Gates Open Research (Jul 2018)

Funding global health product R&D: the Portfolio-To-Impact Model (P2I), a new tool for modelling the impact of different research portfolios [version 2; referees: 2 approved]

  • Robert F Terry,
  • Gavin Yamey,
  • Ryoko Miyazaki-Krause,
  • Alexander Gunn,
  • John C. Reeder

DOI
https://doi.org/10.12688/gatesopenres.12816.2
Journal volume & issue
Vol. 2

Abstract

Read online

Background: The Portfolio-To-Impact (P2I) Model is a novel tool, developed to estimate minimum funding needs to accelerate health product development from late stage preclinical study to phase III clinical trials, and to visualize potential product launches over time. Methods: A mixed methods approach was used. Assumptions on development costs at each phase were based on clinical trial costs from Parexel’s R&D cost sourcebook. These were further refined and validated by interviews, with a wide variety of stakeholders from Product Development Partnerships, biopharmaceutical and diagnostic companies, and major funders of global health R&D. Results: the tool was used to create scenarios describing the impact, in terms of products developed, of different product portfolios with funding ranging from $1 million per annum through to $500 million per annum. These scenarios for a new global financing mechanism have been previously presented in a report setting out the potential for a new fund for research and development which would assist in accelerating product development for the diseases of poverty. Conclusion: The P2I tool does enable a user to model different scenarios in terms of cost and number of health products launched when applied to a portfolio of health products. The model is published as open access accompanied with a user guide. The design allows it to be adapted and used for other health R&D portfolio analysis as described in an accompanying publication focussing on the pipeline for neglected diseases in 2017. We aim to continually refine and improve the model and we ask users to provide us with their own inputs that can help us update key parameters and assumptions. We hope to catalyse users to adapt the model in ways that can increase its value, accuracy, and applications.