Social Audit Ltd
P O Box 111 London NW1 8XG
Telephone/Fax 44 (0)171 586 7771

[email protected]


Dr Munir Pirmohamed
Department of Pharmacology & Therapeutics
University of Liverpool,
Ashton Street Medical School
LIVERPOOL L69 3GE 6 November 1998

Dear Dr Pirmohamed and colleagues,


Many thanks for your letter; I learned a lot from it. It raises many important issues and I will give them the close attention they deserve as soon as I can. In the meantime, I need to share with you these preliminary thoughts. Please do not hesitate to communicate, if you think any of them off-beam or unfair.

1. I note the exclusion clause in paragraph three of your letter, notably that your published findings "… do not necessarily represent the views of any institutions we represent, either collectively or individually or collectively". If this is so important, I think it should have been stated in the BMJ article itself.

2. I accept that the results (and inferences) should be attributed to the authors alone; nevertheless the conduct of the research, the mode of analysis, and the way in which results were presented are very much representative of the highest standards and most influential sources in pharmaceutical medicine. It cannot be irrelevant that one of the authors of the paper is shortly to become Chairman of the Committee on Safety of Medicines, and that two members (25%) of the CSM's Pharmacovigilance Sub-committee are based in the same department at the University of Liverpool.

3. I note you are greatly concerned about what I wrote to the editor of the BMJ, but you quoted it out of context. I wrote to Dr Smith primarily because I had received no word in response to my enquiries of 28 April, nor to the reminder sent to Professor Breckenridge on 6 July. Even now, I am inclined to think that all is not well if it takes six months to get even the references to the works you reviewed. I understand this was a very busy period for you but, if pressure of work really were the main reason for such a delay, it would raise big questions about the effectiveness of the contribution of any hard-pressed academics/clinician to the national drug safety control system.

4. To avoid misunderstanding, I will now publish the notes I sent to Dr Smith and Prof. Breckenridge on 17 July, and shall copy this letter to Dr Smith as well. I accept that I could have been more tactful, but I think it would have made the central point less clear. As most of the references I requested turn out to be 20 - 30 years old, I think the point should be very clear. 

5. Your estimate that ADRs affect between 10 and 20 per cent of in-patients is identical that that produced in Davies DM (ed), Textbook of Adverse Drug Reactions (2nd edition), OUP, 1981 - and is cited to three of the eight supporting references used therein. I believe you are mistaken in thinking there have been no studies of this problem (apart from those cited in your paper) published since 1970. I shall write to you again about this.

6. Given that your estimate of the extent of drug-induced illness in hospitalised patients represents no advance on a review published 17 years ago, I am not at all reassured to learn that this was one of the most extensively peer-reviewed papers you have published. On the contrary, this raises many more questions than it answers - if only because, when I arrived at essentially the same estimates, nearly seven years ago, the CSM (among others) went to some lengths to rubbish my views. It also surprises me that the BMJ should publish a review, critical parts of which are largely based on US data and a few antiquated UK studies, which discussed neither the relevance of US data to the UK, nor the extraordinary paucity of UK data you found.

7. Given the magnitude of the problem - with drug-induced illness ranking as something like the fourth leading cause of death in UK hospitals - the tone of the review did seem to me very bland. Reading between the lines of your paper, drug-induced iatrogenic illness accounts for something of the order of a quarter of a million hospital admissions each year and, in addition, affect at least as many hospital in-patients. It would therefore have been good to see some closer and more critical analysis of the reasons why society has failed to address the problems.

8. I did not appreciate the concluding advice in your letter, that "drugs are a victim of their own success". Given your reliance on antique data, it did not seem justifiable to conclude this about the present and, in the light of what needs to be done, seemed rather complacent too.

8. I was disturbed by your optimistic emphasis of the yellow card system as the cornerstone of postmarketing drug safety surveillance - and would find it very hard to accept this was not influenced by the Department's close connections with the CSM. Under-reporting is of course a problem, has been for over 30 years, and probably always will be. The underlying problem has to be over-reliance on pharmacovigilance to an extent that values the meticulous analysis of lack of evidence over and above good common sense. Thus, on the basis of its analysis of yellow cards, the CSM presently under-estimates the risk of antidepressant withdrawal problems/dependence (as before with the benzodiazepines) by something of the order of 100,000 per cent.

I hope to be able to write to you shortly on several outstanding points.

Yours sincerely,
Charles Medawar


Contents page
List of MCA/CSM Correspondence