An independent charity for science and integrity in healthcare

Read the latest HealthWatch newsletter:  Issue 106, Autumn 2017

This January, as is usual after the Christmas celebrations, the paperback best seller list was packed with books on dieting, about which there is little new to be said. Most of these are about how to lose the weight gained in the previous month, and claim to reveal “secrets” hitherto unknown about diets that will make you thinner and/or healthier. In truth there are very few secrets still to be revealed.

Since World War II we have learned a great deal about human nutritional requirements. There was an urgent need to know how to design rations for the civilian population, and to rehabilitate survivors from starvation in prison camps. Intravenous feeding saved the lives of severely wounded people who could not have been fed by mouth. A by-product of this was that we learnt far more precise information about the daily requirements of micronutrients than could be obtained by analysis of oral diets. In 1969 the Department of Health reported on daily requirements of 10 nutrients, by the time they reported in 1991 this had increased to 40 nutrients, but there has been little change since then.

Lethal diets

If addition of a poison is excluded, the only way in which the diet of a normal adult could be altered so as to cause death in less than a week is to exclude all water. Someone deprived of any non-salty water because they are trapped under debris from an earthquake, or adrift in a lifeboat after a shipwreck, will soon die. How soon depends largely on the temperature and humidity, and hence the rate of evaporative loss of water.

Even if enough water is available total starvation will also inevitably cause death, but after a much longer period. The survival of victims of famine due to drought or warfare is usually threatened by disease as well as hunger, and they may be having small amounts of edible vegetation, so they do not provide reliable information about the lethal effects of starvation alone.

A healthy adult of normal build who takes no nutrients apart from water—such as a hunger striker—will probably die in about 10 weeks. Between 1964 and 1970 there were several reports of severely obese patients who were treated by total starvation for long periods. The longest period recorded is 249 days starvation, during which she lost 64.9 kg. However some people on starvation diets unexpectedly died and at autopsy were found to have severe damage to their heart muscle, so this treatment was abandoned.

The largest reported weight loss (227 kg) was on a low calorie diet (800 kcal/day) and took 2 years as a hospital inpatient [1]. The patient was admitted weighing 310 kg, but an unknown proportion of that was water, since he was initially massively oedematous. Since 1970 it has been accepted that prolonged total starvation is too dangerous, but 800 kcal/day will keep the patient alive and losing weight quite quickly. There is still controversy about how much lower than 800 kcal it is safe to go, since a few patients have died while using very low calorie diets, although there has been insufficient evidence to implicate the diet itself as a cause in these cases. Our research indicates that it is not the prescribed energy intake, but the actual rate of weight loss, that determined the danger. A desirable rate of weight loss in obese patients is 0.5 to 1.0 kg per week.2

Macronutrient imbalance

The dangers of too great a reduction in total energy intake have been considered above, but many commercial diets emphasise alterations in the balance of protein, carbohydrate and fat, from which dietary energy comes. (Alcohol is also a source of dietary energy, but it will not be considered here since the dangers of a high intake of alcohol are well known).

Human beings tolerate very large differences in macronutrient balance. For example vegetarians have a much lower intake of protein and fat than omnivores, but the amount of protein in vegetables is enough to maintain health in adults if total energy intake is adequate. However small children have a higher requirement of protein in relation to body weight because they need extra protein to support growth, and they cannot eat such large quantities of the bulky carbohydrate diet as adults. Hence in chronically undernourished populations the deficiencies show up first in stunted growth of children. Protein supplements are not very effective in improving the health of children who also have an inadequate energy intake: both deficiencies need to be remedied.

The other two macronutrients, carbohydrate and fat, are the main sources of energy. High fat diets (of which the popular “Atkins” plan is an example) have been advocated for weight loss, because if carbohydrate is restricted total energy is almost bound to be restricted also. Very severe reduction of carbohydrate causes ketosis (signalled by an unpleasant smell on the breath), since some carbohydrate is needed for the normal metabolism of fat. There is good evidence that a high fat diet causes a high concentration of lipids in the blood, so if a high-fat diet is used repeatedly or adopted as a long term strategy there is an increased danger of cardiovascular disease and coronary thrombosis.

On the other hand, extreme restriction of fat such as in the most stringent versions of the low-fat diets popular in the 1990’s causes reduction in the intake of fat-soluble vitamins A and D and possible deficiencies in these nutrients.

Micronutrient deficiency or excess

In affluent countries deficiency of vitamins or minerals is very rarely found among people who are having an adequate energy intake from a variety of foods. I have worked in some countries where dietary deficiency of vitamin A, iron, or iodine cause serious illnesses, but I have never seen such cases in the UK. The problems usually arise in adults who are deliberately restricting energy intake in order to lose weight, and are taking inappropriate supplements in order to correct the resulting deficiency of micronutrients. The exceptions to this general statement are old people who are not exposed to sunshine; pregnant women; and adults or children who have metabolic diseases or dietary intolerances.

There is a huge industry that promotes micronutrient supplements on the false premise that if a deficiency of Vitamin X causes ill health then a massive intake must bring extra good health. The reverse is nearer the truth. Especially when the supplement contains a dose of a single micronutrient that could never be encountered in a diet of normal food. For example health food shops offer capsules containing single amino-acids which can never be helpful except in rare metabolic diseases. Aminoacids are the building blocks from which protein is synthesised. A single amino-acid such as leucine or lysine are “essential” amino-acids, so without them protein cannot be synthesised. But a supplement of one of these does more harm than good, because an excess of one amino-acid cannot make protein unless the others are present in the appropriate proportions. The extra aminoacid therefore has to be used to make urea and excreted, resulting in a net loss of protein to the body.

Trace elements such as zinc and copper are required in very small quantities. Transport mechanisms in the gut wall normally absorb the right quantities, but a large supplement of one (for example zinc) may overload the transport system and block the absorption of copper. This can create clinical problems that are very difficult to diagnose. If you do not know about the zinc supplements you have to be quite astute to recognise copper deficiency in a patient who has a normal amount of copper in his diet, but an inadequate absorption of copper caused by an excess intake of zinc.

Importance of kidney function

This brief review has implied that so long as you eat a reasonable amount of ordinary food you will avoid major problems. However this depends on having kidneys working well so that if you take an excessive amount of fluid, or water-soluble vitamins, the kidneys will excrete the excess in urine. The situation is very different in people with impaired kidney function.

The commonest supplement-related life-threatening situation is when someone with damaged kidneys and legs swollen with fluid is advised to take supplements containing potassium. The concentration of potassium in the blood may then rise to a level at which regulation of the heartbeat is disturbed, and this may cause death.

Attempts to achieve rapid and marked weight loss, or consume inappropriate amounts of dietary supplements, should be discouraged. A safer and more useful approach would be to aim for a smaller but sustained rate of weight loss, and to modify eating habits so as to maintain the desired level of weight achieved.

Position paper prepared by John Garrow and approved by the HealthWatch committee January 2009


1) Bortz WM. Am J Med 1969; 47: 325–331.

2) Garrow JS. Obesity and related diseases. Churchill Livingstone, Edinburgh, 1988.

Incidence and Death rates from Cancer

There were 2,845,560 cases of cancer diagnosed in the UK in 2004 and four sites: breast (18%), lung (13%), bowel or colorectal (13%) and prostate (12%) account for over half of all new cases. The number of deaths from cancer, 153,497 in the UK in 2005, was surpassed in frequency only by deaths from heart and blood vessel disease. Cancer is more common now than it used to be because people live longer and cancer becomes more frequent with increasing age. However, the standardised mortality rates (SMR, expressed as cases per 100,000 population) for all cancers fell by over 17% over the 30 years from 1975 to 2005 from 218 to 180 per 100,000 population despite an increase in incidence from just under 300 to over 350 per 100,000.

