Comparisons of hospitals

When deciding on the specifications of an indicator, the Authority investigates and implements approaches to optimise fair comparisons of hospitals across Australia. The Authority implements as many of these approaches as can be supported by the depth and quality of data available. For this report, the Authority consulted with the Independent Hospital Pricing Authority (IHPA) and report-specific advisory committees.

The following five approaches were used by the Authority to calculate Comparable Cost of Care and support fair comparisons between hospitals:

  1. Comparable costs: this process involved a review of the national consistency of cost information and the materiality of any differences between states and territories. Where appropriate, some costs are excluded to support comparability. In other instances, some costs are included because it is not possible to exclude them. In these instances the materiality of this approach was assessed
  2. Units of activity: this process is necessary to standardise costs by accounting for the differences between the complexity of patients admitted at a hospital and the patient’s individual characteristics which lead to higher costs, relative to the patient’s length of stay
  3. Rounding results: the Authority has rounded Cost per NWAU and Comparable Cost of Care in a way that acknowledges any remaining uncertainty in estimates
  4. Suppression of results: where cost data from a hospital were not fully available or comparable, the Authority determined rules that informed decisions to suppress information based on a lack of comparability
  5. Peer groups: this process involves grouping hospitals so that they can be compared to their peers.

Comparable Cost of Care

Emergency department costs

The Authority has previously reported variation in the length of time a patient stays in an emergency department (ED).4For the purposes of completeness and comparability, this measure included the ED presentation costs of acute admitted patients for each hospital.

This allowed fair comparison of the costs associated with the journey of acute admitted patients from the point of entering the ED, until the point of discharge. The ED presentation costs were matched with the individual admitted patient unit record by IHPA using the state record identifier field provided by state and territory governments.

Blood costs

Blood costs are treated differently within and across jurisdictions. Consequently, these costs have been excluded from the measure Comparable Cost of Care.

Teaching, training and research costs

In 2011–12, there was no nationally agreed definition of teaching, training & research (TTR) costs. An independent review by PricewaterhouseCoopers (PwC) indicated some jurisdictions removed these costs before supplying data to the NHCDC.6 Accordingly, where hospitals recorded TTR direct costs, these were excluded from the measure Comparable Cost of Care. At the time of data preparation states and territories were consulted to confirm the exclusion of TTR direct costs from the acute care type. The Authority is aware that the extent to which jurisdictions excluded TTR direct costs from their data submissions varied in 2011–12. Given the quantum of these costs the Authority did not consider jurisdictional variation to materially affect the measures in the report.

Excluded costs

The costs classified as ‘excluded costs’, as defined by the Australian Hospital Patient Costing Standards v2.0, are costs which cannot be allocated to any other costing group. These are immaterial to this report and represent costs which are not consistently reported across jurisdictions.

Due to the variation in reporting practices, these costs were excluded to ensure the fair comparison of public hospitals.

Payroll tax costs

Some hospitals reported the payment of payroll tax in their submission to the NHCDC. Due to the variation in reporting practices, these costs were excluded to ensure the fair comparison of public hospitals.

Medication costs

Some states and territories are not signatories to the Pharmaceutical Reform Agreement. To achieve fair comparison the report has excluded the costs of medications subsidised by the Pharmaceutical Benefits Scheme (PBS).


Depreciation is the allocation of the cost of an asset as an expense over the life of the asset or the period in which it facilitated the generation of income. Depreciation rates are different across jurisdictions and not all jurisdictions provided these costs to the NHCDC.

Due to the variation in practices, this report has excluded depreciation costs to ensure the comparability of public hospitals.

Private patients

The collection of private patient medical expenses is challenging in the NHCDC. Factors influencing the collection of these costs include the use of Special Purpose Accounts and Trust Funds by some hospitals.

In some instances, medical practitioners are reimbursed by these accounts/Trust Funds for the treatment of private patients, or these funds may be used for other hospital expenses. As a result of this practice, some expenses may not be recorded in the hospital’s general ledger and subsequently are not provided as part of the NHCDC submission.7

The impact of costs not being included in the general ledger is that the costs submitted to the NHCDC, and subsequently used in the report, are lower than the actual expenditure.

To improve the comparability of hospitals with differing private and public patient casemix, the Comparable Cost of Care measure inflates the cost of all patients at a hospital based on revenue received by the hospital for private patients (as detailed in the Department of Health’s data collection entitled Hospital Casemix Protocol):

  • charges levied by a hospital
  • benefits paid by private insurers
  • claims made to the Medical Benefits Scheme (MBS).

This approach models private patient costs for a public hospital that are not reported in the general ledger; and therefore not reported to the NHCDC.

This approach is similar to that used by the Cost per NWAU measure. However, the Cost per NWAU measure, after inflating all patient costs, reduces the costs by the modelled private patient revenues. In addition, the NWAU is discounted for private patients. Therefore, patients admitted for a condition/procedure whose treatment is funded by private sources are allocated a lower NWAU than those patients whose treatment is eligible for funding under the National Health Reform Agreement.

Admitted and discharged within fiscal year

The data used for this report are limited to unit records where the patient was admitted and discharged in the 2011–12 financial year. This is based on the independent reviews of the NHCDC for 2010–11 and 2011–12 which highlighted inconsistencies between states, and therefore hospitals, in recording costs for patients discharged during this financial year who had been admitted during the previous financial year.

Some states and territories provided a variation of unit records to the NHCDC submission, for example providing all discharged patients or only discharged patients admitted in the current fiscal year. Therefore the data used has been limited to patients admitted and discharged within the 2011–12 financial year.

Units of activity

AR-DRG v6.0x was used to group data from the APC NMDS for each individual patient unit record. The AR-DRG is based on the codes allocated to diagnosis and procedures that are recorded in the patient medical record for each episode of care.

Each AR-DRG is allocated a defined ‘cost weight’ (calculated using comparable costs), which is a relative measure of a patient’s complexity, calculated as the ratio of the average cost of a given AR-DRG compared to the average cost of all AR-DRGs, for hospitals submitting data to the NHCDC 2011–12. For this report the weights have been calculated:

  • Using the average in scope costs (costs included in the comparable costs)
  • Using patient unit records where patients were discharged this financial year and admitted in the previous or current financial year
  • Using costs where patients were admitted in the current financial year and not yet discharged
  • Excluding hospitals, repotrint ED activity with no ED costs.

This report calculates weights using the AR-DRG v6.0x and adjusts this weight based on individual patient characteristics which are known to lead to higher costs.

The following adjustments were made to the weights:

  • Specialist paediatric
  • Specialist psychiatric
  • Indigenous status
  • Remoteness
  • Intensive care unit (ICU), level III
  • Radiotherapy.

Rounding results

To address any remaining uncertainty in relation to the accuracy of the results, each hospital’s result has been rounded to the nearest $100 for Cost per NWAU and Comparable Cost of Care.

Peer classification system

Peer groups allow hospitals to be compared to other similar hospitals. They minimise the effect caused by hospitals of differing size, service provision and rurality when comparing hospitals.

The peer group version used in this work is based on the peer classification, established by the Australian Institute of Health and Welfare, that existed in 2011–12. These peer groups categorise hospitals according to size and type.

The report focuses on comparing and contrasting information from major and large public hospitals, as these hospitals account for the vast majority of same-day and overnight admissions.

Hospitals in the major peer group are then split into metropolitan and regional groups using the Australian Standard Geographical Classification (ASGC) Remoteness Area, 2006.

The report includes major and large public hospitals (A1.1, A1.2, B1).  The website includes these hospitals plus large regional public hospitals (B2).