1 Tuesday, 15 November 2011 2 (9.30 am) 3 DR BRIAN McCLELLAND (continued) 4 Questions by MR MACKENZIE 5 THE CHAIRMAN: Good morning. Yes, Mr Mackenzie? 6 MR MACKENZIE: Good morning, sir. We move on to a new topic 7 today, topic C2, which is the non-introduction in 8 Scotland of surrogate testing for non-A non-B Hepatitis, 9 and our witness today is Dr Brian McClelland. 10 THE CHAIRMAN: Yes. 11 MR MACKENZIE: Good morning, Dr McClelland. 12 A. Good morning. 13 Q. We don't have to look at your CV again but we know that, 14 I think, between 1979 and 2001 you were director of the 15 Edinburgh and Southeast Scotland Blood Transfusion 16 Service. Is that correct? 17 A. That's correct. 18 Q. And I think you retired from the SNBTS in 2009. 19 A. That's correct. 20 Q. Could we, please, look at your statement you provided on 21 this topic. It's [PEN0170754] and what I would like to 22 do, doctor, is go through your statement and from time 23 to time ask you various questions and also look at some 24 of the documents you have referred to, plus one or two 25 others as well. 1 1 You say in your statement -- your first heading is 2 "Opinions of the importance of non-A non-B post 3 transfusion hepatitis in the UK between 1980-1989". 4 You say that before considering the particular 5 question we asked you, you thought it may be useful to 6 the Inquiry to provide a personal view on the apparent 7 persistence of the belief over the years 1980 to 1989 8 that non-A non-B post transfusion hepatitis was not an 9 important problem in the UK. And one of the themes 10 underlying this history is the view that was taken of 11 NANBH in the UK from around 1980 to the discovery of 12 Hepatitis C in 1989: 13 "Many of the decisions taken or not taken can only 14 be understood in the context of a widely held view that 15 despite an increasing body of evidence to the contrary, 16 this condition was rarely transmitted by blood and was 17 usually not particularly serious." 18 You say: 19 "I have tried to assemble some evidence that 20 illustrates how this view may have originated." 21 We will go on to look at the various papers. 22 Am I right in thinking, doctor, that the papers you 23 list are largely looking at the prevalence of 24 post-transfusion hepatitis in the UK rather than 25 focusing on the seriousness of the disease? 2 1 A. Yes, absolutely. 2 Q. Looking at these papers in turn, please, the first one 3 you refer to is the Medical Research Council Blood 4 Transfusion Research Committee in 1974. This is the 5 year of publication of a report of a study carried out 6 for the UK MRC of hepatitis in recipients of blood 7 components. 8 You explain that: 9 "This study is described in some detail since it is 10 one of only four substantial prospective studies of PTH 11 in the UK." 12 Can you explain, please, doctor, what does 13 "prospective study" mean and how does that differ 14 perhaps with a predictive study? 15 A. Well, in broad terms, actually the term is quite 16 widely -- it is used in a variety of senses, but in 17 broad terms it implies a study where you, as it were, 18 define the questions that you are asking and then carry 19 out work in advance, according to a planned schedule, to 20 test the answers to those questions. 21 The -- the sort of gold standard -- and we will come 22 back to this I'm sure -- of a prospective study is 23 what's called a prospective randomised clinical trial, 24 which is designed in such a way -- it is designed, if it 25 is well designed, to -- to compare at least two groups 3 1 of patients, one group who receives a particular form of 2 treatment or intervention, which could be a diagnostic 3 test or anything in principle, and the other group who 4 either is a control group that receives -- does not 5 receive that intervention or treatment or that receives 6 another one, in which case it could be comparing two 7 treatments. Those studies, if properly designed, should 8 see that the groups of patients who are enrolled to the 9 treatment group and the control group, if you like, are 10 randomly selected, and one of the tests that you apply 11 in analysing a study like that is to see that actually 12 the make-up of gender, age, social class, co-morbidities 13 and so on, is very similar between the two groups. 