Scottish Cancer Registry coverup

Scottish Cancer Registry uses bad epidemiology to cover up leukaemia cluster

[This page dates from June 2007]
map of Dumfries and Galloway  
Health Board showing bands by distance from the sea (64 Kb) Picture credit: Scottish  
Cancer Registry/ Journal of Environmental and Occupational Medicine enhanced by  
Low Level Radiation Campaign

The Dumfries and Galloway Health Board reported a cluster of childhood leukaemia in its area as early as 1993 (or even before). People are making a link with the coast of the Solway Firth, radioactively contaminated by Sellafield and by the testing of depleted Uranium weapons at the Dundrennan firing range, but attempts to obtain figures on leukaemia have been frustrated.
Now officials at the Scottish Cancer Registry have resorted to appallingly bad epidemiology in an attempt to head off public concern. (See this link to the Footprints For Peace latest contribution to the long campaign to stop the DU testing.)

Map: Dumfries and Galloway Health Board area showing bands by distance from the sea; taken from the Scottish Cancer Registry study in the Journal of Environmental and Occupational Medicine.


In 2005 Green Party Members of the Scottish Parliament asked the Common Services Agency (CSA) of the NHS in Scotland for leukaemia data. The request was refused because it was alleged that releasing it would allow patients to be identified. The Information Commissioner and then the Court of Session found that with appropriate safeguards to preserve patient confidentiality, there was no reason why the data should be kept secret. They ordered its release.Click here to see a December 2006 report on the situation after the Court of Session judgement. (Opens in a new window)
In June it was reported that the CSA is appealing to the House of Lords to overturn the decisions of the Information Commissioner and the Court of Session. Click here to see report in the Herald

In the meantime, the Journal of Occupational and Environmental Medicine published a paper by Scottish Cancer Registry officials. It set out to examine the hypothesis that radioactive pollution in the sea is causing an excess risk of leukaemia. Predictably they claim that risk near the sea is no greater than inland. (Click here to see the abstract in a new window.)

What the Scottish Cancer Registry set out to do
Persisting public concerns that there may be an excess of childhood leukaemia close to the Solway Firth coast of Dumfries and Galloway prompted us [i.e. the Cancer Registry] to formally investigate th[e] hypothesis [that low levels of environmental exposure to ionising radiation constitute a risk factor for various health effects, especially leukaemia in children].
What they reported
Despite public concerns expressed through the media, formal analysis of cancer registration data provides no statistically significant evidence of an increased risk of childhood leukaemia in the vicinity of the Solway Firth coastline of Dumfries and Galloway Health Board area.

Area

Standardized Incidence Ratio

1975 - 1989

1990 2002

1975 - 2002

A (grey on map)

0.89

1.57

1.22

B (grey + orange)

1.12

1.18

1.15

C (entire Health Board area)

1.42

1.03

1.23

They show no clear trend with distance from the sea.
There are two reasons for this.
1. Chernobyl ignored.    The main flaw is that the cancer registry authors ignored the fallout from the Chernobyl accident. Since they set out test the hypothesis that the disease is caused by radioactivity, ignoring a major pollution incident which happened right in the middle of the study period is fatal to their credibility. A reputable scientific journal ought to have rejected their work for that reason alone. More precisely and yet more seriously, the study was testing for an effect from radioactivity in the sea, but the Chernobyl fallout fell on the high ground inland, exactly reversing the exposure pattern of interest.
Fortunately, we were able to obtain the figures for the Chernobyl period (see the table below the next paragraph).
2. Study areas fudged    A second problem involves the way the authors analysed the data.
They defined three bands by their distance from the sea. This is shown in the map above, where the band next to the sea is coloured grey, the middle band is orange, and the inland band is yellow. However, they did not apply the simplest, most obvious and most informative test, which would have been to test each band against the others, looking for any gradient in risk. Instead they nested the bands in a complex manner.

  • the grey coastal band they called A; .
  • B consists of the orange area plus band A;
  • the third band, C, is the yellow area plus both of the others.
In our experience this nesting approach has been criticised by COMARE and by the Journal of Radiation Protection
Fortunately the data in the published paper allow us to correct for this fault, as the numbers of leukaemia children in each band can be obtained by simple subtraction.
Our reanalysis
The next table shows the data when the leukaemia cases registered in the Chernobyl period (1986 to 1989) are taken out.
It also shows the three areas compared directly with each other without nesting them.

Area

Standardized Incidence Ratio

1975 to 1986 + 1990 to 2002

1990 - 2002

Grey (coastal)

1.4

1.6

Orange (middle)

1.2

1.0

Yellow (inland)

1.2

0.8

Once the confounding influence of the Chernobyl accident is taken out there is a clear trend of higher risks near the sea. The right hand column shows the data for the most recent years; it appears that the excess risk is accelerating. Click here for a paper with more detail

We are told that test firing of Depleted Uranium at the Dundrennan range began in 1983.


Campaign for the data
The Minister responsible for the NHS in Scotland is under pressure to release the data the Greens have asked for. See this newspaper report
It seems obvious that the Scottish Cancer Registry paper was written in response to the threat of disclosure; the authorities fear that when the data are eventually dragged into the light, the higher risks near the sea will be another nail in the coffin of the official radiation risk model.
See this link to an earlier cover-up in part of the Dumfries and Galloway area.

If you are seeing this page full screen (i.e. without a navigation bar on the left) you can't see how the rest of the site is organised.
This Home page link takes you to the index page, which has links to all the topics we discuss on the site [only use it if this page is full screen]