Douglas Hodges · Prevention of Future Deaths report

On 17 May 2017 I commenced an investigation into the death of Douglas Hodges, aged 83. The investigation concluded at the end of the inquest on 5 October 2017. The conclusion of the inquest was natural causes.

Professor Hodges had a past medical history which included stroke, vascular dementia, congestive cardiac failure, prostate cancer, atrial fibrillation and hypertension.

He was seen at home by his GP on 27 March 2017. His family was concerned that he seemed more confused since starting to take medication (Bicalutamide), prescribed after his prostate cancer diagnosis. His GP thought Prof Hodges may have been suffering the beginnings of a lower respiratory tract infection. He said he was not sure of this – but prescribed antibiotics just in case. He prescribed 500mg Amoxicillin, to be taken 3 times a day, for 5 days.

The GP issued this prescription electronically via the Electronic Prescription Service when he returned to the surgery. It was planned that the prescription would be delivered to Prof Hodges’ home address by Well Pharmacy in Chilwell, Nottingham (‘the pharmacy’), as had happened in the past. The pharmacy, along with many others nationally, uses a software system provided by Cegedim. Other software providers exist, which supply a similar service to other pharmacies.

The prescription was sent to the NHS spine at 15.08, and was downloaded by the pharmacy at 15.33. The GP thought this would be actioned urgently, as an acute prescription, and that Prof Hodges would have his antibiotics that day or the day afterwards.

Investigations have revealed that, after being downloaded, a paper token was sent to be printed at the pharmacy. We were told that paper tokens are kept in a basket (alphabetically by patient’s surname) to be dispensed at a later stage. This may be minutes, hours or days later. Paper tokens are shredded after use.

No fault has been identified with this printer. Another prescription (for a different patient) was printed at the same time. There was no evidence to suggest that other prescriptions at the pharmacy have been sent to the printer but not in fact printed. The paper token has never been found.

My conclusion was that the most likely sequence of events was that the paper token was printed, but somehow mislaid or accidentally disposed of.

Prof Hodges’ antibiotics were never dispensed. No label was created at the pharmacy. A later prescription for him (issued on 31 March) was received and later dispensed by the pharmacy (in fact after he had died). It was not appreciated at that time that the earlier prescription of antibiotics had not been dispensed.

Prof Hodges died in hospital on 3 April 2017, following admission there the previous day. His cause of death (following post-mortem examination) was 1a multiple organ failure, 1b systemic sepsis. Given the short time between his admission to hospital and his death, no source of his infection could be found. He was not thought to have a chest infection. He was treated with antibiotics, administered within an hour of his admission, in line with sepsis protocols.

I found it unlikely that, if Prof Hodges had received the antibiotics prescribed by his GP on 27 March 2017, the outcome would have been different.

I am mindful however of my responsibilities (under paragraph 7(1), Schedule 5 of the Coroners and Justice Act 2009) to act where I am concerned there is a risk of “other deaths”. Put simply, a missed prescription could create a risk of future death in a different case.

Although Prof Hodges had a supportive family, I am mindful in particular of vulnerable patients who do not always have this, who may be reliant on medication being delivered to them timeously.

In relation to NHS Digital :

1. As matters stand, in a community pharmacy setting, there is no way of communicating clinical urgency between prescriber and pharmacy staff at the time the prescription is downloaded. A large number of prescriptions are downloaded every day by pharmacies. Urgent and non-urgent prescriptions look the same on the system. It is only at the labelling stage that any clinicians’ comments can be seen.
2. We were told that a different system (Vision) is used in scenarios where a pharmacy is run by a GP practice. This system is often used in rural settings. The Vision system allows for a prescription to be marked with a red exclamation mark at the point of downloading, where a prescription is urgent.
3. Where prescriptions go via the NHS Spine, such a system is not currently possible.
4. We heard evidence about a recent survey conducted by NHS Digital, in which pharmacists made it clear that this was the change they most wanted to see.
5. It is clear that agreed guidelines will need to be considered, in line with any such change, to define what is meant by the term ‘urgent’.
6. Prof Hodges’ case makes the case for this change very clearly, and I consider there is a real risk of future deaths if this is not addressed.

In relation to Cegedim :

1. During the investigation, an experienced clinical pharmacist who works at the GP practice attended the pharmacy and looked at the system in question. When accessing the ‘help section’ of the software, the information she saw suggested that the red exclamation mark system was available and she questioned why this was not being used. It is not clear why this information is included in the ‘help section’ of software that does not currently support this function. It clearly led to confusion. I ask that Cegedim review this and consider clarifying it.

In relation to Wells Pharmacy :

1. I heard evidence from the patient safety manager of Wells Pharmacy Group, who told us they are trialling a system which includes the following provisions :
a. Downloading prescriptions only up to 2.30pm each day, with the aim of labelling all prescriptions downloaded on that day. We were told that urgency and delivery instructions would then be picked up at the labelling stage.
b. Pharmacists are required to check at the end of each day whether there are prescriptions which have been sent to print but not yet got to the labelling stage.
c. If a patient or representative comes in to collect prescribed medication, only one prescription will be downloaded at a time, to avoid any confusion.
d. Local GPs will be made more aware of likely / realistic timescales between prescription and the medication going to patients.
2. I note that this trial will end in November. Whilst it is entirely appropriate to trial any change like this, there is no guarantee that changes will be implemented after the trial. I would like to know what the intention of the pharmacy is after the trial period ends, and how it proposes to reduce this risk in future.