Introduction

9. Both the scientific basis of the exposure limits set out in the Directive and their impact on MRI usage have been questioned. This chapter looks at the strength of the scientific evidence that informed the establishment of the exposure limits set in the Directive. It then looks at the evidence and scientific advice underpinning the debate on the impact of these exposure limits on the use of MRI in diagnostics and in research. Both factors need to be taken into account in considering the justification for the Directive, or at least its inclusion of MRI.Box 3: Timeline of key events

Date	Event
1993	Original proposal for catch-all Physical Agents Directive
1994	ICNIRP publishes guidelines on static magnetic fields
1998	ICNIRP publishes guidelines on time-varying fields
1999	Proposal for Physical Agents Directive to be separated into four separate measures
December 2002	Danish Presidency publishes proposals for EMF Directive
26 February—2 September 2003	EU Social Questions Working Party considers Directive
17 September 2003	Political agreement reached in European Council on Directive. Static field limits withdrawn
August 2004	Revised ICNIRP guidelines on MR procedures published
March 2004	Consideration of Directive in European Parliament
29 April 2004	Directive adopted

Sources of advice to the Commission

10. The exposure limits established by the Directive are based on those contained in the ICNIRP guidelines of 1998.[15] They are based on an extensive review of all available scientific evidence, which is clearly summarised in the document. As part of its preparation of the Directive, the Commission provided support to international research on the health effects associated with EMF exposure in the late 1990s.[16] Commission officials cited a number of studies, peer reviewed and published, by different European organisations, including the NRPB, that also informed the limits set out in the Directive.[17] In addition, a seminar of experts in Luxembourg in September 2002 discussed whether there was sufficient scientific evidence of potential health risks of EMF to justify legislation. There was agreement here on a number of potentially acute health effects resulting from powerful EMF and other sources.[18]

11. Since the proposals for a Physical Agents Directive were first published in 1993 the Commission has relied upon ICNIRP to inform its work on the Directive. Dr McKinlay, who chaired ICNIRP between 2000 and 2004, provided a list of ICNIRP's publications throughout the 1990s and also of its meetings with the Commission in 2003.[19] However, he made it clear that ICNIRP did not advise regulators on how to use their guidelines: "It is up to governments and super-national governments to decide about regulations. We do not lobby on this. We do not have a view about it, but we do provide scientific advice.".[20]

12. Dr McKinlay also stressed that ICNIRP does not provide guidelines on individual applications: "We do not concern ourselves [at ICNIRP] with exposure to the particular device with a particular frequency".[21] ICNIRP's role is to provide independent and science-based guidelines and recommendations on protection from non-ionising radiation exposure.[22] He told us that, in terms of specific advice to the Commission on MR, there was "None specifically on occupational exposure and little in general on MRI can be recollected."[23] We believe that Dr McKinlay is underplaying ICNIRP's work here. Whilst the 1998 ICNIRP guidelines cover the full range of EMF exposure, ICNIRP has provided specific guidance on medical MR procedures, for both time-varying and static fields. In April 2003 a draft statement of guidance was circulated by ICNIRP to some experts in the MR community.[24] These largely concern patient safety but also include some reference to occupational exposure. These guidelines were eventually published in August 2004. They state that "concerning open or interventional MR devices with field strengths below 1.0 T, staff operating such devices are not exposed at levels higher than the currently recommended limits for occupational exposure. However, there are only limited data on the exposure of surgeons at open MR devices."[25] Nothing is included about the use of machines with field strengths of above 1.0 T, which are now beginning to be used interventionally. These guidelines demonstrate that ICNIRP was well aware that the Directive might have an impact on MR usage, even if it did not have access to the necessary evidence or expertise to establish the extent of this impact.

13. We accept Dr McKinlay's point that it is not the job of ICNIRP to advise on specific devices or applications, and is therefore not in a position to advise on the impact of the Directive. However, the Commission did have had the opportunity to draw on the advice of ICNIRP in order to ensure that it assessed the full potential impact of the Directive. Equally, given its work on the subject, ICNIRP had the opportunity to suggest that possible consequences for MR, including the future development and use of MRI technology, should be fully explored with the MR community. We accept that it was the Commission's responsibility to assess the impact of the Directive and we discuss its efforts in this respect later in this chapter. We conclude that the Commission was right to go to the established international authority, ICNIRP, for advice on which to base its proposals. However, we believe that the Commission did not seek to obtain the maximum benefit from the work undertaken by ICNIRP by exploring the potential impact of the Directive on MRI. Equally, ICNIRP should accept that, if its guidelines are being used as the basis of the Directive, it has some duty to advise, to the best of its knowledge, on those potentially affected by the Directive, to enable the Commission to consult appropriately. This detailed advice does not appear to have been given.

