Q J Med 2002; 95: 131-132
© 2002 Association of Physicians
Biologic |
Nonsense and non-science
With the MMR debacle fresh in our minds, and with the collapse of Railtrack reflecting a ludicrously over-protective posture with regard to risk, it is worth examining the place of science in determining public attitudes to the hazards that apparently beset us in both medical practice and daily life.
Dr Spock is a caricature not because of his ears but because of his insistence on logical behaviour. In applying a consistent and considered approach to problems, he marks himself off from the rest of us—as Hume said, ‘logic is a suit of clothes that Man puts on for special occasions’. Most people becomes aware of the inconsistencies implicit in particular attitudes only when confronted with transparent nonsense. Reasoned objections are not particularly valued. For example, Vaclav Smil's book on the transformation of world food production makes it clear that without the contributions of Fritz Haber and Carl Bosch in the development of the processes that allow the use of synthetic nitrogen, we could only feed 40% of the current population of the world. Slash-and-burn farming feeds one person/hectare, with a 25-year recovery period. The most intensive ‘natural’ farming system known (in Southern China, with recycling and feeding of human and animal waste, carp ponds, green manure and cyanobacterial nitrogen fixation on various rotting materials) could feed around 25/hectare, with an enormous consequent burden of human and animal disease. Yet organic agriculture is sometimes referred to as a ‘solution’ to the poorly-defined problems of modern agriculture, which feeds around 45 people/hectare year after year.
Perhaps because common diseases such as hypertension, diabetes and osteoarthrosis are now managed better, environmental hazards have come to the fore in the public consciousness. Most of the problems perceived arise as a result of hazard identification; quantitative data, and therefore problems that may be properly defined as ‘risks’, are few. Because, in terms of anxiety, the general desire is to demonstrate an absence of harm, absence of data makes the problem far more difficult to resolve than generally better-defined concerns about particular disease outcomes. It is usually possible to determine whether someone has a disease or not, or whether death rates in a particular group exceed those expected. It is much more difficult to determine the effects of limited exposure to a wide range of theoretical toxins where there are no agreed outcome measures following exposure—complaints are often related to symptoms such as sleep disturbance, headache, transient rashes, anxiety or depression, all of which are difficult to quantify.
These data are clearly difficult to obtain in the field of environmental health, where exposures are uncertain and unquantified. This intrinsic failure of data collection is important, since it impacts on the determination of priorities associated with maintaining environmental health. Above all it has allowed the introduction of the precautionary principle (PP) into regulation—a defeat for science.
No system can give an assurance that no harm will follow innovation or intervention; no indemnity can be given against contingent harm. Regulatory systems that control the use of xenobiotics are based on caution (the taking of heed) and precaution (prudent foresight).
Data obtained from animal experiments (in particular) are sometimes used to invoke the PP as a justification of the selective use of incomplete data as a basis for decision-making in regulation. Holm and Harris (1999) defined the PP in this way: ‘When an activity raises threats of serious or irreversible harm to human health or the environment, precautionary measures that prevent the possibility of harm shall be taken even if the causal link between the activity and the possible harm has not been proven or the causal link is weak and the harm is unlikely to occur.’ These authors have pointed out that the definition alters the way we consider evidence, suggesting that we should give more weight to evidence pointing one way rather than the other—even when the epistemic warrant (the support there is for believing that part of the evidence) of the two data sets is equivalent. This distorts reason, and cannot be a valid principle for evaluating data.
So although it is easy to say that decisions about what to do about hazards must be science-based and that interventions must depend on a rational analysis of the data, this is not always what happens. The untrained do not understand the scientific method; the maintenance of the rational pattern of thought necessary in scientific debate is an acquired skill. This can be illustrated in a number of ways. In P.J. O’Rourke's essay on ‘how fluoridation of water turns kids into communists’ he illustrates a number of possible mechanisms. None is scientifically credible but all have a critical feature in common; between a (something that is done to you by others) and z (an undesirable outcome) is a step that is true. So, in one example he uses the arguments: A The Government puts fluoride in water—Fluoride in water prevents dental caries—Child without caries pays fewer visits to the dentist—Child is less exposed to the Readers Digest—Child is insufficiently warned of the dangers of Communism—Z Child becomes Communist.
The carefully trained may spot flaws in this, but in general the public will not—if between an exposure and an adverse outcome there is a demonstrably true step the whole argument will be swallowed (fluoridation does prevent caries). The pattern of scientific thought is hard to maintain in the trained (think of papers you have refereed) and it is not reasonable to expect the non-trained to use it.
There is another difficulty. With ill-defined symptomatologies where outcome measures of improvement may not be readily defined, or conflicting views of what constitutes a benefit as a result of an intervention (do you poison other birds in a reserve to ensure the survival of waders?—a real concern which confronted the author as a regulatory decision) we will always disagree about the necessity for action. In particular, if the intervention is opposed or proposed by a group with a particular interest, there will often be effective misrepresentation of data. By definition, there is no attempt at balance in the consideration of data by pressure groups: they generally ignore data that are non-supportive of their argument. Frequent repetition of an invalid message can carry conviction, particularly if the willingness to be convinced is facilitated by a set of preconceptions that favour misinterpretation. The powerful and often repeated presentations of pressure groups are made against a background where the broadcasting of sober health-related messages is made to a public that is generally unresponsive. It is worth considering that they may be unresponsive because of the conflict of opinion generated by zealots.
This failure is not insignificant; Ebrahim and Smith studied the effectiveness of multiple risk factor intervention in reducing cardiovascular risk factors, total mortality, and mortality from coronary heart disease among adults. They used systematic review and meta-analysis of randomized controlled trials (stopping smoking, exercise, dietary advice, weight control, and administration of antihypertensive drugs and cholesterol-lowering drugs), with more than 900000 participants per year of observation. The pooled effects of multiple risk factor intervention on mortality were insignificant, although a small benefit of treatment might have been missed by their methodology. Whatever view you take of the significance of the defined risk factors, it is important to note that the view these authors express, namely that the development of health protection through fiscal and legislative measures might be more effective than education, has real implications for us all.
Almost anything can be a hazard, but the discovery that a real injury may occur because of an unusual event will not be a surprise to any pathologist with a coronial practise (even if it astonishes lawyers). Most of us would accept that cigarettes are damaging for all, but that does not mean that smokers are not responsible for smoking, that those who remove safety devices from their machines are blameless, or that those who drink and drive should be taken by surprise at their lack of motor skills. A disregard of personal responsibilities by a minority should not influence the attitudes of a complex society. A proper response of those who feel able to advise society as a whole with regard to their diet, the pattern of agriculture, the management of transport or the environment should be defined by a consideration of the collective needs of the many rather than the protests of a few. That definition requires science.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||