The commonest cancer in men is prostate (24%), followed by lung (16%), and colorectum (14%), out of a total of 143,126 male cancers in the UK in 2004. The outlook is however very variable between cancer types. Whereas the outlook for lung cancer is generally poor many older men may harbour prostatic cancer which never causes any problems. In women the commonest cancers are those of the breast (31%) and lung (11%) and then , as in men, the large intestine and rectum (12%) of a total of 141,434 cases.

The frequency of different cancers and the SMR has changed over the years in different ways for different tumours. The incidence of lung cancer initially increased rapidly over the last fifty years in both sexes. This was due to the increase in cigarette smoking but as this has declined in men so has the number of lung cancers reflecting this. The SMR has halved over 30 years (107 per 100,000 deaths in 1971 and 53 in 2005). In women there has been no such fall in smoking habits and the incidence of lung cancer has continued to rise until very recently. The SMR increased from18 to 30 (a 66% increase) in the late 1980s and has not changed since then reflecting the increase in smoking amongst younger women which has not been offset by a decrease in mortality in those over 60.The SMR for female breast cancer fell by 18% between 1996 and 2005 reflecting both earlier diagnosis and more effective treatment. The cause of breast cancer is not known although various risk factors lead to a greater susceptibility such as increasing age, early onset of menstruation, obesity, a first degree relative with the disease and possibly abnormal exposure to female hormones. Oral contraceptives are not associated with an increase and there is no widely accepted evidence that HRT is either. Hormones must however have a major role as removal of the ovaries reduces the risk significantly although the precise way in which this works is not known.

In men the age standardised mortality rate for prostate cancer has fallen from 30 in the early 1990s to 25 (16%) and the majority of deaths are in those over 70 years of age. Mortality rates for cancers of the colorectum have fallen in both men and women over the last ten years but are substantially higher in women. Cancer of the stomach has notably declined but at the same time cancer of the gullet (oesophagus) appears to have increased in frequency. A reduction in stomach cancer is thought to be associated with a reduction in exposure to dietary cancer inducing agents (carcinogens) and better food preservation and dietary habits. The rise in oesophageal cancer has, for some reason, run in parallel with a reduction in gastric ulcers following the recognition of the importance of drugs to reduce excessive acid secretion in the stomach and the eradication of helicobacter pylori, a bacterium which plays an important role in the development of ulcers. Again, the precise connection between these events is obscure. The age standardised incidence for melanoma of the skin has risen rapidly since the 1980s from around 5 to nearly 14 in women and from just over 2 to nearly 12 in men per 100,000 population attributable to excessive sun exposure and the use of sun beds.

The mechanism of development of cancer

The fundamental underlying abnormality in all cancers is damage to the genetic material of the cell, the DNA which codes for the large number of molecules which cells produce. These include a wide variety of regulatory molecules which control cell division and growth in the normal cell by complex inter-acting sequences of chemical reactions so that any normal cell loss e.g. shedding of skin or intestinal epithelial cells or natural cell death is accurately compensated for by cell multiplication. In cancer, this delicate balance between factors which tend to promote cell growth and those that inhibit it, is lost and cell division runs amuck. The genes which are important in controlling normal cell growth become altered by a variety of factors and are transformed into oncogenes. These cause the production of oncoproteins, which are associated with a lack of response to normal growth inhibitory signals, abnormal growth factors and abnormal growth factor receptors on the cell surface. This perpetuates the production of successive generations of cancer cells. The properties of the background nonmalignant tissue in which the tumour is growing are also important and the body’s natural immune system plays a role weeding out abnormal cells.

The daughter cells also have abnormal properties such as being able to invade tissues around the organ (local spread) and to penetrate blood and lymphatic vessels (enabling distant spread). These cells are also able to stimulate growth of their own blood supply and often seem to be able to escape from the body’s immune surveillance system and to resist the normal process of programmed cell death (apoptosis) which is a normal physiological mechanism for keeping cell populations in balance. Finally, it is well recognised that general emotional well being may have a profound influence on the behaviour of cancers in some individuals.

Cancer most commonly arises in epithelium where cell turnover is high – where cells line the surfaces of body tubes and cavities. In general, the more abnormalities in the cell DNA, the more aggressive the tumour is and the appearances of the tumour cells under the microscope resemble their normal counterparts less and less. These tumours are called poorly differentiated as opposed to well differentiated tumours with a better prospect of cure and a longer natural history.

The causes of DNA damage


The factors causing DNA damage are known in some cases. Environmental factors such as smoking are extremely important, not only as a major cause of lung cancer, but also as a very important factor in numerous other cancers such as those of the bladder and larynx in these instances probably acting in concert with other factors which are as yet undefined. If nobody smoked, there would be a huge reduction in cancers of very many types. It is well known that radiation causes cancer as seen in the increase of leukaemias and other cancers amongst survivors of the atomic bombs and following radiation leaks. Following the Chernobyl disaster there has been a huge increase in thyroid cancer which previously was quite uncommon. Mesothelioma which is a cancer arising in the membrane lining the chest and abdominal cavity as well as covering the surface of the lung and intestine is directly caused by exposure to certain types of asbestos. These tumours develop many years after exposure, sometimes only to small amounts of the substance and currently in the UK there is an epidemic of this type of tumour following exposure in the 1950s and 60s.


Viruses may integrate their genetic material into cells and are a common cause of cancer in animals but their role in man is only well established in a few cases such as cancer of the neck of the womb (cervix), the culprit being some forms of wart or human “papilloma” virus. Vaccines have been developed against these strains and should protect against the disease. There is evidence that bacteria may have an important role in the development of cancers of the lymphoid cells of the immune system (lymphoma) in both the stomach and the skin. Viruses causing inflammation of the liver (hepatitis) may eventually lead to liver cancer.


The underlying susceptibility of different people to the development of cancers of different types also varies in exactly the same way as the tendency to develop chronic diseases such as diabetes or heart disease. This is directly related to the genes that they carry. Gradually the genetic profiles making individuals susceptible are being worked out but it is rare for single individual genes alone to be responsible. Some cancers appear to run in families. Carriage of particular genes (BRAC 1 and 2) accounts for susceptibility to the development of breast cancer in many members of some families but this is rare compared with the overall frequency of breast cancer. The vast majority of breast cancers do not arise on this basis although these genes may still play a role together with many others in non-familial cases. Another example is carriage of the dominant gene responsible for the development of large numbers of polyps in the large intestine many of which will go on to cancer but this is again very rare compared with the overall frequency of colonic cancer. In some people there is an absence of the ability to repair DNA errors which inevitably occur occasionally when normal cells divide. These errors are eliminated consistently by most people. In some families the failure of this mechanism is associated with the development of many different types of cancer such as cancer of the lining of the body of the uterus (endometrium) and cancers of the large intestine. Having said this, the precise genetic mutations responsible for most familial clusters of cancer are likely to be multiple and complex.