14. There were further opportunities for advice to be taken during the passage of the Directive through the European Council and European Parliament under the co-decision procedure.[26] The Social Questions Working Group at the European Council considered the Directive during 2003 and suggested some amendments. Following political agreement of a common position on the Directive in September 2003, the Directive was considered by the European Parliament in the first three months of 2004. The Committee considering the Directive proposed some minor amendments, which were accepted. Some concerns about its impact on MRI procedures were raised by the medical equipment manufacturers' representative body, COCIR.[27] This organisation wrote to the Social Questions Working Party in April 2003 to warn that the proposed Directive "could have the effect of restricting or even preventing the use of MRI scanners used in health care" and proposed that MRI equipment be excluded from its scope.[28] When asked about the extent of criticism of the Directive, officials at the Commission eventually acknowledged that "some letters" were received, but we did not get the impression that these were given serious consideration. Passage of the Directive through the European Parliament was swift. The proposal for an exemption in the Directive for MRI equipment was raised in the relevant European Parliament Committee, but was defeated. We note that the European Parliament does not have the time or the resources to conduct a full scientific appraisal of the Directive. Nonetheless, the scientific basis of the Directive was considered in detail by the European Council and one important change was made to it during negotiations, as we set out below in paragraph 21. We discuss in paragraphs 27-40 the extent to which the impact of the Directive was considered.

Strength of the evidence base

15. We have received contradictory views on the strength of the science underpinning the ICNIRP 1998 guidelines. Dr McKinlay stood by them although, as we have seen, he was keen to emphasise that "they are guidelines" and that ICNIRP did not have a view on their use in regulation.[29] Officials from the Commission rejected any notion that the ICNIRP guidelines had been criticised. Mr Biosca de Sagastuy told us that "the ICNIRP guidelines are not contested anywhere in the world. They are the world authority in this field."[30] He pointed to a number of studies by international organisations, including the NRPB, which he said supported the limits set out in the Directive. He drew a distinction between the agreed position of the scientific community, set against the "opinions" of MR manufacturers and medical personnel: "The medical community might have a different opinion but they are not the experts in this field."[31] He drew a parallel with the opposition of the medical community to any restrictions on the use of X-rays until the dangers were identified.

16. This level of certainty was disputed by the MR community. Dr Keevil acknowledged that there was a large body of published research on EMF exposure but asserted that it "reveals a wide margin of uncertainty rather than agreement".[32] Much of the published work on EMF was carried out or commissioned by the NRPB. Following its production of a report on the potential impact of the original 1993 proposals for a Directive, the NRPB carried out a further review for the HSE in 2001 in anticipation of an EMF Directive that year. A report was published in 2002. [33] The NRPB consulted again in 2003. As a result of this work, in 2004 it confirmed that the 1998 ICNIRP guidelines on which the Directive is based should be followed in the UK. The joint submission of evidence describes these 1998 guidelines as being based on a "cautious interpretation of sparse scientific evidence".[34] A recent academic paper reviewing the evidence concludes that the "scientific basis for the exposure levels is incomplete and inconclusive".[35] The quality of the science, as well as the quantity of data available, has also been questioned. A report to the MRC on a meeting of UK stakeholders in January 2006 records that the ICNIRP committee member present accepted that the science underpinning the 1998 guidelines was "poor", although he stood by the limits set. This report also records that there was agreement among the international and national bodies present that some of the ICNIRP guidelines that formed the basis of the Directive were "flawed".[36] These questions are reflected in the HSE-commissioned report of this meeting. One HPA representative is reported as suggesting that "the science is moving faster than the guidance".[37]

EVIDENCE OF ADVERSE HEALTH EFFECTS

17. The strength of the NRPB-commissioned evidence has also been questioned. One aspect of the debate is over the extent to which safe exposure limits can accurately be extrapolated from the available evidence on adverse health effects from exposure. The Chief Executive of the MRC, Professor Colin Blakemore, expressed surprise that his work in 2001 on the Weak Electric Fields Group of the NRPB had been cited as evidence in favour of the proposed exposure limits and therefore of the Directive.[38] He argued that the existence of measurable effects need not imply that these effects were harmful. The group had been asked to speculate about the possible levels of field strength at which there were detectable effects on the body and found that, while it was conceivable that there could be adverse health effects, there was no hard evidence of such effects or that the limits indicated a hazard.[39] He argued that if the current knowledge about radio frequency (RF) fields had been known at the time of the Stewart Report on mobile phones and health in 2000, it would have been difficult for ICNIRP and NRPB not to adopt limits so low as to stop the development of radio frequency telecommunication technology. He said that it was only the uncertainty surrounding clearly agreed thresholds that prevented there being "similar, inappropriate, extremely cautious limits set for radio frequencies".[40] Dr Keevil said that "we are in a grey area, where really there are not proven adverse health effects at these levels of these frequencies".[41] He makes the distinction between biological effects, such as magneto-phosphenes or peripheral nerve stimulation, which are well established, and adverse health effects, which are not, and interprets the Directive as seeking to avoid the possibility of any kind of effect. [42]

