Further Statistical Issues in Project Prioritization in the Pharmaceutical Industry
- PDF / 445,105 Bytes
- 7 Pages / 504 x 719.759 pts Page_size
- 49 Downloads / 200 Views
0092-8615\98 Copyright 0 1998 Drug Information Association Inc.
F’rinted in the USA. All rights resewed.
FURTHER STATISTICAL ISSUES IN PROJECT PRIORITIZATION IN THE PHARMACEUTICAL INDUSTRY STEPHEN SEW, BA, MSc, PHD Professor of Pharmaceutical and Health Statistics, Department of Statistical Science and Department of Epidemiology and Community Health, University College London, London, United Kingdom
Various aspects of portfolio management within the pharmaceutical industry are considered. It is shown that some popular notions of risk management are incorrect. Some additional practical matters are also touched upon. Key Words: Statistical issues; Portfolio management; Pharmaceutical industry
INTRODUCTION My ventures are not in one bottom trusted, Nor to one place; nor is my whole estate Upon the fortune of this present year. The Merchant of Venice
THE PROBLEM OF ANTONIO AS A WISE MERCHANT OF Venice, Antonio knows not to put all his eggs in one basket, nor all his trade in one ship. Although at the low ebb of his fortunes, he believes that his ‘ships have all miscarried’ he reads later that they ‘are safely come to road.’ That all should return, is, of course, extreme good fortune, but that all should founder would have been bad luck and Antonio’s strategy is a wise one for any risk-averse businessman to follow. Some similar considerations, no doubt, are reflected in the development strategies of most major pharmaceutical companies. Although many will have a very high risk tolerance in what is a risky business,
Reprint address: Stephen Senn, 1-19 Torrington Place, London WClE 6BT, United Kingdom. Email: [email protected]
most will wish to spread the risk as widely as possible, not only between different drugs but also between different indications and markets, even if the rise of the ‘corporate raider’ has meant that conglomerates are now more vulnerable than they were in the past and even if a wider focus occasionally leads to blurred vision. In a previous paper (l), some statistical issues in evaluating individual projects in development were discussed. A simple tool which might be used for that purpose, the Pearson index (2), was considered and its weaknesses discussed. This index has also recently been discussed in the pages of this journal by John Gittins (3). The Pearson index consists of calculating an expected net present value and expressing it as a ratio to the expected cost. The net present value of a drug in development is the difference between the discounted eventual reward and the sum of the discounted costs to develop it. The expected net present value is the corresponding figure but with the rewards and costs weighted by appropriate probabilities. For example, for a three-stage project with discounted costs c,, c2,and c3for each stage, discounted rewards if successful, R, and con-
253
Downloaded from dij.sagepub.com at PURDUE UNIVERSITY TSS on June 13, 2015
Stephen Senn
254
ditional probabilities of success for each stage of p I ,p2, and p 3 , the Pearson index is (123:
T
Data Loading...