Behaviour of cancer

Different types of cancer behave in different ways. Some are dominated by local spread for instance, a very common skin cancer called basal cell carcinoma or rodent ulcer. Others may stay confined to a small micro-anatomical area for a long time before invading, for instance in the epithelium in which they originate, before invading into the surrounding supporting structures. An example of this is carcinoma in situ of the neck of the womb revealed by a cervical smear test or carcinoma in situ arising in breast ducts identified on a screening mammogram. Yet other cancers may be locally invasive from the start such as well differentiated breast or prostatic cancer but not spread (metastasise) for a very long time. Even if spread has occurred through the lymphatic vessels with tumour cells lodged in local lymph glands they may never spread beyond this stage, a good example being one of the forms of thyroid cancer. Chronic lymphatic leukaemia may be symptomless for many years. In these cases local treatment or even none at all in the case of chronic lymphatic leukaemia or some forms of prostate cancer is all that is needed. These patients are regularly monitored so that more intense treatment can be instituted if things change. In this situation the only major problem to the sufferer may be the thought of having the disease and fear because of the uncertainty about the future. There will inevitably be some disappointments even within the usually well behaved groups of cancers in the same way that amongst some very aggressive cancers there are occasional unexpected complete cures. This is why the statistics for cancer are usually recorded as a percentage survival over five or ten years or longer. This gives an average statistical estimate of outlook which of course does not necessarily apply to a particular individual. Even amongst lung cancer where say, 93% of people die within 5 years, amongst 5000 cases 350 people will survive for this length of time.

Whereas the survival rates for breast cancer have shown much improvement over the last 30 years (increasing from 52% to 81% at five years and from 41% to 73% at ten years) those for lung have remained constant at 4 to 5% at ten years although there has been some improvement at the 5 year level. Survival rates from colorectal cancers have almost doubled at both 5 and ten years from around 25% to around 50%.


Over recent years there has been a marked improvement in disease-free intervals in terms of years from diagnosis for very many cancers as well as in overall survival and cure rates. One of the changes contributing to this is that the data on the precise site and extent of the cancer in the organ where it has arisen, its degree of resemblance or otherwise to normal cells (differentiation) and whether or not it has spread locally (staging) is now recorded in a standard way so that treatment can be tailored more appropriately. This is possible because of the changes in techniques in diagnostic radiology (MRI and CT scanning) and improvements in the way cancers are analysed and classified under the microscope. A further innovation has been the introduction of the Multidisciplinary Team approach throughout the NHS in which all cases of cancer are discussed on a regular basis at a forum comprising surgeons, oncologists, palliative care physicians, radiologists and histopathologists together with senior specialist nurses in cancer care so that optimum treatment is tailored specifically for each patient following peer review of the diagnosis and discussion of treatment options. Increasingly patients are directly and actively involved in these discussions.


The other major factor associated with improved survival rates is improvement in treatment. Many more cancers can now be significantly modified by modern treatments often given in combination – surgery, radiotherapy, and chemotherapy. This means that whilst not necessarily cured in the long term, patients may have long periods of disease-free health before the tumour relapses.

With regard to treatment in the earlier stages of the disease, huge progress is being made. Firstly, it has been realised that much less radical surgery is just as effective as more extensive removal of normal tissue around the tumour. For instance, removal of the complete breast (mastectomy) is now a rare event unless the tumour is very big or the case is one of recurrence. The same applies to removal of melanomas of the skin (cancer of the pigment producing cells) so that disfiguring scars are not produced. Secondly, surgical techniques are more sophisticated; even internal cancers can be removed successfully using “key-hole” surgery and these techniques are being widely introduced for bowel, prostate and kidney cancer amongst others. This means that there is far less post-operative pain, and patients are mobile by the next or subsequent day following the operation. They are discharged home earlier and suffer fewer complications. It has also been realised that treatment of tumours with drugs or radiotherapy before surgery to kill a lot of the cancer cells improves success rates. In some uncommon tumours of children (soft tissue tumours – sarcomas and kidney tumours) for example, cancers that were previously usually fatal are now frequently cured. Similar significant improvements are also being seen in common adult tumours such as colon and bladder cancer. There have been major developments in the way that radiotherapy is delivered with more sophisticated machines (linear accelerators). This leads to increasing cure rates whilst side effects are reduced. New techniques such as conformal therapy mean that tumours can be targeted accurately and normal tissues spared from radiation.

Many new drugs which are very effective are now routinely used. Some of these are similar to the well established chemotherapy drugs which interfere with cancer cell growth through their effect on DNA. But others work by affecting the hormonal environment making it more difficult for cancer cells to grow. This approach is very effective in breast cancer with the use of drugs such as Tamoxifen and Letrosole Anastazole or Exemestane. In prostate cancer, agents interfeing with the effect of male hormones, hold the tumour in check. Newer drugs are tailor made molecules which block receptors on cells which are important in the complex chain of reactions leading to cell growth. An example is Herceptin which blocks a growth factor receptor on breast cancer cells. Yet other drugs are like antibodies which home in on antigens on the cell surfaces again interfering with cell growth such as Rituximab a drug used in lymphoma treatment.

In contrast to the success of these treatments, no properly-conducted trials have shown 'alternative' or 'complementary' therapies to have any effect on tumour size or progression. Cancers can sometimes shrink without medical intervention, and oncology departments have reported cases where proven cancers have disappeared without treatment, but such spontaneous remission is extremely rare."
Given the failure of alternative or complementary interventions to demonstrate improvements in mortality, and the serious risks to patients who do not seek proper medical treatment, no therapy should be offered by non medically-qualified personnel which claims or implies that it can modify the disease.

Some cancer patients report unsatisfactory contacts with health professionals, with insufficient time spent on explaining issues and talking through strategies and prognosis. This is perhaps unsurprising given the pressure on oncology departments, but is one area where properly qualified clinicians might learn from private 'alternative' practitioners.

The fear of cancer seems to be much greater than the fear of other chronic diseases but most cancers behave in a very similar way. A person may suffer a heart attack, be well for many years and then have another, go into heart failure or suffer a stroke – all these events being due to the same underlying disease process – a narrowing of the arteries supplying blood to different organs. A person with rheumatoid arthritis may suffer acute episodes of pain and immobility and then improve only to relapse later. Gradually the affected joints become more and more damaged. In the same way many cancers are treated successfully initially but may return at the same site years later. Even then, further treatment is often successful. The disease may spread to local lymph glands and still be cured. When it spreads widely, chemotherapy is often helpful in prolonging life. Another reason for the dread that cancer provokes is because people associate it with severe pain and a very unpleasant way of dying. There have been huge strides made in the management of severe pain and with radiotherapy and/or modern drugs often used in combination and carefully related to the patient’s needs, and nowadays there is no reason why anyone should suffer unnecessarily. This treatment is often carried out in specialised centres such as hospices where the doctors, nurses and all the other health workers are especially trained in palliative care. These professionals are expert in dealing with the other symptoms of weakness, poor appetite, nausea or just generally feeling unwell. They are trained to provide psychological, physical and practical support. They behave naturally, are cheerful and friendly and it is often surprising how helpful this is even if the patient is terminally ill.

There are also numerous charitable organisations other than those which are hospital and general practice based providing a wide variety of services, information and advice for cancer patients, relatives and carers. The foremost cancer charity in the UK, Cancer Research provides a wide range of awareness leaflets, a clinical trials database for the public, and comprehensive statistics as well as funding research into all aspects of the disease including educational and psychological support. Cancerbackup provides information, advice, leaflets on treatment and living with cancer as well as 8 walk-in centres based in hospitals in the UK. Macmillan Cancer Support provides practical, medical, nursing and emotional support. Many oncology departments and surgical cancer services have direct links with Macmillan nurses who can provide important continuity of care between the specialised hospital departments and the community. Marie Curie Cancer Care also provides free nursing care to patients in their own homes and funds research programmes. There are also some organisations dedicated to cancers of specific sites such as Breast Cancer Care, and Breakthrough Breast Cancer and details of those dealing with other sites are easily accessed through the organisations listed above.

