There is a “clear need” for pharmacists to be involved in the chemotherapy protocol alteration and promulgation system in South Australia, a deputy coroner says
The inquest into the deaths of four patients with acute myeloid leukaemia closely examined a protocol error which led to several patients being under-dosed with consolidation chemotherapy – an error found out by a pharmacist.
The inquest found that Joanna Pinxteren died as a result of E coli bacteraemia secondary to refractory acute myeloid leukaemia; and that Christopher McRae, Bronte Ormond Higham and Carole Anne Bairnsfather died as a result of acute myeloid leukaemia.
All four deceased had suffered a relapse of their leukaemia following earlier appropriate treatment with chemotherapy.
They had been offered consolidation chemotherapy after entering remission, to reduce the likelihood of relapse.
“The consolidation therapy in the case of each deceased person miscarried because it was undertaken not in accordance with the consolidation therapy protocol that had been determined to be appropriate for those persons,” noted Deputy coroner Anthony Schapel in handing down his findings.
“What happened was that the first cycle of consolidation therapy in each case involved the provision of once daily administrations of cytarabine over the three alternate days when the intended protocol called for administration twice daily over three alternate days.
“The miscarriage of treatment was the result of an error that was contained within the written protocol pertaining to the consolidation treatment for these patients.
A revised RAH protocol had been intended to reflect twice daily administration based on documents relating to the ALLG M15 study; however the protocol which was uploaded said something different entirely.
“The error was due to the inclusion of the word ‘ONCE’ instead of the word ‘twice’, or the expression ‘bd’ which is well understood medical terminology indicating twice daily administration. It had been intended that the protocol should specify twice daily administration,” the deputy coroner said.
The error, which originated within the Haematology Department at the Royal Adelaide Hospital – and was then replicated at the Flinders Medical Centre, which had adopted RAH’s erroneous protocol – was the result of “woefully inept clinical governance” at both hospitals, Mr Schapel said.
Six other patients had also been underdosed as a result: four at FMC and two at the RAH.
It was a pharmacist at the RAH who spotted the error, when he noticed that one doctor had prescribed once daily therapy but another more senior doctor, correctly, had prescribed twice daily.
He discovered the error on Friday, 16 January 2015 as the result of the prescription discrepancy, but Mr Schapel said the error in the actual protocol document was not discovered until Monday 19 January.
FMC would not recognise the error until the end of January 2015.
“If the error had been identified at both hospitals on Friday 16 January 2015, there is a case for concluding that the underdosing of Mrs Pinxteren in respect of her third alternate daily administration of therapy on Saturday 17 January at the RAH probably would have been avoided,” the deputy coroner said.
“Similarly there is a case for concluding that if adequate procedures had been in place for the reporting of adverse incidents across the entire South Australian public health system, the underdosing of Mr Higham on the afternoon of 16 January and on 18 January 2015, during his second cycle of consolidation therapy, probably would have been avoided.”
He also made reference to a surviving patient who had been treated at the FMC – the only patient who was treated in accordance with the erroneous protocol after the error had positively been identified at the RAH on 19 January 2015.
The FMC had autonomy from the RAH in terms of protocol maintenance and content – as did the state’s Queen Elizabeth Hospital, which did not use the erroneous protocol and instead used the eviQ chemotherapy protocol system.
A pharmacist within the Haematology Department at the FMC and another witness both gave evidence that “the accuracy of electronic prescriptions is only as good as the accuracy of the templates that they are based upon, which in turn is only as good as the information that was originally provided when the template was created,” the deputy coroner said.
When in July 2014 the FMC altered its consolidation chemotherapy template to align with what the RAH was doing at that time, the requirement for twice daily administration was erroneously not included.
Kailin Teh, the pharmacist working in the FMC’s Haematology Department queried the new protocol on 21 July 2014.
She emailed a prescriber, a consultant haematologist, at the hospital asking, “’Can you please confirm – From the new protocol the cytarabine is 1g/m2 DAILY on d1, 3, 5 (not BD),” regarding an FMC patient who survived.
The response, which was sent to Ms Teh only, was that “yes she is non-CBF and you are right it is a daily dose of cytarabine”.
“This piece of information was correct if one was talking exclusively about the erroneous once daily stipulation within the protocol, but it was wholly incorrect clinically because the M15 study called for twice daily administration of cytarabine,” the deputy coroner observed.
The deputy coroner said that “the process by which the RAH consolidation chemotherapy protocol was altered was sloppy, ad hoc and prone to introduce error”.
He also criticised the use of email as a communication tool between key personnel at the hospital.
“The introduction of the error into the RAH protocol was directly the cause of the erroneous protocol being applied to the consolidation chemotherapy for Mrs Pinxteren, Mr McRae and Mrs Bairnsfather at the RAH and was also responsible indirectly for the erroneous protocol being applied to the clinical treatment of Mr Higham and Mr Knox at the FMC.”
Ms Teh was not provided with support material regarding the study which personnel at RAH had relied upon when making the intended change to the protocol (which would have included twice daily dosage, rather than the once-daily which was eventually set into place).
Ms Teh said that if she had been provided with this material she would have read it and raised a concern that it did not support the RAH protocol as it was, which evidence the coroner “unhesitatingly” accepted.
“It is an inevitable conclusion, therefore, that if Ms Teh had raised a concern the error would have been detected,” he said.
Uncovering the error
On 16 January 2015 Russell Baldock, a pharmacist at the RAH, was rostered on for prescription verification.
