| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
UKRC 2004 debate |
Health Physics, Gartnavel Royal Hospital, Glasgow G12 OXH, UK
Abstract
This contribution argues the case that, at the present time, the linear-no-threshold (LNT) model provides the only rational framework on which practical radiation protection can be organized. Political, practical and healthcare difficulties with attempting to introduce an alternative approach, e.g. a threshold model, are discussed.
The current system for radiological protection, which is applied to all practices using ionizing radiation, is robust. It has been developed through a substantial effort by the scientific community made over many decades. However, one of the fundamental precepts, that there is proportionate risk of cancer induction extending down to doses at the level of a few millisievert, is a matter of debate. Epidemiological studies of the Japanese survivors of the atomic bombs and patients receiving high exposures from medical procedures do not show a definite link between radiation dose and cancer incidence below 200–400 mSv [1, 2]. Nevertheless, researchers have concluded that there is no convincing evidence to show that the dose–effect response is not a linear extrapolation to zero dose for solid tumours [3].
It may never be possible to resolve the question of risks at low doses completely through epidemiological studies, because the natural incidence of cancer in the population is so high that the sizes of populations which would have to be studied in order to provide statistical validity are too large. Moreover, there are too many confounding factors, such as smoking habits and the "healthy worker" effect, to enable a large enough matched control population to be found [4]. Thus, it is not surprising that there is little evidence of radiation effects at doses below 200 mSv, as such a link would be almost impossible to prove. However, there is one group which does have a much lower natural incidence of cancer, namely children, and among those radiographed while in the womb there does indeed appear to be an excess number of cancer cases from doses as low as 10 mSv, although interpretation of the results is a matter for debate [5].
Political justification
If science does not provide a definitive answer about the risks from low doses of radiation the choice of precepts on which radiological protection should be based must take account of the political, sociological and practical implications. Any change made to the system of radiological protection should be justified by the weight of evidence available and this can best be judged from the views of experts conversant with results of research in the field. Polls held at the L H Gray conference on "Radiation cancer analysis and low dose risk assessment" give an idea of expert opinion [6]. To the question "Is radiation at low doses acting as an initiator?" 80% of the participants answered "yes", and when asked "Does the dose effect relation for cancer induction have a threshold at low doses?" 79% responded "no". Thus the weight of expert opinion at the conference considered that low doses of radiation may carry risk.
Governments have a duty to protect the public, so their advice should err on the side of safety, and cannot afford to ignore the views of the scientific community. The linear-no-threshold (LNT) dose–effect model is a conservative approach. If current practice is continued, individuals are unlikely to be exposed to unnecessary risks. On the other hand, a relaxation of restrictions on low dose exposures has the potential to expose large numbers of individuals to higher doses. If research confirms that there are indeed effects of radiation at low doses, those who have been exposed as a result of the changes are likely to seek redress. This could place a significant burden of liability on governments involved.
There is a "fear factor" associated with radiation in the minds of the public. This has evolved from its destructive beginnings with nuclear weapons and the arms race, and been reinforced by accidents at nuclear power stations such as Chernobyl where large quantities of radioactive material were released into the environment. There is also concern about potential hazards from releases of radioactivity to people living near nuclear establishments, and this has built up a strong environmental lobby. The nature of the unseen hazard, causing hidden damage, which can lead to development of cancer many years later, contributes to the public perception and experts have been unable to allay these fears. If restrictions on the use of radiation are reduced, then the public will be suspicious. Their approval will be difficult to win over, even if the scientific evidence in support of relaxation is undeniable. Any disagreement between experts about the consequences will undermine confidence and the press will seek out dissenting voices to support theories of government conspiracy. Therefore, any decision to relax radiation protection controls can be implemented only if there is overwhelming support from the scientific establishment and experts in the field.
Practical implementation
Another important aspect that must be considered is the practical implication of setting up an alternative system of radiological protection based on a different precept. There are advantages to the current methodology, which may not be easy to replace. The LNT model assumes that risk is directly proportional to effective dose. Not only that, but the dose is cumulative, so that doses can be summed and risk quantified directly. This is not an unreasonable approach based on current evidence. Since doses received by any individual or group can be assessed by measurement or calculation, this provides a methodology with which risks can readily be determined but with large uncertainties. Risks can therefore be used in justification for a wide range of practices involving radiation, including research proposals, planning radiation facilities, restricting radioactive discharges and medical exposures. They can also provide the basis for guidance setting out radiation protection measures for these diverse practices.
