Darran Busby

Coroner's concerns

AI summary

A critical flaw in the electronic patient record system allows radiology results requiring urgent follow-up to be inadvertently filed without clinician review, risking missed diagnoses and treatment delays.

View full coroner's concerns

After it became apparent that the result of Mr Busby's MRI scan had not been reviewed by a clinician, a consultant employed by North Cumbria Integrated Care NHS Foundation Trust ('the Trust') undertook an investigation. He noted that the Trust use EMIS as an electronic patient record. He explained that the Trust used a separate system called ICE to gather the results of tests or scans. ICE is capable of linking to EMIS to input results into the EMIS system. Once a test result has been linked to a patient in EMIS the result enters the EMIS record as a provisional result pending review, and is placed on a work list. The consultant or a deputy then reviews the result, files it with or without comment and records any actions taken. EMIS provides two options: 'file no comment' and 'file and comment'. Results of blood tests which are undertaken to monitor treatment and which are normal may be filed without comment. If there is an abnormality flagged, however, EMIS will default to the file with comment dialogue box even if file no comment is selected. This acts as a safeguard against missing a significant finding. Unfortunately, there is no flag attached in the ICE system for abnormal radiology results, and so no failsafe exists for defaulting to a 'file and comment' if a significant positive or negative finding is reported. In the course of investigating what occurred in relation to Mr Busby's MRI scan, it was determined that clicking more than once on the 'file no comment' button will result in the displayed result being filed, but will also result in filing of the next in the list if that result has no flag indicating the result is abnormal. Thus if a radiology result lies below a normal blood result and a clinician inadvertently double clicks to file the first result, the radiology result is also filed without comment and without the result being displayed. Furthermore, multiple clicks up to 6 (and perhaps even beyond) will lead to multiple filings. In the result it is possible that a clinician inadvertently clicking 'file no comment' more than once on one result would cause results which require urgent follow up being filed without a clinician being involved. I am concerned that this might lead to lost opportunities to treat patients whose scans reveal, for instance, early malignancies. It might also mean that scans which reveal the need for urgent action will be overlooked. I am therefore concerned that future deaths will occur. I was impressed by the candour of the report provided to me and the efforts that the Trust have already taken to resolve this issue. However I noted that the evidence I received was that "In order to fix this issue it is likely it will require action by the publishers of EMIS to prevent accidental filing of results. To attempt to mitigate this issue whilst a permanent fix is sought I have worked with colleagues from Pathology and Radiology to attempt to have all radiology results (where the greatest risk lies) flagged within the ICE system as abnormal, so that any attempt to file the result prompts via the file and comment dialogue box. Unfortunately at the time of writing this letter the flag, which is triggered in ICE for any radiology report originating within Cumbria Neuroscience, does not carry through to EMIS and we continue to seek a local solution to mitigate this newly identified risk." In the circumstances I have concluded that it is necessary for action to be taken to prevent future deaths.

Responses

3 respondents

EMIS Other

9 Mar 2022 PDF

Action Planned

EMIS is reviewing and will update the EMIS Web Hazard Log and Safety Case to reflect identified concerns; highlighting established system and training mitigations. EMIS is reviewing training material relating to the filing of results, to include reference to those results that may require a more detailed review, such as radiology results. (AI summary)

View full response

Dear Mr Cohen, With regard to the Regulation 28 report dated 13th January 2022 (the “Report”), we have reviewed and considered the circumstances relating to this very unfortunate incident. Whilst we note that the failure to review the MRI scan report was not causative or related to Mr Busby’s death, we treat such matters with the utmost seriousness and an internal review has been undertaken, focussing on the issues of:
1) No flag attached for abnormal radiology results;
2) Inadvertent ‘multiple clicking’ of the ‘file no comment’ button in EMIS Web will result in the filing of more than one result if subsequent results are not flagged as abnormal. We would like to note the collaboration of North Cumbria Integrated Care NHS Foundation Trust (“the Trust”) throughout this investigation.
1) No flag attached for abnormal radiology results In relation to this issue, we reviewed the functionality available within EMIS Web and confirmed that it is working as designed, and as per NHS Digital specification. EMIS Web flags results based on the presence of a normal / abnormal clinical code (relevant to the investigation type) within the message received from the Trust’s reporting system (in this case ICE). In this case, the message containing the MRI result did not have a code which indicated ‘normality/abnormality’ for the MRI. The MRI report was tagged with a ReadV2 code “56D – Other diagnostic radiology” terminology code, which is not specific enough to provide any receiving system (in this case EMIS Web) with the relevant details to flag the report as abnormal and present the relevant warnings. The Report further notes that the Trust had subsequently, unsuccessfully, attempted to add flags to all pathology and radiology results. The above explains why this action was not successful as it requires associated coded clinical terminology, rather than custom flags, for results from any source system.

