PHS61 - Using Big data for an Economic Evaluation of reducing the Innappropriate prescribing of atypical Antipsychotics among Dementia Patients

Abstract

NPS MedicineWise Australia developed and implemented a multi-faceted academic detailing program to improve general medical practitioner prescribing of antipsychotic therapy. A significant component of this program focused on the appropriate use of medicines for older people with dementia. Evidence has demonstrated that older adults in community and residential aged care facilities are prescribed antipsychotics for the management of their dementia symptoms despite lack of evidence of benefit. This study is an economic evaluation of the NPS MedicineWise educational program in terms of reduced falls and stroke incidence due to reduction in antipsychotic prescribing including olanzapine, quetiapine and risperidone A cost-benefit analysis was performed to evaluate the costs and benefits of preventing hospitalisations due to strokes and falls as a result of reduced antipsychotic prescribing due to the program. A decision tree model was constructed using TreeAge Pro 2014. Data sources were from our clinical audit, general practitioner survey and interrupted time series analyses of administrative data from the Australian Pharmaceutical Benefits Scheme (PBS). A meta-analysis was conducted to provide outcome data on the reduction of antipsychotics. Probabilistic sensitivity analysis was performed to address any parameter uncertainty using Monte Carlo simulation. Following implementation of the NPS MedicineWise intervention, a reduction in prescription of atypical antipsychotics among older adults was observed in the PBS. Modelling of this reduction translated into a reduction in the incidence of both stroke and fall hospitalisations among this cohort resulting in costs avoided for Government. This is a comprehensive economic analysis conducted in Australia to have utilised PBS data and large scale evaluation results to examine interventions focussing on the appropriate use of medicines in older adults. The results will be used to further inform future public health interventions aimed at managing the health outcomes of this vulnerable and growing population.