Safety and quality

Staphylococcus aureus bacteraemia

The SAB data are provided by state and territory health authorities for public hospitals and by individual private hospitals and hospital groups. Data are only reported for hospitals where available—generally, this is for hospitals covered by a healthcare-associated infection surveillance program. Day hospitals are not usually covered by a healthcare-associated infection surveillance program as healthcare-associated infections are not generally diagnosed during a day hospital stay.

Three pieces of information are reported for public hospitals that have information on SAB:

the number of cases of SAB associated with care provided by the hospital (both admitted and non-admitted care) during the reported time period
the number of days of patient care under surveillance during the reported time period for the hospital (for hospitals with at least 5,000 days of patient care)
the SAB rate for the reported time period expressed as the number of cases of SAB per 10,000 days of patient care (for hospitals with at least 5,000 days of patient care).

For some private hospitals, only the SAB rate is reported. Patient days are not reported if they are regarded as commercial-in-confidence. Counts of cases are also not reported so that the number of patient days cannot be derived.

National benchmark

There is a national benchmark of no more than 2.0 cases per 10,000 days of patient care for acute care public hospitals in each state and territory. This benchmark was agreed by the Council of Australian Governments in the National Healthcare Agreement.

Calculation of measures

The SAB rate for a hospital is calculated by multiplying the number of SAB cases associated with care provided by the hospital by 10,000 and dividing by the number of days of patient care provided by the hospital that were under surveillance.

The number of cases of SAB reported is the number of healthcare-associated cases that were associated with care provided by each hospital. Cases include both methicillin- or multi-resistant Staphylococcus aureus (MRSA) and methicillin-sensitive Staphylococcus aureus (MSSA). Cases associated with care provided in healthcare settings other than hospitals are excluded.

Definition of a case of SAB

A case (or patient-episode) of SAB is defined as a positive blood culture for Staphylococcus aureus. For surveillance purposes, only the first isolate per patient is counted, unless at least 14 days has passed without a positive blood culture, after which an additional episode is recorded.

A Staphylococcus aureus bacteraemia will be considered to be healthcare-associated if: the first positive blood culture is collected more than 48 hours after hospital admission or less than 48 hours after discharge, OR, if the first positive blood culture is collected 48 hours or less after admission and one or more of the following key clinical criteria was met for the patient-episode of SAB:

SAB is a complication of the presence of an indwelling medical device (e.g. intravascular line, haemodialysis vascular access, cerebrospinal fluid shunt, urinary catheter)
SAB occurs within 30 days of a surgical procedure where the SAB is related to the surgical site
An invasive instrumentation or incision related to the SAB was performed within 48 hours
SAB is associated with neutropenia (<1 x 10⁹/L) contributed to by cytotoxic therapy.

Exclusions:

Cases where a known previous positive test has been obtained within the last 14 days are excluded. For example: If a patient has SAB in which 4 sets of blood cultures are positive over the initial 3 days of the patient's admission only one episode of SAB is recorded. If the same patient had a further set of positive blood cultures on day 6 of the same admission, these would not be counted again, but would be considered part of the initial patient-episode.

Note: If the same patient had a further positive blood culture 20 days after admission (i.e. greater than 14 days after their last positive on day 5), then this would be considered a second patient-episode of SAB.

Definition of a day of patient care

A day of patient care is a day, or part of a day, that a patient is admitted to hospital to receive treatment or care. The number of days of patient care reported for a hospital is the sum of the number of days spent in hospital by all patients who were discharged from hospital during the reported period.

The calculation of total days of patient care for the reported period for a hospital:

includes the total days of patient care of those patients discharged during the reported period including those admitted before the reported period
excludes the days of patient care for those patients admitted during the reported period who were not discharged until after the reported period

If part of a hospital was not included in SAB surveillance arrangements (e.g. children's wards, psychiatric wards), then the days of patient care for that part of the hospital are excluded. This means that the reported days of patient care may not be the total number of days of patient care provided by the hospital in the reported period.

Coverage

Some hospitals are not covered by SAB surveillance programs and the SAB pages note that data are not available for them. Some hospitals have incomplete coverage of SAB surveillance within the hospital. Within-hospital coverage may be affected by the exclusion of certain wards from SAB surveillance (such as psychiatric wards and children's wards). SAB surveillance data for Queensland includes only patients aged 14 years and over.