Investigation by EMIS, in collaboration with the Trust, has determined that a local EMIS Web protocol to flag 56D-coded terms would not be possible. EMIS Web does not have the ability to configure rule- based logic on a particular code or result type to drive the desired protocol alert.

The understanding of functional limitations of third party systems (ICE & Cris), reviewed as part of this investigation, is based upon information provided by the Trust during the investigation process.

The UK messaging standards for laboratory and radiology reports are currently based on EDIFACT specification which are some 20 years old. EMIS is aware that NHS Digital are currently building a modern set of FHIR messaging standards which, when implemented, will enable report level flagging of normal / abnormal findings which will significantly improve the safety of these messages; EMIS is collaborating with NHS Digital on these standards. There are no current timelines for introduction from NHS Digital.

2) Inadvertent ‘multiple clicking’ of the ‘file no comment’ button in EMIS Web will result in the filing of more than one result if subsequent results are not flagged as abnormal The report states that the Trust, during the course of their investigation, determined that “clicking more than once on the 'file no comment' button will result in the displayed result being filed, but will also result in filing of the next in the list if that result has no flag indicating the result is abnormal.”

In relation to this issue, we reviewed the functionality available within EMIS Web and confirmed that it is working as designed and in accordance with NHS Digital requirements and specification.

Within EMIS Web, there are two views that a user can select when undertaking results filing activities: ‘Inbox overview’ and ‘Detailed overview’.

Inbox overview (dummy data screenshot shown below in fig 1) shows all report tasks in a list view including the date the report was received, the patient’s name and the report type.

Fig 1

Within this view, it is not possible to inadvertently file multiple results through ‘multiple clicking’ of the ‘file no comment’ button, as the next result in the list is not automatically selected. The ‘file no comment’ button is not available again (after being pressed once) until the next result is selected manually by the user. Detailed overview (dummy data screenshot shown below in fig 2) shows all report tasks in a list, and the full report is seen on screen (when one result is selected).

Fig 2 Within this view, after clicking the ‘file no comment’ button, the result in context will move down the list to the next available result. As another result is then in context, a user can click ‘file no comment’ again, and this may be done without the user reviewing the displayed report in appropriate detail.

The user therefore can click through any number of results in quick succession. If clicking fast enough, the system will not have time to load the result on screen before the user has clicked the ‘file no comment’ button again. The user therefore may not see the full report.

Irrespective of which method is used for results filing activities, it is important to note that filing actions do not remove the results from the user’s screen. Results must be manually archived for this to occur. Results remain visible to the user and can be unfiled and re-actioned if needed. The system will indicate to the user which results have been filed by a green tick, meaning the user can re-review those results prior to archiving (as per dummy data screenshot shown below in fig 3).

Fig 3

A user can choose to archive using the ‘archive’ button on the EMIS Web ribbon or, when the user gets to the end of the results list, the system will suggest archiving, to which the user must apply a manual acceptance. Whichever method the user chooses to archive the results, they must click through a screen alert to confirm intended action (as per screenshot shown below in fig 4). It is important to note that the screen alert is defaulted to ‘No’ to prevent accidental clicking through.

Fig 4

In terms of results filing, and against the information provided in the Report, our investigation shows that the system is working as expected and as per NHS Digital specification. Conclusion Based upon the information provided in the Report and the subsequent investigation, EMIS do not believe there are any software developments to be undertaken to mitigate risks relating to either of the issues raised in this case. In respect of the abnormal flag for radiology results, EMIS do not believe that mitigation through EMIS Web development is plausible until emerging NHS Digital standards are implemented, given there are no clear identifiers on these types of results that would allow a system to differentiate between a normal or abnormal result. In relation to any inadvertent filing activity, EMIS believe that there are sufficient failsafe measures within the system, alongside appropriate diligence from the user, to prevent such occurrence. It must be the responsibility of the clinician to review, file and subsequently archive results at a speed and with a level of diligence that fits the clinical nature of the results and the patient involved. As a result of this case and your findings, EMIS is undertaking a number of actions to support our users in the prevention of future harm:
1. EMIS is reviewing and will update the EMIS Web Hazard Log and Safety Case to reflect these identified concerns; highlighting established system and training mitigations that can reduce the risk of future patient harm.
2. EMIS is reviewing training material relating to the filing of results, to include reference to those results that may require a more detailed review, such as radiology results – to prevent users becoming inappropriately reliant upon clinical decision support such as ‘abnormal flags’.

