Chapter 1: Developing a New Medicine Decision Analysis

In Book 12: Operations Research Applications for Decision Analysis and Prediction

ISBN: 978-620-5-49772-2

Author: Vojo Bubevski (Independent Researcher)

Abstract 

This chapter illustrates a Decision Analysis for the development of a new drug by a pharmaceutical company. It is currently the year 2020 and a new medication is about to go through three phases of testing, and it must be successful in all phases before it can be marketed, starting in 2027. The company can use any of five strategies: i) do everything in-house; ii) build a pilot plant for manufacturing the drug; iii) contract out the production and build a larger plant for manufacturing the drug; iv) contract out the production and license manufacturing to a third party; and v) license everything to a third party right away. Many monetary and probability inputs are given for the decision analysis, and they are used in the Cash Flow and NPV calculations for the various strategies. A decision tree is applied, with links to the input probabilities and the NPVs. The performed sensitivity analysis quantifies the changes of the output NPVs as impacted by the changes of the input independent variables. The decision analysis results help management to make decisions as appropriate.

Keywords: Risk & Decision Analysis, Pharmaceutical Industry, Financial Risk, Sensitivity Analysis; Decision Tree; Monte Carlo Simulation; Stochastic model.

The Results

The optimal decisions are identified by considering the Path Probability and Path Value of the end nodes of the tree. That is, the end node with maximal Path Probability and Path Value identifies the optimal decision path. The decision tree tools resolve the optimal decision path and present it as a policy suggestion, i.e., the Optimal Decision Tree.

Considering the Optimal Decision Tree in Figure 3, the optimal strategy is to contract out the production during Phases I and II. Then, assuming success in Phases I and II, build a large plant for manufacturing the medicine during Phase III and beyond.

The Optimal Decision Tree path results are the following:

1.      The tree starting Decision Node Phase I/II in 2020, has an Expected Value of $11,777.72 MM;

2.      The Phase I/II Contract Decision has a Decision Indicator of TRUE and a Branch Value of -$3,628.66 MM, which is the cost to the pharmaceutical company for the Years 2020 – 2022;

3.      The Phase I/II Contract Decision has the Chance Node with an Expected Value of $11,777.72 MM, which is the same as the Decision Node Phase I/II Expected Value;

4.      The Yes branch coming from this Chance Node has a Branch Probability of 41.25% and a Branch Value of zero, which ends the Phase I/II Contract Decision in 2022;

5.      The Phase III Decision Node has an Expected Value of $33,720.15 MM;

6.      The Phase III Manufacture Decision has a Decision Indicator of TRUE and a Branch Value of zero;

7.      The Phase III Manufacture Decision has a Chance Node with an Expected Value of $33,720.15 MM, which is the same as the Decision Node Phase III Expected Value;