Interpretability and comparability

Each hospital's SAB rate is compared to the national benchmark for states and territories (public hospitals) of no more than 2.0 cases per 10,000 days of patient care for each state and territory.

Dial images are shown alongside statistics for SAB on the website to illustrate the performance of an individual hospital compared to the national benchmark.

For hospitals with fewer than 5,000 days of patient care in a year, the number of SAB cases reported is shown, however the SAB rate is not shown because rates can fluctuate widely for hospitals with small numbers of days of patient care.

The national benchmark was agreed by the Council of Australian Governments in the National Healthcare Agreement for acute care public hospitals for each state and territory. For illustrative purposes this benchmark has also been used when displaying data for private hospitals.

SAB rates have been rounded to two decimal places.

The images mean that the rounded SAB rates at a given hospital were:

	as low as possible (0 cases)
	between 0.01 and 0.49 cases per 10,000 days of patient care
	between 0.50 and 0.99 cases per 10,000 days of patient care
	between 1.00 and 1.49 cases per 10,000 days of patient care
	between 1.50 and 1.99 cases per 10,000 days of patient care
	2.00 cases per 10,000 days of patient care (on benchmark)
	between 2.01 and 4.00 cases per 10,000 days of patient care
	more than 4.00 cases per 10,000 days of patient care (more than twice the national benchmark)

Sometimes it is difficult to determine if a case of SAB is associated with care provided by a particular hospital. Counts therefore may not be precise where cases are incorrectly included or excluded. However, it is likely that the number of cases incorrectly included or excluded would be small.

If a case is detected by a hospital but is associated with care provided by another hospital then it is reported (where known) by the hospital where the care associated with the SAB occurred.

In almost all cases, SAB will be diagnosed when the patient is an admitted patient. The intention is that cases are reported whether they were associated with admitted patient care or non-admitted patient care in each hospital. However, no denominator (total days of patient care) is available to describe the total admitted and non-admitted patient activity of hospitals. Therefore the accuracy and comparability of the SAB rates among hospitals and over time is limited because the count of days of patient care reflects the amount of admitted patient activity, but does not reflect the amount of non-admitted patient activity. The amount of hospital activity that days of patient care reflect varies among hospitals and over time because of variation in case mix and admission practices. Since the cases of SAB reported includes cases associated with non-admitted hospital care this can increase the rate for hospitals that have a higher volume of activity and/or number of cases associated with non-admitted hospital care.

Hospitals provide a range of different services. Hospitals that undertake more invasive procedures may have a higher potential for SAB cases than hospitals that undertake fewer invasive procedures.

It is possible that hospitals that do not report SAB cases (because they are not covered by SAB surveillance arrangements) have a lower potential for SAB, especially if these hospitals undertake fewer invasive procedures than those hospitals that are covered by surveillance arrangements.

Hand hygiene

Back to 'About the data - Question 10'

Hand hygiene data are provided by state and territory health authorities for public hospitals and by individual private hospitals. The data are derived from audits of hand hygiene 'moments' that are conducted up to three times per year under the National Hand Hygiene Initiative.

Hospitals that provide information on hand hygiene report:

The estimated hand hygiene rate based on the sample of 'moments' observed in an audit.
The number of observed hand hygiene 'moments'.
The estimated rate is compared to the national benchmark and is reported as:
- higher than the benchmark
- similar to the benchmark, or
- lower than the benchmark.

Interim national benchmark

An interim benchmark of 70% for hand hygiene reporting on MyHospitals has been advised by the Australian Commission on Safety and Quality in Health Care.

Calculation of measures

The estimated hand hygiene rate for a hospital is a measure of how often (as a percentage) hand hygiene is correctly performed. It is calculated by dividing the number of observed hand hygiene 'moments' where proper hand hygiene was practised in a specified audit period, by the total number of observed hand hygiene 'moments' in the same audit period, and multiplying by 100. The rate is rounded to one decimal place.

Hand hygiene audits

Hand Hygiene Australia (HHA) defines the national hand hygiene auditing guidelines, sets standards of data quality used by hospitals and coordinates the National Hand Hygiene Initiative.

Audits are performed by individuals who have undergone HHA auditor training.

Hand hygiene is measured by observing a specified number of 'moments' (based on the number of occupied acute beds in a hospital, or the number of procedures performed annually in a private day hospital) to determine whether healthcare workers carry out hand hygiene as required.

