This Article Abstract Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Clewer, P R Articles by Jackson, P C Search for Related Content PubMed PubMed Citation Articles by Clewer, P R Articles by Jackson, P C

Reporting overexposures and unintended exposures in diagnostic procedures

Department of Medical Physics and Bioengineering, Southampton General Hospital, Southampton, UK

Correspondence: Mr Philip Clewer, Medical Physics and Bioengineering, Southampton General Hospital, Tremona Road, Southampton SO16 6YD, UK. E-mail: philip.clewer{at}suht.swest.nhs.uk.

	Abstract

Top Abstract Introduction Overexposures and unintended... The future Why report? Conclusion References

 
Legislation requires the reporting to regulatory authorities of incidents in which patients have been exposed to ionizing radiation to an extent "much greater than that intended". The authorities have published guidance on what is considered to meet this requirement. However, there is still some confusion regarding, particularly, the necessity to report some unintended doses. It is believed that there is a disproportionate amount of resource spent investigating some unintended exposures because all such exposures will have an effective overexposure factor of infinity, irrespective of the magnitude of the dose and the associated risk. This paper proposes changing the definitions of "overexposure" and "unintended exposure" and the adoption of a reporting process based upon risk assessment. All records and data would be collected and, if required, reported, but investigation of individual incidents would take place only for incidents carrying a greater risk than 1 in 10 000.

	Introduction

Top Abstract Introduction Overexposures and unintended... The future Why report? Conclusion References

 
For many years now, the medical radiation protection community has lived with the legal requirement that when it is suspected that a patient has been exposed to ionizing radiation to an extent "much greater than that intended" due to equipment malfunction or defect, the employer must notify the Health and Safety Executive (HSE) forthwith. The requirement was stated in the Ionising Radiation Regulations 1985 (regulation 33(2)) and in the replacement Ionising Radiations Regulations 1999 (regulation 32(6)). The HSE published guidance [1] on what should be considered to be "much greater than intended". The HSE's document included a table of "guideline multiplying factors" and stated that if the ratio of the suspected exposure to the intended exposure is greater than or equal to the appropriate multiplying factor, it should be presumed that HSE expects to be notified of the incident. The multiplying factors for diagnostic examinations were 3, 10 or 20, depending on the magnitude of the intended effective dose. For instance, an incident where a patient undergoing a chest X-ray receives an exposure of 20 times or more than intended, the HSE should be notified. For any CT examination, the "reporting threshold" was set at three times the intended exposure. The factors also applied to situations where equivalent multiples of patients received doses greater than intended since such situations might arise with equipment malfunctions.

In 2000 the introduction of the Ionising Radiation (Medical Exposure) Regulations 2000, known as IR(ME)R, brought a similar requirement for notification of incidents when a patient is exposed to ionizing radiation to an extent "much greater than intended", but due to any reason other than equipment malfunction or defect (regulation 4(5)). In this case, the notification is to the Department of Health (DoH). The guidance [2] issued with IR(ME)R stated that application of the HSE guidance on which doses are likely to be much greater than intended is appropriate. Therefore, for example, procedural failures that lead to a patient receiving 20 times the intended exposure from a chest X-ray should be reported to the DoH.

Paragraph 6.8.2 of the same guidance stated that "patients who undergo a procedure that was not intended ... should be considered as having received an unintended dose of radiation". The introduction of the phrase "unintended dose of radiation" has caused some confusion because it appears to imply that unintended doses are separate to events described as "much greater than intended". This has raised some anomalies that lead to the question, "what is the purpose of reporting these incidents?" – a question to be addressed later in this paper.

It appears that the DoH intended to publish guidelines that would be aimed specifically at incidents reportable under IR(ME)R rather than reportable under IRR99. These guidelines have not been forthcoming, but are still expected [private communication with DoH, 2005 and 2006].

For the purposes of this paper the authors will take the intended effective dose to be the "typical effective dose" published in 1999 by the National Radiological Protection Board (now the Radiation Protection Division of the Health Protection Agency) [3].

	Overexposures and unintended exposures

Top Abstract Introduction Overexposures and unintended... The future Why report? Conclusion References

 
If a patient is referred for a plain film chest X-ray, then the intended dose is 0.02 mSv.

The guideline reporting threshold for an adverse event for such an examination is 20 times the intended dose. Therefore if the patient received 19 times the intended dose there is no requirement to report the incident, although the Employer may wish to do so. This means that the patient could receive an effective dose of 0.36 mSv over and above the intended 0.02 mSv, giving a risk of a fatal cancer of around 1 in 55 000, using the 5% per Sv risk factor model [4].

