Matthew Christopher Roberts · Prevention of Future Deaths report

On 7 April 2016 the Senior Coroner commenced an investigation into the death of Mr Matthew Christopher Roberts. The investigation concluded at the end of the inquest I held on 3 February 2017.

The conclusion of the inquest was that Mr Roberts died on 7 March 2016 from a complication of necessary medical treatment, the medical cause of his death being (1a) major haemorrhage at tracheostomy site (1b) acute arteritis of the innominate artery (2) coma and hypoglycaemia due to mixed drug and insulin toxicity.

1. Mr Roberts, who was born on 28 December 1993, was a talented music student studying at the University of Chichester. In late 2015 he suffered what appeared to be a first episode of psychosis. This led to him taking an intentional overdose of insulin in early December 2015. After a brief psychiatric hospital admission under Sussex Partnership NHS Foundation Trust (SPFT) Mr Roberts was discharged to his parents’ home in Wiltshire on 24 December 2015.

2. Mr Roberts was supervised and treated, initially, by the Avon and Wiltshire Mental Health Partnership NHS Trust’s ‘North Wiltshire Intensive Service’ team and subsequently by the Wiltshire Early Intervention in psychosis service (the EI team). The transfer from the Sussex services to the Wiltshire services over the Christmas period was described as “seamless”. Contact was made with him very shortly after his return and throughout his period in Wiltshire Mr Roberts was seen or spoken to by team members from the Avon and Wiltshire mental health services almost daily.

3. It was recognised that, in time, Mr Roberts would return to Chichester University in the SPFT’s catchment area, and so on 22 January 2016 the Wiltshire EI team made preliminary contact with the relevant Sussex EI team in Bognor who indicted they would accept a referral.

4. By 12 February 2016 there had been a noticeable improvement in Mr Roberts’ mental state and in view of the perceived lowered risk of suicide the plan, formed in liaison with the University mental health services, was that Mr Roberts would soon return to Chichester and recommence his University studies.

5. It was recognised that this plan involved some therapeutic risk taking as his return to University would, potentially, be a stressful event for Mr Roberts. A risk management plan was therefore devised. It was determined that Mr Roberts’ mental state and risk of suicide was such that he would still require NHS supervision alongside engagement with the University counselling service on his return to Sussex. Indeed the Wiltshire EI team manager described the involvement of the local Bognor EI team as a “crucial” aspect of the risk management plan.

6. Mr Roberts also recognised himself that he needed such support as on 12 February 2016 he asked for the Team which covered the Chichester area to be informed of his impending return and for an appointment to be arranged for the Thursday or Friday of the following week (i.e. 18/19 February)

7. On Tuesday 16 February 2016 the team manager of the Wiltshire EI team made a telephone referral to the SPFT Bognor EI team when he spoke to the Bognor team manager. He informed the Bognor EI team that Mr Roberts was returning to Sussex the following day (although in the event he actually returned on Thursday 18 February).

8. The outcome of that telephone conversation was that the Wiltshire EI team manager understood that Mr Roberts would be offered an appointment with the Bognor EI team on Thursday 18 or Friday 19 February 2016. The Bognor EI team manager did not, however, share this understanding. The Bognor team manager told the inquest that when he offered to arrange an appointment for Mr Roberts “this week” he had meant that telephone contact would be made with Mr Roberts within the next 7 days (as was the service standard) and then at that telephone contact a future date for a face to face appointment would have been arranged with Mr Roberts.

9. Following the telephone conversation between the two EI team managers some supplementary information was faxed by the Wiltshire EI team manager to the Bognor EI team. That fax was time stamped as arriving at Bognor at around 13.30 hours on Tuesday 16 February 2017. The fax included (amongst other documents): correspondence from a Consultant Psychiatrist to the GP dated 21 January 2016 that set out the psychiatrist’s assessment of Mr Roberts’ current mental state, progress in treatment and diagnostic formulation; a letter from the Wiltshire EI team manager to the University counsellor, dated 12 February 2016, setting out the perceived level of risk and containing the risk management plan; a risk assessment document. The cover page to the fax was clearly marked to show the fax consisted of 22 pages.

