Chris Busby and BSRRW in Stockholm, 10 August 2010

A place for invalid arguments, faulty statements and general waste.

Chris Busby and BSRRW in Stockholm, 10 August 2010

Postby Lantzelot » 30 Jan 2011, 03:28

This is a condensed version of the report (in Swedish) about the BSSRW meeting in Stockholm (
While the Swedish version covers all speakers at the meeting, this English translation focuses on the presentation by the British Professor Chris Busby.

On Tuesday, 10 August 2010, I attended a seminar in the Sibelius Hall in Finlandshuset in Stockholm, organized by the nonprofit organization BSRRW (Baltic Sea Region Radioactivity Watch), which during the summer 2010 made a tour around the Baltic Sea to propagate against nuclear power. The day before, they conducted a ceremony outside the Swedish parliament where they paid tribute to all victims (past, present and future) for the nuclear industry. Pictures of similar activities from the tour can be found on

Their reasoning is: The Baltic Sea is the world's most radioactive sea, and the Swedish nuclear power emits more radioactivity than the other countries, and therefore nuclear power should be phased out. A summary of the reasoning can found at:
Of course they never put the emission levels in relation to the emissions from the Chernobyl accident (10000-100000 times higher), or the natural background levels which are considerably higher. It is true that the Baltic Sea have a higher level of radioactivty than other seas, at least if you look at Cesium-137 (the dominant fallout from Chernobyl). But despite this, the dose contribution from anthropogenic sources (Chernobyl, atmospheric nuclear tests, nuclear power, etc.) only consist of one third of the contribution from natural sources. And of these anthropogenic emissions, nuclear power contributes with 0.04% of the total dose to the most critically exposed group.
The data have been taken from the Helsinki commission (HELCOM) reports (see, for example BSEP No. 117 found on the following page:, but BSRRW do not read the full report, or they intentionally choose not to refer to its conclusions. If they did, they would find that the level of radioactivity is hardly the Baltic Sea's largest environmental problem. PCBs and dioxins are a much bigger problem, not to speak of nitrogen and phosphorus concentrations, and oxygen depletion.

The meeting started with some of the main speakers and organizers at the podium. They presented a press release (found at the top of this link: and made some strong statements about the nuclear threat and the great betrayal of the Swedish parliament, which in 17 June 2010 voted for allowing new nuclear power in Sweden (although with only two votes margin). This, BSRRW stated, was in conflict with basic human rights. In the audience there were about 20 people, mostly women in upper middle age with Nuclear Power No Thanks-buttons, but the number varied during the day. A few people dropped by and were replaced by others.

The main attraction of the day (at least for me) was Professor Chris Busby, he is the main reason that I decided to attend the seminar. Busby has found cancer clusters in many places in Britain, and he pops up behind various organizations and investigations related to radiation risks. In the government sponsored commissions that he has participated in he usually does not manage to agree with the other members. Therefore a minority report is sometimes issued, including his and a few others conclusions, that are very different from the ones in the main report. He has also created his own risk model (ECRR) of radioactivity which he claims is much better than the one by ICRP (International Comission for Radiation Protection ( ICRP is an international expert panel whose recommendations are only advisory, but most countries follow their recommendations on radiation protection limits). Busby claims that the establishment is working against him and his ideas, and it is rare that his papers pass the peer review in scientific journals. Some entertaining articles (both written by freelance journalist and author Emil Schön, who also write for Friends of the Earth and Milkas) can be found here (in Swedish):
and here:

In my personal opinion, Busby has a few ideas that are interesting and that could very well be of importance for others to look into, but it is very difficult to judge them unless you are an expert in radio-biology. However, his way of behaving, and the tone of some of his articles (which are usually not published in a normal fashion because they are not usually accepted, and of course this is a conspiracy against him), give a strong suspicion that he is not honest, and most of his research has been criticized by people in the field. So it could be easy to dismiss him as a nut-crack.

The problem is that he says what many anti-nuclear groups want to hear, and gives them a scientific background to refer to. And many people are afraid or worried because of what he says. If I as a nuclear physicist have difficulties to assess the validity of his claim, then it is even harder for a person with no scientific background to evaluate it. I therefore wanted to demonstrate to the people at the meeting that there are reasons to doubt what he says, and that he handles data in a dishonest way.