If a patient who is not intended to undergo any X-ray examination actually undergoes a chest X-ray for some reason, e.g. misidentification, the effective dose to that patient is 0.02 mSv with a risk of around 1 in 1 000 000 of a fatal cancer. Although the overexposure factor could be said to be infinity, the actual risk to the individual patient is much lower than the 19 times-the-intended-dose incident, but this incident is theoretically reportable because it was an "unintended exposure".

From a patient perspective this may seem confusing in that the error with a lower risk of harm is reportable, whereas the error with a relatively higher risk of harm is not subject to that requirement.

It is clear that an error is reportable if the overexposure factor is above the relevant threshold for reporting, e.g. someone has a pelvic radiograph (effective dose 0.7 mSv) when a chest radiograph (effective dose 0.02 mSv) was intended. In this example the overexposure is 0.72/0.02 ( = 36) (assuming they go on to have the intended examination), i.e. greater than 20 and therefore reportable. However, if the sequence of events was the other way around the overexposure factor would be 0.72/0.7 ( = 1.03). This would not be over the reporting threshold as far as an overexposure is concerned, but the chest radiograph would have been unintended and so the incident is reportable for this reason rather than overexposure.

For instance, if a patient undergoes an incorrect examination, in practice this will lead to a "greater than intended" dose rather than an unintended dose since both the "incorrect" and "correct" examinations will have been performed. In this case, reporting to the DoH would be dependent on the magnitude of the overexposure.

The DoH currently regards any case of wrong examination, such as "left ankle" instead of "right ankle", to be an "unintended exposure" and therefore reportable, regardless of magnitude. This means that hospitals have to report unintended doses even if they result in a dose and risk much lower than the dose and risk to patients who experience overexposures that do not exceed the reporting threshold. It appears that some hospitals are unclear as to the reporting requirements – some do report the "wrong ankle" examination whereas others do not.

	The future

Top Abstract Introduction Overexposures and unintended... The future Why report? Conclusion References

 
In 2003, the HSE published proposals for a document, HS(G)226, to replace PM77. One of the significant proposals was that the guideline multiplying factor for all diagnostic procedures should be reduced to 1.5. Therefore all occasions when a repeat X-ray exposure is required due to equipment defect or malfunction would require reporting to HSE. This proposal attracted considerable comment and when the final document was published in March 2006 [5], the guideline multiplying factors for diagnostic radiology and nuclear medicine were divided into three bands as follows:

• 1.5 times intended dose – interventional radiology, radiographic and fluoroscopic procedures involving contrast agents, nuclear medicine with intended effective dose > 5 mSv and CT examinations
  
• 10 times intended dose – mammography, nuclear medicine with intended effective dose 5 mSv but > 0.5 mSv, everything else not referred to elsewhere
  
• 20 times intended dose – radiography of extremities, skull, dentition, shoulder, chest, elbow, knee and nuclear medicine with intended effective dose 0.5 mSv.
  

Hence the factors for middle and low dose examinations will remain unchanged, as do those for therapeutic procedures, but the higher dose diagnostic procedures have been given a guideline multiplying factor of 1.5. Therefore any repeat exposures in this category are reportable. The DoH has indicated that, certainly for the time being, they will not adopt the 1.5 times factor, but will continue with the 3 times factor for high dose diagnostic procedures [private communication with Department of Health, 2006].

The above arrangements are appropriate to situations where there has been equipment failure, but should the same approach be applied to the individual patient?

	Why report?

Top Abstract Introduction Overexposures and unintended... The future Why report? Conclusion References

 
Perhaps it would be useful to return to the question "why do we report incidents?".

IR(ME)R implement most of the requirements of the Medical Exposures Directive [6], but the Directive does not require member states to implement a reporting system. However, it does require in Article 11 that member states should ensure that "all reasonable steps to reduce the probability and the magnitude of accidental or unintended doses of patients from radiological practices are taken...". Perhaps this is in part a solution to the confusion; incident reporting can lead to prevention of similar occurrences by building up statistical evidence of where problems lie and where attention should be focused.

The Directive also states that "the main emphasis in accident prevention should be on the equipment and procedures in radiotherapy, but some attention should be paid to accidents with diagnostic equipment". This implies a need to concentrate on preventing radiotherapy incidents because of their relatively high doses, and therefore risks, compared with diagnostic procedures.