18. Some have pointed to the fact that MRI equipment has been in use now for over 20 years and there has been no evidence of any adverse health effects resulting from EMF exposure alone. According to evidence provided by the Medicines and Healthcare Products Regulatory Authority (MHRA), there have been two reported cases of physiological effects being experienced as a result of MRI exposure. (It is not clear how long these effects lasted.) The adverse incidents reported—some 144 since 1995—are primarily due to accidents or failings in the procedures that have resulted in contact burns, damage from projectiles or from internal medical devices. There are generally over five times as many radiology adverse incidents as there are MRI incidents each year, although this in part reflects the current availability of each type of equipment.[43] Commission officials confirmed that the incidents of adverse health effects were caused by accidents rather than exposure.[44] The ICNIRP guidelines are not designed to cover the prevention of such accidents: these are already covered by established stringent safety procedures in hospitals, which medical practitioners have a responsibility to enforce, as Mr Biosca de Sagastuy acknowledged.[45] Mr Biosca de Sagastuy explained that there was no evidence of adverse health effects due to exposure because the levels of exposure experienced by medical workers were lower than those contained in the Directive.[46] In response, Dr Keevil argued that "hundreds of millions of patients have been exposed to MRI over the past 25 years, at gradient field amplitudes up to 100 times the occupational exposure limit, with no evidence whatsoever of harm." He acknowledged that this exposure is to patients rather than workers, but contended that there is no reason to assume differences in the susceptibility to exposure between the two.[47]

19. The Directive refers to the "risk to the health and safety of workers due to known short-term adverse effects in the human body".[48] The MR community argues that these known short-term health effects of EMF are not necessarily adverse. It is asserted that the data provided by experiments and animal research have been used to make extrapolations to excessively high levels of exposure which can be considered safe. The joint submission observed that that the "leap from cautious, guarded statements in the ICNIRP guidelines to 'known adverse health effects' in the Directive would certainly not have survived objective scientific review."[49] Professor Blakemore also questioned the link between biological effects and adverse health effects: "It would be very unfortunate if MRI, with all its proven benefits, were to be curtailed, simply because thresholds for biological effects can be defined, but without clear evidence that such effects are hazardous."[50] The British Institute of Radiology states that "The proposed limits were based on hypothetical rather than established adverse effects on health, yet the effect would be an increase to both staff and patients exposure to the well-established hazards of ionising radiation from alternative, X-ray based imaging techniques."[51] Another witness engaged in operating clinical MRI equipment, Dr Calverd, agreed: "The supposed basis of the Directive's exposure limits to time-varying magnetic fields is an arbitrary multiplier applied to a reported threshold for some subtle, transient physiological effects."[52] The joint submission asserted that "the limits are not presented in the Directive as precautionary values, but as established thresholds for onset of adverse effects."[53] The Wellcome Trust complained that "The limits are absolute—there is no scope for time averaging, or for less restrictive limits for brief exposures."[54] In short, there was widespread support in the MR community for Dr Keevil's assertion that the premise on which the Directive is based—that there are known adverse health effects—is a false one.[55]

20. The Directive is based upon the provision of "protection against known adverse health effects". This description, as we have seen, can be taken as implying that these effects occur at the limits set out. This impression has been encouraged by some statements from the Commissioner with ultimate responsibility for the Directive, Commissioner Spidla, who said in November 2005 that "The Directive is designed to protect workers against excessive exposure to MRI and EMF which scientific experts agree is dangerous for health".[56] This interpretation was not the intention of the Commission and officials did not agree with this statement.[57] Rather, the limits are intended to be precautionary: the exposure levels are set so as to ensure that workers are protected from any possibility of adverse health effects, even though these are not necessarily proven. A precautionary approach, which we discuss in detail in chapter 4, is used when there is uncertainty in the scientific evidence currently available. There is also an assumption implied in this approach that steps will be taken to accumulate the necessary evidence and level of certainty to inform a review of decisions reached. The lack of available evidence of adverse health effects at present is not reason in itself to avoid taking preventative action, but it should require a convincing scientific case to be made in favour of statutory regulation, including a balancing of the risks of harm against the costs, pending the establishment of a fuller evidence base.

STATIC FIELDS

21. ICNIRP published guidelines on exposure to static electromagnetic fields in 1994. The Directive proposed in 2002 included limits for static fields which were based in part upon these guidelines. Representations made to the Commission during 2003, notably the COCIR submission, focussed upon the impact of the provisions relating to static fields. In September 2003 ICNIRP informed the Commission at an informal meeting that these guidelines were to be reviewed in the near future and that static fields should therefore not be included in the Directive.[58] The Commission accepted this advice, as did the European Council, in which a number of Member States, including the UK, had argued for their withdrawal. The provisions relating to static fields were withdrawn during negotiations in Council on 17 September 2003.[59] Dr McKinlay said that the advice from ICNIRP to the Commission was that static fields should be excluded on the grounds of an imminent review. The implication here is that the existing evidence base was insufficiently strong, or reliable, to be the basis of a Directive. Indeed, it was the lack of evidence, rather than an impending review, that was cited by Council as the reason for withdrawal.[60] We understand that this was an issue of some controversy: the removal of static fields was, to some, an unfortunate weakening of the protection to workers afforded by the Directive. A review of the science on static fields has now been carried out and ICNIRP is currently considering revised guidelines. The introduction of limits for static fields will be considered again in the 2009 review of the Directive.[61] We find it puzzling that static fields were included in the initial proposed Directive when the principal source of scientific advice for the Commission, ICNIRP, was about to review its own guidelines and advised against using existing guidelines as a basis for the Directive. This suggests that communication between the two organisations was not as effective as it could have been, but it does demonstrate that the legislative process was responsive to new scientific advice.

