Prescribed List Reforms - Consultation Paper 10 - General Use Items utilisation, expenditure and integrity

The key areas of concern, given the retention of GUIs in Part D of the PL, continue the risks that were outlined by EY's 2020 Review of the General Miscellaneous Category of the Prostheses List and articulated in our previous responses to the consultation regarding the EY review and to Consultation Paper 5 on Bundling of Benefits for General Use items.
We remain of the view that the existing structure encourages and supports inefficient utilisation of GUIs; gaming of the structure of GUIs that result in maximising use of formulations of listed devices that have higher listed benefit amounts over those with lower listed benefit amounts; and risk of exposure to claims for listed items that were either not qualified for use of PL rebate codes or not used in the surgery to which the claim pertains.

Mechanisms and structures that incentivise efficient utilisation and minimises waste would deliver the most meaningful benefits to the system and patients.

Remove GUIs from the PL as originally intended with the PL reforms.
Short of this, closely monitor and publish utilisation levels of each GUI rebate code at a procedure type and DRG level.
Implement transparency measures within the HPP that clearly denote what items are covered by any individual PL rebate code.

Items that have a higher risk of over-utilisation and waste are those which have potential wide applicability to a range of procedures and have varying formulations with differing benefit levels and for which accuracy of documentation of use in procedure records is usually vague or absent (especially as it pertains to measures of quantity used and amounts not used [or wasted]).
Examples include - Haemostatic devices, Internal adhesives, Adhesion barriers, and Dura defect repair items
Other items include those which generally have non-PL-listed alternatives that are funded through existing alternative mechanisms - e.g., Infusion pumps.

These were outlined by EY in their review report and others were specified in our responses to previous consultations as stated in the answer to Q1.
IHACPA's review also identified instances of high variability in utilisation which is an indicator standard clinical practices may not necessarily require use of certain GUIs.

Examination of episode-level data can elucidate such patterns.
IHACPA can undertake this analysis for the whole of sector.
For example, the data would find that following the introduction of Dermabond Prineo on the PL, utilisation of the item in knee replacement procedures rapidly increased to plateau at about 50% of episodes. Thus, in half of episodes, Dermabond Prineo was not required as a clinical need to close the wound. Closer examination of the data would find that utilisation was not determined by patient characteristics but rather by clinician and hospital preference. Some facilities had almost no utilisation of the item for their knee replacements while others used the item in almost all cases of knee replacement. However, there is no observable difference in patient outcomes that would align with such variability in utilisation.

IHACPA had previously commenced this work in advising regarding bundling GUIs.
However, IHACPA failed to include in their calculations for the denominators on utilisation, episodes (on a similar DRG basis) which had zero utilisation of specific GUIs. Thus, IHACPA failed to acknowledge the shift of utilisation across the system from cases with zero utilisation of specific GUIs to one or more GUIs.
If IHACPA was tasked with measuring and monitoring all such patterns of utilisation of GUIs, then mechanisms could be installed to disincentivise and/or mitigate the impacts from inappropriate over-utilisation and prevent the system from degrading to greater inefficiency in use of GUIs.
One such mechanism may be to determine and set existing efficient utiilisation rates specific to existing casemix and evaluate what the utilisation ought to have been for a subsequent period of utilisation and casemix if utilisation had not inflated. If utilisation inflated, then the benefit amounts could be adjusted downwards to, in effect, cap benefit outlays at the previously determined and set efficient utiilisation levels.