Friday, December 15, 2006

A heavy financial burden

According to one leading obesity specialist, our expanding waistlines could spell financial disaster for the NHS. Professor Naveed Sattar, writing in this week's British Medical Journal said, 'The problem of rising prevalence in obesity may get much worse - rates could climb still further, bankrupting the health system and leading soon to reductions in life expectancy.' This remark is part of an article, co-written with other health academics, that suggests a range of measures to try to prevent obesity including improved training for medical professionals and a single government agency with responsibility for the problem. Suggestions that have received a lot of attention in media reports include putting a helpline number on all clothing over a certain size and putting warning labels on food products with more than 700 calories (or 250 calories for snacks).

While this issue of health costs may be a serious one, it's simply not the subject of the BMJ article - although it is the headline in the BMJ's press release. The possibility of obesity overwhelming health services is merely asserted. In any case, the NHS has always faced the problem of matching limited funds to almost unlimited possible ways to spend it. The worst case scenario is that the obesity 'epidemic' becomes so bad that resources are pulled away from other fairly serious conditions to meet demand. That's quite a different idea to the health system collapsing.

However, we've been here before. There have been numerous claims in the past that a particular health problem was going to swamp health services without ever happening. For example, in the nineties, we were told that hundreds of thousands of people would die from variant-CJD contracted from cattle with 'mad cow disease'. Total deaths since the disease was identified over 10 years ago have been measured in the dozens - so few, in fact, that they have always been out-numbered by deaths from the older 'sporadic' form of CJD which was generally regarded as 'rare'. More recently, there have been fears that a bird flu pandemic would cripple health services - but no such pandemic has occurred while millions of pounds have been spent on drugs which, in all likelihood, will have little impact on the progress of any such disease.

It is true that people in the UK are getting fatter and that above a certain weight, the fatter you are the more likely you are to suffer from certain chronic conditions. However, the fears of financial disaster seem to be based on the assumption that people will just carry on getting fatter and fatter, and an over-estimation of just how much ill-health will result. Already, US health researchers have greatly lowered their estimates for ill-health arising from 'excess' weight because those categorised as 'overweight' seemed to have a higher life expectancy than those in the 'normal' category.

The estimates also understate the degree to which medical interventions may improve treatment of obesity-related problems in the future, in ways which do not cost an arm and a leg. For example, some research suggests that taking even quite moderate exercise can substantially reduce the risk of health problems in the very overweight, regardless of whether they actually lose weight.

What the article really illustrates is that the people with the greatest knowledge of how to treat a patient with obesity-related health problems may not be the best people to comment on politics, economics or society's attitude to food and health more generally. But as the media coverage of obesity shows us time and again, the authors are not alone in needing some perspective on the issue.

Read on:

'Obesity- can we turn the tide?', British Medical Journal, 16 December 2006

spiked-issue: Obesity

Friday, December 01, 2006

A rasher panic

'Bacon link to bladder cancer risk', says BBC News, reporting on research from Harvard University published in the American Journal of Clinical Nutrition. Researchers found that people who ate bacon at least five times a week were 59 per cent more likely to develop bladder cancer than those who never ate it. The study was conducted on data from 136,000 people over the course of 22 years. A similar finding was made for those who ate skinless chicken. The authors suggest that chemicals called nitrosamines, found particularly in processed meats, and heterocyclic amines may be to blame.

This study shouldn't put people off those bacon butties just yet. Bladder cancer is a relatively unusual form of cancer. In the UK, only about four per cent of all cancers diagnosed are bladder cancers, although that still means 10,000 cases per year and 4,800 deaths from the disease.

For starters, the relative risks involved are low. In 1994, the US National Cancer Institute suggested that 'in epidemiologic research, [increases in risk of less than 100 per cent] are considered small and are usually difficult to interpret. Such increases may be due to chance, statistical bias, or the effects of confounding factors that are sometimes not evident.' The number of deaths involved during the course of the study (a total of 808 cases out of 136,000 people followed over 22 years - little more than 0.5 per cent) also suggests the risks are small.

But even if these risks were accurate, most deaths from bladder cancer are in old people. In fact, age is a far better predictor of ill-health than anything else. So, to avoid a small risk of bladder cancer when we've already had our three-score and ten years, should we forego the pleasure of eating what many people would argue is their favourite food? As is so often the case these days, this sets the extension of life above the enjoyment of life.