EMIS will continue to review and investigate any cases of a similar nature, and review effectiveness of any current and ongoing mitigations. EMIS have identified, during collaboration with the Trust, a number of operational improvements within their results management practices that the Trust could implement to improve patient safety. Additionally, EMIS will work with the Trust to modify local technical configurations to support the workflow. EMIS will support the Trust to implement these should they wish to. We trust that the details outlined above are of help. Finally, as a company, we work very hard to support health care services across the UK and patient safety is of paramount importance to us. We were saddened to read of the issues relating to this particular incident and we would like to pass our condolences on to the family. If you have any further queries then please contact our Senior Clinical Director, (via , in the first instance. Kind regards

Dr

Chief Medical officer, EMIS Group

North Cumbria Integrated Care NHS Foundation Trust NHS / Health Body

11 Mar 2022 PDF

Action Taken

The Neurology team has stopped using the “file no Comment” button in favour of the “File and Comment” button. The Trust has notified colleagues in Primary Care and anticipate implementing a RAD system in April 2022. (AI summary)

View full response

Dear Sir

Regulation 28: Prevention of Future Deaths Report Response Deceased: Mr Darran BUSBY

I write following the inquest held on 13 January 2022, before Mr Robert Cohen, into the death of Darran Busby who sadly died on 14 August 2021 with a medical cause of death of 1a) Hanging. During the inquest Mr Cohen issued a Regulation 28 report to both the Trust and EMIS. The Regulation 28 report reflects that Mr Cohen was appreciative of the Trust’s frank evidence. The evidence however identified a failure to review the outcome of a head MRI undertaken in April 2021 as a result of functionality issues within the EMIS system. Mr Cohen recognised that the failure to review the MRI, or any other aspect of the Trust’s care and treatment, was not causative or related to Mr Busby’s death. However, Mr Cohen raised concern that the functionality of the EMIS system may, inadvertently, lead to future deaths if action is not taken, and his statutory duty to report to the Trust was engaged. The specific concerns raised within the Regulation Report were:

 There is no flag attached in the ICE system for abnormal radiology results, and so no failsafe exists for defaulting to a 'file and comment' if a significant positive or negative finding is reported.

 Clicking more than once on the 'file no comment' button will result in the displayed result being filed, but will also result in filing of the next in the list if that result has no flag indicating the result is abnormal. Therefore making it possible that a clinician may inadvertently click 'file no comment' more than once on one result would cause results which require urgent follow up being filed without a clinician being involved. I am grateful to Mr Cohen for raising these concerns as it is imperative to the Trust that prospective safety issues are identified and rectified to ensure our services are safe and effective. As a result of the Trust’s progress I am now in a position to provide an update in this matter ahead of the timescale of 11 March 2022. The Trust’s Digital Services has since engaged with EMIS in support of testing a workable solution, and have made available all resources necessary to support the work on this issue. In addition to the Trust’s engagement with EMIS, the Trust ensured that services utilising the same functionality within EMIS, outlined within the Regulation 28 report, were identified. The Trust identified two services: the community ward at Cockermouth Hospital and the Neurology Service. The ward at Cockermouth Hospital reverted back to using the ICE Order Comms system (ICE was outlined in the Trust’s evidence to the inquest), following an initial assessment of the functionality. Whilst this option