Thus the LNT model provides a solid foundation for determination of radiation protection in practice. The application of a graded scale for protection measures based on the advice of radiation experts is accepted. An alternative system derived from a threshold model would be difficult to develop and the basis for it is unclear. The implication with such a system is that low doses can be ignored. However, this is not appropriate unless it is certain that low doses are safe. Decisions would need to be taken about what constituted a low dose and a new strategy developed for implementation of radiation protection. Considerable effort has gone into the present system and it should be maintained unless there is clear, unambiguous evidence that it is incorrect.
The acceptance of a low dose threshold would imply that there are no risks from low doses of radiation, such as those received from radioactivity in the environment arising from radioactive discharges. Constraints relating to the doses received by workers and members of the public from discharges would no longer make any sense. However, any relaxation of regulation would receive strong opposition from the environmental lobby. Therefore, restrictions on the amounts of radioactivity released would still need to be implemented, but application of the threshold model would remove the basis on which radiological impact on the environment is assessed. The foundation on which a replacement methodology could be built is full of uncertainty.
Medical exposures
It is important for the healthcare profession to consider the potential impact of a threshold dose model on medical exposures. The principle of justification could be relaxed to make radiation examinations more widely available, but is this prudent and would it improve healthcare? A robust system is in place in the UK under the Ionising Radiation (Medical Exposure) Regulations 2000 [7]. Diagnostic medical exposures should only be carried out if they will affect patient management. Relaxation of the principle of justification is more likely to result in an increase in unnecessary investigations than to improve the provision of diagnostic imaging services. X-rays might be taken to keep patients happy, probably many by non-specialists, without the same level of training in image interpretation. Any increase in demand on medical imaging facilities will impose a greater burden on the limited funds available for healthcare services. This is likely to affect adversely the overall standard of service provision and add to the spiralling costs of healthcare.
Protagonists of the low dose threshold theory might argue that whole body CT screening would have tremendous health benefits, but this is unlikely. Screening could be carried out for groups at high risk of lung cancer or coronary artery disease, but would not be cost effective healthcare. Examination of the large amounts of image data provided by CT scans would require significant inputs of radiologist time. If abnormalities were found, many would be benign, but would still need to be followed up. CT screening does not make the best use of the limited imaging resources available and could carry a significant health risk. The UK healthcare system needs more resources to allow it to become more efficient, but not to provide more imaging investigations of questionable value.
It might also be possible to relax the optimization of X-rays, if there is less concern about dose. The current UK system for provision of medical imaging services works well. Every effort is made to optimize the procedures for image quality and radiation dose as part of recognized good practice. If the numbers of investigations and facilities are increased, it is likely that X-rays will be taken by non-specialists, who have less experience in optimization, unless the numbers of radiologists and radiographers are expanded. If optimization is not carried out, the healthcare system could be accused of placing patients at unnecessary risk.
The current system of justification and optimization for medical exposures ensures that a high standard of investigation is provided for all patients that need it. Any relaxation would be unlikely to improve patient care and could increase the demand unnecessarily on an already overstretched service. Thus political and practical considerations provide sufficient reasons to continue current practices for medical exposures.
Conclusion
In conclusion, the scientific evidence does not show that there is no risk from radiation at low doses. It is at best inconclusive. Political justification cannot be given to a system based on a low dose threshold that could lead to increased exposures with greater associated risk. Moreover, it would be hard to convince the public of the wisdom for relaxing controls on the use of radiation, and this would be likely to lead to heavy criticism from the press and the environmental lobby, leading to loss of public confidence. Practical implementation of a threshold model would be more difficult because there would be no simple method with which to assess risk. The basis for risk benefit assessments carried out currently in justifying research proposals, planning radiation facilities, restricting radioactive discharges and justifying all practices using radiation would be removed, without any clear alternative methodology being provided in its place. The time is not yet nigh when the LNT approach to implementation of radiation protection should be abandoned, either in healthcare or in other practices using ionizing radiation.
Audience participation
The question put to the audience based on this contribution was "Do you believe that, for medical exposures, political and practical considerations provide sufficient reasons to continue the current methods used to implement principles of Justification and Optimization, even in the face of strong scientific evidence for a threshold at low doses?"
Responses: Yes 69%; no 22%; don't know 9%.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