TIME-VARYING FIELDS

22. The exposure limits in the Directive for time-varying fields were based upon more recent guidance. The 1998 ICNIRP guidelines provide a comprehensive review of available evidence on the health effects of time-varying fields which is used to inform the limits set for public and occupational exposure. They summarise the evidence of biological effects and the potential for adverse health effects from exposure to EMFs at a range of frequencies up to 300 GHz. In line with other safety guidelines, ICNIRP takes a measurable exposure level which is known to cause an effect, and divides this level by a factor in order to provide an exposure limit which is safe. The argument is over whether, as the International Electrotechnical Commission and others maintain, this factor is too great.

23. We have referred above to the questions raised about the science underpinning these 1998 guidelines. In setting out the evidence base, the 1998 guidelines do acknowledge some degree of uncertainty:

"In establishing exposure limits, the Commission [ICNIRP] recognises the need to reconcile a number of differing expert opinions. The validity of scientific reports has to be considered, and extrapolations from animal experiments to effects on humans have to be made. The restrictions in these guidelines were based on scientific data alone; currently available knowledge, however, indicates that these restrictions provide an adequate level of protection from exposure to time-varying EMF."[62]

The guidelines also state that:

"There is insufficient information on the biological and health effects of EMF exposure of human populations and experimental animals to provide a rigorous basis for establishing safety factors over the whole frequency range and for all frequency modulations. In addition, some of the uncertainty regarding the appropriate safety factor derives from a lack of knowledge regarding the appropriate dosimetry."[63]

The guidelines make clear that the limits will be periodically reviewed in the light of further advances in identifying adverse health effects. In 2004, ICNIRP acknowledged that its 1994 and 1998 guidelines were "written many years ago, and they are now under review."[64]

24. We asked what advice ICNIRP gave to the Commission on the certainty of the scientific basis for the 1998 guidelines. In response, Dr McKinlay referred to uncertainties surrounding the interpretation of scientific data and the selection of "safety factors", which were discussed at numerous seminars and conferences. He also referred to the caveats contained in the guidelines (such as those quoted above) but did not indicate that any specific advice was given to the Commission regarding levels of certainty.[65] The reason that ICNIRP made no recommendation for withdrawal in respect of time-varying fields is that the revision of variable fields "seemed rather a long way off".[66] Dr McKinlay was reluctant to accept any role in advising the Commission on the suitability of available evidence for a Directive on time-varying fields on the grounds that this was a matter for the regulators. This contrasts with the position on static fields, when ICNIRP did advise on the reliability of the available evidence. It is, of course, for the Commission to take decisions upon what to include in a Directive, but, in making this decision, it relies upon the advice of ICNIRP to assess the level of certainty in the evidence base. ICNIRP is well placed to advise the Commission on the strength of the evidence base, rather than just the date of the next review. Having advised on the exclusion of static fields from the Directive, it would be inconsistent and slightly disingenuous of ICNIRP to evade all responsibility for advising the Commission on the strength of the evidence base regarding time-varying fields.

INTERNATIONAL STANDARDS

25. At an international level, different standards are in operation, as set out in Box 2. The IEEE sets standards for EMF exposure. Mr Biosca de Sagastuy argued that ICNIRP and the IEEE use different models of the human body to calculate maximum exposure limits but that "they follow the same basic restrictions as ICNIRP" and "there is very little difference" between the two.[67] Dr Keevil argued that, in the frequency range relevant to the MR community, the basic restrictions are expressed in different ways and that the limits differ "by a factor of almost 20".[68] Furthermore, manufacturers and users of MRI equipment follow the different International Electrochemical Commission standard (IEC 601-2-33), which includes specific requirements for the safety of magnetic resonance equipment for medical diagnosis. This standard is currently being amended to include occupational exposure to EMFs in the frequency ranges relevant to MRI. We understand that the occupational exposure limits adopted by IEC will be somewhat different to those of ICNIRP.