Assessment of new treatments

It is now widely accepted that new treatments must be properly compared with each other and with established treatments before they can be put into general use. This can only be done properly and without bias in controlled trials where different treatments are directly compared in two groups of identical patients. Health Watch strongly supports this approach for both orthodox and unorthodox treatment and helps to monitor situations where inappropriate claims are made for treatments which have not been subjected to this vigorous process. It is also important to identify trials in which claims about the efficacy of a drug have been made on an incorrect basis either because the data have been misinterpreted, there is a bias present, or sometimes, and most unfortunately, because the results have been falsified. The National Clinical Research Network in the UK has increased the number of patients able to enter clinical trials of new treatments to a record high, with more patients than in the U.S. or any other European country. This will lead to further improvements in treatment in the future.

Screening and prevention

In cancer, as with all diseases, prevention is better than cure. Screening programmes to detect early cancer are well established for cervical and breast cancers in the UK and many authorities agree that they have made a significant impact on survival. Although there is no universal agreement and the evidence is not as clear cut as some would have it, these screening programmes are set to continue. The evidence for the effectiveness of screening for bowel cancer is good and the programme is due to be implemented in the UK. Currently there is insufficient evidence to merit the introduction of a programme for screening for prostatic cancer. Sadly there is widespread agreement that there is no effective method of screening for lung cancer.

So what else can be done to prevent cancer developing? Apart from giving up smoking which would lead to a huge reduction in the numbers of many types of cancer, a sensible attitude to sun exposure would curb the rise in skin cancers, particularly melanoma which has increased recently. We depend on careful research and observation to identify links between cancers and external agents as in the example of asbestos and mesothelioma quoted above. Otherwise a generally healthy life-style with plenty of exercise, a good balanced diet including vegetables and fruit which contain anti-oxidants, (substances which are important in reducing damage to cells) and avoidance of obesity are very important factors.


The causes of many cancers are still unknown and may well prove to be multifactorial. It is also unlikely that there will be a single cure. In future, it will be possible to further subdivide cancers of a particular organ or site previously regarded as a homogenous group. Using a combination of techniques it should become feasible to accurately characterise the cancer cells forming a given tumour. This might include their appearances under the microscope (morphology), the abnormal genes, the different growth factor receptors on the cancer cells, the biologically active compounds produced by the malignant cells, and the local tissue environment of the tumour cells as well as the tumour extent. Hopefully this will lead to a much more specific, effective and individually tailored treatment regime for individual cases.

Prepared by Dr Leo Horton and approved by the HealthWatch Executive committee 9/1/2008.

Useful Contact Details

Cancer Research UK: PO Box 123,Lincoln’s Inn Fields, London WC2A 3PX Tel: 020 7240 2000,

Cancerbackup: 3, Bath Place, Rivington Street, London EC2A 3JR Tel:0808 800 1234

Breast Cancer Care: 5-13, Great Suffolk Street, London SE1 0NS Tel: 0845 0920 800

Breakthrough Breast Cancer: 3rd Floor, Weston House, High Holborn, London WC1V 7EX Tel:0800 100 200,

Marie Curie Cancer Care: 89, Albert Embankment London, SE1 7TP Tel: 020 7599 7777 29, Albany Street, Edinburgh, EH1 3QN Tel: 01314 563 700 Block C Mamhilad House, Mamhilad ParkEstate, Pontypool, Torfaen NP4 0HZ Tel: 01495 740 827 60 Knock Rd., Belfast,BT5 6LQ Tel:0289 088

Macmillan Cancer Support: 89,Albert Embankment London SE1 7UQ Tel: 020 7840 7840

One of the principal objectives of HealthWatch is to promote the testing of treatments and the conduct of clinical trials. There are three false arguments used to claim trials are unnecessary for certain types of treatment.

  1. If patients feel better after the treatment, nothing else matters.
  2. Trials of truly holistic treatment are impossible: there can never be a control patient with whom the result can be compared because each patient is different and the treatment is tailored to the individual.
  3. Clinical trials cost huge sums of money that small independent practitioners cannot afford.

We believe that due to a lack of proper testing, patients are offered treatments that are less effective, less safe and more expensive than they need to be, both in conventional and alternative medicine.

If a patient feels better after a treatment, it is very satisfactory but does not mean the treatment is effective. The patient might have improved without any treatment, or there might have been even more improvement with a simpler, cheaper or safer treatment. The best way to find out, almost always, is to do a suitable clinical trial.

Every good practitioner, whether conventional or alternative, should practise holistic medicine; treatments should always be chosen to suit the whole patient in their particular circumstances and environment, and not simply the pain or lump that they may have. Similar pains or lumps may be treated differently in different patients. Despite this, it is still possible to do valid clinical trials using the guidelines set out below.

The onus of providing evidence of efficacy is on those who promote a treatment and not, as is sometimes stated, on the scientific community in general. It is true that large scale multi-centre drug trials can be very costly, but most of the expense is in administration and in the biochemical tests that are not an essential part of the comparison, but are to check for any unexpected harmful effects of the treatment.

Some minimum conditions for a proper trial to compare two treatments:

Protection of research subjects

It is illegal to conduct any research involving human subjects without the prior approval of a Research Ethics Committee. It is the duty of these committees to check that the research is properly designed, and that previous research on the topic to be studied has been adequately considered. Any conflict of interest (e.g., the research is sponsored by an organisation with a commercial interest in treatment being tested or the researchers have a financial interest such as company shares) must be declared. Any possible harm to trial participants must be as little as possible, and volunteers must be given a clear description of what the trial involves and what other treatments might be, so they are able to give fully informed consent that they are willing to participate. They need to be told that they are free to opt out of the trial at any time, without giving any reason and without suffering any detriment.

Choice of treatments

If there is no treatment that is known to be safe and effective for the condition under study, then a new treatment may be compared with a placebo treatment. If, on the basis of existing evidence, two (or more) treatments are equally appropriate and it is not known which is better, they should be compared, using a group of patients for each treatment. If the new treatment is to be compared with one of several already established treatments, the best should be selected as the comparator treatment. Each treatment should be clearly described, including its formulation, dose, route and frequency of administration, full product characterisation and details of manufacturer.

Aims and objectives of the trial

These must be adequately described, e.g., to compare the effectiveness of two treatments for a specific medical condition, over a given period of time, in a specific group of patients. The measurements of the desired treatment effects (outcome measures) must be appropriate, precisely defined and, where possible, objective and reproducible.

Patient allocation

To be as sure as possible that the groups are otherwise similar, and to avoid the researchers choosing which patient is given which treatment (avoidance of selection bias), allocation of patients to treatment groups must be random. Assigning patients by alternation or by dates of birth is not acceptable. Ideally, true randomisation is best achieved with the aid of a third person not directly involved in the trial (often contacted by phone) who, after having checked that the patient meets the entry criteria, allocates the randomly selected treatment group for that patient.

Registration of Clinical Trials

The trial should be on a register. This ensures other researchers working in the same field are aware of its existence and means the results will be available for review even if the outcome is unfavourable to the procedure or product. Registration reduces the trend for positive trials to be published and hence available to reviewers, while negative trials are not, thus avoiding publication bias. The register held by the European Medicines Evaluation Agency (EMEA) of the European Union could form a part of such a register, but must be freely available to researchers and not maintained ‘Commercial in Confidence’ as it currently is.