As hospital wards provide different types of healthcare, such as surgical care or medical care, the audit process involves selecting wards for auditing.

Some of these are categorised as high risk (e.g. intensive care units, dialysis, and high dependency units) for hand hygiene auditing purposes. HHA uses three ward selection methods:

Option A - high risk wards with rotation of other wards
Option B - high risk wards with auditing of all other wards
Option C - intensive care unit with auditing of all other wards

Public hospitals with fewer than 25 occupied acute care beds and public hospitals that primarily provide mental health or rehabilitation care are not required to report hand hygiene data. These types of hospitals can voluntarily provide hand hygiene data for inclusion on the MyHospitals website; however, comparison with the national benchmark is not shown. Participation is voluntary for all private hospitals, including day hospitals. A comparison with the national benchmark is not shown for day hospitals with fewer than 2,000 procedures per year or for other private hospitals with fewer than 25 occupied acute care beds.

For most hospitals where the number of observed hand hygiene 'moments' was fewer than 100, data are not reported.

Hand hygiene 'moments'

HHA's hand hygiene 'moments' are based on those defined in the World Health Organization Guidelines on Hand Hygiene. Some minor modifications have been made for Australian health care conditions. A 'moment' is when there is a perceived or actual risk of pathogen transmission from one surface to another via a person's hands. HHA specifies five critical 'moments' when hand hygiene should be performed by healthcare workers. The ideal is to perform hand hygiene correctly in each of these five critical 'moments'. The five 'moments' are:

before touching a patient
before a procedure
after a procedure or body fluid exposure risk
after touching a patient
after touching a patient's surroundings.

For more information on auditing, 'moments' and ward selection methods see the HHA '5 moments for hand hygiene' manual (external link, opens in a new window) [http://www.hha.org.au/UserFiles/file/Manual/HHAManual_2010-11-23.pdf].

Hand hygiene rate estimates and confidence intervals

Because the hand hygiene rates are based on audits from a sample of hand hygiene 'moments', in a sample of hospital wards the calculation is only an estimate of the true rate for that hospital, and is associated with a 95% confidence interval.

A 'confidence interval' is a statistical term describing the range ('interval') within which we can be sure ('confident') the true rate falls. Confidence intervals indicate the reliability of the estimated rate and are calculated using data provided by hospitals.

When only a small number of 'moments' are audited, the confidence interval is larger, meaning we are less sure of the true rate. When a large number of 'moments' are audited, the confidence interval is smaller, meaning we are more sure of the true rate.

The table below shows how rates are compared to the national benchmark:

Estimated hand hygiene rate	Confidence interval of the rate overlaps with the national benchmark:	Reported as:
Lower than national benchmark	No	Lower than national benchmark
Lower than national benchmark	Yes	Similar to national benchmark
Equal to national benchmark	Yes	Similar to national benchmark
Higher than national benchmark	Yes	Similar to national benchmark
Higher than national benchmark	No	Higher than national benchmark

Each hospital's estimated hand hygiene rate is illustrated using a dial that indicates the hospital's rate with respect to the national benchmark. The three options are as displayed in these images:

	lower than the national benchmark
	similar to the national benchmark
	higher than the national benchmark

Interpretability and comparability

States and territories commenced the National Hand Hygiene Initiative over a staggered period. Therefore, hospitals that started later may have lower rates than hospitals that started earlier.

Differences in ward selection methods may also affect comparability of the data. Some states and territories use the HHA ward selection methodologies, and some use locally modified HHA ward selection methods. The ward selection method used can sometimes change from one audit period to the next. In general, for public hospitals:

New South Wales predominantly uses HHA ward selection method Option A and B, with a range of other ward selection methods applying in about 20% of hospitals
Victoria uses HHA ward selection method Option A
Queensland uses a localised approach that varies between facilities. However, most public hospitals use a ward sampling method in line with Option B
Western Australia predominantly uses HHA ward selection method Option B
South Australia uses a modified version of HHA ward selection method Option C, which contains aspects of Options A and C, whereby all intensive care units and high-risk wards are sampled plus a selection of general wards depending on the size of the hospital
Northern Territory uses HHA ward selection method Option A
The Australian Capital Territory uses HHA ward selection method Option A and C
Tasmania uses HHA ward selection method Option A