One thing I think we should raise with CSCA is the issue of flagging urgent or unexpected reports. We will have problems implementing this though as we do not have a fully integrated RIS/EPR like Meditech. The RIS doesnít currently support reporting flags, and even if it did, any flags created in the RIS would need to be passed to the EPR report repository. Unfortunately IHE doesnít help here as it only deals with DICOM structured reports, and the RIS only produces simple text reports. A supplement is proposed for HL7 ďstructuredĒ text (e.g., HL7 Clinical Document Architecture (CDA)) reports, which could contain flags, but this is not available yet.
On the suggestion for report acknowledgement feeding back to the RIS, I stick to my original view that itís not strictly necessary. If the main use is an audit trail as you suggest (and I agree), this can be achieved through the EPR alone, as discussed above.
I suppose to cover all bases it would be nice to have a 'Request read receipt' option in the RIS when issuing reports, which could work in the same way as it does with email. I must admit can't really envisage a scenario where I would use it however. If I was sufficiently worried to ask for a read receipt (and keep checking to see if the report had been read), I would have phoned the clinician in the first place.
I still think that once the report has been issued, agreed local practice should be that the onus passes onto the clinician to check the reports 'queue' regularly. As discussed, IT can help here if the system can automatically send alerts to the requesting clinical team if the report remains unacknowledged after an agreed period of time.
Iím aware that medico-legally this is a grey area, as Iíve heard cases of radiologists being reprimanded for not ensuring a report is communicated in a timely fashion. I agree with this, but only if the report would have made a difference if it had been communicated urgently, i.e within the agreed period of time above.
I agree with Rhidian's questioning the usefulness of a tick box for 'urgent' reports. As he says it then transfers a responsibility back to the reporter to check that the box has been ticked! I fear that ringing through a genuinely concerning report to the relevant clinician remains our responsibility. With junior doctor full & partial shift systems we all know that the person who makes the original request is rarely the one assumes ongoing responsibility for that patient. Thus I wonder how effective merely ticking a box will be... Despite this, I think the concept remains fundamentally a good one and like the idea of the RIS sending out alerts when a report remains unchecked/unticked beyond a predetermined time. At least some kind of signature-type acknowledement will then have occured by the recipient, albeit electronic rather than on paper.
We have recently introduced a standard for all reporters that, if a report is of immediate clinical importance, we either phone the doctor and then record the fact as the last line of the report, or, in the case of GPs, say whether we have phoned the GP or faxed the report and the date of contact. It doesn't overcome the issues under discussion but is a pragmatic response to a real risk issue.
Nick, I may not have expressed my self clearly as I DO think a report acknowledgement tick box is very useful, but this is primarily for the benefit of the clinical team, to indicate that someone has taken initial responsibility for checking the report. What I was questioning was the need for this acknowledgement to be passed back to the reporting radiologist. As you say the responsibility should not be left with the radiologist to check that the box has been ticked.
Andrea, I agree colour coding reports is useful, except when you are viewing them on a greyscale PACS monitor!
On John's point we also record in the report whether we have communicated the result in person to the clinical team. Recording it doesn't help the patient of course, but it helps cover our backs!
I am all in favour of phoning an abnormal report, but if this becomes the norm and the minimum expected standard (ie. it is our responsibility) what happens for instance if we can't get an answer? I saw a chest drain half fallen out this week. I bleeped the doctor, and went back to reporting. An hour or two later I remembered he never answered, so I rang the ward. So is it his fault or mine if I hadn't remembered?
I have a similar issue with colour coding reports. What constitutes an abnormal report? We will end up marking as abnormal, colour coding or phoning up an awful lot of reports if we fear legal comeback if we don't. A surgeon once asked for abnormal reports to be highlighted in red because the "Conclusion: Colon Cancer" at the end of a report wasn't clear enough. At some point the referrer has to take the responsibility for looking at and acting on the report, and if we bend over backwards too much to help him/her we will end up with all the responsibility for a situation over which we have little control.
I certainly feel the functionality should be in the EPR, not the RIS, and once the report is in the EPR the responsibility should lie with the clinician to check for unread reports.
It is extremely rare in my experience with the shift system in place that one can actually find a: the person requesting the investigation b: someone who knows the patient and is looking after him/her.
However I want to enclose some comments from a colleague who is the lead cancer radiologist for our sha (DR Garvey)
"manual of cancer measures 2004 highlists a need for a robust system to ensure that an unsuspected diagnosis of cancer is communicated over and above the written report"
"from MDU.. we know that a significant proportion of claims arise from system failure...failure of communication is a common system failure...Members are difficult to defend if not complying with national standards such as the manual of cancer measures"
" difficult to find expert radiologist who would support a member who did not have a system for rapid notification of unsuspected cancer"
" Merseyside Cancer network peer review will require compliance with such a system"
I am very much against removal from clinicians of the responsibility to check their requested reports, but I enclose this to indicated where things are heading... and systems in npfit will need to support this.
posted on Friday, April 15, 2005 - 10:11 am
I'm fascinated by this thread, as there appear to be two elements to it:
1. Radiologists covering their backs 2. the desire to see a safe and robust system.
Both of these are of course eminently desirable. I suspect that Acknowledge on its own is probably not enough. What would give the greatest reassurance to all would be a further electronic entry, made by the referer (or whoever acknowledges the report) along the lines of...."Based on this report I have carried out the following actions (admit pt, further tests etc etc etc"
This would then complete the loop: referer asks question (makes request), question answered (radiologist reports) referer acknowledges AND acts on report findings. Only then, could you be sure that the results have been understood correctly and acted upon appropriately.
Where would such an entry reside? -probably on the PAS (or PCI in Meditech speak)
Would clinicians see this as 'big brother' watching over them? -perhaps
I suspect that the responsibility to communicate reports of immediate clinical importance will be one radiologists will be unable to avoid, whatever protocol/ procedure/ technology is put in place.