Pin
Get email alerts
Request correction
Investigation and inquest
On 18 September 2018, I commenced an investigation into the death of Shanté Andrée Marie Turay-Thomas. The investigation concluded at the end of the inquest on 13 January 2020. I made a narrative determination made at inquest, a copy of which I attach.
Circumstances of the death
Shanté was allergic to nuts and on 18 September 2018 told her mother that she had eaten hazelnuts. She died soon after of acute anaphylaxis.
Coroner’s concerns
1. At the time of her death, Shanté was not receiving specialist care for her allergies. However, her general practitioners (GPs) failed to appreciate this. They assumed that she was being treated for her allergies at the transitional asthma clinic to which she had been referred following her paediatric discharge. This was an incorrect assumption.
The GPs had not identified Shanté (who had a high BMI and was severely allergic) as being at particularly high risk from her allergies and asthma, and had no awareness that they were the sole providers of Shanté’s allergy care.
2. Shanté’s GPs knew that she should carry two adrenaline auto injector (AAI) pens at all times, and they may have mentioned this to her, but they failed to record this and they did not emphasise it to her.
They failed to emphasise to Shanté and her family that the reason for carrying two pens is primarily because in the event of severe acute anaphylaxis, the very strong likelihood is that both pens will need to be administered, one five minutes after the other, to keep the patient alive until the arrival of an emergency ambulance.
The GPs did not explore with Shanté the reason for her erratic requests for a pen. They did not explore with her where she kept her pens. They did not test her understanding of medical advice.
3. The Emerade AAI accompanying leaflet does include the advice that two pens should be carried at all times, but the advice is not re-iterated on the outside of the box. Consideration will need to be given to whether this is the appropriate advice in all cases, but it seems worthwhile to review the issue as a whole.
(I assume the same is true of the EpiPen and the JEXT, but I heard no evidence about these at inquest.)
4. The Emerade AAI is sold singly. It could be sold in boxes of two as the norm and only singly in the alternative.
5. When Shanté’s AAI was changed from an EpiPen to an Emerade, her GPs failed to reconsider the prescription and to increase her dose from 300mgs to 500mcgs.
6. Following the scriptswitch, the GPs failed to ask Shanté to come in to the surgery for training in use of the Emerade. This would also have presented an ideal opportunity to explore Shanté’s understanding of the use of her pens and to ensure that she understood she needed to carry two at all times.
7. The GPs relied upon the advice given by Enfield Clinical Commissioning Group (CCG) that the scriptswitch was simply the replacement of one branded product with another branded product of the same drug/device. This gave false reassurance. The CCG joint formulary committee introduced a new drug for GPs, but then gave the wrong advice to accompany this.
8. The CCG failed to draw prescribers’ attention to the need, following scriptswitch from EpiPen to Emerade, to reconsider the dose and to prescribe the higher dose of 500mcgs for patients at higher risk (which would have included Shanté).
9. The CCG failed to inform prescribers that the Emerade pen requires different training to the EpiPen because different AAIs do not operate in the same way. In fact, the CCG gave the opposite advice.
10. I did not hear evidence that there is any NICE (National Institute for Health & Care Excellence) guidance on the point. It seems that the whole area would benefit from NICE guidance.
11. It would also benefit from review by NHS England, particularly in terms of the guidance given to CCGs and regarding consideration of incentives (CQUIN indicators etc.) to educate and train in the management of allergies.
12. The Emerade AAI (and I assume the EpiPen and JEXT) leaflet does not specifically advise that training from a healthcare professional is needed in how to use this particular AAI as opposed to any other.
13. I heard that the gold standard of training for use of any AAI is to give the patient the relevant pen (whichever that patient is prescribed) containing a placebo rather than adrenaline and, following appropriate instruction, ask the patient actually to administer a dose.
I heard at inquest that the incidence of this standard of training (in any setting) is rare. That may be for good reasons, but it seems that revisiting best practice training at a national level would be helpful.
14. When Shanté became ill following the ingestion of nuts, her mother rang NHS 111 and got through to the London Central & West (LCW) service. However, the call handler incorrectly recorded Shanté’s location: he failed to untick a box and so her grandmother’s address was recorded as her location, rather than her mother’s address where she was staying at the time.
In an example of good practice, this error was recognised by the clinician who later took over the call. However, what nobody at LCW realised was that the Adastra computer system would not then update in real time for any screens save that of the particular clinician inputting the information.
The staff at LCW have since been made aware of this and have been trained to walk over and look at the primary screen to check the address, but it is not clear to me that there is now a national understanding of that element of the system.
15. During the course of the 111 call, a number of errors were made. These were the errors of LCW individuals. When LCW audited the call in the first instance, the audit identified the problem with the address, but failed to recognise how badly the call had gone in other ways. Without effective audit and recognition of failings, it is difficult to see how there can be effective improvement.
16. The individuals making these errors were working within the context of NHS Digital’s categorisation of anaphylaxis as needing a category 2 ambulance rather than a category 1 ambulance, on the Adastra computer system that supports the LCW 111 service.
This was the wrong categorisation and not the categorisation that the call would have received if 999 had been called and the London Ambulance Service contacted in the first instance. Acute anaphylaxis is immediately life threatening and must be treated as a category 1.
I heard at inquest that NHS Digital has since changed its categorisation. However, I also heard that for those areas (I think approximately half the country, though this is not completely clear to me), where the 999 service and the 111 service are supported by different computer systems rather than the same system being common to both services, there could remain inconsistencies of categorisation between 999 and 111.
Even where there are inconsistencies in categorisation, the 999 service will not re-categorise following a 111 clinician’s categorisation, unless a 999 clinician has spoken to the patient, so inappropriate 111 categorisation will not be safety netted by the 999 service. This must be recognised and factored in.
17. In terms of national training for 111 call handlers, the NHS Digital distance learning pack contains advice that is in part inadequate and in part wrong. It does not give the crucial information that one dose of adrenaline, by whichever device it is administered, is very unlikely to be sufficient in the case of acute anaphylaxis. It contains a photograph to illustrate the use of an AAI, but in the photograph the device is held incorrectly.
18. I am unclear as to whether the Adastra 111 algorithm automatically prompts administering a second AAI five minutes after the first if there has been no improvement, but it should.
19. One of the errors made by the first 111 call handler was a failure to ask to speak direct to the patient. This was the error of an individual.
However, this is not the first time that the issue has been brought to the attention of NHS Digital. At inquest, I asked the witness who appeared on behalf of NHS Digital, and indeed had been chosen by NHS Digital as the person best able to assist the court, if this had been an issue in the past. He said no. However, on 18 December 2018, Peter Harrowing, HM Assistant Coroner for Avon, sent a prevention of future deaths report to NHS Digital following the inquest touching the death of David Longden.
It was only when I asked the witness appearing on behalf of NHS Digital specifically about Coroner Harrowing’s report in respect of Mr Longden, pointing out that Coroner Harrowing had raised the need for NHS Digital to place greater emphasis on the call handler speaking to the patient, that the witness remembered that he had indeed seen that report.
I choose to characterise this as a memory lapse rather than as an intention wilfully to mislead the court. (A witness who lies whilst giving evidence on oath at inquest may be found in contempt of court and may even be prosecuted for the crime of perjury.) Nevertheless, if NHS Digital does not have a grasp of this sort of detail, specifically brought to its attention by a coroner in a prevention of future deaths report, it is difficult to see how there can be effective improvement.
20. The issues within this prevention of future deaths report are predominantly national issues, but I heard at inquest that there is no person with named accountability for allergy services and allergy provision at NHS England or the Department of Health as a whole.
Information checked against the published report and official responses