Newsletter no 72, January 2009

Contents

  1. First: the starting body weight of the volunteers was about 100 kg, so they all needed to lose about 30 kg. After a year those on placebo lost on average 6 kg, and orlistat had lost on average 10 kg, which is indeed 68% more. However, to put the same data in a different way, after a year those on placebo still had 24 kg to lose while those on orlistat had 20 kg to lose, (which is only 17% less) and both groups were, on average, still seriously overweight.
  2. Second, the range of weight change within each group was huge. Each group contained about 370 individuals. The standard deviation of weight loss in each group was about 10 kg. So, instead of saying, “the orlistat group lost 10 kg”, one could more accurately say, “The mean loss for the orlistat group was 10kg, plus or minus 10 kg.”
  3. Third, while on average the orlistat group lost 10 kg from the baseline weight, they actually lost 3.5 kg in the 4-week run in while taking placebo, so they lost (on average) only 6.5 kg on orlistat.

    There are many other examples of a blurred line between efficacy and profit. Most pharmacists will stock a range of cough medicines that purport to be most effective for tickly coughs, with others for dry ones or loose ones, but whose active ingredients are similar (and sometimes identical). How can we count on a profession which is essentially a sales machine?

    If pharmacists are going to be entrusted with a greater role in advising the public on healthcare matters then the RPSGB must be able to assure the public that all members understand the importance of scientific trials and are qualified to give advice about efficacy that is based on sound and unbiased evidence. Truth and integrity cannot be adulterated to suit the harsh realities of the high street. That is not to deny that retailing is a tough and highly competitive business, and without clever marketing a lot of pharmacies would go out of business. However pharmacists must come clean about the differences between goods they stock as retailers and real medical products which they can honestly endorse on the basis of sound research evidence. If a pharmacist is directly involved in selling untested products then he or she should be expected to say clearly that the product has not been tested to the standards normally required of pharmaceuticals.

    Pharmacists are highly trained and generally do a very good job, in fact health services would struggle desperately without them. But they cannot have their cake and eat it, or one day the NHS, the medical profession, and indeed the customers, will rumble them.

John Garrow
Emeritus Professor of Human Nutrition
University of London


Reference
1. Sjöström L, Rissanen A, Andersen T et al. Randomised placebo-controlled trial of orlistat for weight loss and prevention of weight regain in obese patients. Lancet 1998; 352: 167–173.

Top of page


TECHNOLOGY : THE ELECTRO-INTERSTITIAL SCANNER: CAN IT DO WHAT I THINK IT CLAIMS TO DO?

In the last issue of this newsletter (HealthWatch Newsletter issue 71, October 2008) James May told how he investigated an electro-interstitial scanner (EIS) after one of his patients came to him with a detailed report for which she had paid a pharmacist offering the service £60. On request, the manufacturers sent him an 18 page booklet about the scanner, complete with 70 references to the scientific literature. James passed me this with the question, “Can the EIS do what it seems to be claimed to do?”

So, what is it claimed to do? The manufacturer said in an email, “The EIS system has a number of applications. It can be used by practitioners to get a picture of the functioning of the organs of the body allowing them to direct further examination as necessary. I would stress that the system in itself is not a diagnostic. It is also useful to monitor the effectiveness of treatment and adverse side effects. It is also a very valuable tool to guide patients to make necessary lifestyle and nutritional changes and to monitor the effect of these changes.” [1]

In subsequent emails the manufacturers have again stated that the EIS is not licensed for diagnostic use, and no-one to whom they sell or lease the machine should use it for diagnosis. The information in the booklet states that the scanner has been used to monitor acid-base balance, tissue oxygen, and the effects of chemotherapy; as a marker for unipolar depression and underactive thyroid; screening and follow-up and monitoring of various pathologies including diabetes, hepatitis, circulatory problems, duodenal ulcer, chronic bronchitis and asthma, and cancer. So, while it is clearly not intended to be marketed as a diagnostic tool, I could be forgiven for misinterpreting the information—if it can be used to monitor these conditions, might it not also be useful to diagnose them?

