Medication not given: administration of time critical medication in the emergency department

This is the first in a series of investigations exploring why medications intended to be provided to patients are not given. Patients who need medications can suffer harm if these are not provided.

The investigation explored the systems and processes in place to support staff to recognise, prescribe and administer time critical medications in the emergency department (ED). Time critical medications are medications that must be given at specific times of the day to ensure they are fully effective. The investigation also explored the role played by electronic prescribing and medicines administration (ePMA) systems in supporting care in this area.

To explore these issues, the investigation used a patient safety event involving a patient aged 85 with Parkinson’s. The patient usually took his own medication (self-administered) at home – two doses of Parkinson’s medication four times a day, at set times – to help control his symptoms.

The patient had attended hospital for an outpatient appointment where he mentioned he had back pain following a fall at home the previous day. He was advised to attend the ED and went there immediately after his outpatient appointment.

The patient spent 3 days in the ED. During this time he should have received a total of 18 doses of his Parkinson’s medication, which was a time critical medication. However, seven doses were not given and three doses were given late. This meant that only 8 of 18 doses of Parkinson’s medication were provided to the patient on time.

The patient was transferred to a medical ward where his Parkinson’s symptoms deteriorated and he lost the ability to swallow. The patient died 4 weeks after his admission to the ward. The causes of death identified on his death certificate included bronchopneumonia (severe chest infection), Parkinson’s, and frailty of old age.

This series of investigations will explore patient safety events that took place in NHS organisations to understand the local factors that may impact on patients not receiving medications as planned.

HSSIB spoke with a number of national stakeholders who told the investigation that there would be value in understanding more about how situations occur where patients are not given their usual medication when in the ED. The investigation shares findings from the patient safety event and highlights opportunities for local level learning in NHS EDs and across healthcare more widely to help improve patient safety in this area.

• The patient required time critical medication for Parkinson’s but did not receive, or received late, 10 of 18 doses during his time in the ED. The coroner reported Parkinson’s as a factor leading to the patient’s death.
• There were no defined roles or responsibilities in the ED to ensure patients who required time critical medications were identified, and medications prescribed, as soon as possible.
• The ED had no dedicated pharmacy support to help staff in providing care to patients who required time critical medications.
• The patient spent 52 hours in the ED; for 44 of these he was cared for in a corridor because of demand on ED services. Corridor care created additional challenges for ED staff and specialty teams and may have limited opportunities to store the patient’s medication which he had brought from home.
• The Trust did not participate in the Royal College of Emergency Medicine Quality Improvement Programme on time critical medication.
• The ePMA system did not include a function to alert staff about patients who required time critical medications to be prescribed or administered.
• An outage in the ePMA system meant the patient required both an electronic and paper prescription chart. This may have caused additional confusion about the patient’s medication.
• Staff had adapted their practice to ensure they could effectively use the ePMA in the ED setting. This was because of challenges in accessing computers and medication rooms in the ED environment.
• The patient self-administered some doses of his Parkinson’s medication, but this was not planned and self-administration by patients was not widely supported by clinical staff or local guidance.
• Staff were not able to check neurology clinic letters because there was a backlog in these letters being uploaded to the electronic patient record system.
• Staff were not able to check information with the patient’s GP practice or Parkinson’s specialty team at the time the patient’s medication was prescribed in the ED, as this was outside of these services’ working hours.
• Staff received contradictory information from the patient’s son and the GP summary care record about the dosage of medication the patient required. The GP summary care record was taken as the most accurate record, but the information it contained was incorrect.
• Once the patient’s medication information was entered onto the ePMA system, no further attempt was made to contact the GP practice or Parkinson’s specialty team to confirm it was correct.
• Some information about the patient’s medication within the GP patient record was transferred to the GP summary care record, but other information was not.

HSSIB makes the following safety observation

1.1.1 This investigation focuses on the systems and processes in place to support staff in recognising, prescribing and administering time critical medications in the emergency department (ED). It also explores the role of electronic prescribing and medicines administration (ePMA) systems in supporting or hindering care in this area.

