Bioelectromagnetics

Mmm this is strange. I have a couple of times asked Roger for the details about his in vitro experiment with the Harmonizer device. He has not replied. Since the information is near at hand, I'm tempted to suspect that he'd rather I did not look at it.

But I will. It is right here.

Analysis

Now, of course the first and foremost criticism of this experiment is that there is 1 (one) sample of every group. This alone is enough to reject the trial, since, while you may do math on the results, you cannot make any stastistical conclusions, simply because you cannot determine standard distributions for any of the groups. Since the test uses continous data of good resolution and credibility, it would not have taken much to overcome this obstacle: A mere 3 samples per group would have been enough.

Another serious problem is described here:
One Personal Harmoniser-exposed sample and one unexposed sample were placed in separate mu-metal boxes and then exposed overnight to a Philips 301 mobile phone on standby, by means of a separate 30cm gold wire leading into each box. For the same period one Implosion-exposed sample and one unexposed sample were also placed in separate mu-metal boxes, into which separate but adjacent gold wires were introduced, but kept in a separate room and not near the cellphone. The final (fifth) sample was enclosed in a mu-metal box and then placed inside a double skinned mu-metal container for the duration of the experiment and maintained at 20 ° C.
So the samples were placed in three different locations during the test. This means that a number of confounding factors could exist, most pertinently:

- Temperature differences.
- Humidity differences.
- Exposure to other ambient fields.
- Exposure to electrostatic discharge.
- Exposure to light.
- Exposure to mechanical disturbances.

This is a secondary reason to dismiss the trial.

A number of smaller problems exist, e.g.:

- The degree of exposure from the cellphone is unknown.
- It is not clear how cells were selected for counting, it is just stated that at least 500 were counted from each sample. Even it the counting is blinded, this procedure is doubtful.
- Devices were exposed to the harmonizer by connecting them via a wire that was brought into contact with the device. The wire remained in the sample (a similar wire was in the other samples). Obviously this means that even if a better designed test should verifiy this result, the test does not show anything about the Harmonizer device as such, since it does not exclude the possibility that touching the wire to any metal object would have the same effect.

Finally, both the begining and the end of the report contains a lot of irrelevant data cited from uncertain sources, and a lot of unsubstantiated speculation.

The information about the resonance frequenct of the harmonizer device is implausible. The frequency cited (625MHz) is based on half-wave resonance of a wire the lenght of an uncurled device, but since the device is curled in a double spiral, this assumption is wrong. It would be fairly easy to measure the actual resonance frequency, but it seems rather irrelevant.

My conclusion

The test is basically flawed, because the sample size is too small to enable the conclusions made, and because there are many serious potential confounders. Also, the design is not well suited to test for the desired effect, because even without the flaws, it would not be possible to reach an unambigous conclusion based on the results.


Mmm, Roger, you are awfully fond of Trypan Blue counts and samples in mu-metal containers with gold-wires. Are those the only kind of tests your laboratory can do ;)?

Hans
 
cogreslab said:
The first cited study (electric blankets) was surely negative, but the Lagorio et al study says:

"At the end of 2000, the epidemiological evidence to support the association between exposure to extremely-low-frequency magnetic fields and the risk of childhood leukemia is less consistent than what was observed in the mid 90s". This isn't a negative finding, surely?

Maybe it's a reading comprehension issue on your part .

To help you I have annotated (in bold italics) the abstract to facilitiate your understanding

We review the epidemiological evidence on childhood leukemia and residential exposure to 50/60 Hz magnetic fields we are reviewing a series of existing studies. The possibility of carcinogenic effects of power frequency magnetic fields (ELF-EMF), at levels below units of micro tesla (microT), was first raised in 1979 by a case-control study on childhood cancer carried out in Denver, USA A putative EMF-cancer link was identified in 1979. In that study, excess risks of total cancer and leukemia were observed among children living in homes with "high or very high current configuration", as categorised on the basis of proximity to electric lines and transformers brief description of the methodology - suggestion that the classification of exposure was sub-optimal. Many other epidemiological studies have been published since then, characterised by improved--although still not optimal--methods of exposure assessment since this date other studies have beee run with gardually improving means of assessing exposure (no it's still not perfect). At the end of 2000, the epidemiological evidence to support the association between exposure to extremely-low-frequency magnetic fields and the risk of childhood leukemia is less consistent than what was observed in the mid 90s the more recent studies (i.e. those in which the esposure has been more accurately been established) show epidemiological evidence at odds with the premise that EMF causes cancer. At the same time, a growing body of experimental evidence has accumulated against both a direct and a promoting carcinogenic effect of ELF-EMF. there's more evidence, from non-epidemiological sources, refuting a link between EMF and cancerSuch "negative" experimental evidence hampers a causal interpretation of the "positive" epidemiological studies we cannot do a causal analysis on the epidemiologal studies on the grounds that (1) the epidemiological evidence is evaporating now that the exposure is being determined properly and (2) there seems to be no link when we look at non-epidemiological studies

If you consider that positive, no wonder you see evidence everywhere!
 