was explored for the Neurology Service it not a feasible solution for the service due to the potential of introducing other risks such as transcription error when transferring data from ICE to EMIS. The Trust has looked at ways of using codes to flag results via the ICE system but when the interface is linked to EMIS, the data comes directly from the source system (in this case, this would be either the Telepath Laboratory Information Management System or the GE (recently replaced by Philips) Radiology Information system). Both these systems bypass the ICE Order Comms system and interface directly with EMIS Web. Therefore, adding codes to ICE would not impact on any functionality for flagging. Whilst this information has been shared with EMIS to inform their consideration of solutions to this issue the Trust has sought other appropriate remedies. The Trust is implementing a Rad Alert system, which will operate separately, though alongside ICE and upon recognising an alert code in a radiology report it will email the referring consultant/GP to advise them of a significant radiology finding. In the event the email is not acknowledged within a given time period (variable according to the severity of the alert) the system will alert the rad alert admin in order that alternate clinicians can be emailed. This should prevent a recurrence of this incident regardless of whether the report is being reviewed on EMIS or on ICE as it is a separate way of highlighting the significance of the report to the referrer. It is anticipated that the RAD system will be implemented in April 2022. In the interim Dr and the Neurology team have increased vigilance when reviewing results, and have accepted the key recommendation from the Digital Services to stop using the “file no Comment” button in favour of the “File and Comment” button. This approach will introduce a direct action by the clinician that means a result cannot be filed inadvertently as a popup box always appears. This introduces extra mouse clicks and is therefore more time consuming but does provide the assurance that the results cannot be filed without appropriate review until a more robust system based solution is in place. The Trust has notified colleagues in Primary Care as users of EMIS through discussion with the CCG Chief Clinical Information Officer, to minimise any similar adverse action within GP provision. The Trust will continue to work with EMIS and support their work in identifying a solution to this issue. In addition the Trust has provided feedback to EMIS to support their consideration of communication and escalation to system users if such issues are highlighted to enable them to better engage with users to understand the risk and develop early solutions. I hope the above information provides assurance to the Chief Coroner that the Trust has an ongoing commitment to finding a resolution to this issue and in turn, mitigate the risk of future deaths to the users of the Trust’s services. Whilst it is anticipated that the implementation of the RAD systems will mitigate the risk from the Trust perspective with implementation commencing in April 2022, we will have undertaken testing and confirmation of this solution by the 1 July 2022; we will also continue to engage with EMIS and would like to propose that we provide a further update on this date.

Signed:

Date: 11 March 2022

Chief Executive

NHS North Cumbria Integrated Care NHS / Health Body

6 Jul 2022 PDF

Action Taken

The neurology team has stopped using the 'file no comment' function and increased vigilance when reviewing results. A new standard operating procedure was developed and reports containing the text "significant radiological finding" have been flagged in the ICE system. A RAD alert system is being piloted to email consultants/GPs about significant radiology findings. (AI summary)

View full response

Dear Sir,

Regulation 28: Prevention of Future Deaths Report Response Deceased: Mr Darren BUSBY (Case Ref: 2281920)

As indicated within our communication on 11 March 2022, I would like to provide you with a further update as to the Trust’s progress to address the concerns raised within the regulation report:

 There is no flag attached in the ICE system for abnormal radiology results, and so no failsafe exists for defaulting to a 'file and comment' if a significant positive or negative finding is reported.

 Clicking more than once on the 'file no comment' button will result in the displayed result being filed, but will also result in filing of the next in the list if that result has no flag indicating the result is abnormal. Therefore making it possible that a clinician may inadvertently click 'file no comment' more than once on one result would cause results which require urgent follow up being filed without a clinician being involved.

As previously advised, the Neurology team has increased vigilance when reviewing results, and have ceased using the file no comment function and instead continue to use the “File and Comment” button. This ensures a direct action is required by clinicians, resulting in a pop up box always appearing and inadvertent misfiling being prevented.

The Trust continues to ensure that the improvement of review and action of abnormal radiology and histology results is a priority and as such has established a task and finish group to oversee this area. A new standard operating procedure has been developed and from 18 May 2022 reports containing text “significant radiological finding” have been flagged as abnormal in the ICE system with a red exclamation mark alongside.

We previously advised that the Trust was implementing a RAD alert system, which will operate separately, though alongside ICE and upon recognising an alert code in a radiology report it will email the referring consultant/GP to advise them of a significant radiology finding. Rad-Alerts contain a safeguard so that prompts are issued until an alert is acknowledged. If this is not responded to within 72 hours, it will be followed up with a phone call. The RAD alert system is currently built and available within the test area of the system and is in pilot phase.

I hope the above information provides assurance to the Chief Coroner that the Trust has put in place immediate mitigation of the risk and is continuing to progress with the long term permanent solution.

Signed:

Date: 6 July 2022

Report sections

Investigation and inquest

On 23 August 2021 an investigation was commenced into the death of Darran Busby. The investigation concluded at the end of the inquest on 13th January 2022. The conclusion of the inquest was Suicide 1b 1c

Circumstances of the death

4 On 14th August 2021 Mr Darren Busby was at home with his family. He and ended his life. Prior to his death Mr Busby had complained of headaches and had been referred for an MRI scan of his head. He had had the MRI scan in April 2021. It emerged in the course of my investigation that the outcome of that MRI scan was never reviewed by a clinician. I did not conclude that this was causative of, or related to, Mr Busby's death. However, the circumstances in which it was not reviewed give me cause for concern that there is a risk of future deaths unless action is taken.

Similar PFD reports

Shared signals

Coroner's concerns

Responses

Report sections

Similar PFD reports

Related inquiry recommendations

Report details

Responses identified

Sent to

Source links