Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Too much radiation leaked during sham exposure
- David Bücher (13 March 2006)
-
Sham exposure too high
- Clas Tegenfeldt (22 April 2006)
-
What about confounders?
- Dr. Vinay Bothra (26 April 2006)
-
Signal level
- Rein B Meijboom (27 April 2006)
-
Are some people sensitive to mobile phone signals? Author's Response
- G James Rubin, Anthony J Cleare, Simon Wessely (3 May 2006)
-
The issue of bias and experimental method
- Jonathan M. Woolfson (15 February 2007)
-
Authors' response to Woolfson
- G James Rubin, Anthony J Cleare, Simon Wessely (21 February 2007)
-
Big money and sham science
- Jonathan M. Woolfson (29 April 2007)
|
|
|||
|
David Bücher, student of mathematics Darmstadt, Germany
Send response to journal:
|
The specific absorption rate (SAR) of the radiation leaked during sham exposure (stated to be approximately 2 mW/kg, [1]) might be considered as too high: it is even above the SAR during the exposure phase of the study by Zwamborn et al. 2003 [2], which was less than 0.1 mW/kg. 1. MCL MTHR GSM and TETRA handset exposure systems for human volunteer studies. www.mcluk.org/MTHR_exposure_systems 2. Zwamborn APM, Vossen SHJA, van Leersum BJAM, Ouwens MA, Mäkel WN. Effects of Global Communication system radio-frequency fields on Well- Being and Cognitive Functions of human subjects with and without subjective complaints. TNO-report FEL-03-C148 (2003). www.tno.nl/tno/actueel/tno_nieuws/2003/onderzoek_tno_naar_effect/tno_fel_report_03148.pdf Competing interests: None declared |
|||
|
|
|||
|
Clas Tegenfeldt, EMF consultant Sweden
Send response to journal:
|
A dual band GMSM telephone can have an transmitter power of 0,002 W up to 2 Watts. The SAR of some GSM telephones reach 2 W/kg, if we assume this happens when the phone transmitts 2 W then we would have 0,002 W/kg when the phone transmitts at its lowest level. Thus the sham exposure in the report is about the same as when someone is talking in a GSM telephone near a basestation (thus the lowest transmitted power). This happens for indoor GSM systems as well as macro cells (if you are some hundred meters from the basestation). The difference reported in the study is 1,4 W/kg and 0,002 W/kg for exposure and sham respectively, that is less than a factor of thousand I used above. It is therefore HIGHLY questionable that the SHAM exposure can be regarded as such! If there would be a clear cut threshold for symptoms provoked by the exposure that happened to be between the 0,002 and 2 W/kg exposures then - maybe - the study would have given some information. But, it is also possible, and also more probable, that the sham exposure (like talking in a GSM phone at lowest power) is above the threshold and therefore CANNOT be considered a SHAM exposure. The reported fact (see fig. 2 in report) that symptoms increased even during "sham" exposure is a dead giveaway that the "sham" exposure wasn´t sham! The subjects gave a baseline for symptoms after 30 minutes in the enviroment, but still: their symptoms started to increase even during sham for 50 minutes and then started to decline after stopping the exposure! Rubin et. al. makes a point of this "interesting fact" but have completely missed the obvious! If one assumes that all three exposure conditions are exposured (no sham) the graph (fig.2) make more sense. It is also highly questionable that the sham exposure was due to technical constraints, it is NOT a hard task modifying a GSM phone (or other RF exposure device) such that the leakage is well below their stated "SHAM" of 0,002 W/kg. From a technical viewpoint the "sham" exposure of 0,002 W/kg is totally unnecessary and unmotivated. The motivation for the sham exposure is "heat", but is that a justifiable motivation? One can simply add a thermal resistance powered by a DC voltage instead!!!!!! Such a resistor costs just about what one second of laboratory time costs... Competing interests: EMF consultant |
|||
|
|
|||
|
Dr. Vinay Bothra, SpR Public Health, Cheshire & Merseyside Health Protection Unit Health Protection Unit, Moorgate Point, Moorgate Road, Knowsley Industrial Park, Kirkby L33 7XW
Send response to journal:
|
Editor – The article “Are some people sensitive to mobile phone signals? Within participants double blind randomised provocation study”1 was recently selected for critical review by our local journal club group. It is interesting to note that this study did not follow the usual randomised control trial (RCT) methodology. The unique advantage of RCT is that the potential for bias in allocation to study groups is removed and randomisation takes account for the effect of both known and unknown confounding factors. In the above study, the authors pre-selected subjects into ‘sensitive’ and ‘control’ groups, and this undoubtedly raises questions regarding the control of known and (especially) unknown confounders. In our view, this trial could have been designed to have a study sample of ‘sensitive’ subjects (only) and divide them into two groups – one offered the intervention (mobile phone signal) and the other ‘control’ group given non-pulsing carrier wave signal or sham treatment (no signal). Both groups could then be followed-up and the results analysed for the same outcome measures. This design would take account of unknown confounders, make the trial simpler and yet answer the same research question posed in this study. The issue of a small study sample could be resolved using a crossover trial. Here the intervention comparison is “within-subject” rather than “between-subject” and the sample size needed is smaller. This study design would have been appropriate in this case as the effect of the intervention can be assessed “quickly” and carry-over effects managed by introducing wash-out periods between the interventions. Reference 1. Rubin GJ, Hahn G, Everitt BS, Cleare AJ, Wessely S. Are some people sensitive to mobile phone signals? Within participants double blind randomised provocation study. BMJ 2006; 332: 886-9. Competing interests: None declared |