MRC_Hans said:
Mmm this is strange. I have a couple of times asked Roger for the details about his in vitro experiment with the Harmonizer device. He has not replied. Since the information is near at hand, I'm tempted to suspect that he'd rather I did not look at it.

But I will. It is right here.

Analysis

Now, of course the first and foremost criticism of this experiment is that there is 1 (one) sample of every group. This alone is enough to reject the trial, since, while you may do math on the results, you cannot make any stastistical conclusions, simply because you cannot determine standard distributions for any of the groups. Since the test uses continous data of good resolution and credibility, it would not have taken much to overcome this obstacle: A mere 3 samples per group would have been enough.

Another serious problem is described here: So the samples were placed in three different locations during the test. This means that a number of confounding factors could exist, most pertinently:

- Temperature differences.
- Humidity differences.
- Exposure to other ambient fields.
- Exposure to electrostatic discharge.
- Exposure to light.
- Exposure to mechanical disturbances.

This is a secondary reason to dismiss the trial.

A number of smaller problems exist, e.g.:

- The degree of exposure from the cellphone is unknown.
- It is not clear how cells were selected for counting, it is just stated that at least 500 were counted from each sample. Even it the counting is blinded, this procedure is doubtful.
- Devices were exposed to the harmonizer by connecting them via a wire that was brought into contact with the device. The wire remained in the sample (a similar wire was in the other samples). Obviously this means that even if a better designed test should verifiy this result, the test does not show anything about the Harmonizer device as such, since it does not exclude the possibility that touching the wire to any metal object would have the same effect.

Finally, both the begining and the end of the report contains a lot of irrelevant data cited from uncertain sources, and a lot of unsubstantiated speculation.

The information about the resonance frequenct of the harmonizer device is implausible. The frequency cited (625MHz) is based on half-wave resonance of a wire the lenght of an uncurled device, but since the device is curled in a double spiral, this assumption is wrong. It would be fairly easy to measure the actual resonance frequency, but it seems rather irrelevant.

My conclusion

The test is basically flawed, because the sample size is too small to enable the conclusions made, and because there are many serious potential confounders. Also, the design is not well suited to test for the desired effect, because even without the flaws, it would not be possible to reach an unambigous conclusion based on the results.


Mmm, Roger, you are awfully fond of Trypan Blue counts and samples in mu-metal containers with gold-wires. Are those the only kind of tests your laboratory can do ;)?

Hans

It is always useful to have an independent person taking a critical look, and I am genuinely grateful, Hans, for your taking the trouble to do this.

We always try to construct study designs so that they can be replicated. We used Trypan blue because it is a well tried means of assessing viability (on pubmed there are over 2000 studies using that technique, i see).

The issues regarding ambient exposures, humidity and electrostatics were addressed by using a mu metal container. This not only keeps out light but also stray fields. The temp. and humidity was measured as being the same throughout (though we did not include this in the report), so I don't think there was any serious confounding on that score.

As to the sample size, I would say that there are many studies using that number of cells. One problem if you use a larger number is that the ambient effects around the microscope e.g. the halogen light start having an effect too, which can be a bigger confounder. I recall Bob Liburdy (yes, the very same) once presenting a fascinating study of how even different microscope objectives could have different magnetic field strengths. Also the sheer time it takes to count can mean the cells ahave degraded somewhat by the end of the exercise.

Yes, we use mu metal containment a lot to cut down stray radiation, since without it the sort of confounders you mention might be an uncontrolled influence. We use gold wire now and again, and we also use test tubes (you forgot to mention those - Just a joke).
 
To Don: "the more recent studies (i.e. those in which the esposure has been more accurately been established) show epidemiological evidence at odds with the premise that EMF causes cancer".

NO. The most recent still persistently confirm this effect. The two largest (NCI and UKCCCR) both confirm this effect.

Tell me of any well-conducted studies which do not confirm this effect.
 
cogreslab said:
It is always useful to have an independent person taking a critical look, and I am genuinely grateful, Hans, for your taking the trouble to do this.

Mmm, OK, you're welcome. I design and review tests for a living, so, anytime ;).

We always try to construct study designs so that they can be replicated. We used Trypan blue because it is a well tried means of assessing viability (on pubmed there are over 2000 studies using that technique, i see).

Yes, no doubt. This was mainly a joke, but not quite. In using this method, you make the unfounded a priori conclusion that the experiment has an impact on the viability on lymphocytes. Of course you also need to show that this has any relevance to the in vivo situation.

