• Did INTERPHONE commit …#1 ?

This is the “modified” version of earlier blog. It is published in meantime, while awaiting the response from IARC to my letter (see next blog at  https://betweenrockandhardplace.wordpress.com/2011/07/20/%e2%80%a2did-interphone-commit-2/).

****************

INTERPHONE scientists have recently published two separate papers on the same topic:

Is there relation between tumor location and the area of the brain that was most exposed?

The first article was published in the American Journal of Epidemiology (AJE) (http://www.ncbi.nlm.nih.gov/pubmed/21610117) and the second article was published in the Occupational and Environmental Medicine (OEM) (http://www.ncbi.nlm.nih.gov/pubmed/21659469).

The AJE study was based on the INTERPHONE data from: Denmark, Finland, Germany, Italy, Norway, Sweden, and Southeast England, whereas the OEM study is based on INTERPHONE data from: Australia, Canada, France, Israel and New Zealand. The Japanese data and the Japanese scientists did not participate in either of the publications.

The AJE study is negative. In the opening sentence of the Discussion section of the article the authors said:

“…Our results do not support the hypothesis of gliomas among mobile phone users being preferentially located in the parts of the brain with the highest radio-frequency exposure…”.

The OEM study is weakly positive. In the Discussion section of the article the authors said:

“…Overall, there was weak evidence of stronger associations of glioma and meningioma when a comprehensive estimate of RF dose rather than just mobile phone use was used in the case-control analysis…”.

The strength of the INTERPHONE analysis was to be the size – the largest study with the most cases of tumors. But now, by the split of the data, this advantage has been lost.

In AJE article the authors admit that the size matters and stated in the discussion that:

“…To our knowledge, this is the largest study on detailed glioma localization published to date, with 888 glioma cases from 7 countries. Further research with similar methods but a larger number of long-term users is warranted…”. (Interestingly they did not mention that more cases were availabloe within INTERPHONE).

Whereas in the last sentences of the OEM article the authors stated that :

“…Uncertainties around these results require that they are replicated before they can be considered to be real. The best way to replicate them would be to repeat this analysis in data from the other eight Interphone countries and in future studies with longer latency periods and higher cumulative exposures…”.

What should have been published, and what scientists and decision makers and public at large awaited, was  the analysis of the whole INTERPHONE data consisting of the 888 cases in AJE + 553 cases in OEM + Japan cases of glioma. For now we do not know what such analysis would say?

To the careful reader it appears that the split of data between two publications goes along the lines of what opinions scientists involved in Interphone Consortium have expressed before, in their publications and in their news media interviews. Also, the differences in attitude of the authors towards the possibility of health effects of mobile phone radiation come clearly out in the text of both articles.

Furthermore, neither the authors of the AJE paper nor the OEM paper have provided, in their published articles, any scientific justification for the split of the data and publication of two papers.

The authors of the AJE paper state only that: “…The current analysis was based on data from 7 European study centers within the Interphone Study, an international collaborative case-control study with the main objective of assessing whether mobile phones increase the risk of brain tumors…”

The authors of the OEM paper state only that: “…Analyses used data from the five Interphone countries – Australia, Canada, France, Israel and New Zealand – that agreed to transfer their data to Barcelona when EC, the Interphone Principal Investigator, relocated there…”

I have recently asked the senior authors on AJE and papers to let me know what the scientific reasons for the split of the data were, but I did not receive explanation.

The AJE senior author, in his e-mail, explained the reasons for the separate publication as follows:

“…Quite simply, a large ship turns slowly. With the experience of slow and arduous process of reaching consensus in the large group and in contrast much faster progress with the North European group we wanted to take the next step and move from interview data alone to more in-depth analysis sooner rather than later. I learned only later that another publication was also being prepared in parallel by Elisabeth (who was fully aware of our analysis, as I had circulated our manuscript to her before submission)…”

The OEM senior author has provided an explanation for the separate publications in her comment to me as follows:

“…The reason for the different approaches was mainly circumstantial. I left IARC for Barcelona in 2008 and there was no one left at IARC to coordinate international analyses of Interphone. For reasons of data protection and logistics, only 5 of the countries got permission for the database to be transferred to CREAL for further analyses; those are the countries included in the OEM paper…”

Interestingly, and importantly, the final conclusions in both published studies fit the pattern of what opinions the authors have spoken in the past. Namely:

The AJE article, that was co-authored by scientists who in the past expressed a view that there is no causality link between brain cancer and mobile phone radiation, has concluded that there is no correlation between the location of brain cancer and exposure to mobile phone radiation.

The OEM article, that was co-authored by scientists who in the past expressed a view that there might be causality link between brain cancer and mobile phone radiation, has concluded that there might be correlation between location of the brain cancer and exposure to mobile phone radiation but due to small sample larger studies are needed to confirm this observation.

Of course it is possible to “somehow understand” the reason behind the decision to publish two separate articles – it is the bad experience of INTERPHONE scientists associated with the first whole INTERPHONE paper published in 2010. Writing of this 2010 paper has taken nearly 4 years because within INTERPHONE were clashing two factions – those who think that there is no causal relationship between glioma risk and RF and those who think that there might be. So, in preparation of the second set of data for publication, apparently to avoid lengthy process, the scientists who are known for their “no causality opinion” have gathered together and published AJE study and the scientists who think that there might be a causality link have published OEM paper.

However, as said above, even if “understanding” the frustration of INTERPHONE scientists and their “quest” for quick solutions, it was non-scientific reason for splitting the data. Science did not justify such action.  Unfortunately, science “lost”, and two separate studies were published.