1.2.1 The Institute for Safe Medication Practices in the USA defines time critical medications as ‘those where early or delayed administration of maintenance doses of greater than 30 minutes before or after the scheduled dose may cause harm or result in substantial sub-optimal therapy’ (Institute for Safe Medication Practices, 2011). The term ‘maintenance dose’ refers to the amount of a medication needed to maintain a steady state in the body.

1.2.2 In England, a nationally agreed list of time critical medication is not in place; instead, organisations define these themselves. In 2010, the National Patient Safety Agency – the predecessor of the NHS England national patient safety team – issued a patient safety alert, ‘Reducing harm from omitted and delayed medicines in hospital’ (National Patient Safety Agency, 2010). This outlined how organisations were to ensure medications were correctly administered without delay. The alert stated that medication for Parkinson’s should be included in the list of critical medications where timeliness of administration is crucial. The Healthcare Safety Investigation Branch (HSIB) has previously considered the complexity in defining high-risk (including time critical) medications (Healthcare Safety Investigation Branch, 2020).

1.2.3 This initiative created a national impetus and healthcare organisations were required to have local guidance on this safety issue. A review of previously issued safety guidance stated that consideration needed to be wider than just the ‘medication type, including consideration of ‘higher risk patients’… and ‘higher risk situations’’ (NHS England, n.d.a).

1.2.4 The Royal College of Emergency Medicine (RCEM) has defined six types of time critical medications using the acronym MISSED:

• ‘Movement disorders – Parkinson’s/Myasthenia
• Immunomodulators – including HIV medication
• Sugar – diabetes medication (insulin)
• Steroids – Addison’s and adrenal insufficiency
• Epilepsy – anticonvulsants
• Direct oral anticoagulants (DOAC) and warfarin’. (Royal College of Emergency Medicine, 2023a)

1.2.5 However, the RCEM list is not exhaustive. Other national bodies and local organisations also include the following medications as time critical (Care Quality Commission, 2023; East Midlands Emergency Medicine Educational Media, 2024):

• antibiotics
• pain management medications, including opioids
• resuscitation medication.

1.2.6 In the patient safety event described in this report, the Trust included a list of critical medications, which should not be delayed or omitted. The list was:

• antimicrobials (antibiotics, antivirals, antifungals)
• anticoagulants
• antiepileptics
• cardiac medication
• corticosteroids
• clozapine
• desmopressin
• diabetic medication (oral agents and insulin)
• intravenous fluids
• immunosuppressants post-transplant
• opioid analgesics
• Parkinson’s medication.

1.2.7 The NHS Specialist Pharmacy Service (SPS) launched a national scoping review on the use of time critical medications in October 2024 with an aim to collect examples of ‘what has worked well and what has not worked so well’ within NHS organisations in how they manage time critical medications, with the intention of creating a future national medicines safety improvement programme (NHS Specialist Pharmacy Service, 2024). NHS SPS also has other resources available via its website to support in the understanding of time critical medications and the challenges that these medications can pose to patient safety (NHS Specialist Pharmacy Service, 2023).

1.3.1 Parkinson’s is a condition which affects parts of the brain. It is a progressive condition which means people will experience worsening symptoms over time. Parkinson’s affects a person’s levels of dopamine, a chemical which sends messages to the brain to inform muscles when to move. These messages go down to the muscles via the nervous system. When a person has Parkinson’s the amount of dopamine in the brain reduces which makes it more difficult for the messages to get to the muscles. The main symptoms of Parkinson’s are:

• Tremors – where the body uncontrollably shakes, these tremors can be small and affect fine motor function, for example in the fingers which makes holding objects difficult, or can affect gross motor function, for example tremors in the legs which make walking difficult.
• Slow movement.
• Stiff and inflexible muscles – this can lead to problems with movement, including the ability to speak and swallow. (NHS, 2022)

1.3.2 Parkinson’s affects more men than women. Approximately 153,000 people in the UK have Parkinson’s and worldwide it is the fastest growing neurological condition in terms of prevalence. Parkinson’s UK states that there is limited research into aspects of race/ethnicity and Parkinson’s. It is thought that white males are likely to be more affected. Also, people from ‘Black or Caribbean backgrounds are more likely to be affected by slowness of movement, thinking and memory problems and dementia than other ethnic groups. People from Asian backgrounds seem more likely to freeze and experience uncontrolled movements (dyskinesia)’ (Parkinson's UK, 2020).