The issues regarding ambient exposures, humidity and electrostatics were addressed by using a mu metal container. This not only keeps out light but also stray fields.

The mumetal screen is not very effective against magnetic fields. If you get a 75% reduction, you're lucky, and that is provided you never dented them or dropped them on the floor. EM fields are coupled straight into the container by your gold wire.

The temp. and humidity was measured as being the same throughout (though we did not include this in the report), so I don't think there was any serious confounding on that score.

Mmmm, a good rule of auditing is that if it is not in the report, it is not there. Unless you took special precautions there vill have been small differences. Even within the same room, there can be considerable differences.

As to the sample size, I would say that there are many studies using that number of cells. One problem if you use a larger number is that the ambient effects around the microscope e.g. the halogen light start having an effect too, which can be a bigger confounder. I recall Bob Liburdy (yes, the very same) once presenting a fascinating study of how even different microscope objectives could have different magnetic field strengths. Also the sheer time it takes to count can mean the cells ahave degraded somewhat by the end of the exercise.

Uhh, it was not THAT sample size that was my primary objection. It was the number of containers. You had inly 5. You cannot make stastistics from an n=1 experiment. (This is a fault I note in most of your reports, btw.

Yes, we use mu metal containment a lot to cut down stray radiation, since without it the sort of confounders you mention might be an uncontrolled influence. We use gold wire now and again, and we also use test tubes (you forgot to mention those - Just a joke).

Well, as long as you do not put too much trust in them.
You know, your opening statement is nice and positive, but after that, I am disappointed. You are not addressing the main problem here: The one-sample design. This makes your experiment useless. Also note my remarks about the design being unable to produce an unambigous result, even if the errors are corrected.

Now, the error you make is by no means unusual; htere are tons of experiments out there where the sample size does not support a stastistical evaluation. But, unfortunately, it makes your experiment worthless.

Hans
 
cogreslab said:
To Don: "the more recent studies (i.e. those in which the esposure has been more accurately been established) show epidemiological evidence at odds with the premise that EMF causes cancer".

NO. The most recent still persistently confirm this effect. The two largest (NCI and UKCCCR) both confirm this effect.

Tell me of any well-conducted studies which do not confirm this effect.
So you DO have a reading comprehension problem. I will repeat the salient parts of the post in the hopes that your reading comprehension improves.

The post was annotated (in bold italics) with the meaning derived from the previous text. The the more recent studies (i.e. those in which the esposure has been more accurately been established) show epidemiological evidence at odds with the premise that EMF causes cancer are their words, not mine (or rather their words rendered for the hard of understanding)

Note:
My not taking an opposing position to the results (or, given your seeming inability to understand the results of studies from the texts, your interpretation of the results) of the NCI and UKCCCR reports is not an indication that I believe that they confirm the effect, merely that I have not had a chance to read them yet
 
From the NCI website.....

http://cis.nci.nih.gov/fact/3_46.htm
There was no clear conclusion drawn from the data. Although the data showed some association between appliance use and leukemia, there was no consistent pattern of increasing risk with increasing exposures. The scientists speculate that the magnetic fields from electrical appliances are unlikely to increase the risk of childhood ALL
Let me paraphrase that for the hard of understanding.

- No clear conclusion
- Although there is some association between appliance use and leukemia it isn't causal
- Electic (I think I meant electric here, but of course they could be electic and electric) appliances don't cause ALL
 
From here http://www.feb.se/Bridlewood/UKSTUDY.HTM a summary of the results of the UKCCCR study. To cherry pick some quotes:

Results of the UK Childhood Cancer Study, published in The Lancet medical journal, showed no link between electromagnetic radiation and childhood cancer.

``This is a very powerful study for the levels found in the UK. No other study has investigated so many cases of cancer in children over such a long period of time,'' said Professor Nick Day of the University of Cambridge, who led the study.

``The magnetic component of EMFs has been under suspicion for some time, but this major study provides firm evidence that exposure to the levels of magnetic fields found in the UK does not augment risk for childhood disease,'' said Richard Doll, the chairman of the UK Childhood Cancer Study

Did I look at the wrong study ?
 
To Hans:

By one sample, do you mean from one isolated culture or from one exposure? In fact we counted (12 I think) hemacytometer squares and quantified these separately. Normally we use three runs and on different dates, but for the endogenous field studies the results were so surprising we did six separate cultures. Then we did six again. The effect was surprisingly robust.
 
The Don said:
From here http://www.feb.se/Bridlewood/UKSTUDY.HTM a summary of the results of the UKCCCR study. To cherry pick some quotes:





Did I look at the wrong study ?