This should include a clear description of the types and characteristics of patients eligible to take part (entry criteria), their previous treatments or medication(s), details of who will administer the treatments and how, and how the treatment effects will be recorded. For example, in a trial comparing methods of pain relief, one needs to devise a method of scoring the intensity of pain, which can then be compared with the intensity of pain before treatment and between treatments. Ideally, to prevent any possible bias, neither the patient nor the person who assesses the effects of treatment should know which treatment is being used (known as blinding), though this is not always possible. Other factors that could influence the effects of the treatment being tested, such as diet, exercise and other medications, need to be recorded, as do side effects and reasons for dropping out of the trial.

Data analysis

Patients who fail to complete treatment for any reason should not be excluded from the final analysis, as comparisons should usually be on an intention to treat (ITT) basis. In some trials it is necessary to report results both based on ITT analysis and also using only completers. It is always necessary to report how many patients refused consent, how many patients met the entry criteria, how many were randomised to each of the two (or more) comparison groups, and how many dropped out from each of the treatment groups and their reasons for doing so, if known. Any apparent differences in results must be analysed to determine the probability of their arising by chance. A probability of less than one in twenty is often regarded as a reasonable indication of a real difference. For this reason, the more patients who enter a trial, the smaller the differences that can be detected. Small numbers can reliably detect only big differences, and make it more likely that the two groups are not properly comparable in important respects other than the treatment.

This is a very simple summary of essential features of a well-designed clinical trial. HealthWatch is very willing to give advice to anyone seeking to evaluate the efficacy of any method of treatment and or type of benefit or side effect.

Further reading about trial design

From trial outcomes to clinical practice. Drug & Therapeutics Bulletin. 34:38-40. (1996)

Randomised controlled trials in single patients. Drug & Therapeutics Bulletin 36:40 (1998)

Excellent advice on the reporting of clinical trials is on

This position paper was revised by John Garrow and Walli Bounds and endorsed by the executive committee of HealthWatch on 12 January 2005.

Functional foods can be defined as foods that contain one or more added ingredients to provide a positive health benefit, over and above the normal functions of food to provide nutrients, satisfy physiological and psychological hunger and provide pleasure from eating.

This definition excludes vitamins and minerals added to foods to replace losses in manufacture. For the purposes of this paper I will also exclude vitamins and minerals added to breakfast cereals, which were originally intended to provide the micronutrients that would have been obtained from a traditional cooked breakfast. Similarly I will exclude vitamins A and D added (by law in UK) to margarine to provide the vitamins that would have been obtained from butter.

The concept of functional foods was developed in Japan in the 1980s, with a formal definition of “foods for specified health use” (FOSHU) in 1991, accompanied by a regulatory system to approve the statements made on labels and in advertising, based on scientific evaluation of the evidence of efficacy and safety [1]. At the turn of the century the European market for functional foods was some £830 million and at the time was predicted to increase to £1.6 billion by 2010 [2] – this may well have been an underestimate.

As with so many areas, while there may be good evidence of potential benefits of some functional foods, and well-conducted trials show improvement in biomarkers and risk factors, as yet there is little evidence of improved health or increased lifespan in most cases.

Food fortification to prevent deficiency

When a micronutrient (vitamin or mineral) deficiency is widespread in a population, a common approach is to enrich or fortify a staple food. The problem here is that if enrichment is voluntary, so that consumers have a choice of whether to buy the fortified or unfortified product, it is likely that the most vulnerable groups of the population will not be reached. However, if enrichment is mandatory then political problems of freedom of choice arise. It is noteworthy that despite the excellent evidence that fluoride reduces dental decay very significantly, fluoridation of water supplies is not universal in Britain, because of (unfounded) fears of “mass medication”.

Iodine and goitre prevention

The earliest example of food enrichment to meet a public health problem was the use of iodine to prevent goitre; in 1900 it was added to chocolate in Switzerland, as a way of meeting the iodine needs of the children - most vulnerable group. However, the most popular vehicle for iodine enrichment is table salt. In some countries iodized salt is required by law either throughout the country or in vulnerable regions; in others it is optionally available. In the Netherlands and Australia, by law, bread must be baked using iodized salt.

There is no doubt that iodization of salt is effective in preventing goitre (and the more serious problem of goitrous cretinism in infants born to iodine deficient mothers). Four years after the introduction of iodization in Guatemala in the early 1950s the prevalence of goitre had fallen from 38% to 5%. Similarly impressive reductions in the prevalence of iodine deficiency disease have been reported from other countries following iodization.

There is, however, a problem with widespread iodization. Adults who have compensated for inadequate iodine deficiency by developing goitre (i.e. enlarging the thyroid gland so that it produces an adequate amount of thyroid hormone) are at risk of hyperthyroidism when their iodine intake increases.

Folic acid and the prevention of neural tube defects

There is excellent evidence [summarised in references 3 and 4] that 400μg of additional folic acid daily halves the number of babies born with spina bifida and other neural tube defects – and this does not include the relatively large number of affected births prevented by therapeutic termination of pregnancy following ante-natal screening. In USA (and a number of other countries) flour has been fortified by law with folate for more than a decade, and there has been a decrease of 27 – 50% in the number of babies born with neural tube defects [4].

Until now, fortification of flour has not been mandatory in UK, but the advice has been that women who are planning a pregnancy should take supplements, or choose folic acid enriched foods that are available. The problem is that half of births are not planned, and the critical period for folate intake (when the neural tube closes) is between 21 – 28 days of gestation – before the woman knows she is pregnant. Despite more than a decade of publicity for folic acid supplements, there has been no decrease in the number of pregnancies affected by neural tube defects in UK or other EU countries where supplementation is recommended and fortification is not mandatory [4].

The Food Standards Agency in UK has now recommended that flour should be enriched with folic acid, and has published a list of options [5]. However, intakes of folic acid above about 1000μg /day pose a hazard to two groups of the population: the elderly, in whom it might mask early signs of vitamin B12 deficiency due to atrophy of the gastric mucosa with increasing age; and people with epilepsy, since high intakes of folic acid antagonise some anticonvulsant medication. Therefore a balance must be struck in deciding the level of fortification, to ensure benefit to unborn children without putting vulnerable people at risk of excessive intake. The FSA advice is that if flour is to be enriched with folate then voluntary enrichment of other foods should cease [4,5].

…and cardiovascular disease?

There is excellent evidence that an elevated blood concentration of homocysteine is a risk factor for atherosclerosis and stroke. There are eminently plausible biological mechanisms to implicate homocysteine as a causative factor in these diseases, and in people genetically at risk of elevated homocysteine relatively high intakes of folate lower the blood concentration significantly [6].

Possibly as a result of mandatory enrichment of flour in USA and Canada, there has been a significant fall in stroke mortality [7], although it is not clear how much this is due to reduced incidence and how much to a reduced case-fatality rate [4]. Randomised controlled trials suggest that folate is beneficial in reducing stroke incidence – the HOPE2 trial showed a 25% reduction in stroke mortality with supplements of folate (and other vitamins involved in homocysteine metabolism) [8]. In another intervention trial, although there was a significant reduction in plasma homocysteine, there was no reduction in death from myocardial infarction, other cardiovascular disease, or, indeed, all cause mortality [9].

This raises two sets of questions:

• Is the theory wrong – are the biological mechanisms that implicate homocysteine incorrect? Is homocysteine a marker of atherosclerosis rather than a cause? It may be that elevated plasma homocysteine is a result of atherosclerosis affecting the kidney, so that its excretion is impaired.