14 There are, you know, a number of very well evolved 15 criteria for -- to determine the quantity of design and 16 execution of the prospective randomised control trial. 17 At a sort of lower -- below the gold standard, if 18 you like, there are studies which set out to look at 19 what happens in the future when you initiate an action 20 in the present, and those are broadly called 21 "prospective studies", but they don't all by any means 22 tick the -- all the boxes required for a very high 23 quality study. 24 Q. Yes, and predictive study, what is that? 25 A. It is not a term I have ever used. I don't really know 4 1 what it means. 2 Q. I see. We might give it some context if we see it in 3 some of the literature we look at over the course of 4 today. That may help. 5 A. I'm happy to try and do that. I would make the 6 distinction more between a prospective study, which is 7 the sort of characteristics that I have sketched out, 8 and an observational study, which would be essentially 9 looking at what has happened, something that you are 10 already doing and you collect the information about what 11 has been done for a number of years in the past and you 12 then try to draw some inferences about the relationships 13 between one event, for example, the use of a treatment 14 or a test, and another event, which is the development 15 or non-development of an illness in the patient. 16 But those studies are always very difficult to 17 interpret because of the risk of confounding factors, 18 such as associations between a particular type of 19 patient and the probability of getting a particular 20 treatment, and it's extremely difficult, even with the 21 use of quite complicated statistical techniques -- in 22 fact I would say it's impossible -- to draw conclusions 23 about the cause and effect from these retrospective type 24 of studies. 25 Q. Yes. 5 1 A. But they are very useful for generating hypotheses, for 2 saying it looks as though something is happening because 3 of something else, then you go on to do a properly 4 designed prospective study to test that hypothesis. 5 Q. Yes. I understand. 6 Returning, please, to your statement and the MRC 7 study, you explain that: 8 "From mid 1969 to the end of December 1971, patients 9 at the Central Middlesex Hospital ..." 10 Participated in giving a pre-transfusion blood 11 sample for ALT and viral studies. 12 We see that of the 2,184 patients who were 13 transfused during the study period, follow-up was 14 completed on 768 who received an average of 3.7 units of 15 blood per patient. 16 Over the page of your statement, we see: 17 "Routine testing of donor blood for Hepatitis B only 18 began during the last five months of the study period: 19 "Raised ALT values were found after transfusion in 20 158 patients." 21 Six of whom underwent liver biopsy: 22 "None showed histological features typical of acute 23 viral hepatitis." 24 Then you quote: 25 "The authors stated that these 158 patients were 6 1 investigated for conditions other than viral hepatitis, 2 eg drug induced liver injury. It was arbitrarily 3 decided that where such other potential causes existed, 4 the patient would not be regarded as suffering from 5 viral hepatitis. On this basis, eight patients 6 (1 per cent) were judged to have had post-transfusion 7 hepatitis. Sustained elevation of ALT without other 8 clinical features of hepatitis was present in 35 9 patients." 10 You then again quote: 11 "The authors concluded that 'the overall incidence 12 of icteric and anicteric hepatitis in the present survey 13 (1 per cent) is low compared with the incidence found in 14 prospective studies in Japan (65 per cent) ... USA 15 (18 per cent) ... and Germany (14 per cent).' However 16 if PTH had been defined to include all the patients with 17 ..." 18 That's the end of quote and you go on to say: 19 "However, if PTH had been defined to include all the 20 patients with persistently elevated ALT, the PTH rate 21 would have been 35/768 or 4.5 per cent. If PTH had been 22 defined to include patients with any elevation of ALT 23 following transfusion, 158 of the 768 patients 24 (21 per cent) would have been defined as having PTH." 25 That paragraph perhaps just illustrates the 7 1 difficulties at the time, doctor, in trying to 2 accurately conclude the true rate of post-transfusion 3 hepatitis based on elevated ALT levels.