|||
|
|
|||
|
Rein B Meijboom, None 2803BP Gouda
Send response to journal:
|
"The rooms, ... , were not shielded against outside electromagnetic fields. " This sentense states that the signal applied is added to an unknown level which is not constant. I assume the reaction of humans to electromagnetic fields is to some extent dependent on its level. So what has been tested? Rein Meijboom Competing interests: None declared |
|||
|
|
|||
|
G James Rubin, Lecturer King's College London, Institute of Psychiatry, Weston Education Centre, Cutcombe Rd, London SE5 9RJ, Anthony J Cleare, Simon Wessely
Send response to journal:
|
We thank the authors for their comments and are grateful for this opportunity to respond. Perhaps the most serious issue is that raised by Bücher and Tegenfeldt regarding the leakage from our exposure equipment in the sham mode (SAR<0.002W/Kg). The implication is that an exposure at this level may have been sufficient to trigger symptoms in our sensitive group, thus obscuring any difference between this and our active GSM condition (SAR=1.4W/Kg). We believe that this concern is misplaced, for three reasons: 1. Our sham leakage did not pulse: Although theories to explain why modern mobile phone signals might cause adverse health effects remain speculative, most commentators who have expressed concern have suggested that pulsing signals such as GSM may give rise to greater health effects than non-pulsing signals [e.g. 1, 2], possibly through their interaction with biological electrical activities at a cellular level [1]. As stated in our paper, the leakage in our sham condition was of a non-pulsing, continuous wave signal. If this concern about pulsing mobile phone signals is valid, then our GSM signal should have resulted in greater symptom severity than the non-pulsing leakage. Under the same rationale we question the relevance of the TNO study mentioned by Bücher [3, see also re-analysis in 4]. That study tested participants who reported sensitivity to GSM mobile phone masts. Our study tested for sensitivity to GSM handsets. In areas normally accessible to the general public, exposures to the signals emitted by masts are very much weaker than the exposure received from using a handset, hence the difference in signal strengths used by the two studies. More importantly, of the three active conditions used in the TNO experiment (GSM-900, GSM- 1800, UMTS) only UMTS (a “3G” signal) had any significant effect on the subjective well-being of the volunteers, and this despite these volunteers having been recruited by the basis of their reported sensitivity to GSM rather than UMTS. These results await replication, but if valid suggest that not all forms of signal have adverse health effects. We would re- iterate that our control condition leaked a non-pulsing continuous wave signal, and not a pulsing signal such as UMTS. 2. The importance of a dose-response effect: Tegenfeldt questions whether the “threshold” at which mobile phone signals cause subjective symptoms lies above or below our 0.002W/Kg sham leakage. But is it valid to assume that the relationship between mobile phone exposure and symptoms is governed by a threshold, or should we expect a dose-response relationship instead? Data from those observational studies which have reported an association between mobile phone exposure and symptoms suggest that the dose-response model is the more accurate. For example, several studies have reported that increased frequency of mobile phone use and increased duration of calling time correlate with increased headache prevalence and severity, while increased use of hands- free kits is associated with decreased headache prevalence and severity [5, 6, 7]. Others have reported that individuals whose phones have an SAR>0.5W/kg are more likely to report symptoms than those whose phones have an SAR<0.5W/kg [8]. Even some observational studies on mobile phone mast emissions [e.g. 9] suggest what has been described as “statistically significant dose-response associations between the measured exposure level and [numerous symptoms]” [10]. The mean power of the signal in our active GSM condition was over 700 times stronger than that in the sham condition. Given this evidence of a dose-response relationship, it seems reasonable to assume that our active and sham conditions should have elicited very different levels of symptom severity, if GSM sensitivity does indeed have an organic basis. 