The final business case included a service profile for the new RCH. This profile was determined by a service planning review undertaken in 2003-2004 and formed the basis of a planned facility configuration for the new hospital. The new RCH will have 50 more beds than the existing hospital and the design of the new RCH provides flexibility for future expansion.
The RCH service plan defined the services to be provided by the RCH, based on demographic and activity projections to 2016. The service review incorporated service demand modelling and a review of factors including:
• RCH service delivery levels and admissions data
• service demand projections based on a DHS service planning tool and forecasting model
• ABS data
• consultation with consumers, stakeholders and staff.
The service planning for the new RCH used the data available at the time as well as the results of a forecasting model provided by DHS. As with any forecasting model there were some known limitations in the model used to forecast service demand for the new RCH.
These limitations related to issues around:
• assumed maximum lengths of stay for patients
• assumed bed occupancy rates
• moderation (or capping) of the allowable levels of increase and decrease in annual admission rates.
DHS advised us that forecasting models are only the start of the service planning process, and that in applying the forecast model, DHS indicated that it reviews actual experience and consults with clinicians to identify areas where the output of the model needs to be adjusted.
DHS and RCH could not provide evidence confirming how any advice from clinicians and other experts was taken into account to address limitations in the forecasting model to enable appropriate adjustments to be made to forecasts of admitted patient activity for the new RCH.
However, notwithstanding the lack of evidence of the basis for adjustments to the model forecasts, it is clear that adjustments were made.
A similar issue arose in relation to the forecasting of future activity for non-admitted patients (such as those patients using outpatient clinics). The DHS forecasting model was not used for these forecasts and another method was applied. RCH could not provide evidence that RCH clinicians were given the opportunity to review whether the forecasts for non-admitted activity were appropriate.
We also observed issues around the extent to which DHS understands the development, operation, and internal logic of the forecasting model. The model used by DHS is a proprietary model owned by a consulting firm and is used for statewide service demand forecasting by DHS. DHS advised that the model has been validated in other jurisdictions but was unable to provide specific details about the development and operation of the forecasting model.
While it is clear that service planning for the new RCH used the data available at the time, in the absence of relevant evidence, we have been unable to conclude on how well that data was applied in developing service demand forecasts for the new RCH.