There is an impressive list of clinical trials of the EI scanner for monitoring and follow-up of various conditions, but we are not given any results or references to these trials. The manufacturers’ booklet contains some good sound (text-book) physiology on interstitial fluid volume, but most of this is not relevant to the scanner. Likewise, on following up the references cited in the booklet, while most were to peer-reviewed medical and scientific journals, few were actually relevant to the scanner.

However, let’s come back to the original question: whether it can do what it claims to do, not what I think it might claim to do.

The underlying principle of the scanner is that it measures the electrical conductivity of the body. This is well-established as a means of measuring body fat. Electrolytes (sodium, potassium, chloride, bicarbonate and phosphate, etc) in lean tissues conduct an electric current, fat does not. For body fat determination a low voltage high frequency (typically 50 kHz) alternating current is used, and the impedance (equivalent to resistance) between the hands and feet is measured. The greater the impedance, the more fat is present. Such devices are widely used in nutrition research, and have begun to appear in gyms for people to use themselves.

A high frequency alternating current penetrates all body water compartments: the bloodstream, intracellular water and water between cells (the interstitial fluid). The EIS, by contrast, uses a low voltage direct current, which the manufacturers state only measures interstitial fluid electrolytes and the volume of interstitial fluid. Unfortunately they do not cite a reference for this so it is not clear whether research supports this use of direct rather than alternating current in this application.

The EI scanner uses six electrodes, attached to the left and right forehead, and both hands and feet. Readings are taken using 21 different pairings of the electrodes as a way of modelling the body. The manufacturer’s booklet and website [2] suggest that sophisticated mathematical analysis of the results permits estimation of the volume and contents of the interstitial fluid in discrete organs and tissues. I cannot follow the mathematical arguments in the paper cited [3] but I doubt that you could differentiate between say the liver, pancreas and gall bladder, or indeed between the large and small intestines, with only six electrodes—even in 21 different combinations.

We are given a series of references to papers showing that, with suitably placed electrodes, conventional bio-electric impedance can be used for estimating cardiac output, confirming the results of mammography (because of the very different composition of normal breast and tumour tissue) and monitoring hydration during and after surgery, etc. All very sound, but not actually relevant to the EI scanner.

We are then referred to papers on electrochemical measurement of various compounds in interstitial fluids. This is all good sound chemistry, and with suitable equipment, and techniques including spectrophotometry and electrochemical detection, it is indeed possible to measure different compounds, both electrolytes and non-electrolytes such as glucose, and individual small hormones (e.g. thyroid hormone) and neurotransmitters. But not, however, simply by using electrodes placed on the skin.

There is some discussion of the pH of tissues, and especially of cancerous tissue. Again, pH can be measured using electrodes placed in tissues or the blood vessels supplying and draining them, but not by using electrodes on the skin. There are references to two papers on tumour pH. One is a fluorescence microscopy study in rabbit ears with implanted tumour cells [4], and the other [5] involves NMR spectroscopy and magnetic resonance imaging of human tumour cells grown in mice. I presume these are supposed to provide a theoretical basis for how changes in pH claimed to be found on EIS are related to the presence or absence of cancers.

In conclusion, it is difficult to decide quite what the EI scanner is supposed to do.

The manufacturers are clear that it is not a diagnostic tool, yet all the information in their booklet suggests that it can be diagnostically useful. I am not clear from the booklet whether it purports to be able to measure glucose, cholesterol, proteins, individual ions, oxygen and carbon dioxide or not. There is a table showing values for these in different body fluids, but there is no claim that they are measured. However, James May’s patient came to see him with an eight-page report with what appeared to be a full biochemical breakdown, advice about the function of all organs, and advice about recommended foods. The manufacturers say that this report should not have been produced. I agree with them. They do not claim that these measurements are made, and I very much doubt that the EI scanner could make them.