3. The similarity between our results and those of other provocation studies: No study should ever be viewed in isolation. We would therefore draw readers’ attention to the seven other blind or double-blind provocation studies that have previously tested whether people who report sensitivity to mobile phone handsets can detect the signals coming from these handsets or react to them with increased symptom reporting [11, 12]. Several of these have used real handsets switched on or off as the exposure conditions, and none has described any issue with leakage from their equipment. As such the concerns of Bücher and Tegenfeldt would not seem to apply to them. Yet the overall results of those experiments are the same as our own. While it is possible that different methodological problems have affected each study, a more parsimonious explanation is that there is simply no effect there to be seen. Our third correspondent, Dr Bothra, questions whether the design of our study was appropriate. Firstly he asks whether by “pre-selecting” our participants into the sensitive and control groups we may have introduced bias into the results. This seems something of a moot point. We did not “select” our participants to be sensitive or control; they came to us either reporting problems with mobile phones or not. Dr Bothra then makes further suggestions before concluding that a within-participants study design “would have been appropriate in this case.” Our study did indeed use a within-participants design, with each volunteer being exposed to each of our three experimental conditions. Finally, Meijboom highlights the fact that our rooms were unshielded and as such we did not know the background level of electromagnetic fields. This is correct. However, each of our participants spent 30 minutes adjusting to the background electromagnetic field levels in our rooms prior to each testing session. Only two reported severe symptoms following this period, and both were excluded. As such, what was being tested was what participants reported in real life – that exposure to a mobile phone signal against the backdrop of everyday electromagnetic fields results in increased symptom severity. We question the need to create an artificially “clean” environment in order to observe symptoms that seem readily apparent to the volunteers during their everyday lives in the real world. References 1. Hyland G.J. Physics and biology of mobile telephony. Lancet. 2000;356: 1833-1836. 2. Independent Advisory Group on Non-Ionising Radiation. Health Effects from Radiofrequency Electromagnetic Fields. 2003. National Radiological Protection Board: Oxfordshire. Doc NRPB 14 (2) 1-177. 3. Zwamborn, A.P.M., Vossen, S.H.J.A., van Leersum, B.J.A.M., Ouwens, M.A., & Makel, W.N. Effects of Global Communication System radio- frequency fields of well being and cognitive functions of human subjects with and without subjective symptoms. 2003. The Hague, Netherlands Organisation for Applied Scientific Research (TNO). 4. Health Council of the Netherlands. TNO study on the effects of GSM and UMTS signals on well-being and cognition. 2004. The Hague: Health Council of the Netherlands; publication number 2004/13E. 5. Chia S.-E., Chia H.-P. & Tan J.-S. Prevelance of headache among handheld cellular telephone users in Singapore: A community study. Environ Health Perspect. 2000;108:1059-1062 6. Oftedal G., Wilen J., Sandstrom M. & Mild K.H. Symptoms experienced in connection with mobile phone use. Occup. Med. 2000;50:237- 245. 7. Hocking B. Preliminary report: Symptoms associated with mobile phone use. Occup. Med. 1998;48:357-360. 8. Wilen J., Sandstrom M. & Mild K.H. Subjective symptoms among mobile phone users - a consequence of absorption of radiofrequency fields? Bioelectromagnetics. 2003;24:152-159. 9. Hutter H.-P., Moshammer H., Wallner P., & Kundi M. Subjective symptoms, sleeping problems, and cognitive performance in subjects living near mobile phone base stations. Occup. Environ. Med. 2006, 63:307-313. 10. http://www.powerwatch.org.uk/news/20040809_spain_dt.asp [accessed 2 May 2006] 11. Rubin G.J., Das Munshi J. & Wessely S. Electromagnetic hypersensitivity: A systematic review of provocation studies. Psychosom. Med. 2005:67; 224-232. 12. Wilen J, Johansson A, Kalezic N, Lyskov E, Sandstrom M. Psychophysiological tests and provocation of subjects with mobile phone related symptoms. Bioelectromagnetics 2006:27;204-214. Competing interests: None declared |
|||
|
|
|||
|
Jonathan M. Woolfson, teacher Italy
Send response to journal:
|
The context for the research reported by this article is the condition known as electrical sensitivity. A little background research into that phenomenon would have revealed to the designers of this experiment that - sufferers react in very different ways to a wide range of different electromagnetic frequencies and field strengths - that many more symptoms than those reported in the article are commonplace - that those most debilitated by the condition (and therefore theoretically most likely to register ‘positive’ results in a trial such as this) would not be able to travel to the test site because of generalized surrounding environmental electromagnetic pollution; consequently the self-selecting group of sensitive people were not actually that sensitive - that even less sensitive people often suffer also from chemical and other environmental intolerances, meaning that they would have experienced reactions to the presence of such residues as wood resin, paint or cleaning fluids in the