The message of the paper by Grosvenor et al. [1] is clear and reassuring; namely that there is some added value to be obtained by insisting that radiologists report the chest radiographs of patients arriving on a medical admissions unit. I want to comment on two incidental issues raised in the paper, both relating directly or indirectly to the Ionising Radiation (Medical Exposures) Regulations [IR(ME)R] [2]. The first concerns the apparently simple task of ensuring that referrals for plain film radiography are properly justified in the IR(ME)R sense of the word. Historically, plain film requests have been assessed by the radiographer carrying out the examination. If in his or her experience the referral sounds reasonable, the exposure goes ahead; if it seems in any way dubious, the advice of a radiologist or more senior radiographer will be sought. We all know that this works well, but that the occasional unnecessary exposure slips through. The task for those involved with the implementation of IR(ME)R has been to devise a system of authorization or justification of plain film examinations which complies with the spirit of the legislation without at the same time bringing our busy departments shuddering to a halt. This is not the place to discuss the details of IR(ME)R implementation, and I only mention it here because this paper demonstrates that the difficulties associated with the justification of plain film requests reflect inherent limitations in the utility of imaging guidelines. The authors found that no less than 44% of requests for chest radiographs were clinically sound, but were not specifically covered by the College’s guidance “Making the best use of a department of clinical radiology” (MBUR). This is no criticism of the guidelines; it simply illustrates the fact that while it is possible to devise a set of reasonably comprehensive protocols covering role of computed tomography (CT) in abdominal and gastrointestinal disease, a book of guidelines, which included all the acceptable indications for a chest radiograph would be too heavy to lift and impossible to use in practice. So, whatever solution we adopt for justifying plain film requests in our departments, it is not sufficient simply to give the radiographers copies of MBUR and tell them to get on with it. For what it is worth, I think the only practical answer is to write a protocol that allows radiographers to continue working as they have always done, doing everything we sensibly can (and that “sensibly” is important) to minimize unnecessary exposures. At the risk of sounding heretical, the doses involved in the occasional questionable plain film are small, and we need to concentrate our policing of IR(ME)R on the high-dose procedures where the most damage can be done. Hopefully, we were already vetting these “special” investigations on clinical grounds long before the advent of IR(ME)R. The other issue I want to raise is that of evaluating and recording the outcome of an examination. The authors of this paper rightly emphasize Regulation 7(8), which states “The employer shall take steps to ensure that a clinical evaluation of the outcome of each medical exposure is recorded...” The associated note on good practice enlarges on this: “If it is known prior to the exposure taking place that no clinical evaluation will occur, then the exposure would not be justified and could not lawfully take place”. In other words, we shouldn’t be exposing patients to radiation if we know that no-one will be looking at, and acting on, the result of the examination. The authors of this paper interpret the regulations to mean that the burden of reporting an acute chest radiograph, and recording the result in the patient’s notes, falls on the clinician in those situations where an immediate report by a radiologist is not available. There is no doubt that there are good clinical governance reasons for taking this line; there is little value in a radiologist’s report if it is only available after the patient is recovering, or dead. However, I am not convinced that this is necessarily a direct requirement of IR(ME)R. The regulations do not state that the evaluation has to be made immediately, or within any particular timescale, specifying only that it should occur “in accordance with the employer’s procedures”. Unless those procedures include a timeliness