During his work he checked the script for an AML patient, which prescribed twice daily administration of cytarabine on three alternate days, namely 19, 21 and 23 January 2015.
“As part of the prescription filling process it was necessary for Mr Baldock to check the prescription against the relevant chemotherapy protocol which he did,” the deputy coroner noted.
“On examining that protocol he realised that the protocol only called for daily administration of cytarabine. This protocol of course was the erroneous protocol that had been on the SA Pathology server since mid-July 2014.
“Mr Baldock also established that in respect of the patient’s first cycle of consolidation chemotherapy she had been administered once daily cytarabine on three alternate days which had been in accordance with the existing erroneous protocol.”
Another script already existed, one which had been compiled on 12 January, which called for only once daily administration on the three alternate days beginning Monday 19 January 2015.
Mr Baldock decided to clarify the twice-daily script with its prescriber, who gave evidence that the script had been compiled not from the erroneous protocol, but from the original ALLG M15 study document which set out the correct dose. This prescriber had spent considerable time away from the hospital and was not aware that the hospital protocol stipulated once-daily administration.
The error was not established until Monday, though the coroner said it could “easily” have been established on the Friday.
It was immediately rectified in the RAH system and the pharmacy was instructed to identify all patients who had been wrongly administered consolidation chemotherapy in accordance with the erroneous RAH protocol to that point in time.
The deputy coroner said that the fact that the error was not uncovered at the FMC until later in the month was due to “poor clinical governance at both hospitals and what appears to be the fact that government tertiary hospitals in this State act in a silo fashion insofar as each of them does not appear to know what the others are doing, even though they are working within the same sphere of medicine”.
The consultant haematologist at FMC gave evidence that the protocol error was brought to his attention on 30 January 2015, by one of the pharmacists at the RAH.
Kailin Teh, the FMC pharmacist, said she found out about the protocol error when this prescriber informed participants at one of the facility’s weekly haematology meetings; she was then asked to identify the affected patients, which she did. The coroner stated that this meeting had taken place on 12 February 2015.
The prescriber said in his evidence that Ms Teh had been instructed to do this on 4 February, but the deputy coroner said that it seemed “highly unlikely” to him that she would have neglected to perform such an exercise for an entire week, finding that the ward meeting was not informed about the error until 11 February.
“There is no evidence that any person at the FMC became aware of the error within the FMC protocol template until 30 January 2015,” he said. “I find that the error within the FMC protocol template was not formally communicated to any persons in authority at the FMC until a ward meeting of 11 February 2015. I find that this was manifestly too late.”
The court heard from a number of experts regarding how the underdosing could have affected patients, including Professor Allan Boddy, who at the time of the inquest was a Doctor of Philosophy and Professor of Pharmacy (Cancer Therapeutics and Personalised Medicine) in the Faculty of Pharmacy at the University of Sydney
“In principle, administration of a lower total dose, and limiting exposure to anti-leukaemic concentrations of cytarabine to a single short period with once daily dosing would provide a less than optimal treatment,” Prof Boddy said in a report for the court.
“It is possible to say that the regimen of 1g/m2 once daily is likely to have an inferior anti-leukaemic effect compared to 1g/m2 twice daily, as the design of dosage regimens is based on probabilities of optimal treatment.”
Mr Schapel found that “a prescription for consolidation therapy that involved once daily administration of cytarabine over alternate days was not a standard therapy and was not recognised as a standard therapy”.
However, “I do not believe that it is possible for this Court to conclude in any of the five cases which this Court has examined that any remission period or period of overall survival in the case of the four deceased was significantly foreshortened”.
He made a number of observations, including that there “appears to be no sound basis” for a system where haematology departments in different hospitals have complete autonomy in relation to the content of chemotherapy protocols”.
“The chemotherapy error at both the RAH and the FMC would have been avoided if both hospitals, like TQEH, utilised the EviQ chemotherapy protocol system,” Mr Schapel said.
“To my mind there is a clear need for uniformity as between the chemotherapy prescription systems across the board in South Australia.
“Another matter that arises from this inquest is that there is a clear need for pharmacists to be involved in the chemotherapy protocol alteration and promulgation system.
“It so happens that at both the RAH and the FMC pharmacists were either instrumental in identifying the error in the case of the RAH, or in the case of the FMC at least instrumental in identifying an issue, an issue that clearly should have been resolved in favour of what the pharmacist had suspected was an error.”
He said the Safety Learning System method of reporting adverse incidents “does not work” and “does not work across the entire SA Health system”.
“The fact of the matter is that it was utilised in the first instance not at all,” he said.
“It was ultimately utilised late in the piece and it did not in any way serve to ensure that when the chemotherapy error was identified at the RAH it was also identified simultaneously at the FMC.”
He made recommendations including that a state-wide chemotherapy protocol system be developed in relation to the treatment of haematological illnesses, a system which should not be individualised in respect of particular hospitals, but apply to all hospitals who provide haematological services.
This should include a state-wide committee to govern protocol development and alteration, comprising the heads of each hospital’s Haematology Department, the chief haematological pharmacist from each hospital and in the case of AML, a specialist consultant with expertise in that disease.
“There should be a State-wide electronic prescription system that is uniform in its operation within all Haematology Departments in public hospitals in South Australia,” he said.
“Electronic prescription templates created by pharmacists should involve checking against the outcome of meetings that have taken place in accordance with the system described above. They should be checked against the written evidence in support of the protocol alteration. The final electronic prescription should be approved by the committee before it is uploaded onto any prescription system.”