• Is it too early to see results? Folic acid enrichment has only been mandatory in USA and Canada for a decade, and the people currently dying from cardiovascular disease have longestablished atherosclerosis. It is presumably younger people who will be protected if high intakes of folate are protective. Furthermore, the RCTs [8,9] were secondary prevention trials, in people who already had cardiovascular disease or had already suffered a heart attack.

Intestinal bacteria – prebiotics, probiotics and synbiotics

It is a sobering thought that we host ten times more bacteria in the intestinal tract than there are cells in the human body. Some of the 100 or more species are pathogenic, some are harmless commensals, and some (especially the lactic acid producing bacteria) are beneficial, either producing a variety of compounds that prevent the growth of pathogenic organisms or fermenting resistant starch and dietary fibre to provide short-chain fatty acids that are the preferred fuel for intestinal mucosal cells and may have anti-cancer activity.

The main lactic acid producing bacteria of interest are Lactobacillus and Bifidobacteria spp., and there is evidence for beneficial effects of lactic acid bacteria in controlling allergies, preventing or curing constipation and generally maintaining gastro-intestinal health [10,11].

Can we modify intestinal bacteria to increase the population of beneficial organisms?

There are two approaches to this: consumption of beneficial bacteria (collectively known as probiotics) and consumption of non-starch polysaccharides (and especially short-chain oligosaccharides), which provide a substrate for fermentation by the probiotic bacteria, and are therefore known as prebiotics. The combination of the two, probiotic bacteria and prebiotic carbohydrates, is known as a synbiotic.

Consume the bacteria – probiotics

Elie Metchnikoff was co-recipient (together with Paul Ehrlich) of the 1908 Nobel Prize in Physiology or Medicine for his discovery of cell-mediated immunity and phagocytosis. From about 1900 until his death at what was then the advanced age of 71 he was interested in ways of increasing longevity and, inspired by the reported longevity of people in the Balkans who drank fermented milk, he advocated the consumption of lactic acid-producing bacteria in soured milk and yoghurt.

Inspired by Metchnikoff, the Japanese microbiologist Minoru Shirota (who was 83 years old when he died in 1982) isolated a strain of Lactobacillus casei that could not only survive the acid conditions of the stomach and thrive in the intestinal tract, but could out-compete potentially harmful organisms. In 1935 he started selling a yoghurt-like drink containing this organism under the name Yakult.

According to the Yakult company’s website, “each tiny bottle [65 mL] of Yakult contains 6.5 billion active bacteria of the strain Lactobacillus casei Shirota” [10]. Yakult was introduced to the European market in 1994, from a factory with a production capacity of 720,000 bottles per week. This one production site has a current capacity of 10 million bottles a week, and the company claims that in 2007 Yakult “is enjoyed daily by 25 million people in 28 countries around the world and this number continues to grow” [10]. Danone [11] produce two probiotic yoghurts: Activia, which contains Bifidobacteria and Actimel, which contains L. casei. Sales of Actimel are reported to have reached €1.4 billion in 2006 [12]. Both companies sponsor a considerable amount of research into the effects of changing intestinal bacterial flora, and are generous sponsors of scientific meetings.

We cannot know whether the longevity of Metchnikoff and Shirota was due to their consumption of lactic acid bacteria, and any RCT will presumably take 70 – 80 years (or longer) to produce results. However, what we do know is that you have to keep drinking the probiotic yoghurt. The beneficial strains of Lactobacillus casei and Bifidobacterium spp. do not seem to be able to establish stable colonies in the intestinal tract [13].

Feed the bacteria – prebiotics

It has long been known that something present in human milk (but absent from cows’ milk) that promotes the growth of Bifidobacteria and other beneficial species, and that the intestinal bacterial flora of breast fed babies differed from that of bottle fed babies. What was originally called the bifidus factor is now known to be a mixture of prebiotic oligosaccharides – carbohydrates that are not digested in the small intestine, but provide a substrate for bacterial fermentation in the large intestine. A number of foods are now available that contain added prebiotic carbohydrates, including drinks, yoghurts, biscuits, breakfast cereals and margarines. Does encouraging the growth of lactic acid bacteria have any effects on health?

There is a good evidence for the effects of some prebiotics in alleviating constipation. They are also useful in treating hepatic encephalopathy, which is due to ammonia intoxication as a result of liver failure – here the effect is due to acidification of the gut contents, so that ammonia diffuses from the blood stream into the gut, and is trapped as ammonium salts, which cannot cross back into the bloodstream [14]. The evidence is less good for the prevention of colon cancer, intestinal infection, and recurrence of inflammatory bowel disease, but a number of trials have suggested that prebiotics can prevent colonisation of the intestinal tract with pathogens such as Clostridium difficile and Helicobacter pylori [14], although there are beneficial effects against the development of experimental colon cancer in animals [15]. There is also evidence that prebiotics can modulate various properties of the immune system, especially the gut-associated lymphoid tissues (GALT) [16], and enhance calcium absorption [17].

Although inulin and other oligosaccharides have a lipid-lowering effect in experimental animals, the results of human trials have been mixed. Williams and Jackson [18] suggest that this is because it is not possible to feed human volunteers amounts of prebiotics equivalent to those used in experimental animals because of the gastrointestinal discomfort experienced by most people consuming more than about 15 g/d.

Perhaps the situation is best summed up by this quotation from Cummings and Macfarlane [19] “prebiotic carbohydrates clearly have significant and distinctive physiological effects in the human large intestine, and on the basis of this it is likely that they will ultimately be shown to be beneficial to health” (my italics).

Changing dietary fat and lowering cholesterol

For half a century or more we have known that elevated serum cholesterol, and especially low density lipoprotein cholesterol, is a major risk factor in atherosclerosis and coronary heart disease. There are excellent biological mechanisms to explain this. Low density lipoprotein that has not been cleared by the liver (either because the liver has enough cholesterol for its needs, and so down-regulates the uptake mechanism or because the lipoprotein has been oxidised in the circulation, and so is not recognised by the liver receptor) is taken up by macrophages, which infiltrate the arterial wall, die and lay down fatty plaques that occlude the arteries [20].

The results of a series of experiments in the 1950s and 1960s in which volunteers were fed diets with different amounts of saturated or polyunsaturated fats to replace mono-unsaturated fat showed that serum cholesterol increases proportionally to 2 x the intake of saturated fat and decreases proportionally with the polyunsaturated fat intake. Similar studies showed that dietary cholesterol intake was less important, increasing serum cholesterol proportionally to the square root of cholesterol intake [21,22].

An interesting whole country experiment has been conducted in Mauritius. In 1987 the government, concerned by the high incidence of coronary heart disease, changed the formulation of the main cooking oil from one based on palm oil (and hence high in saturated fat) to one based on soya bean oil (providing 30% polyunsaturated fat). Total fat and cholesterol intakes were unchanged, and after 5 years average serum cholesterol concentrations had fallen significantly [23]. However, there has been no report as yet of any change in cardiovascular mortality (checked with a MedLine search on 15/8/2007). Perhaps even 20 years is not long enough to see any beneficial effects on younger people.

Plant sterols and stanols to lower serum cholesterol

Average daily intakes of cholesterol from the diet are between 300 – 600 mg /day; in addition to this, some 2000 mg of cholesterol is secreted each day in the bile, much of which is reabsorbed. This means that anything that will reduce cholesterol absorption from the small intestine will have a much larger effect on whole body cholesterol (and hence serum cholesterol) than would be expected from the dietary intake alone.