CONCLUSIONS ON EVIDENCE BASE

26. We are not in a position to evaluate the validity of the evidence base on which the 1998 ICNIRP guidelines are based, to compare them with other international guidelines, nor to assess whether the limits are excessively cautious. However, we have found that significant uncertainties around the scientific basis of the guidelines exist. Indeed, the ICNIRP guidelines themselves describe a number of studies which offer conflicting evidence or are inconclusive. There is undoubtedly a large and growing evidence base on EMF exposure and the MR community agrees that the NRPB literature review carried out in 2004 is widely regarded as a definitive summary of the state of the science.[69] It should also be observed that ICNIRP and NRPB/HPA have highlighted uncertainties in the evidence base, and the need for further research, particularly on any long term effects of exposure to static fields.[70] The ICNIRP guidelines also acknowledge a degree of uncertainty that was not relayed to us by the Commission. It may be that officials were concerned that they gave a misleading impression: the witnesses from the Commission submitted a note of clarification following our visit which acknowledges "differences in some details" amongst various published international assessments whilst asserting that, in respect of established health effects, there was no scientific evidence to challenge the underlying concepts adopted by ICNIRP and the Directive.[71] We welcome the fact that the scientific advice on which the Directive is based is all published: this transparency has assisted debate. However, officials we met at the Commission misrepresented the level of certainty in the scientific evidence underpinning the Directive. This approach was unhelpful, and can only undermine confidence in the way in which scientific evidence was used by the Commission to support the Directive.

Impact of the Directive

27. There is also considerable debate over the impact of the Directive on the usage of MRI for medical research and for diagnostic purposes. The views we heard in the Commission on this were diametrically opposed to those of the MR community. The uncertainty is caused in part by the difficulty in measuring the extent to which current usage of MRI actually exceeds the prescribed exposure limits. The Directive prescribes action limits, which when exceeded require monitoring to be carried out to check compliance with exposure limits. This requires detailed calculation in each individual case. It can be difficult to measure what values have been exceeded in MRI as an MRI scanner involves the combination of three different EMF, as set out in Box 1. The exposure of a worker will depend on many factors including: the design of the MRI equipment; the strength and frequency of all the EMF fields used in the system; the precise location of the worker relation to the EMF fields; the speed of motion of the worker; and the sequences that the scanner is running (scanners have different sequences for different medical applications which switch the time-varying fields on and off at different rates).

IMPACT AND RISK ASSESSMENTS

28. The other main cause of uncertainty stems from the failure of impact assessments, at both EU and UK levels, to identify the full range of sectors that might be affected and to examine in detail the extent of this impact.

EU Commission

29. The Commission published a risk assessment in 1993 when the original Directive was first proposed. This did not identify any implications for MRI. When the new proposed Directive came before Council some ten years later the UK asked for a new risk assessment to be carried out. At the time, this was not mandatory. (Since 2004 there is a commitment for all major policy defining documents and legislative proposals to be accompanied by impact assessments.[72]) This 1993 assessment was not thought in all quarters to be of great quality. It was criticised by a manufacturers' representative body as "insubstantial and does not bear close examination nor take into consideration [existing] product related legislation …".[73] The request for a new assessment was rejected by a majority at the European Council: Member States believed that existing risk assessments could be extrapolated easily to reassess costs and that national assessments had already been carried out to evaluate the impact.[74] Mr Jansen from the Commission argued that it was impractical to have another impact assessment, which takes one year to complete, each time amendments were made while the proposals were before Council and the European Parliament.[75] This misses the point. It was not a case of looking at the effects of minor amendments to the original proposals: the Directive was firmly based on guidelines that were published in 1998, some five years after the original proposals on which the initial assessment was made. It was essentially a different piece of legislation. In the ten years since 1993 the technologies supporting medical imaging, as in many other areas, progressed significantly. Machines became more powerful and the potential medical benefits as well as potential negative health effects rose accordingly. Commission officials agreed, but said that the need for a new assessment was "only to show that there are even bigger risks than was originally thought."[76] Of course the potential for harm might increase with ever more powerful machines, but we find this response revealing and indicative of a mindset in the Commission that could not envisage any adverse consequences of the Directive. We were alarmed to discover that the European Council was prepared to rely on a ten year old risk assessment to inform legislation in an area of rapidly developing science and technology. We welcome the moves taken to ensure that new proposals are accompanied by new impact assessments, as long as these are taken to include revived Directives such as this one.

The UK

30. In the UK, the HSE commissioned the NRPB to produce a report on the potential impact of the 1993 proposals for a Physical Agents (EMF) Directive. In response to an anticipated revival of the proposal for a single Directive on EMF the HSE then commissioned a further review of the published evidence by NRPB in 2001. This report concluded that "many of the exposure measurements that are reported complied with the relevant reference levels, however a number of devices and applications have been identified where the reference levels or basic restrictions may be approached or exceeded".[77] However, MRI equipment was not listed among the sources of EMF in question. The report called for further work in some areas.

31. The Regulatory Impact Assessment (RIA) on the Directive produced by HSE in November 2003 stated that its only knowledge of EMF over-exposure was as a result of very infrequent accidents or incidents and that the effect of the Directive on such incidents was likely to be minimal. It noted that the RIA was "unable to identify any health and safety benefits from the Directive".[78] This in itself is a fairly damning assessment of the case made for the Directive as a whole. The RIA estimated that around 250 pieces of equipment and 1250 workers were potentially affected by the Directive.[79] In spite of the fact that the RIA estimates that between 200 and 500 organisations concerned with MRI equipment would be affected, no representative organisation of MRI equipment manufacturers, the medical or research communities is listed among those consulted.[80] The focus was primarily on the cost implications for major industries.

