EWCA Civ 290
Another case of misuse of statistics by an Expert
Mr. Miah was 26 years of age and suffered from severe athetoid cerebral palsy. This was brought about in the final stages of labour when he was born. It was a catastrophic case of cord prolapse. He was delivered as a new born baby with severe brain damage.
It was common ground that had he been delivered just five minutes earlier brain damage would not have resulted (assuming the delivery was otherwise uneventful).
The crucial issue, therefore, was whether the umbilical cord had already prolapsed when his mother was examined by the midwife. By that time the mother’s membranes had ruptured and it was agreed between experts that at or around the time of rupture of the membranes, was the most common time for a cord prolapse to occur.
If cord prolapse had taken place before the end of that examination by the midwife it was common ground that she would have been negligent in failing to identify it.
The trial judge found as a fact that the midwife could remember the events of the birth, notwithstanding that those events had taken place some 25 years ago.
The judge found that the cord prolapse had occurred after the midwife had undertaken her examination of Mr Miah’s mother.
To what extent does lapse of time prevent a judge from making findings as to recollection of events by a witness? Counsel for Mr. Miah argued that the judge had been wrong to accept that after such a length of time, the midwife had in reality been able to remember the events, rather than reconstructing the events of the birth.
What discretion does the judge have to prefer the evidence of one expert to another? It was argued for Mr Miah that the judge was wrong in preferring the evidence of the health authority’s expert over that of Mr Miah’s expert.
The trial judge had had to “weigh up” the probabilities by reference to evidence other than that from the midwife. This included weighing up the evidence of the experts where there were differences between them. The judge also had to consider such records as existed.
Both experts accepted that the time of rupture of the membranes presented the greatest risk of cord prolapse. Mr Miah’s expert was viewed by the Court of Appeal has having given well founded evidence save in one respect in which he was criticised for mis-use of statistics. He relied on an article which set out the results of a clinical study (the Murphy & McKensie article) to support a proposition, which on objective analysis, it did not address one way or the other. However in this case the mis-use of statistics, which was frankly acknowledged by the medical expert concerned, was not material to the decision.
The judge had been entitled to take the view that for anyone to have missed the prolapsed cord would have been grossly negligent. He had been able to assess the midwife as a witness over a considerable length of time and so could form a view as to her memory of events and of whether she was the sort of person who would miss a prolapsed cord.
The judge at trial held in favour of the health authority and against Mr Miah as Claimant. The Court of Appeal whilst expressing sympathy with Mr Miah said that the decision of the trial judge was not one with which they would interfere.
This provides another example of mis-use of statistics in clinical expert evidence. However on this occasion whilst the error was unfortunate, it was not material to the eventual decision.
This again, however, emphasises the dangers of medical experts relying on statistics.
A key problem, of course, is that numerous scientific papers published in learned journals are packed solid with statistics. Those involved in maintaining and enhancing standards in medicine often propound the reliance on solid science – and that must include statistics derived from studies. Evaluating the weight to be applied to one set of statistics against another and assessing the significance of different clinical studies has commonly been a role for the medical expert in providing evidence. However this decision and the earlier cases involving Sir Roy Meadow show the current dangers of medical experts making use of statistics.
But is it really the case that expert statisticians need to be engaged in these cases in addition to medical experts? That could simply add further layers of dispute, cost and confusion – since there is every possibility of one statistician disagreeing with the views or findings of another statistician. It could be said with some justification that judges are even less well-equipped than medical experts to evaluate statistics. The use of statistics in the context of expert evidence represents an area of medico-legal practice that needs to be robustly addressed so as to avoid inconsistent approaches by experts which would not be in the public interest.