New French study on childhood leukemias near nuclear power plants

Geocap study confirms findings in Germany, Great Britain, and Switzerland

A new French study [1] (GEOCAP) of childhood leukaemia near nuclear power plants (NPPs) has found a statistically significant increase in leukaemia in children below age 15 in 2002-2007 within 5 km of 19 French NPPs. The French finding is persuasive as it was determined in two separate ways. First, by a comprehensive nationwide case-control study (odds ratio, OR=1.9). And second, by a conventional incidence study (standardized incidence ratio, SIR=1.9).

Many newspapers in France (but none in the UK) carried this story prominently, but in fact it is the fourth European study showing this result. After the shocking results of the KiKK study [2] in 2007, further studies with the same or similar findings were carried out in Germany [3], Great Britain [4], and Switzerland [5].

The authors ended their article stating: “Overall, the findings call for […] collaborative analysis of all the evidence available from multisite studies conducted in various countries”. But they did not do this themselves although it’s relatively straightforward. Dr Alfred Körblein and I did this with the new French study and three other studies in Britain, Germany and Switzerland and we found a 37% increase which is statistically significant.

In more detail, our pooled analysis of acute leukaemia in children under five in Great Britain (1969-2004, 0-25 km, SIR=1.30), Germany (1980-2003, 0-30 km, SIR=1.41), Switzerland (1985-2009, 0-15 km, SIR=1.40), and in France (1990-2007, 0-20 km, SIR=1.37) yields an overall SIR of 1.37 in the 5 km zone. Table 1 below shows the numbers of observed and expected cases and SIRs in the distance zones.

Table 1 : Standardised Incidence Ratios (SIRs) in Four Countries (under-fives, 0-5 km)

Country
Observed
Expected
SIR= O/E
p-value (one sided)
Germany
34
24.09
1.41
0.0328
Great Britain
(COMARE 14)
20
15.4
1.30
0.1464
Switzerland
11
7.87
1.40
0.1711
France
14
10.2
1.37
0.1506
Pooled Data
79
57.5
1.37
0.0042

 Statistical significance

The French authors state they only found increases in the period 2002-2007 but not between 1990-2001 and not over the whole study period 1990-2007. However this is incorrect: buried deep in the study (at the very end, in the last section of their Table 2) it is revealed that there is in fact a 37% (rounded to 40%) increase for children below 5 for the whole period 1990-2007. But because the increase is not “statistically significant”, the authors discount it.

This brings up an important point, and that is that the treatment of statistical significance by official scientists in epidemiology studies on radiation exposures is often problematic. The difficulty is that, as was done here, if a cancer increase is not “statistically significant”, scientists incorrectly dismiss the finding by saying there was “no increase” or “no significant increase”. Other choice phrases are that “no suggestion” or “no evidence” of a leukemia increase is found.

This is poor scientific practice: what the French scientists should have stated is there was an increase but it did not pass their statistical test: ie there was a >5% possibility that the result was a chance finding. This is what statistical significance means. The problem is that the absence of statistical significance can simply be due to small numbers as well as to lack of effect: disentangling the two influences is not possible. This means statistical tests must be used with caution as the use of a given cut-off for statistical significance (usually 5%) can lead to incorrect conclusions: in statistics, this is called a type II error.

The solution is to look for other similar studies to boost the numbers in order to attempt to obtain findings which ARE statistically significant. Which is what Dr Körblein and I did – see above.

Dose Estimation

The authors made considerable efforts to estimate possible radiation doses from the NPP radioactive air emissions in various zones near NPPs. They did this to see if there was a link between the emissions and the resulting leukemias. They found no link but there are severe problems with their methodology and their conclusion. One problem is that these are average doses in various near zones: they are not doses to individuals and there could be wide dose variations. Another problem is that no consideration was given to estimating embryonal or fetal doses in pregnant women: this is where the real danger lies. See my article (Fairlie I (2010) Hypothesis to Explain Childhood Cancer near Nuclear Power Plants. Int J Occup Environ Health 16:341–350) which explains this.