Cholesterol is absorbed together with other lipids in the diet in lipid micelles, and is then esterified with fatty acids before being incorporated into chylomicrons that enter the circulation. Any cholesterol in the intestinal mucosal cells that is not esterified is rapidly exported from the cell back into the intestinal lumen, by an active transport process. Analogues of cholesterol, such as the plant sterol β-sitosterol and the stanols, both compete with cholesterol for inclusion in the lipid micelles and also inhibit the enzymes that esterify cholesterol, so that less enters the circulation.

There is abundant evidence that consumption of plant sterols and stanols lowers low density lipoprotein cholesterol, and the effect is additive to that of statins, the drugs that inhibt cholesterol synthesis [24,25]. A variety of sterol- or stanol-containing low fat spreads, yoghurts, drinks and cream cheeses have been marketed. There are also chewable sweets containing Benecol stanol esters [26]. Lowering serum cholesterol certainly reduces the risk of myocardial infarction, and there is no evidence that plant sterols and stanols enter the circulation or contribute to atherogenesis (apart from rare children with sitosterolaemia, a genetic defects of the intestinal sterol exporting protein, who develop severe atherosclerosis at a young age).

Functional foods containing plant sterols and stanols seem to “do what it says on the label”. However, a review by Muldoon [27] of intervention trials to lower serum cholesterol (before the introduction of statins or plant sterols and stanols) showed that while there was a pleasing 15% reduction in death from cardiovascular disease, there was no reduction in overall mortality, and a 75% increase in death that was not illness related (i.e. accident, murder and suicide). This has never been satisfactorily explained.


The concept of superfoods was developed in the USA in 2003-4 and was introduced in Britain by an article in the Daily Mail on December 22nd 2005 [28]. Superfoods are ordinary foods that are especially rich in nutrients or antioxidants and other potentially protective compounds, including polyunsaturated fatty acids and dietary fibre. Scanning through a handful of websites [28 - 31] thrown up by a Google search for “superfoods” gives the following list:

almonds, apples, avocado, baked beans, bananas, beetroot, blueberries, Brazil nuts, broccoli, Brussels sprouts, cabbage, carrots, cocoa, cranberries, flax seeds, garlic, ginger, kiwi, mango, olive oil, onions, oranges, peppers, pineapple, pumpkin, red grapes, salmon, soy, spinach, strawberries, sunflower seeds, sweet potato, tea, tomatoes, watercress, whole grain seeded bread, whole grains, wine, yoghurt.

There are very few surprises in this list. Most of these are foods that nutritionists and dietitians have talked about for years as being nutrient dense – i.e. they have a high content of vitamins and minerals / 100 Calories. The nuts, seeds and olive oil are an exception, but they are all good sources of polyunsaturated fatty acids.

The labelling and marketing of the foods as superfoods seems disingenuous (or a clever marketing strategy), but if such marketing leads people to eat more fruit and vegetables and reduce their saturated fat, salt and sugar intake then it can only help to reinforce the message that the nutrition and public health communities have been preaching for more than a quarter of a century. Position paper prepared by David A Bender and approved by the HealthWatch committee 11/7/2007


1. Shimizu T. Health claims on functional foods: the Japanese regulations and an international comparison. Nutrition Research Reviews 2003; 16: 241-252.

2. Ashwell M. Functional foods: a simple scheme for establishing the scientific validity for all claims. Public Health Nutr 2001; 4(3): 859-62.

3. Department of Health. Report on Health and Social Subjects 50: Folic acid and the Prevention of Disease, Report of the Committee on Medical Aspects of Food and Nutrition Policy. London: Stationery Office; 2000.

4. Scientific Advisory Committee on Nutrition. Folate and Disease prevention, 2006. Available to download from: (accessed 15/8/2007).

5. Food Standards Agency. Options for improving folate intakes of women of reproductive age and preventing neural tube defects. 2007. Available to download from: (accessed 15/8/2007).

6. Homocysteine Studies Collaboration. Homocysteine and risk of ischemic heart disease and stroke: a meta-analysis. JAMA 2002; 288 (16): 2015-22.

7. Yang Q, Botto LD, Erickson JD, Berry RJ, Sambell C, Johansen H, et al. Improvement in stroke mortality in Canada and the United States, 1990 to 2002. Circulation 2006; 113(10): 1335-43.

8. Lonn E, Yusuf S, Arnold MJ, Sheridan P, Pogue J, Micks M, et al. Homocysteine lowering with folic acid and B vitamins in vascular disease. N Engl J Med 2006; 354(15): 1567-77.

9. Bonaa KH, Njolstad I, Ueland PM, Schirmer H, Tverdal A, Steigen T, et al. Homocysteine lowering and cardiovascular events after acute myocardial infarction. N Engl J Med 2006; 354(15): 1578-88.

10. (accessed 15/8/2007).

11. (accessed 15/8/2007).

12. (accessed 15/8/2007).

13. Oozeer R, Goupil-Feuillerat N, Alpert CA, van de Guchte M, Anba J, Mengaud J, et al. Lactobacillus casei is able to survive and initiate protein synthesis during its transit in the digestive tract of human flora-associated mice. Appl Environ Microbiol 2002; 68(7): 3570-4.

14. Marteau P, Boutron-Ruault MC. Nutritional advantages of probiotics and prebiotics. Br J Nutr 2002; 87 Suppl 2: S153-7.

15. Pool-Zobel B, van Loo J, Rowland I, Roberfroid MB. Experimental evidences on the potential of prebiotic fructans to reduce the risk of colon cancer. Br J Nutr 2002; 87 Suppl 2: S273-81.

16. Schley PD, Field CJ. The immune-enhancing effects of dietary fibres and prebiotics. Br J Nutr 2002; 87 Suppl 2: S221-30.

17. Griffin IJ, Davila PM, Abrams SA. Non-digestible oligosaccharides and calcium absorption in girls with adequate calcium intakes. Br J Nutr 2002; 87 Suppl 2: S187-91.

18. Williams CM, Jackson KG. Inulin and oligofructose: effects on lipid metabolism from human studies. Br J Nutr 2002; 87 Suppl 2: S261-4.

19. Cummings JH, Macfarlane GT. Gastrointestinal effects of prebiotics. Br J Nutr 2002; 87 Suppl 2: S145-51

20. LaRosa JC, Hunninghake D, Bush D, Criqui MH, Getz GS, Gotto AM, Jr., et al. The cholesterol facts. A summary of the evidence relating dietary fats, serum cholesterol, and coronary heart disease. A joint statement by the American Heart Association and the National Heart, Lung, and Blood Institute. The Task Force on Cholesterol Issues, American Heart Association. Circulation 1990; 81(5): 1721-33.

21. Keys A, Anderson JT, Grande F. Prediction of serum-cholesterol responses of man to changes in fats in the diet. Lancet 1957; 273(7003): 959-66.

22. Hegsted DM, McGandy RB, Myers ML, Stare FJ. Quantitative effects of dietary fat on serum cholesterol in man. Am J Clin Nutr 1965; 17(5): 281-95.

23. Uusitalo U, Feskens EJ, Tuomilehto J, Dowse G, Haw U, Fareed D, et al. Fall in total cholesterol concentration over five years in association with changes in fatty acid composition of cooking oil in Mauritius: cross sectional survey. BMJ 1996; 313(7064): 1044-6.

24. Law M. Plant sterol and stanol margarines and health. BMJ 2000; 320(7238): 861-4.

25. Katan MB, Grundy SM, Jones P, Law M, Miettinen T, Paoletti R. Efficacy and safety of plant stanols and sterols in the management of blood cholesterol levels. Mayo Clin Proc 2003; 78(8): 965-78.