32. The RIA carried out by the HSE, having identified that MRI equipment would be affected by the Directive, failed to explore any further, in spite of the fact that concerns about its impact were being raised with HSE in the four months prior to the publication of the report (see paragraphs 56-60). The Chief Executive of the HSE, Mr Podger (who was not in post at the time), acknowledged its failings: "The truth is that that regulatory impact assessment was done very quickly because, as you know, the proposal only suddenly appeared out of the blue in September 2002".[81] We do not accept that the proposal "came out of the blue": it had been known that it was forthcoming since 1999 when the original all-encompassing proposed Directive was divided into four parts, one of them being EMF. We have noted that the HSE itself commissioned the NRPB in 2001 to undertake work on the exposure limits covered by the Directive. The HSE also was aware of the Directive in 2001 through its membership of the Interdepartmental Liaison Group on non-ionising radiation.[82] We conclude that the HSE did not apply the necessary expertise to its assessment of the impact of the Directive. We recommend that the Health and Safety Executive ensures that regulatory impact assessments on EU proposals are conducted in a comprehensive manner, on a sector by sector basis, with care being taken to address the broader impact, rather than just the costs, of the legislation.

VIEWS OF THE MR COMMUNITY

33. The views of the medical community on the impact of the Directive on MRI can be summarised as follows:

• "It will be difficult to monitor patients requiring close supervision during imaging - e.g. anaesthetised or sedated children, very sick patients and uncooperative psychiatric patients - since staff will not be able to stand close to the scanner. (It is suggested that the limits will be exceeded if a worker stands within 1-2 metres of the bore during imaging.[83])
• Movement of staff near the scanner may be restricted even when it is not operating. The static magnetic field is present at all times, and movement through it will expose staff to a time-varying field that may breach the relevant limit. This will also affect testing of magnets during manufacture and maintenance of installed systems.
• Most interventional MR procedures will become illegal, as clinicians will not be permitted to stand close enough to the scanner to perform them.
• Some functional MRI studies will become impossible—e.g. studies on deaf-blind subjects, where staff 'sign' into the palm of the patient during imaging;
• There is likely to be an increase in X-ray and CT imaging in place of MRI, resulting in increased radiation risk to staff and patients, where the risks are known."[84]

Dr Keevil told us that "It is true to say that the vast majority of clinical diagnostic MR imaging would not be directly affected" but that there were whole new areas, such as interventional MR, which would be "effectively blocked by this".[85] He states that there are a growing number of cases (some 40,000 in the UK each year) in which staff are required to remain in the vicinity of the scanner during imaging. This is typically necessary for some children and particularly anxious or seriously ill patients.[86]

34. Manufacturers also believe that there will be an adverse impact of the Directive. Siemens state that the limits contained in the Directive are "in conflict with MR practice and equipment design".[87] The European medical equipment manufacturers' Committee, COCIR, has said that the limits contained in the Directive "severely hamper the normal installation, use and maintenance of MRI equipment."[88] At a European Congress of Radiology in March 2006 Mr Hans Engels, Head of Safety at Philips, said that the new limits would hamper several specific situations in the hospital and interventional MR. It would also affect the manufacturing process of the MR community.[89]

VIEWS FROM THE COMMISSION

35. At the Commission, officials giving evidence rejected any idea that there could be the impact described above. Mr Biosca de Sagastuy told us repeatedly that in their view, and that of the scientific experts, the Directive would have no impact on the continued use of MRI in hospitals.[90] He told us that the machines used for surgery were very low powered and that for invasive procedures, MRI machines were only used "for a very limited amount of time, a maximum of five minutes and no more."[91] Mr Biosca de Sagastuy assured us that magnetic resonance was discussed at length in Council, with experts present, and that no problems for health personnel were foreseen.[92] Officials had been satisfied in visits to hospitals that there would be no impact on current use of MRI equipment.[93]

36. Officials did acknowledge a potential impact on maintenance work and research. Mr Biosca de Sagastuy said that the Directive "could have an impact on the maintenance procedures, yes", for example in the testing by technicians of equipment.[94] There could also be consequences for the use of some of the newer, more powerful machines being developed for research, but this would require further investigation.[95] Professor Blakemore did not dispute that there may be an impact on clinical practice but thought that research was likely to be most directly affected. He also observed rightly that, in time, research has a tendency to translate into clinical practice.[96] This was not a point that was acknowledged by officials from the Commission, who agreed with the suggestion that the impact of the Directive had been greatly exaggerated by the MR community.[97]

37. The views we heard in Brussels were evidently new to many in the MR community. Dr Keevil expressed astonishment at the evidence we heard and told us that he found it "amazing those individuals could say that there is no impact".[98] He informed us that there were two 1.5 T interventional MR systems in use in the UK—far more powerful than the 0.4 T machines that Mr Biosca de Sagastuy referred to, and that his measurements suggested that his team were over the relevant action value in the Directive "by a factor of about 40".[99] He argued that in interventional procedures there is continuous use of machines for "up to around 20 minutes" rather than the five described by Mr Biosca de Sagastuy.[100] He reports that movement of staff through a static field exposes them to a slowly time-varying field which induces currents that "almost certainly" exceed the limits at 3 T and "quite possibly" also at 1.5 T.[101] We can only express alarm that, two years after the adoption of the Directive, officials responsible for the detailed work on it have an understanding of the use of MR equipment that is so far removed from that of the practitioners themselves.