So, for me, these two recent publications from the INTERPHONE should be retracted and replaced by a single analysis. And let me give and example that explains why I think so.

If a scientist would perform 13 experiments in laboratory and then would pick 7 or 5 results and publish them in separate article(s), such scientist would be justly accused of wrong handling of the data.

Are epidemiologists allowed to act differently from other scientists?

It is a very “unhealthy” situation of the search for RF “health” effects. Especially, that everyone looks up to epidemiologists to provide the valid and reliable scientific evidence.

Advertisements

14 thoughts on “• Did INTERPHONE commit …#1 ?

  1. Pingback: Interphone’s new study… | BRHP – Between a Rock and a Hard Place

  2. Pingback: Censorship of science at STUK/Finland – Ministry doesn’t care… | BRHP – Between a Rock and a Hard Place

  3. Pingback: • Did INTERPHONE commit…#3? « "Between A Rock and A Hard Place"

  4. Dariusz,

    For a journalist your comments are very important. While the press is often vilified for not fully reading the scientific results, in this case the more careful the reading the more thorough the confusion.

  5. Dear Elisabeth.

    Thank you for the clarifying messages. I did not go into details of both INTERPHONE articles published in AJE and OEM. My attention, and worry, got the fact that the data was arbitrarily split and published separately. Independently of how “good” or how “bad” the explanation for reasons of separate publications is, the fact remains – two papers were published and both papers will remain as valid peer review references. They should not be. Unless, of course, “someone with guts” will retract them and replace with summary analysis.

    It seems that the polarization of stances among some of the researchers in this filed, into two very vocal groups – those saying always NO and those saying always YES, has strong impact on what is published and how things are presented in conclusions of peer-review papers. It seems that reliable science has less and less to say and the scientists who are trying to present more balanced view are being ostracized by both “NO” and “YES” groups. In the end the quality of science suffers.

    And I think that the non-scientific reasons and the struggle between “NO” and “YES” camps are also to blame for the AJE and OEM “publications”.

    Best wishes and hopefully see you in China!
    Dariusz

  6. Graham

    The reason for the different approaches was mainly circumstancial. I left IARC for Barcelona in 2008 and there was no one left at IARC to coordinate international analyses of Interphone. For reasons of data protection and logistics, only 5 of the countries got permission for the database to be transfered to CREAL for futher analyses; those are the countries included in the OEM paper.
    .
    Now that Joachim Schuz is at IARC,this will is less problematic. We have jointly recontacted the 13 country investigators and are working on a plan for further international analyses.

    Best regards

    Elisabeth

  7. Elisabeth,

    Many thanks for this informative reply. Were the other 8 papers excluded because the PIs didn’t want their data analysed in this way, was it because the additional workload of acquiring and analysing the data with the team involved in your paper would have been too much, or was there another reason these papers were not included?

    Best Regards,
    – Graham

  8. I have another query regarding possible serious scientific incompetence that especially applies to these last two papers regarding brain tumour location with respect to the “radiation plume” of the handset.

    Neither paper addressed or even discussed the real ELF magnetic field pulses that surround handsets, especially GSM ones (that this data would largely come from), that result from time-varying currents taken from the battery. ELF magnetic fields have been classified as IARC 2B since 2002. The levels close to the handset can exceed 50 microteslas and the levels though most of the user’s head by more than 1 microtesla. Incidence of Childhood leukaemia is reported as doubled at about 0.4 microtesla. So mobile phone handset use studies, especially as regards EMF issues (RF or ELF) should always address both. The authors have known this since before Interphone started as this was in the published scientific literature by 1997.

    It seems totally inappropriate, to me, in these analyses regarding tumour location to completely ignore another acknowledged (IARC 2B) possible major confounder that is present in the total EMF exposures from the handsets and that would be exposing much of head at significant levels.

    Alasdair (Philips), Powerwatch.

  9. This is a very good analysis by Dr Leszczynski . This should be taken very serious by the scientific community. Τhis type of approach(ignoring data, etc.) to EMF problem can be seen in published studies but also in consensus reports etc.. We have to establish rules so such behaviors are eliminated or minimized.

    Stelios A Zinelis

  10. Dear Dariusz,

    Just a couple of clarifications:

    – your quote from the OEM paper discussion refers to only one set of analyses in the paper, not to the overall results: there were case-control and case-case analyses that strengthen each other. If you look at the summary of the paper – in the “what this paper adds” section, you can read “Our results suggest that there may be an
    increase in risk of glioma in the most exposed area of the brain among long-term and heavy users of mobile phones. These results are uncertain (in light of the uncertainties associated with tumour centre localisation, radio frequency dose estimation and sample size) and require replication before they can be taken to indicate a cause effect relationship”.

    – the last sentence of the OEM paper actually states: “The best way to replicate them would be to repeat this analysis in data from the other eight Interphone countries and in future studies with longer latency periods and higher cumulative exposures.”

    We are definitely in agreement that these analyses should be conducted on the entire 13-country dataset and this has been discussed several months ago by the Interphone Study Group. There was no omission of data based on results, since, at the time the analyses were undertaken by the different groups, no one had any idea what the results would be.

    Now that results of the individual initiatives are available, methods of both papers will be shared with the study PIs, comments considered and plans made to proceed with similar analyses on the entire dataset.

    With best wishes

    Elisabeth

  11. This kind of output is realy surprising to hear about if we consider that the INTERPHONE study is one of the biggest and more expensive study made on mobile phones…
    This is too bad.

  12. Thanks Mike. Nice to hear that retired scientists 😉 still follow the field. Dariusz

  13. You make some good points Dariusz. Well done… pity the Interphone scientists do not act as ONE. Mike Repacholi

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s