In practice there are two reasons for reporting incidents, be they overexposures or unintended exposures.

The first of these, as mentioned above, is to produce a national database of statistics on incidents, showing where the problems lie and which areas of practice require more attention to reduce the likelihood or magnitude of incidents in future. However, it is believed that this could be done more efficiently by employers being required to supply their own statistical information regularly, perhaps annually.

The second reason for reporting incidents is that significant errors can be fully investigated by an independent body and appropriate actions considered in relation to employees and/or employers. These may lead to an investigation by the HSE or DoH, perhaps leading to an improvement notice or prosecution. Such incidents obviously require an individual notification to the appropriate regulatory body.

The following is proposed as a way of combining ease of reporting for statistical purposes and the need for "enforcement" reporting:

• Requirements for reporting should be redefined in terms of effective dose. Exposure is too vague a term for use in this context. Although there may be some controversy regarding the relationship between risk and effective dose, this would at least take into account the tissues and organs of the body that are irradiated. In fact, the PM77 does use effective dose in its table of guideline multiplying factors while the regulations refer to overexposures.
  
• "Overexposure" should be redefined as "overdose of radiation" and related to the occasion of a patient who it is intended should receive a dose receiving more than intended. By using the effective dose quantity, an overdose would include incidents where the patient initially received an examination of the wrong part of their body and then the correct examination. It would be intended that they receive an effective dose, but because they receive an additional dose they have been given a greater dose than intended. Therefore a patient who attends for a radiograph of the pelvis but receives a chest radiograph in error, followed by the intended pelvis radiograph, or vice versa, is the subject of an overdose of radiation, whichever way round the events occur – although the magnitude of the overexposure factor is very different.
  
• The system of published guideline multiplying factors for reporting of overdoses of radiation could be retained, but the authors suggest that an alternative system is adopted for reporting unintended doses as outlined below.
  
• "Unintended exposure" should be redefined as "unintended dose" and related to the occasion of a patient who it is intended should not receive a radiation dose receiving a dose. This would clearly identify those patients who should not have received any dose of radiation but who have for some reason, usually mistaken identity, done so.
  
• A threshold should be set, in terms of effective dose above which unintended doses should be reported to the regulatory authority individually and urgently. Below the threshold the incident should be logged by the employer and information sent to the regulatory authority at regular intervals as nationally agreed. Setting a threshold would enable appropriate effort to be directed towards investigating and preventing re-occurrence of the more serious incidents in a timely manner. Of course, incidents involving doses below the threshold could still be reported individually if the employer believes there is merit in doing so. It is accepted that the public may view the setting of a threshold as condoning error and employers not being held to account to a higher authority. However, this situation might be addressed by the Healthcare Commission having access to the adverse event reporting of radiation events.
  

If this philosophy is accepted, the next consideration is the setting of the reporting threshold for unintended doses and reviewing the threshold for reporting overdoses. Many values could be proposed, but before doing so it may be instructive to look at risks associated with other areas of life. In "Living with Radiation" [7], the National Radiological Protection Board published comparative risk data for the annual risk of death in the UK from some "common" causes. This included a 1 in 100 000 annual risk of being murdered, a risk of any 40-year-old individual dying from any cause as 1 in 700 and the annual average risk of death from cancer being 1 in 400.

If the guideline multiplying factor of 20 for low dose examinations, which includes intended effective doses up to 0.5 mSv as described in PM77 [1], were to be retained then the threshold for reporting unintended doses could be set at 20x0.5 mSv = 10 mSv. However, this does seem relatively rather a high dose for reporting individual incidents of this nature and is above the dose level of a CT chest examination. Another approach would be to set the threshold at 20 times a chest X-ray dose. This would set the threshold at 0.4 mSv and require individual reports of incidents involving many more types of diagnostic examination. This type of approach is subject to change in the effective dose of examinations over time and subjective assessment of the acceptability of a threshold. Alternatively, we could consider who is actually receiving this dose of radiation. A hospital patient receiving an unintended dose should not have undergone a medical exposure and is no different to any other member of the public. In 1993, the National Radiological Protection Board published guidance [8] relating to the latest international recommendations and included a recommendation for a public dose constraint for a single new source [of ionizing radiation] of 0.3 mSv per year. This dose, equivalent to a lateral lumber spine X-ray examination [4] and with a risk of a fatal cancer of 1 in 67 000 using the 5% per Sievert model, could be suggested as an appropriate reporting threshold for incidents where someone who should not have received a dose at all has been irradiated. This approach is perhaps less than acceptable in that it appears that a patient should not accept any greater risk of harm than a member of the public. This approach is not acceptable as, de facto, all forms of medical intervention have some risk. Apart from the unavoidable risk associated with the intended intervention, there is also a risk of an adverse event involving the patient but unrelated to the patient's condition.