The trouble is one which haunts those in this forum: you are not looking at the actual studies, you are looking at the websites reporting them. These often put spin on the results, and that is what is happening here, I suspect. Have you actually read the texts of these two studies? We have copies in our library.
 
Oh, and Dr Henshaw....

He has the theory that high voltage power lines cause illness.

His theory is that the mechanism by which they do this is that high voltage power lines produce "corona ions". These "corona ions" are highly charged particles of pollution. Because they are highly charged, they are more likely to be deposited and so more likely to have an effect.

The NRPB looked into this and concluded that the effect was small
http://www.nrpb.org/press/press_releases/2004/press_release_3_04.htm.

Dr Henshaw is madder than a wet hen about this http://www.electric-fields.bris.ac.uk/Response.htm and feels that there really is some effect here.

Some points of correction however:

- CogresLab said that Dr Henshaw's funding was withdrawn. he does however seem to be active (and listened to) in the field
- The NRPB has not squashed the results - it just disagreed with their importance


Of course how the talisman could protect from these corona ions is anyone's guess
 
said Richard Doll, the chairman of the UK Childhood Cancer Study
Wow! You know who this is? GOD! The man (or one of the group) who discovered the epidemiological evidence that smoking causes lung cancer. If he says something is or isn't so in epidemiological terms, you can rely on it.

The Blessed Virgin Mary.
 
cogreslab said:
Have you actually read the texts of these two studies?
Yes I have thank you. Given your problems with reading comprehension, understanding statistical analysis and your low baseline of scientific knowledge, have you understood them ?
 
Rolfe said:
Wow! You know who this is? GOD! The man (or one of the group) who discovered the epidemiological evidence that smoking causes lung cancer. If he says something is or isn't so in epidemiological terms, you can rely on it.

The Blessed Virgin Mary.

Hail, Mary, full of grace.

"But he thinks scientists these days are too eager to rush into print "In 1950 we didn't say: 'Look this is an association which looks fishy,' we said: 'The evidence is that smoking causes cancer of the lung' - and we were sure because we didn't publish until we'd checked the London results on patients in Leeds, Bristol, Cambridge, and Newcastle."

Thought that thought might have some bearing on these, lo!, many pages.
 
The haemocytometer, please could you say how it is laid out (eg neubauer, fuchs rosental), or if you prefer how many squares it had (size/volume as well).


Assuming there were not 10 squares, how did you pick the 10 squares to be included in the study.
 
Rolfe said:
Wow! You know who this is? GOD! ......

...... The Blessed Virgin Mary.
without being too culturally insenstitive isn't God the one who got you knocked-up. Perhaps Richard Doll could make good 2,004 years of child maintenance payments :)
 
Don: you can get to the actual text of the UKCCCR study via the Lancet link.

Take a look nat the figures, anmd you will see ORs of 1.4 to 1.68 when the magnetic field exceeds 0.4uT.

The conclusion that "EMF" epi literature offers substantial evidence is of course totally wrong since there has simply not been enough work on the electric component.
 
BillHoyt said:
"But he thinks scientists these days are too eager to rush into print "In 1950 we didn't say: 'Look this is an association which looks fishy,' we said: 'The evidence is that smoking causes cancer of the lung' - and we were sure because we didn't publish until we'd checked the London results on patients in Leeds, Bristol, Cambridge, and Newcastle."

Thought that thought might have some bearing on these, lo!, many pages.
Precisely. I've heard him (on the radio) being very critical of some present-day researchers, stuck in the "publish or be damned" culture. He had some very sensible things to say about dose-response relationships and the existence of a plausible mechanism of action being necessary before you could plausibly point the finger.

But remember his experience. Cigarettes cause lung cancer. No doubt a lot of the establishment weren't too happy about that discovery. But nobody could suppress it (though they argued against it, of course), and Professor Doll got a knighthood. That's what happens when you're right.

The Blessed Virgin Mary.
 
We use Sigma's Brightline hemacytometers. These are scored to give four very large squares in each corner and 16 smaller squares on each flank, then in the centre a smaller grid again. We use the four corner suares which have I think 16 squares within them, plus the middle one. There are two wells on the hemacytometer, so we count both. This gives us ten readings for each count, which are then summed and averaged. We take this aggregated average for statistical purposes. It's fairly common procedure (see Ali, a great little book on the Separatioin of Human blood Cells, page 34). Sigma used to describe the protocol in their catal;ogue, but not in recent years if I recall.

The scoring is similar to the "improved" Neubauer ruling.

(edited after I nipped into the lab to check the hemacytometers).
 

ISF - Join now!

Every member here is approved by hand. No bots, no spam, just people who care about evidence and honest debate.

Membership is free!

Create your free account

Back
Top Bottom