He started talking about Chernobyl and the fact that the Baltic Sea is the world's most radioactive sea, and showed some pictures from the HELCOM reports. Then he mentioned the study by Martin Tondel from Linköping, who in his PhD dissertation from 2007 claimed a direct link between the fallout of Cs-137 from Chernobyl and increased cancer in the Gävleborg county and several other places in the northern part of Sweden (i.e. north with respect to Stockholm). The Tondel study has been criticized both by the former SSI (Swedish Radiation Protection Institute, now part of the Swedish Radiation Safety Authority, SSM, and by several medical doctors.
Busby decided to make his own study, but he has no access to raw data in the same way as Tondel had. "They know what I would do with the data, so they refuse to give it to me!" he said. Therefore, he has used data from the cancer registries of the Swedish National Board of Health and Welfare, which can be downloaded from their Statistics Database. This is not as elegant as Gapminder (, but is still pretty easy to use, and it makes a lot of information easily accessible to the public.
Anyone is welcome to try it for themselves (in Swedish, though there are a number of reports in English):

Busby's argument is that the radioactive fallout (from Chernobyl) in the sea exposes people who live near the sea to a higher dose of radiation. The waves on the shore releases radioactive particles from the sea water, local people then breathe them into their bodies, and the internal dose of radioactivity gives breast cancer faster than an external dose. He therefore chose to compare the number of cases of breast cancer before and after the Chernobyl accident for every county of Sweden. The picture below shows Busby preliminary study, this is the actual table that he showed.

Chris Busby's table with breast cancer data in Sweden before and after the Chernobyl accident in 1986. Click on the picture for full size.

Busby found that the number of cases of breast cancer increased in the counties facing the Baltic Sea (Halland and Västergötland are facing the Kattegatt Sea on the Swedish west coast, but they are considered to belong to the Baltic Sea Region), while there was no change, or even a decreased cancer incidence, in the counties that do not have contact with the sea. One can of course question this in many ways. For example, the population of Uppsala county is not concentrated near the coast, and therefore one can question the reasoning that it should be counted in the category "BALTIC".

After Busby's presentation I asked him why he had taken the average of two years (1984-1985) before the Chernobyl accident, and then the average of four years (1988-1991) after the accident. Why not equal time intervals before and after? I do not remember if I got a clear answer to this, but he claimed that he had taken years at random for testing. Later he said it was the four-five years after the Chernobyl accident that are the ones that were significant according the Tondel study and that it was the results of this study that he wanted to check.

Then I asked if he seriously believes that the Chernobyl accident was beneficial for women in the Jämtland county since the number of breast cancer cases had decreased by 19%. "No of course not, how can you say something like that!" he said indignantly. Somewhere here, our exchange of words became rather intense and stressful, because the organizers wanted to start the next presentation, but I was determined to continue my interrogation. The questions I presented (unfortunately not in a calm and elegant way, I have to admit) were:
* If it is not relevant that cancer dropped by almost 20% in Jämtland, why is it relevant that they have increased by 15-20% in the Baltic Sea Provinces?
* Why is not the data for Gotland county included in the table? There are also other counties missing, why are they not shown?
* Why do you not explain to the audience that the number of breast cancer cases in Sweden has increased almost constantly since 1970, nearly doubling over the period 1970-2008, so that people understand the background effect? Did you subtract this upward trend from your data before deducing the results?

Busby is clever, and he gave an impression of being completely unaware of the facts that I asked about. It could be credible, he probably does not understand Swedish, and most likely somebody who knows Swedish helped him with extracting data from the Statistics Database. That he does not have full knowledge about the geography of Sweden is also a feasible explanation, so maybe he really did not know where Gotland is. But it would be very strange that a number cruncher like him would not even look for what is available in the database, and in this case to check every county in the list. And Busby is no fool, so it's hardly a coincidence that some information is missing: Of the counties that are missing in his table it turns out the majority of them have trends that, when using Busby's way of reasoning, contradicts his theory. Additionally, all counties except Örebro and Västmanland were identified and named by him on a map of Sweden that was shown in his presentation on the slide before the one with the table (available here), so he should be aware of all the counties and their location.

