PDB26 Evidence Requirements in Australia for Drug Reimbursement: The Survival of Cost Minimisation Analysis

Objectives:
Over a decade ago, Briggs and O'Brien declared the ‘death' of cost-minimisation analysis (CMA). The principle of CMA, however, as described by Drummond is still being considered valid by regulators such as the Australian Pharmaceutical Benefits Advisory Committee (PBAC), which recommends which drugs should be reimbursed under the Pharmaceutical Benefits Schedule (PBS). This study sought to examine the evidence needed for drug reimbursement in Australia for diabetes drugs when presenting a CMA.

Methods:
Current PBAC guidelines were reviewed specifically from a cost minimisation point of view. Public summary documents (which summarise PBAC deliberations) for all reimbursement decisions related to diabetes drugs over the past 5 years were extracted. Data pertaining to clinical claims, economic analyses and decision were analysed.

Results:
Public summary documents were reviewed for eight diabetes drugs: insulin glulisine, insulin detemir, liraglutide pioglitazone, rosiglitazone, saxagliptin, sitagliptin, vildagliptin. Of these, five reimbursement submissions were based on CMAs, two on cost effectiveness analyses (CEA) and one on both. The CMA submissions demonstrated non-inferiority when compared to the nominated comparator either by presenting head-to-head clinical trial evidence, adjusted indirect comparisons or pooled individual patient data (IPD) analyses. All were recommended for reimbursement in Australia. Notably, budget impact of each was limited (<AU$10 million over a 5-year period). The three CEA submissions claimed superiority of their products over the nominated comparators either in terms of efficacy or safety but none were successful. Notably, each was associated with a considerable budget impact.

Conclusions:
Far from being dead, CMA remains alive and well in Australia. The PBAC accepts CMAs on the basis of non-inferiority and perhaps also if reimbursement is associated with a limited budget impact. Acceptable evidence is not restricted to head-to-head trials, but also includes adjusted indirect comparisons as well as IPD pooled from multiple clinical trials.