov.au