CONCLUSIONS ON IMPACT

38. In reaching firm views on the impact of the Directive, the Commission relied on the views of the scientific experts, primarily ICNIRP. As we have seen, ICNIRP has no responsibility to advise on the impact of their guidelines on medical practice and its advice on MR procedures has been limited as far as occupational exposure is concerned. Part of the reason why it would have been difficult for any impact assessment to make authoritative judgments about the effects of the Directive is the lack of published research specifically directed at this issue. Dr Keevil told us that "the MR community believes that there is substantial evidence that exposure in MRI exceeds the limits".[102] He submitted to us a paper summarising this evidence which is awaiting peer review and publication.[103] He also referred to current research being undertaken at Royal Marsden Hospital on 0.5 T scanners. However, he could not point to a body of peer reviewed, published research confirming the views of the MR community.[104]

39. It is difficult to reconcile the substantial differences of opinion on the impact of the Directive between the Commission officials we took evidence from and medical practitioners. There is general agreement that there will be an impact on maintenance procedures and on future research involving more powerful machines, although the extent of this impact remains uncertain. There are strong suggestions from MR practitioners to suggest that the limits established in the Directive will affect the conduct of existing MRI procedures. This evidence is not strong at present: the necessary research has not been conducted to provide an authoritative view. This in itself is not surprising, as this type of research could only be expected to be carried out at the request of regulators.

40. In the light of this paucity of evidence, we were surprised by the sometimes dismissive attitude we found in Brussels towards the views of the medical practitioners. The need for more evidence on the potential impact has now been acknowledged by the Commission, as we record in chapter 7. However, the fact that there is such uncertainty over the Directive's impact some two years after its adoption of the Directive reflects poorly on the influence of scientific advice in the policy making process in Brussels. This uncertainty was caused partly by the failure of the Commission to conduct a proper, up-to-date impact assessment, and partly by the Commission's reluctance to take seriously and investigate the concerns of the MR community when they were raised. Alternative views were ignored or dismissed rather than investigated and confronted by further evidence. As a result, the research necessary to determine the impact of the Directive on MR use is only now beginning to be undertaken. It is deeply regrettable that the impact of the Directive on MRI procedures was not established before the Directive was adopted. This case study illustrates the potential consequences of the failure of policy makers to seek comprehensive scientific advice early in the policy formulation process and to commission the necessary research to inform this process where uncertainty or gaps in knowledge exist.

Justification for a Directive

41. The rationale for the Directive was to provide uniform health and safety standards for workers across Europe which could also be used to inform manufacturers and promote fair competition. Some Member States argued that existing guidelines were sufficient and that a Directive would increase administrative burdens on employers. For the MR community, the disadvantage of enshrining exposure limits in a Directive are that the limits are absolute and to be enforced inflexibly, regardless of the circumstances. The question as to whether the advantages outweigh the disadvantages is ultimately a political judgment, but it is one that depends to a certain extent on scientific advice.

42. ICNIRP guidelines are already widely accepted and followed. It is up to individual countries to use them as a basis for national guidance and to enforce them. Similarly, there are internationally agreed IEC guidelines for equipment manufacturers. As far as MR is concerned, we have seen no evidence that the absence of a Directive was either distorting the competitive environment or leading to adverse health effects. The UK Government argued against a Directive from the outset. The Minister, Lord Hunt, confirmed the view of the HSE: "we felt that there was no need for the directive because we already had these guidelines".[105] While internationally respected guidelines can be, and are, reviewed and updated regularly in line with a growing evidence base, an EU Directive is a far less flexible tool. We share the view of the Stewart Report on mobile phones and health of 2000, which stated that: "We are not convinced of the need to incorporate ICNIRP guidelines into statutes. We believe that they are liable to change as more scientific information on possible health effects becomes available."[106] On the impact of EMF exposure on health, the science is in places uncertain and is being updated all the time. The reasonable demands of health and safety officers for precise, measurable limits based on scientific certainty do not sit easily with the normal scientific discipline of learning by constant evidence gathering and review. Guidelines that can be readily updated when necessary are a useful tool for uniting these two strands. We acknowledge the desire of the Commission to put EMF exposure limits on an equal footing throughout the EU and we have not examined the full scope of the Directive and its impact. However, for MRI at least, we do not believe that there was a strong enough case for enshrining exposure limits in a Directive. We agree with the Government that existing guidelines are sufficient. The Directive will, at best, impose burdens on employers and, at worst, inhibit the use of valuable diagnostic procedures and important research.