trial location - that failure to shield the room from other electromagnetic fields was a sure-fire way of skewing the results! - that the onset of symptoms can occur many hours after exposure, and recovery can take anything between one day and three weeks: this renders the reported results practically meaningless. Why did the authors of this article not gather information on this wider context for their work before designing this experiment? An investigation of this condition which seriously entertained the possibility that it is ‘physical’ rather than ‘psychological’ would have had to take account of these essential factors rather than simply attempting to test the reports of a minute sub-group of sufferers about their response to one particular kind of electromagnetic pollutant. It is difficult not to feel that the experiment was indeed planned to demonstrate that mobile phones do not harm humans – hardly a surprise, then, that that is exactly what it did. I invite the authors to speculate on what their experiment would have looked like had some of the funding for it come from a powerful, multi-million pound industry whose primary interest was in stopping people from using mobile phones. Competing interests: None declared |
|||
|
|
|||
|
G James Rubin, Lecturer Mobile Phones Research Unit, King's College London, SE5 9PJ, Anthony J Cleare, Simon Wessely
Send response to journal:
|
We thank Woolfson for his interest in our paper. As we have conducted more than a little background research into electrosensitivity (1, 2), we are happy to revisit this debate with him. He is correct that electrosensitivity is a heterogeneous condition, particularly when it comes to the sources of electromagnetic fields that are reported as triggers. However, although several surveys of electrosensitivity sufferers have reported that mobile phones are one of the commonest types of trigger (3, 4), with one UK survey suggesting that up to 69% of sufferers find them ‘moderately’ problematic or worse (5), we were clear that our study did not test all possible forms of electrosensitivity, but simply whether some people are sensitive to mobile phone signals. In this context, we are also quite content with our choice of primary outcome measure: although electrosensitivity sufferers report a wide variety of symptoms (3), the symptom most commonly attributed to mobile phone use is headache (6). Woolfson suggests that severely debilitated electrosensitivity sufferers will not have taken part in our study. This is apparently problematic because they are the people “most likely to register ‘positive’ results in a trial.” According to Woolfson, our test yielded ‘negative’ results because our “self-selecting group of sensitive people were not actually that sensitive.” Whilst this might be a valid argument, we fear that Woolfson has overstated his case. Firstly, while some people with severe electrosensitivity are not willing to take part in this form of study, others are. As we noted in our paper, 18 of our participants experienced ‘definite’ impairment or worse in their everyday lives as a result of their condition and many experienced reactions during our study that were so severe that we had to halt the testing early for them. We therefore disagree with Woolfson’s assessment that these volunteers “were not actually that sensitive.” Secondly, Woolfson’s argument suggests that only people who are highly debilitated by electrosensitivity would be able to detect a difference between our experimental sessions. Again, we have to disagree. The question to ask is not ‘were the participants highly sensitive,’ but rather ‘were the participants sensitive enough.’ Our participants clearly were sensitive enough. As a group they reported usually experiencing noticeable symptoms after only 6.5min of beginning a mobile phone call (our experimental sessions lasted for 50mins), they reported normally experiencing headaches in over 70% of calls (headache was our primary outcome) and they were confident that they could tell which of our sessions were active (although this confidence turned out to be misplaced). With regards to the issue of chemical exposures and background EMF levels in our testing rooms, we would reiterate again that all of our participants spent 30mins at the start of each testing session sitting quietly in the room, after which we asked them to report if they were experiencing any symptoms. Only two reported baseline symptoms that might have masked any effects of our exposure, and both were excluded from the remainder of the study. With regards to whether we should have screened our room against background EMFs, we refer Woolfson to our answer to Meijboom, above. We would add that of the 36 provocation studies into electrosensitivity that we are now aware of, some have been conducted with screening and some without: in practice this makes little difference to the results (1). As to whether “symptoms can occur many hours after exposure,” we agree that they can, but note that most people with electrosensitivity say that their symptoms occur within only a few minutes of exposure (3). The participants in our study reported that they usually experienced symptoms within a mean of 6.5min from beginning a call. Similarly, although we are sure that for some people “recovery can take anything between one day and