When he hesitated about Gotland county, I said that results from this county does not confirm his theory and that Gotland, at least in my opinion, is facing the Baltic Sea.
He replied that "These are the data that are available to me, and this is what I have been able to do. If you can do something better or demonstrate what you say you are welcome to do so."
"Sure," I replied, and held up some papers with my own comparison of breast cancer cases, i.e. from the Statistical Database. Someone applauded and I heard a "Bravo!" from some person in the audience. Busby looked embarrassed, 1-0 for me.

My last comment before the discussion was interrupted for the next speaker was something like (this is how I remember it):
"I am not a radio-biologist and can not assess the quality of much of what you say. You have many interesting theories that I and many others try to take seriously, and there are many people in various environmental organizations who rely on you and refer to what you say as established truth. In this case there is a strong suspicion of cherry picking (i.e. choosing the data that confirms one's theory and ignoring those that contradict it, a very dishonest way of handling data for anybody who wants to call himself or herself a scientist). If you've cheated so clearly in a case that is so easy for any Swede to double check, how will we be able to trust anything that you say regarding the other points where it is not so easy for a layman to check your claims? "

Somewhere in our argument a medical doctor in the audience asked if Busby had taken into account that mammography (breast cancer screening) had been introduced in several counties during this time period (it affects the results because more cases of breast cancer will be detected for a few years following the introduction of the procedure). This Busby had no information about, and he had therefore not corrected for this effect. The fact that statistical variations can be quite large in the counties with relatively small populations, was also something that he did not seem to have considered. The data presented in the table are normalized to the number of cases per 100 000 women, and neither any uncertainties nor any attempt to estimate the uncertainties are reported by Busby. Instead, BSRRW, with Busby's approval, issued a press release upon their arrival in Stockholm, which states (the text can, as mentioned above, be found here:
The radioactive contamination of the Baltic Sea is now astonishingly high and is causing serious health problems in coastal populations, said Prof Chris Busby, one of the experts advising the organization. Breast cancer rates in the coastal counties of Sweden increased by 15% in the 5 years following the Chernobyl accident, whereas, in the inland counties the rates actually fell! He said. This too is a matter for the European Courts and National Courts in the affected countries since their citizens are suffering harm.

The graphs that I had made from the statistical database are available as a PDF file for anybody who interested:
The left column shows the number of cases of breast cancer in some selected counties (Stockholm, Uppsala, Kalmar and Gävleborg county are displayed together, and Jämtland, Värmland, Dalarna and Gotland are displayed together), as a function of time over the period 1970-2008, normalized for every 100 000 women in order to compare different counties. The right column gives the same data but in the absolute number of cancer cases for each county. I printed both options to be able to explain if anybody was curious about the data, and also in order to make a rough estimate of the uncertainty in the data (the square root of the absolute number of cases). The years that Busby selected for his study are marked with grey background.
Clearly, the number of breast cancer cases have increased quite steadily since 1970, the red symbols are for the whole nation. There are some variations, and there seems to be a peak around 1989, but without more detailed information one can not draw any conclusions about the causes. The peak is relatively small for the entire nation (but it looks more dramatic for individual counties) and can have many causes. The graphs also show some other drastic fluctuations in different counties, both up and down.

After the BSRRW meeting I made a detailed comparison with Busby's table. In my old plot I had chosen the "Number of new cancer cases per 100 000 persons (crude rate)", so the values in the plots varies from Busby's table because he probably chose the category "Age-standardized incidence per 100 000 population as of 2000". There are different ways to standardize data on the statistical database, and I'm not familiar with what is most relevant, but the trends are roughly the same even if the individual values may differ.
In the picture below, I have done the same exercise as Busby, but I have included the counties that are missing in Busby's presentation. They are divided into the categories BALTIC and INLAND, in the same way as Busby did. I have also included values for the whole nation (Riket).

My version of Bysby's table. I have added the "Baltic" and "Inland" counties that Busby had not included. Click on the picture for full size.

It should be noted that some values do not match identically with Busby, it may be due to rounding errors or that he has accidentally written down the wrong values, or used some normalization that I am not aware of. I have downloaded the data directly into an Excel file from the Statistical Database and taken the values as given by Excel, with rounding to the nearest integer. In any case, the deviations are never more than one or two units. However, the data for Uppsala county differs very much. Maybe Busby has mixed data from different normalizations, or there may be typographical errors. One can also imagine other explanations...

