George Monbiot's November 2011 attacks in The Guardian
Childhood leukaemia in Wales

In Liverpool in May George Moniot accused Busby of profound statistical mistakes in investigating high rates of childhood leukaemia on the Irish Sea coast of north Wales. When LLRC invited Mr. Monbiot discuss the issues he refused. We posted a page showing that the "profound statistical mistakes" were in fact made by the Welsh Cancer Intelligence and Surveillance Unit. The evidence is in the Journal of Public Health, March 2006.
When the Mayor of Oxford issued a similar invitation last month Monbiot refused again, accusing Busby of being "a liar". The meeting went ahead nonetheless on 3rd November. A video camera stood in for Monbiot as Busby explained the issues. (YouTube clip.)
Monbiot has now (22nd November 2011) used the platform given him by The Guardian to renew the assault. Under a heading 'leukaemia cluster' in north Wales [is] baseless scaremongering he repeated his garbled version at greater length, quoting copiously from a 2008 article by Dr. John Steward, Director of the Welsh Cancer Intelligence and Surveillance Unit (WCISU) and colleagues. (I return to this later and give the link.)

In 2004 Steward had been faced with a problem - a paper we presented at CERRIE that revealed a large excess of cancer and leukaemia in children in the coastal towns of north Wales. The children had been found by a reporter for Harlech Television (HTV).
WCISU investigated its own cancer registration data for the same period (2000–2003) and in a self-published paper in 2005 (with update) confirmed a statistically significant excess in the Menai area (Relative Risk = 4.3) and Caernarfon (RR = 12.8). Monbiot has not mentioned this rather startling fact. The 12.8-fold excess in Caernarfon is greater than the notorious Seascale (Sellafield) cluster. Maybe he missed it, and you do have to read Steward's 2008 article carefully to see it because he was mostly concerned with discrediting the old Wales Cancer Registry (WCR) which WCISU replaced in 1997.
The older WCR figures are important because they reveal that the high rates along the contaminated coast of north Wales in 2000–2003 are part of a long and worsening trend. We said this in our 2004 paper. It has serious implications for public health policy and the future of nuclear power. Contrariwise, Steward's 2005 paper represents them as a non-significant blip. We worked out that he done it by grossly inflating the populations of the towns, so increasing the number of cases expected and reducing the significance of the actual cases observed. A letter was published in Journal of Public Health, as mentioned above. In response, the UK Government committee COMARE agreed that Dr. Steward had used wrong populations and confirmed the existence of high rates of child leukaemia in Caernarfon and Menai. Monbiot has not mentioned this. Instead he has cherry-picked his quotes from the 2008 article in which Steward again seeks to play down the significance of the early leukaemia excess. He does this by attacking the quality of WCR's data, and by criticising both Green Audit's analysis and the utility of the now defunct "Areas of Residence" which provide the data on populations. Oddly, this article was not (as far as we know) submitted to the Journal of Public Health, which is where it ought to be since it is part of an ongoing debate. Instead it was published in the Journal of Radiological Protection whose Editor-in-Chief is Richard Wakeford, an ex-employee of BNFL, now a professor at Manchester University's Dalton Nuclear Institute. Wakeford is a bitter critic of Busby and LLRC. Yet more oddly, the article fails to cite the Journal of Public Health letter which revealed Steward's mistake with the populations. A full analysis is beyond the scope of this page. The President of the Royal College of Physicians' Faculty of Public Health wrote to Professor Busby regretting the lack of debate and promising to ask Richard Wakeford to publish Busby's reply in a future edition of JRP. Wakeford has recently promised to do so. (Professor Busby has observed that Wakeford's own epidemiology is not above reproach - see this link.)

Richard Bramhall: 27th November 2011