Bacon link to bladder cancer risk, BBC News, 29 November 2006

Bladder cancer, Cancer Research UK

Thursday, November 23, 2006

Wi-fi fears

'Health fears lead schools to dismantle wireless networks,' reports the Times (London) this week. Parents and teachers have been requesting that wi-fi networks connecting computers in classrooms be removed because of fears that the radiation from them could be affecting health. 'We listened to the parents’ views and they were obviously very concerned. We also did a lot of research. The authorities say it’s safe, but there have been no long-term studies to prove this,' said Tim Cannell, headmaster of a preparatory school in Sussex. Nor is this concern unique: in January this year, for example, the president of Lakehead University in Canada refused to allow a campus wireless network for the same reasons. Fred Gilbert said it was 'a matter of taking precautions and providing an environment that doesn’t have a potential risk'.

There is no evidence that the tiny amounts of radiation involved in wi-fi networks - measured in milliwatts - could have any biological effect. Concerns about wireless networking are obviously a spin-off from other concerns about mobile phones and power lines - but, if anything, the concern is even more misplaced.

As numerous studies have shown, there is basically no evidence that having a mobile phone right next to your head will cause any harm. But wireless cards and routers aren't even placed right next to our heads. Not only is the output from a wireless network card or a router tiny to start with but the intensity of the output weakens dramatically as you move away from it, so having a wi-fi router some distance from you is even less likely to be harmful.

You can see this effect with a 40-watt light bulb. Put your eye close to it and the light is likely to feel painfully bright. As that same light spreads out, however, it becomes much weaker. On the walls of the room, the light from a 40-watt bulb will be pretty feeble. So, even being a few feet from a source of energy makes a huge difference to the power you receive from it - and the same thing applies to wi-fi equipment in classrooms.

This concern has got nothing to do with any evidence that there might be harm caused by such devices. The headmaster's statement about the lack of long-term studies is obviously true - wi-fi is new technology. But we don't need epidemiology to tell us that tiny amounts of energy are harmless.

Rather, the notion of invisible forces silently damaging us - or worse, our vulnerable children - is a powerful metaphor for the way that modern society has increasingly left us feeling isolated and open to attack. Unlike the microwave radiation, this health scare, like so many others in recent times, is all in the head.

Health fears lead schools to dismantle wireless networks, The Times, London, 20 November 2006

Health concerns limit wireless Internet at Lakehead University, IT Business (Canada), 23 January 2006

Friday, November 17, 2006

Where's the beef?

A study linking red meat intake and some forms of breast cancer in pre-menopausal women was reported around the world this week. Researchers at Harvard Medical School analysed data from the Nurses Health Study II following women aged between 26 and 46 years over a 12 year period. From over 90,000 women followed-up, there were 1,021 cases of breast cancer. Cases of breast cancer related to progesterone and oestrogen receptors were almost twice as common in those who ate 1.5 portions of red meat per day than in women who ate three or fewer portions per week.

Breast cancer is most commonly a post-menopausal condition. Even if the figures stated were an accurate reflection of an increased relative risk, the overall risk for any pre-menopausal woman is still small. For example, in this study, only around one per cent of the women involved developed breast cancer.

As Cancer Research UK noted in response to the report: 'According to this study, a woman would need to eat more than one-and-a-half portions of meat a day, every day, to significantly increase her risk of hormone-sensitive breast cancer before the menopause. But the overall risk of pre-menopausal breast cancer is low when compared to getting the disease after the menopause. So even at the highest rates of meat consumption this is overall still a relatively small increase.'

Indeed, as the Harvard researchers note: 'Previous epidemilogic data on red meat intake and breast cancer risk have been inconclusive.' Rather than accept that this link is unlikely to be of any practical significance, they appear to have sliced and diced the data further until finding something, then tacked a theory on afterwards.