What is most interesting with the table is the counties that are missing in Busby's table. Södermanland county confirms the other trend, but the fact that Gotland, Östergötland and Västerbotten have been excluded seems to be more than an unfortunate coincidence.
Even more interesting is what happens when you look at the inland counties that are missing in Busby's table, three of the four counties disagree with Busby's theory. Now there is no doubt that he has deliberately excluded data that does not fit with his theory (so called cherry picking), a very dishonest way to handle data if you want to be taken seriously.

Note that my table does not in any way prove or disprove any theory regarding the increase (or decrease) of the number of cases of breast cancer associated with Chernobyl. Without much more detailed information, and various corrections for different sources of error (so called confounders) so you can not draw many conclusions from these data. I also have no knowledge of how to implement this type of study in a proper way.
My point is that Chris Busby as well cannot establish any credible link. However, my comparison shows that he has tried to "prove" his theory in a dishonest way by selecting data that fits his theory, and ignoring the data that contradict it. And if it, by any chance, would turn out that Chris Busby actually happened to omit certain counties without any malicious intent, it is very irresponsible of him to travel around and spread the results of such a poor study and argue that it confirms Martin Tondel's PhD thesis from 2007 (see also the press release: If he so recklessly draws conclusions in this case (or is deliberately lying, the reader is kindly adviced to use his or her own judgement), what says that his earlier studies, with a variety of cancer clusters in the UK and elsewhere, have been performed in a more serious way?

The presentation by Busby also dealt with depleted uranium, and a study he recently published (July 2010) about cancers in Fallujah in Iraq. He showed, among other things, a table on the number of births in different age groups and compared the sex ratio. According to Busby, male fetuses tend to be more susceptible to genetic disorders (chemical, radiological) than female fetuses. Normally, there are 1050 boys born for every 1000 born girls, this is pretty constant all over the world. But in Fallujah the Busby group found that the number of male births in the years after the intensive U.S. bombings (2004) was much reduced, only 860 boys per 1000 girls (-18% versus normal). In the next age group, which includes children born in 2000-2004, however, the ratio is almost the reverse, 1182 boys per 1000 girls (+13%). So I asked him what happened in Fallujah in the year 2000. Busby was wondering what I meant. "Because there were much more boys than girls born in the years after 2000, how should I interpret this?" I wondered.
No, I could not argue like that, he said. There are children who died after they were born, or they have moved away since then. The medical doctor, who was on the same track, was pretty intense at this stage, and I tried to give her support so that we could get clear answers from Busby. It turned out that they did not have data on how many children that had been born in the different periods of time (as stated in the article), they only have data on the number of children in different age groups that were there at the time of the survey, undertaken by going around and knocking on doors and doing interviews. "Epidemiologically, he has therefore nothing to show" said the medical doctor.
Anyone interested can read the article here, it has received plenty of attention in the media all over the world:
Note that the article only says in passing that depleted uranium is discussed as a suspected cause. In this case, not even Busby dares to state a connection (although his way of presenting it at the BSRRW meeting gave a different impression). The problem is that so many other people see his study as a proof of how dangerous depleted uranium is. To conduct studies by knocking doors and performing interview does not need to be a bad method (in any case it is better than nothing), but then one should be open with that the method has limitations. Furthermore, one should be careful with drawing drastic conclusions and spread them in public. Once again, Chris Busby has shown his inability to do so.


During the next talk I did not hear so much, because Busby sat down with me and we continued to discuss in whispering tone, which unfortunately disturbed some people in the audience. I showed my plots, and he explained how difficult it is to obtain data and that he certainly knows what he is talking about. I tried to get into his head that he should not be spreading this type of information to the public when it is based on such questionable methods, but whatever I said did not seem to give any lasting impression on him. Then I tried to find out more about his risk model. "You may purchase our report", he said. Meanwhile several people came up to us and wanted to see the plots, some of them left their email addresses to me as they wanted to hear more about it. 2-0 for me. In the end I bought the book with Busby's risk model (ECRR-2010 report: for 25 euro, 2-1 for Busby. Any assessment of this report will be made in some future post, if and when time permits.

In later talks during the meeting, some persons praised Busby's for being a person who stands up and speaks the truth. 2-2 for Busby, I guess a draw is all I get...

Mattias Lantz
Posts: 39
Joined: 11 Dec 2008, 21:08
Location: Wako-shi, Saitama, Japan

Return to The Deep Repository

Who is online

Users browsing this forum: No registered users and 2 guests