David A Bender
Senior Lecturer in Biochemistry, University College London

References

1. E-mail from Roddy Macdonald of Biomedical Technology Ltd to James May, 29th August 2008.
2. Manufacturer’s main company websitehttp://www.ldteck.com (accessed 16/11/08).
3. Bryan K, Numerical recovery of certain discontinuous electrical conductivities. Inverse Problems (Institute of Physics)1991 7: 827–40.
4. Martin GR & Jain RK. Non-invasive measure,ment of interstitial pH profiles in normal and neoplastic tissue using fluorescence ratio imaging microscopy. Cancer Research 1994; 54: 5670–4.
5. Reference to a 1979 PhD thesis by Gillies RJ, linked to Prof Gillies’ home page http://bmcb.biology.arizona.edu/gillies.html (accessed 16/11/08).

 

Top of page


BOOK REVIEW : The Tiger That Isn’t:
SEEING THROUGH A WORLD OF NUMBERS

by Michael Blastland and Andrew Dilnot
Published by Profile, 10 July 08. £7.99 Paperback. ISBN 978 1846681110

When I was a child, my bedroom had a swirly beige-and-white wallpaper. I often thought I saw faces and even tigers (how did the authors know?) in the random swirls; nowadays, fundamentalist Christian parents in the USA play their teenagers’ pop records backwards and fancy they can hear the antichrist.

Chaque un à son gout. Ignorance and a keen imagination can take a person along some strange paths. Blastland and Dillnot are, respectively, creator and presenter of Radio 4’s statistics series More or Less. And their book comes with endorsements: Rory Bremner says it made statistics far, far too interesting, and David Dimbleby says it explains to lay people how to make sense of numbers and how we can avoid having the wool pulled over our eyes. New Scientist considers that every journalist should get paid leave to read and re-read this book until they’ve understood how they’ve been spun.

I agree. This book is written with wit and pizzazz. There are chapters on the pitfalls of the distortions of spin-doctors. Mrs Thatcher’s governments changed the definition of ‘unemployed’ 23 times (or was it 26? accounts differ). The Daily Telegraph screamed that one boy in four is a yob when boys were asked if they had injured anyone, however slightly, by hitting or jostling, including their family and schoolfriends. Labour apportioned £300m for childcare over five years. Is that a bonanza? Well, divide by a million—the number of pre-school children—and you get £300 a child. Over five years, that’s £1.26 per child per week. How much childcare can you buy for that?

Around midnight on bonfire night 2003 in Wishaw, West Midlands, a mobile phone mast came down. It had been carefully unbolted first, and next day villagers prevented T-mobile from re-erecting it. In the twenty houses within 500 metres, there had been nine cancer cases, and to the villagers the cause was obvious. And that is one reason why it’s important to understand clustering, a natural phenomenon.

Then there’s regression towards the mean, which means that extremes (of temperature, rainfall, behaviour, or many other things) are likely to be followed by something more average. So, is it better to reward good performance or to criticise or punish bad performance? It is easy to conclude that punishment works better, as it is more likely to be followed by improvement.

This book is entertainingly written and a pleasure to read. It takes many examples from healthcare spinning, including manipulation of NHS waiting lists; and from crime statistics—the police reported a drop in violent crime but hospital A&E records showed that this was an artefact. The book will stay on my shelf alongside the classic Innumeracy by John Allen Paulos (he uses his middle name to avoid being taken for the late Pope), Reckoning with risk by Gerd Gigerenzer (Penguin 2002), and Irrationality by the late Stuart Sutherland, republished in 2007.

Caroline Richmond
Medical journalist and author, London


Top of page


 

Opinions expressed in letters and articles published in the HealthWatch Newsletter belong to the authors and do not necessarily reflect the views of HealthWatch. The editor reserves the right to amend text if necessary but will, where possible, consult the author to ensure accuracy is maintained. Letters and articles for publication are welcomed and should be sent to the Editor by e-mail to: newsletter@healthwatch-uk.org

Copyright © 2009 HealthWatch.