However, even the results stated should come with a health warning. Relative risks of 2.0 (ie, a doubling of risk) or below need to be treated with extreme caution. There are plenty of reasons why the results might be inaccurate. This was a study based on food questionnaires completed in 1991, 1995 and 1999. Were these questionnaires filled in correctly? Our eating habits tend not to be fixed but vary over time - will these changes be captured accurately in four-yearly snapshots? How was this red meat consumed - did cooking methods make a difference? Did the women who ate more red meat tend not to eat other foods as a result which might be protective? There are so many possible questions with such a study that only relatively high relative risks should really be taken seriously.

Further, there is a striking difference in overall calorie intake for the different groups. The women who ate the least red meat consumed, on average, 1524 calories per day, while the most frequent red meat eaters consumed 2359 calories per day. Unsurprisingly, the latter group weighed, on average, well over a stone more than the former group. Are calorie intake or body weight factors?

There are so many uncertainties in this study that it is probably best to ignore it. Even if it is accurate, it can only make people feel guilty about eating what is, after all, nutritious and tasty food with almost certainly no effect on their life expectancy. What purpose was served by that?

Read on:

Red meat intake and risk of breast cancer among premenopausal women, Archives of Internal Medicine, November 2006

Friday, November 03, 2006

Trans fat

Health officials in New York are planning to prohibit the city’s restaurants from serving food with more than a minute amount of partially-hydrogenated fat, or trans fat. This is based on research which suggests that these fats, frequently used in cooking oils and baked goods like cakes, biscuits and doughnuts, can have a negative effect on cholesterol levels while not having any particular nutritional value other than calories. Trans fats seem to raise levels of ‘bad’ LDL cholesterol, in much the same way as saturated fat, but also seem to lower levels of ‘good’ HDL cholesterol. Dr Walter Willett from Harvard University has estimated that if artificially-produced trans fats were removed from the American diet, up to 228,000 heart attacks could be prevented each year.

There are many ingredients in our diet which add little to the nutritional value of food but which are added to improve the flavour, texture or other qualities: including sugar, salt (beyond the relatively small amount we need to survive), spices, oils, butter and so forth. Trans fat falls into this category, improving the texture of certain foods while being more stable than many of the alternatives. If we are going to ban it, we should make sure we're doing it for a good reason.

The problem is that the claims made about this 'artery-clogging' fat seem out of all proportion to reality. It certainly appears to be the case that eating more trans fat and saturated fat increases cholesterol levels – but the link between cholesterol and heart disease is a lot more tenuous than we're usually led to believe. In fact, as Malcolm Kendrick points out elsewhere on spiked, 'no clinical trial on reducing saturated fat intake has ever shown a reduction in heart disease'.

The link between trans fat and heart disease seems to be equally weak. As Steven Milloy has pointed out on his website, most of the studies done seem to be produce results which are not statistically significant or the association is weak. It is certainly not the kind of data that would justify the removal of a useful food product from restaurants as New York's health officials propose – and makes a mockery of Willett's figures on heart attacks. The irony is that when saturated fat became dietary public enemy number one thirty years ago, one of the alternatives widely suggested was trans fat.

We should stop listening to those health campaigners who have made a career out of scaring us about our food. Trans fat is highly unlikely to shorten our lives, but scare stories like this one will certainly make them more miserable.

Fear of margarine: the trans fat myth, Junkscience

The Great Cholesterol Myth, by Malcolm Kendrick

Monday, October 16, 2006

Yet another thing for pregnant women to avoid...

'Mercury may contribute to premature birth' reports New Scientist on research from Harvard School of Health suggesting that eating too much fish increases the risk of giving birth early. 1024 women in Michigan were tested for mercury levels in hair samples. Women who gave birth at least two weeks early were three times as likely to have mercury concentrations of 0.55ppm or above than those who carried their baby for the full term. In turn, women who ate more fish - particularly canned fish - were more likely to have higher mercury levels.

The total number of women who gave birth early in this study was just 44, so drawing any firm conclusions would, perhaps appropriately, be very premature. But more important is the wider issue of how to advise women on what is the best thing to do. Even the lead researcher for this report can see the problems. 'The messages are really very conflicting because fish is both a benefit and a potential source of hazard,' said Dr Fei Xue. Fish is particularly in vogue at the moment because of the high levels of omega-3 oils it contains. But now, eating fish is apparently dangerous, too. Dr Xue recommends fish oil supplements in place of fish as a compromise.

