A bastion of the establishment—indeed, a past president of the Royal College of Paediatrics and Child Health—is brought down by a plucky mother and her family after her wrongful conviction for murder. It reads like a Hollywood script. But many people involved in child protection are deeply worried about the recent decision by the General Medical Council (GMC) to strike Roy Meadow off the medical register.
Meadow was struck off after having appeared in court as an expert witness in the murder trial of a mother who was alleged to have deliberately killed her two young boys. The case against Sally Clark was subsequently found unsafe after she had served more than two years in prison.
The Lancet recently argued that the verdict of serious professional misconduct against Meadow was not only unjust, but would damage the future of child protection services in Britain by discouraging paediatricians from acting as expert witnesses in such cases.
Around 30,000 children under the age of 16 are on child protection registers. About 40 per cent are on the register for neglect, 20 per cent each for physical and emotional abuse and 10 per cent for sexual abuse.
Current estimates in England and Wales are that 30-40 infants die each year as a result of covert homicide—about 10 per cent of the current annual total of sudden unexpected deaths in infancy. Clearly the vast majority of deaths that are categorised as sudden infant death syndrome (or SIDS) do not involve abuse. But, by definition, in most cases the experts do not know why the child died—as a result it's very difficult to counsel a family or a mother about the chances of a repeat tragedy within the same household.
It is on this issue that the GMC case, as well as the media criticism of Meadow, turns. In court during the Clark case, Meadow quoted from a report published by Her Majesty's Stationery Office titled "Sudden Un-expected Deaths in Infancy, The CESDI SUDI Studies 1993-1996," in which the chances of a child dying from SIDS were broken down by various household characteristics, such as presence of a smoker, single-parent family and socioeconomic standing.
Using the figures supplied in the booklet, Meadow quoted that the chances of two children dying of SIDS in a married middle-class household, such as Clark's, came to 1 in 73m. This now infamous figure from the booklet was arrived at by squaring the 1 in 8,543 chance of a single SIDS death in a household like Clark's.
The argument between experts and statisticians hinges on whether this figure might have misled the jury, given, for example, that Clark or her husband could have been carrying the gene for some rare disorder that would have meant that the chances of having two children die concurrently from some natural medical cause would not have been so remote. Indeed, if some common but as yet unknown connection had been in play, once one child had died the odds of another fatality within the same family would have been lowered dramatically.
Meadow was struck off the medical register by a GMC disciplinary tribunal because, it argued, he had abused his position as a doctor by giving evidence that was outside the limits of his professional competence. But if there was no genetic problem and so no connection in terms of natural causes for the deaths, then the figure Meadow quoted would have been about right. The issue turns on the "unknown" of a possible hypothesised connecting cause; even today, the precise cause of the two babies' deaths remains unidentified.
However, the issue for the GMC seems to be that Meadow did not qualify and "contextualise" the statistic enough while on the witness stand. Some might argue that this was a deficiency not of a particular expert witness, but of the court system in general—why, after all, was other expert testimony not called by the court to properly discuss and question Meadow's figure?
It's also useful to recall that Clark's double murder conviction was not deemed unsafe because of the statistics. It was actually a failure to disclose the results of microbiological tests on samples from the postmortem on the Clarks' second son, Harry, that led the appeal court to quash Clark's conviction in 2003. The case turned more on pathology results and resulting argument than on Meadow's statistics.
The judge in the original trial specifically warned the jury against relying on questionable statistics, and one court of appeal judge commented that, "If there had been no error in relation to statistics at the trial, we are satisfied that the jury would still have convicted on each count."
The court of appeal that finally deemed the Clark conviction unsafe did not criticise Meadow, but questioned the fact that there had been no consideration as to whether statistical evidence should have been admitted at all.
Peter Fleming, a cot death expert from Bristol University, was apprised by Clark's lawyers during the original case that Meadow had quoted the estimate of 1 in 73m to the prosecution team. Fleming alerted them to the fact that this risk had possibly been taken out of context, and that a second SIDS death might indeed be related to environmental, genetic or biological factors. But Meadow was not cross-examined in court about Fleming's qualification. It seems odd that Meadow is now being blamed for not addressing the point, even though it was not even put to him by Clark's legal team when they knew of it.
The adversarial nature of the criminal court system makes it the wrong place in which to try and resolve issues related to child abuse; the protection of children, parents and paediatricians is rarely served by the current scheme. Both Sally Clark and Roy Meadow are casualties of the system.