The HSE's approach to risk and benefit in the nuclear power industry may give some guidance. The HSE has looked at the boundary between a tolerable risk (of death) and an unacceptable risk to both workers and the public from the use of a nuclear power station. For the public who have the risk imposed on them "in the wider interest of society", the boundary is said to be 1 in 10 000 per annum [9]. It would be difficult to maintain that the use of X-rays for medical diagnosis is not "in the wider interests of society" and so there will always be a risk of unintended exposures. Using the 5% per Sievert model, a risk of 1 in 10 000 is equivalent to an effective dose of 2 mSv, similar to that from a CT scan of the head and not very different from the average annual "background" dose to a citizen of the UK. This is still a reasonably high level and is likely to mean that no plain film unintended doses would require reporting. However, putting this into context, a lifetime risk of 1 in 10 000 of a fatal cancer is still very small in comparison with the natural risk of a fatal cancer of about 1 in 4 [10].

Table 1 summarizes the possible unintended dose reporting thresholds and reasons for their adoption.

	Conclusion

Top Abstract Introduction Overexposures and unintended... The future Why report? Conclusion References

 
The present system of reporting overexposures and unintended exposures can lead to confusion and the benefits are unclear. The current use of the linear no threshold dose–risk model means that all radiation doses should be as low as reasonably achievable. However, it does seem an inappropriate use of resources to give the same effort in investigating and reporting unintended doses to, say, the ankle as a CT scan of the pelvis. This paper proposes alternative definitions to "overexposure" and "unintended exposure" and the adoption of a reporting threshold for unintended doses. Unintended doses above this threshold would be the subject of an individual report to the relevant regulatory authority. Unintended doses below the threshold would be recorded by the employer and a statistical report made to the regulatory authority at appropriate intervals.

Received for publication January 24, 2006. Revision received June 5, 2006. Accepted for publication June 16, 2006.

	References

Top Abstract Introduction Overexposures and unintended... The future Why report? Conclusion References

 

1. Health and Safety Executive. Fitness of Equipment Used for Medical Exposure to Ionising Radiation, HSE PM77. London, UK: HSE, first edition 1992, second edition 1998
2. Department of Health. The Ionising Radiation (Medical Exposure) Regulations (2000 together with notes on good practice). London, UK: available in electronic format only at www.dh.gov.uk/assetRoot/04/05/78/38/04057838.pdf [Accessed 3 July 2006]
3. National Radiological Protection Board. Guidelines on Patient Dose to Promote the Optimisation of Protection for Diagnostic Medical Exposures, Documents of the NRPB, vol. 10. no. 1. Didcot, UK: NRPB, 1999
4. International Commission on Radiological Protection. 1990 Recommendations of the International Commission on Radiological Protection, ICRP Publication 60. Oxford, UK: Pergamon Press, 1991
5. Health and Safety Executive. Equipment used in Connection with Medical Exposure. Guidance Note PM77 (third edition) March 2006 on HSE website: www.hse.gov.uk/pubns/guidance/pm77.pdf [Accessed 3 July 2006]
6. Council of the European Union. Council Directive 97/43/Euratom of 30 June 1997 on health protection of individuals against the dangers of ionising radiation in relation to medical exposure. The Council of the European Union, 1997
7. National Radiological Protection Board. Living with Radiation. National Radiological Protection Board. Didcot, UK: NRPB, 1998
8. National Radiological Protection Board. Board Statement on the 1990 Recommendations of ICRP, Documents of the NRPB, vol. 4, no. 1. Didcot, UK: NRPB, 1993
9. Health and Safety Executive. Reducing Risks, Protecting People. Norwich, UK: HMSO, 2001
10. Cancer Research UK website: http://info.cancerresearchuk.org/cancerstats/mortality/ [Accessed 3 July 2006]

This article has been cited by other articles:

				M V Williams Improving patient safety in radiotherapy by learning from near misses, incidents and errors Br. J. Radiol., May 1, 2007; 80(953): 297 - 301. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Clewer, P R Articles by Jackson, P C Search for Related Content PubMed PubMed Citation Articles by Clewer, P R Articles by Jackson, P C