The real problem is the obsession with any potential threat to the health of the fetus at the expense, it would seem, of all other considerations. A recent article by Anna Browning for BBC News summed up the confused and contradictory situation very well. Among the potential hazards she lists are: weight gained during one pregnancy affecting the next one; drinking alcohol; blue cheese; soft cheese; shellfish; caffeine; bagged salad; coleslaw; cigarettes; flax seed; paté; liver; raw eggs; hot baths; lack of folic acid; too much exercise. She could have added eating tuna and avoiding foods with a high glycaemic index, but in truth the list of dos and don'ts is almost endless. Many of them also contradict wider advice about what is good for us.

Women are bombarded with advice about what they should and shouldn't do to the point where they are increasingly reduced to the role of grim incubation. The advice assault doesn't start with conception or end with birth. On the one hand, to become pregnant apparently demands that you are a paragon of healthful virtue, taking regular exercise, eating wholesome food and avoiding fags and booze (despite the obvious link between alcohol and the process of conceiving). After the birth, the pressure from health workers to breastfeed, with all the attendant implications for diet and lifestyle, means that what should be a natural and joyful process has been increasingly turned into an anxious and miserable chore. The irony is that levels of infant mortality and birth defects are far lower than in the past. For example, infant morality in the period 1901-05 in England and Wales was 138 per 1000 births. For 2005, it was just 5.1 per 1000 births.

Heavy drinking and smoking are best avoided, but the risk of any harm from these other factors seems very small - surely no higher than fretting over every mouthful and sip of what passes the lips. Moreover, we should put mothers back at the centre of the equation as human beings with a variety of needs and roles of which carrying a child is just one. If having a baby means that life must come to a crashing halt, fewer women will be prepared to take the risk, denying themselves the undoubted pleasures of parenthood.

Mercury may contribute to premature birth, New Scientist, 13 October 2006

Confused, guilty and pregnant, Anna Browning, BBC News, 2 October 2006

Tuesday, September 26, 2006

Pack it in

'The pack-a-day habit threatening our kids' health,' intones the British Heart Foundation (BHF) to promote its Food4Thought campaign. According to the BHF, eating one pack of crisps per day will lead children to consume almost five litres of cooking oil in the course of a year. Their press release notes that half of British schoolchildren 'admit' to eating a pack of crisps everyday while almost one-in-five eat two packs or more. Other nutritional shockers include the finding that three quarters of mothers feed their children ready meals or takeaways more than three times a week and only 13 per cent of boys and 12 per cent of girls reported eating the recommended five portions of fruit and vegetables daily.

This campaign is as heavily laden with spin as the crisps are with oil. The reference to consuming a 'pack-a-day' has a strong whiff of cigarettes about it. Since in the popular imagination, 'cigarettes=death', the implication is that eating so many crisps will have a similar impact. Yet, while the picture of a little girl pouring a gallon container of oil down her throat looks repulsive, the comparison is no less grotesque.

Oil is a perfectly normal and healthy part of the diet. Eating nothing but crisps would be quite likely to produce a greasy and rather anaemic looking child, but as long as there is some variety in children's diets (and not just between cheese and onion and smokey bacon), there shouldn't be a problem.

Another way of putting that 'almost five litres of oil' figure would be that children consume about two-and-a-half teaspoons of oil per pack. In energy terms, the oil contributes about 100 calories. Not exactly devastating. But the ruse of adding up a year's consumption is ludicrous. For example, if a child drinks a litre of water per day, that means they consume 365 litres per year - enough to fill four baths. If they attempted to drink all of it at once they'd certainly drown. Yet no-one is suggesting that drinking water is bad for you.

There is no such thing as 'bad' foods, only bad diets. Even then, the link between eating fat and ill-health has never been backed up by the evidence. When major studies have been conducted into the effect of changing diet to a low-fat or low saturated fat intake, the results have been extremely disappointing for those seeking to establish such a link. Rather than targeting health campaigns at children which cause unnecessary worry, finding the root causes of heart disease and better ways to treat it more effectively would be the right path to take. Unlike our children's diets, it's always proven to be more fruitful.

The pack-a-day habit threatening our Kids' health, British Heart Foundation, 22 September 2006