26. (accessed 14/8/2007).

27. Muldoon MF, Manuck SB, Matthews KA. Lowering cholesterol concentrations and mortality: a quantitative review of primary prevention trials. BMJ 1990; 301(6747): 309-14.

28. page_id=1798 (accessed 14/8/2007).

29. (accessed 14/8/2007).

30. (accessed 14/8/2007).

31. (accessed 14/8/2007).

On 1st September 2006, UK regulations came into force which permit homeopathic medicines to carry indications on their labels [1]. Hitherto, only such products on the market before 1971, when the 1968 Medicines Act came into force, could carry such claims under a ‘licence of right’ (in common with all other medicines at the time). All homeopathic products marketed after 1971 are not allowed to carry indications for the diseases they claim to treat. There are currently about 3,000 homeopathic licences, and it is no surprise that the vast majority are licences of right. This contrasts rather sharply with the situation of orthodox medicines, for which virtually no pre-1971 licences exist today.

The new regulations stem from a desire to resolve this obviously anomalous situation, driven by a European Directive. The Medicines and Healthcare products Regulatory Agency (MHRA), issued a consultation on its proposals in February 2005. Four options were offered. Essentially these were:

  1. to do nothing
  2. to revoke all licences of right
  3. to allow efficacy claims based on non-clinical trial data
  4. to do the same as (3) but also to review all licences of right on a voluntary basis.

The MHRA states that the consultation responses were in favour of the last option, and this is now embodied in Statutory Instrument 2006 number 1952. The new regulations were laid before Parliament four days before the summer recess, and came into force on 1st September 2006, over five weeks before the new session, giving no opportunity for debate. Interestingly, the MHRA says that there were no strong public health reasons for taking any action, and that the only reason for rejecting the first option was the expectation of agitation by the homeopathy companies.

Although one purpose of the regulations might have been to encourage manufacturers to transfer efficacy claims from serious conditions to minor conditions only, a by-product is to allow such claims without the need to provide any supporting evidence. Instead, the MHRA will accept what it calls “nonscientific data” - its own words. In its explanatory notes, the MHRA admits that homeopathic products “have difficulty in demonstrating efficacy in clinical trials”. This is no different from saying “they do not work”. Data now acceptable can come from homeopathic “provings”. It cannot be over emphasised that “provings” have nothing at all to do with efficacy, and are carried out by giving healthy people undiluted homeopathic stocks. These may be of plant, animal, or mineral origin. The symptoms elicited by this process are imagined to indicate the diseases which the ultra-dilute finished product is able to treat, on the principle of “like cures like”. This, the so-called “law of similars”, is not supported by any scientific evidence. Indeed, since the new regulations appeared, a meta-analysis has appeared which shows that there is no consistency or reliability for studies of provings published from 1945 to 1994 [2]. Thus the already scientifically invalid basis for the MHRA’s proposals is also invalidated by the homeopaths themselves. The other main principle of homeopathy is the “law of infinitesimals”, the idea that medicines become more potent the more they are diluted. There is of course no evidence to support this either, and it is in conflict with all that we know about pharmacology, therapeutics, and indeed physics and chemistry themselves.

In addition, the new regulations accept as evidence, proof that the product has been used for the claimed indication “within the homeopathic tradition”. Obviously neither this requirement, nor “provings”, is anywhere near a definitive test of efficacy. The regulations do not list any other types of evidence as acceptable. Potential sources of data were listed as including homeopathic pharmacopoeiae and materiae medicae, and bibliographies, such that they would be accepted by homeopathic practitioners. In other words, all that is necessary is to convince homeopaths, and it is not necessary to win over anyone with a more scientific view of medicine.

The MHRA prefaced its consultation by stating quite clearly that clinical trial evidence was lacking. Interestingly, not one of the various homeopathy organisations which responded enthusiastically to the consultation even suggested that such a view might not be correct, despite vociferous claims from many homeopaths that clinical trials do show efficacy. It is not too difficult to find such positive trials if one ignores the matter of methodological quality. This might explain why, despite two centuries of use, HealthWatch position paper: The regulation of homeopathic medicines in the UK 2 the clinical evidence for homeopathy actually gets weaker over time [3]. The key question which any scientist, and the MHRA in particular, should ask is: “After 200 years why are we still arguing about the efficacy of homeopathy?”.

The MHRA published a summary of responses to the consultation4, but omitted to mention that three medical Royal Colleges strongly criticised homeopathy. The Royal College of Physicians stated that, “it is important that unsubstantiated or false claims of efficacy are absolutely prohibited”. The other critical Royal Colleges were those of General Practitioners, and of Physicians (Edinburgh). Certain other groups, such as the National Eczema Society, voiced very similar objections. However the Royal College of Radiologists very warmly supported option (4), while all along mistaking homeopathy for herbal medicine. The Royal College of Nursing was even more enthusiastic, but the response was written by a homeopath. Several other organisations issued critical statements, including the Royal Society, The Academy of Medical Sciences, the Biosciences Federation, the Medical Research Council, and the Royal Society. The British Pharmacological Society (of which the MHRA’s chief executive and chairman are both members) said:

The British Pharmacological Society believes that any claim made for a medicine must be based on evidence, and that it is the duty of the regulatory authorities, in particular the MHRA, to ensure that no claims can be made for the efficacy of any form of medicine unless there is good evidence that the claim is true. Despite many years of investigation, we have no convincing scientific evidence that homeopathic remedies work any better than placebo. Pharmacologists have noted frequently that most homeopathic products are diluted to the extent that they contain no molecule of active ingredient, that is, no medicine, which is highly misleading to consumers who are unlikely to recognise the expression “30C” for example. Furthermore, there are serious concerns, even in cases where they are used for minor ailments, that officially endorsed use of such remedies may put patients at risk of delayed diagnosis. The Society is therefore surprised that the national rules scheme for licensing homeopathic products, which came into force on 1st September (Statutory Instrument 2006 1952), will regard non-scientific data as evidence of efficacy.

By now, readers may be wondering what is driving such a bizarre move. The answer might be found in the MHRA’s own Regulatory Impact Assessment [5]. It is stated there that not to act thus would “inhibit the expansion of the homeopathic industry”. This is the first time that the MHRA has admitted to a commercial remit. It is not in its mission statement, but this is:

We enhance and safeguard the health of the public by ensuring that medicines and medical devices work and are acceptably safe. No product is risk-free. Underpinning all our work lie robust and fact-based judgements to ensure that the benefits to patients and the public justify the risks.

Les Rose BSc CBiol MIBiol FICR MAPM

HealthWatch committee member


Adapted with the publication’s permission from Rose, L British health care regulation moves away from science. Biologist 2007; 54 (1): 3-5.


1. MHRA Press Release. ?IdcService=SS_GET_PAGE&useSecondary=true&ssDocName= CON2024653&ssTargetNodeId=389. Accessed 5th December 2006.

2. Dantas F, Fisher P, Walach H et al. A systematic review of the quality of homeopathic pathogenetic trials published from 1945 to 1995. Homeopathy 2007; 96: 4-16.

3. Editorial: The end of homoeopathy. Lancet 2005; 366: 690.

4. MLX312 Summary of Responses. idcplg?IdcService=SS_GET_PAGE&useSecondary=true&ss DocName=CON1004429&ssTargetNodeId=373. Accessed 5th December 2006.

5. MLX312 Full Regulatory Impact Assessment. me=CON1004429&ssTargetNodeId=373. Accessed 5th December 2006.