10. The following day, on 17 February 2016 a daily ‘zoning meeting’ was held by the Bognor EI team. At this meeting all new referrals would be allocated to team members and the inquest was informed that there should then have been initial consideration of the patient’s clinical needs and the level of risk the patient presented to allow the team to formulate a plan and determine a response time for making the first telephone contact with Mr Roberts.

11. The evidence of the Bognor EI team manager (confirmed by his immediate line manager) was that there was not always opportunity to read the information sent by a referrer before such a zoning meeting, and so the practice within the team was to read the referral information at a later time. At the latest this would be shortly before the first face to face contact with the patient.

12. Mr Roberts’ case was allocated to the Bognor EI team manager. However the team manager had not read any of the faxed documents, therefore he relied solely upon the verbal information provided by the referrer to determine the urgency with which telephone contact should be made. The detailed written information sent by the Wiltshire EI team was not considered by the Bognor EI team to inform the decision-making regarding the patient’s needs and the current level of risk.

13. Indeed, in Mr Roberts’ case the Bognor EI team manager had no recollection of ever reading the faxed material and I concluded at the inquest that it was never read by anyone associated with the Bognor team. On considering the fax actually received, which was disclosed part way through the inquest, it was apparent that it was incomplete – it was automatically stamped to show it as containing only 17 pages in total (not the 22 noted on the cover sheet) It was immediately obvious on even the most casual review that the Consultant Psychiatrist’s letter must have been missing its opening page or pages. However neither the Bognor EI team manager nor his immediate line manager were aware of this before I drew it to their attention at the inquest hearing.

14. There had been no attempt made by the Bognor EI team to contact Mr Roberts when, on 22 February 2016 (the sixth day following his referral), Mr Roberts took a large overdose of insulin and his prescribed psychotropic medication.

15. Mr Roberts was admitted to ICU at St Richard’s Hospital, Chichester in a coma having suffered significant hypoxic brain damage. In the course of his ICU treatment a tracheostomy tube was inserted on 2 March 2016. Unfortunately Mr Roberts died some days later on 7 March 2016 when a rare but recognised complication of tracheostomy insertion arose. He suffered acute arteritis of the innominate artery leading to a major haemorrhage at the tracheostomy.

16. What precipitated Mr Roberts taking the overdose on 22 February 2016 is unknown and I concluded that it could not be said that Mr Roberts’ overdose would probably have been prevented had there been an earlier appointment with him made by the Bognor EI team. Nevertheless, it is of concern that there was no policy, system, procedure or practice in place that would have ensured that the Bognor EI team promptly considered relevant clinical information upon being sent by a referrer and used that information to inform their determination of the patient’s needs and the current level of risk, and hence the urgency with which their first contact should be made.

17. Finally, as it was understood that the Avon and Wiltshire Trust would be conducting their own RCA, there had been no internal review of the events surrounding Mr Robert’s case conducted by SPFT. The final RCA report from Avon and Wiltshire Trust, which was completed in June 2016, was not provided to the SPFT and in any event this report focused on the role of the Wiltshire services. Hence those shortcomings I set out above were not identified until the inquest; an adequate review by SPFT should have revealed matters far sooner.

(1) That there was no relevant policy, procedure or practice requiring faxes to the Bognor EI team be logged and scrutinised on receipt so that it might be noted if faxed pages were missing and potentially important information not received.

(2) That there was no policy, procedure or practice, requiring a member of the EI team to read written information provided by a referrer before the zoning meeting and initial risk assessment. Additionally it was practice, on occasions, for the information to be left unread until shortly before the first face to first appointment with the patient. Hence the determination of patient’s needs, the current level of risk and the urgency with which the first contact should be made with a patient was not informed by all the available information being fully considered.

(3) That there was no relevant policy, procedure or practice whereby the Bognor EI team would clearly confirm with the referrer the date on which contact with a newly referred patient would be made.

(4) That SPFT did not appear to have undertaken any formal review of the death of someone known to the organisation and, although SPFT were aware a RCA was being conducted by Avon and Wiltshire NHS Trust, SPFT had not received nor sought that final RCA report from Wiltshire. An opportunity to learn relevant lessons from the above events had therefore been delayed until the inquest, almost a year after events.