three weeks,” for most it is much quicker than this, tending to be only slightly slower than onset time (3). For the large majority of participants in our study, recovery was reported as usually occurring within 2 hours. We did check for possible carry over effects by asking participants to score their symptom severity at the start of each testing session, but saw nothing to suggest that this was a major issue. Finally, Woolfson invites us to speculate about what our study would have looked like if the funding source had been different. Anke Huss and colleagues have recently conducted a systematic review examining the effects of funding on the results of studies of controlled exposure to radiofrequency radiation with health-related outcomes (7). Their conclusions? That studies such as ours with ‘mixed funding’ are just as likely to observe a significant effect as studies funded solely by charities or public bodies. However, ‘mixed funding’ studies did tend be the most methodologically robust of all those they examined. 1. Rubin GJ, Das Munshi J, Wessely S. Electromagnetic hypersensitivity: A systematic review of provocation studies. Psychosom Med 2005;67: 224-232. 2. Rubin GJ, Das Munshi J, Wessely S. A systematic review of treatments for electromagnetic hypersensitivity. Psychother Psychosom 2006;75: 12-18. 3. Roosli M, Moser M, Baldinini Y, Meier M, Braun-Fahrlander C. Symptoms of ill health ascribed to electromagnetic field exposure - a questionnaire survey. Int J Hyg Environ Health 2004;207: 141-150. 4. Schuz J, Petters C, Egle UT, Jansen B, Kimbel R, Letzel S et al. The "Mainzer EMF-Wachhund": Results from a watchdog project on self- reported health complaints attributed to exposure to electromagnetic fields. Bioelectromagnetics 2006;27: 280-287. 5. Eltiti S, Wallace D, Zougkou K, Russo R, Joseph S, Rasor P et al. Development and evaluation of the electromagnetic hypersensitivity questionnaire. Bioelectromagnetics 2007;28:137-151. 6. Oftedal G, Wilen J, Sandstrom M, Mild KH. Symptoms experienced in connection with mobile phone use. Occup Med 2000;50: 237-45. 7. Huss A, Egger M, Hug K, Huwiler-Muntener K, Roosli M. Source of funding and results of studies of health effects of mobile phone use: Systematic review of experimental studies. Environ Health Perspect 2007;115: 1-4. Competing interests: None declared |
|||
|
|
|||
|
Jonathan M. Woolfson, teacher Italy
Send response to journal:
|
I am grateful to the authors for their thorough answer to my original query. Since this issue is currently in the headlines in the UK, it seems a timely moment to respond. In my former message I invited the authors to speculate about what their experiment might have looked like had some of its funding come from a powerful, multi-million pound industry whose primary interest was in stopping people from using mobile phones. My own sense is that the authors will have trouble in persuading people that scientific research funded by big business is immune to inevitable, albeit mainly indirect, pressures; but this is not really my point, which is rather about possible defects in the authors’ experiment design, defects probably shared by many such experiments (see Rubin’s answer to Meijboom). It seems to me that the authors start with the following premise: ‘We don’t believe this is really a physical phenomenon; let’s try and design an experiment that proves that it is’. My invitation was to speculate about what an experiment would look like that started with the opposite premise (ie, in my speculative scenario, an experiment partly funded by powerful supporters of the banning of mobile phones): ‘We believe this really is a physical phenomenon; let’s try and design an experiment that proves that it is NOT’. In the experiment designed by Rubin et al. many people are apparently reacting to exposure, but since they are also apparently reacting to being told that they MIGHT be exposed, the results from actual exposure appear to be undermined. It seems characteristic of the experimenters that, beyond all the other confounding factors that have been raised in this discussion, it never crosses their minds that if you tell people 'I may or may not be about to inflict considerable discomfort on you, but I am not going to tell you', they may exhibit anxiety symptoms, some of which may be confused with the symptoms of electrosensitivity. I propose that only if this ‘suggestibility’ factor, which appears to suit the purposes of their funders so conveniently, is removed, could we really find out what is going on. That means that subjects must not be told that they are about to take part in such an experiment. I recognise that this is neither practicable nor, above all, ethical, but I hope that thinking about it may help us to understand why the experimenters are coming up with answers that the everyday experience of tens of thousands of people worldwide (the former president of the World Health Organization, Dr Gro Harlem Brundtland, included) prove to be so depressingly wrong-headed. Of course this is not true of all such researchers in this area, and the authors will know that countries as diverse as Russia, Sweden, Germany and Switzerland have all approached the health risks of electromagnetic fields with a very different spirit. I wonder why this is the case. Competing interests: None declared |
|||