At the 2012 EAHP congress in Milan a seminar was devoted to the value of new drugs in oncology. This article is a reflection of a presentation given during that seminar by Dr Angelo Palozzo from Padova, Italy. It is accompanied by a personal opinion by Professor Krämer from Mainz, Germany and a commentary by Drs Cajaraville and Tames from San Sebastian, Spain.

In this article, advanced pharmacoeconomic principles are applied. The reader is encouraged to read earlier introductory articles on pharmacoeconomics in EJHP (such as the cover story on pharmacoeconomics in issue No 3, 2008 or on innovative payments systems in issue No 3, 2009).

The author argues that hospital pharmacists can contribute to a better understanding of the real cost-effectiveness of a new drug by applying pharmacoeconomic principles in analysing registries of patients who have been treated with expensive drugs.

Introduction: effectiveness versus efficacy

Post-marketing studies help to assess differences between ‘efficacy’ and ‘effectiveness’. In clinical trials the studied drug or treatment must adhere to strict operational conditions good clinical practices (GCP), patient's inclusion and exclusion criteria) and efficacy or toxicity are well-defined in the treatment groups. Performing the same treatment in a general population may yield considerable differences from a randomised controlled trial because the treatment conditions are less well-defined (patients, diagnosis, treatment variations) and more patients are involved.

Randomised clinical trials (RCTs) involve a limited number of carefully selected patients and treatments are closely monitored with regard to schedules and compliance of patient doses as well as intervals. There are strict clinical and diagnostic controls and pharmacological controls are always performed. In real-world clinical practice the whole population is treated (the numbers are larger) and target groups are not selected (ie, paediatric, geriatric). As a result, less frequent adverse drug reactions can be detected (ie, 1 in 1000 treated patients); side …