16   Q 886 Back

17   Q 712 Back

18   European Parliament, Procedure file, COD/1992/0449C, www.europarl.eu.int/oeil/file.jsp?id=215622 Back

19   Dr McKinlay was also Vice-Chair from 1996-2000; ev 77 Back

20   Q 889 Back

21   Q 887 Back

22   ICNIRP, The Global Focus for Non-Ionising Radiation Protection, www.icnirp.org Back

23   Ev 79 Back

24   ICNIRP, Medical MR Procedures: Protection of Patients, Volunteers and Staff, April 2003 [not published] Back

25   ICNIRP, Medical Magnetic Resonance (MR) Procedures: Protection of Patients, 2004, Health Physics 87, p 197 Back

26   The co-decision procedure requires the European Council and European Parliament to agree on legislation.  Back

27   COCIR is the European Co-ordination Committee of the Radiological Electromedical and Medical IT Industries. Back

28   COCIR, Letter to Social Questions Working Party, 11 April 2003 Back

29   Qq 866  Back

30   Q 801 Back

31   Qq 678, 744-45 Back

32   Ev 66 Back

33   Ev 34 Back

34   Ev 42 Back

35   Hill DLG, Keevil SF, Impact of electromagnetic field exposure limits in Europe: is the future of interventional MRI safe? Acad Radiol (2005)12: 1135-1142 Back

36   EMF Workshop, 5 January 2006, Note by Professor Derek Hill and Professor Jo Hajnal Back

37   Galson Sciences Ltd, EU EMF Physical Agents Directive EC/40/2004 Implementation into UK National Legislation, Report of Roundtable Discussions, 5 January 2006, p 4 Back

38   Q 833 Back

39   Q 840, ev 59 Back

40   Q 840 Back

41   Q 819 Back

42   Ev 66, Q 825 Back

43   Ev 61 Back

44   Qq 727-32 Back

45   Q 690 Back

46   Q 733  Back

47   Ev 65 Back

48   Directive 2004/40/EC of the European Parliament and of the Council, Official Journal of the European Union L 159 of 30 April 2004, Article 1, para 2  Back

49   Ev 44 Back

50   Ev 60  Back

51   Ev 40 Back

52   Ev 38 Back

53   Ev 44 Back

54   Ev 52 Back

55   Q 819 Back

56   Press notice, 18 November 2005, www.esmrmb.org/hauptframe.php?pid=409 Back

57   Q 753 Back

58   Q 888, ev 78 Back

59   Ev 70 Back

60   European Parliament, Procedure file, COD/1992/0449C, www.europarl.eu.int/oeil/file.jsp?id=215622 Back

61   As above Back

62   ICNIRP Guidelines, Health Physics, 74, No. 4, 1998, pp 494-5 Back

63   As above, p 509 Back

64   ICNIRP (2004), Health Physics 87, p 197 Back

65   Ev 77 Back

66   Q 893 Back

67   Q 804 Back

68   Ev 67 Back

69   Ev 44 Back

70   eg NRPB-W24, September 2002 & NRPB-W59, July 2004, ICNIRP (2004), Health Physics 87, p 211 Back

71   Q 715, see footnote Back

72   See www.ec.europa.eu/governance/impact/docs/progress_report_council_sec_1377.pdf Back

73   WEM/ORGALIME Position Paper, Draft Directive for the protection of workers from exposure to electro-magnetic fields and waves (EMF), 20 March 2003, p 2 Back

74   Q 696 Back

75   Q 701 Back

76   Q 704 Back

77   NRPB-W24, September 2002 Back

78   Health and Safety Executive, Proposal for a Physical Agents (Electromagnetic Fields) Directive, Regulatory Impact Assessment, November 2003, para 16 Back

79   As above, para 20  Back

80   As above, para 12 Back

81   Q 879 Back

82   This is discussed further in paragraph 71. Back

83   Note on EMF Workshop, 5 January 2006, Professor Derek Hill & Professor Jo Hajnal Back

84   Ev 44 Back

85   Q 808 Back

86   Dr S F Keevil, Impact of the Physical Agents (EMF) Directive on medical magnetic resonance imaging, 2006, [not published]. Back

87   Ev 63 Back

88   COCIR position paper on the Council Common Position on the EMF Directive, 2003 Back

89   Presentation given at ECR 2006, Session on Safety, Considerations in MR (European Radiology Supplements A-286) Back

90   Qq 691, 717, 739 Back

91   Q 691 Back

92   Q 715 Back

93   Q 779 Back

94   Qq 736-37 Back

95   Q 692 Back

96   Q 813 Back

97   Q 693 Back

98   Q 806 Back

99   Qq 808, 810, ev 68 Back

100   Ev 69 Back

101   Ev 65, 69 Back

102   Ev 68 Back

103   Dr S F Keevil, Impact of the Physical Agents (EMF) Directive on Medical Magnetic Resonance Imaging, 2006, [not published].  Back

104   Q 812 Back

105   Q 902 Back

106   Independent Expert Group on Mobile Phones, Mobile Phones and Health, May 2000 [The Stewart Report], para 6.36 Back