AN EVIDENCE-BASED APPROACH TO PREDICT LONG-TERM PATIENT OUTCOMES AND KEY VALUE DRIVERS
Author(s)
Baid H, Priedane E, Al-Dakkak I, Kumar A, Gagnon J
Pope Woodhead & Associates Ltd, St. Ives, UK
OBJECTIVES: Given the unique issues associated with clinical trials for rare indications, there is usually limited information about long-term patient outcomes and key value drivers at the time of product launch. The objective of this study was to develop a predictive model framework to provide clarity on potential long-term outcomes of interest and associated key value drivers. METHODS: In this study, we focused on a product used to treat a rare genetic syndrome that inhibits the body’s ability to break down fat (lipids). The first step was to gather information on patient outcomes from the product clinical trials and via a structured literature review. Pre-determined criteria were then used to evaluate and rank the evidence for quality, plausibility, and relevance. The outcomes were filtered further to focus on those most likely linked to disease progression and mortality. A patient flow was constructed to link short-term outcomes to potential long-term ones with and without treatment. Probabilistic modelling was applied to determine the likelihood of each complication/outcome to occur. RESULTS: The model provided a probabilistic comparison of different outcomes with and without treatment. The outputs of the predictive model were then entered into an early health economic model to assess key value areas of interest to payers, healthcare systems, and patients. The model framework helped form a reasonable hypothesis that could predict the potential impact of the product on this rare disease. CONCLUSIONS: Predictive models have their limitations, and unexpected outcomes can occur in a real-world setting. Nonetheless, this model provides an early indication of key areas that a new rare disease therapy should focus on to provide improved value. As a next step, monitoring will focus on how closely real-world outcomes compare with those from the predictive model to allow for further refinements.
Conference/Value in Health Info
2017-11, ISPOR Europe 2017, Glasgow, Scotland
Value in Health, Vol. 20, No. 9 (October 2017)
Code
PSY12
Topic
Epidemiology & Public Health
Topic Subcategory
Safety & Pharmacoepidemiology
Disease
Diabetes/Endocrine/Metabolic Disorders