But the main difficulty is that estimating such doses to local people is fraught with uncertainties, as found by the 2004 Report of the UK Government’s CERRIE committee – see www.cerrie.org. This explained that dose estimates from environmental releases depended on several computer models and the assumptions they contained. In addition, my recent post on the higher radiation doses from radioactive spikes at NPPs (see above) adds to these uncertainties. Indeed these problematic estimations are alluded to by the French Government’s radiation advisor IRSN. Their press release on the new French study – (see http://www.irsn.fr/FR/Actualites_presse/Actualites/Pages/20120112-Publication_scientifique_Inserm-IRSN_leucemies_chez_enfant.aspx) says that these dose estimations are“… based on a rather elaborate approach (consideration of a range of 12 radionuclides, using real data on emissions, local weather data on strong winds and precipitation, modeling the transfer in the different compartments of the environment, considering the different routes of exposure (inhalation, ingestion, external exposure due to the deposit, use of specific dose coefficients of the relevant body ..)“

Quite right, and as a result it is difficult to place much confidence in either the zonal dose estimates or the conclusion that radioactive emissions are not the cause of the increased cancers.

Conclusion

The new French study provides valuable additional evidence for an increased leukaemia risk near nuclear power plants. The new data fits squarely within the epidemiological evidence of increased childhood leukemias near NPPs in three other European countries. Over the four multisite studies, a consistent pattern of increased incidences of childhood leukemias near NPPs has clearly emerged.

Indeed, these increases in childhood leukemias have been found not just in Europe but all over the world. The 2008 Laurier et al study [6] taken together with Laurier and Bard’s 1999 study [7] indicate a stunning total of over 60 studies world-wide which have examined child cancers near nuclear facilities, with most of them finding cancer increases.

Unfortunately, the French authors refuse to recognise this overwhelming body of evidence. They state “In general, no excess risk has been evidenced by the multisite studies although persistent localized excesses of childhood acute leukemia (AL) have been reported around a few specific sites….”. This refusal to accept the obvious is incongruous and actually slightly funny: many will grin ironically when they read it…and perhaps a grin is the best response.

References

  1. Sermage-Faure C, Laurier D, Goujon-Bellec S, Chartier M, Guyot-Goubin A, Rudant J, Hémon D, Clavel J. Childhood leukaemia around French nuclear power plants – the study, 2002-2007. Int J Cancer. 2012 Jan 5. doi: 10.1002/ijc.27425. [Epub ahead of print].
  2. Kaatsch P, Spix C, Schulze-Rath R, Schmiedel S, Blettner M. Leukaemia in young children living in the vicinity of German nuclear power plants. (2008) Int J Cancer; 122: 721-726.
  3. Kaatsch P, Spix C, Jung I, Blettner M. Childhood leukaemia in the vicinity of nuclear power plants in Germany. (2008) Dtsch Arztebl Int. Oct;105(42):725-32.
  4. Committee on Medical Aspects of Radiation in the Environment (COMARE)  (2011) Fourteenth report. Further Consideration of the Incidence of Childhood Leukaemia Around Nuclear Power Plants in Great Britain. London: Health Protection Agency. http://www.comare.org.uk/press_releases/documents/COMARE14report.pdf
  5. Spycher BD, Feller M, Zwahlen M, Röösli M, von der Weid NX, Hengartner H, Egger M, Kuehni CE. Childhood cancer and nuclear power plants in Switzerland: A census based cohort study. International Journal of Epidemiology (2011) doi:10.1093/ije/DYR115. http://ije.oxfordjournals.org/content/early/2011/07/11/ije.dyr115.full.pdf+html
  6. Laurier D, Jacob S, Bernier MO, Leuraud K, Metz C, Samson E, Laloi P. (2008) Epidemiological studies of leukaemia in children and young adults around nuclear facilities: a critical review. Radiat Prot Dosimetry. 132(2):182-90.
  7. Laurier D, Bard D (1999) Epidemiologic studies of leukaemia among persons under 25 years of age living near nuclear sites. Epidemiol Rev. 21(2):188-206.
Share:Share on FacebookTweet about this on TwitterShare on LinkedInShare on Google+Digg thisPin on PinterestShare on RedditShare on StumbleUponShare on TumblrEmail this to someone
This entry was posted in News and comments, Uncategorized. Bookmark the permalink.