1.3.3 The symptoms of Parkinson’s can be controlled with different types of medication. Prior to attending the ED the patient in this investigation was taking a Parkinson’s medication containing levodopa and benserazide called co-beneldopa. Parkinson’s medication often needs to be taken several times a day, and the timing of the medication is important. A delay of more than 30 minutes in taking any medication containing levodopa can lead to a deterioration in symptoms, and in serious cases, neuroleptic malignant syndrome, which is a life-threatening condition (Parkinson's UK, n.d., National Institute for Health and Care Excellence, 2018).

1.3.4 Parkinson’s UK has produced recommendations for NHS organisations to support timely, safe and appropriate medicine management for people with Parkinson's (Parkinson’s UK, 2024). This includes considerations for EDs.

1.4.1 In 2023, HSIB published a report titled ‘Harm caused by delays in transferring patients to the right place of care’. The investigation highlighted that patients requiring admission to hospital experienced long wait times in EDs and that the ED was used as an overflow ward as a result of delayed discharges from the hospital into social care (Healthcare Safety Investigation Branch, 2023). Patients over the age of 65 years are also more likely to be affected by prolonged waits for admission in EDs than younger patients (Independent Investigation of the NHS in England, 2024).

1.4.2 People with Parkinson’s may attend an ED, either for reasons related to Parkinson’s or for other reasons not related to Parkinson’s. Attendance at an ED at any time risks overlapping with the time a person is due to take their Parkinson’s medication.

1.4.3 The ED is responsible for managing a patient’s immediate complaint, but also needs to ensure their continued safety by ensuring that all time critical medications are reviewed and where appropriate prescribed and administered in a timely manner.

1.4.4 RCEM have produced a ‘safety flash’ about time critical medications in ED (Royal College of Emergency Medicine, 2023c). This suggests:

• EDs should display a poster at reception or triage that empowers patients to inform staff if they take any time critical medications.
• Hospitals should ensure minimal delays between ED referral and specialty review. Time critical medications should be prescribed for patients at least up to the time of referral for speciality admission.
• There should be an ED pharmacy service that is responsible for managing medication availability and governance.

1.4.5 RCEM have produced a position statement about the clinical responsibility for patients within the ED when patients physically remain in the ED but have been referred to a specialty team (Royal College of Emergency Medicine, 2023b).

1.4.6 Although this report focuses on a patient with Parkinson’s, the learning may be relevant for any medical condition that requires time critical medication.

1.5.1 An electronic prescribing and medicines administration (ePMA) system is defined as follows:

‘The utilisation of electronic systems to facilitate and enhance the communication of a prescription or medicine order, aiding the choice, administration and supply of a medicine through knowledge and decision support and providing a robust audit trail for the entire medicines use process.’ (NHS Connecting for Health, 2009)

1.5.2 In practical terms, an ePMA system in a hospital supports the safe, effective, and cost-effective use of medicines from a patient’s arrival at hospital until their discharge. ePMA systems may exist as standalone systems or systems that may integrate with an organisation’s wider electronic patient record systems. Neither of these electronic systems include the dispensing of medications from the hospital pharmacy.

1.5.3 HSIB has previously investigated the role of electronic prescribing and medicines administration systems and safe discharge (Healthcare Safety Investigation Branch, 2019).

1.6.1 The GP summary care record is a shared record that is available to all healthcare organisations. Information from certain fields in the GP’s electronic patient record is sent to a national database. Details of medications prescribed are included in the national database, alongside the patient’s identifying information. Other information is also included that is not relevant to this investigation. Specific details relating to GP consultations, notes, letters and test results are not included in the national database and this information is only available via the GP.

1.7.1 In autumn 2023 the RCEM began a Quality Improvement Programme (QIP) focusing on time critical medications, in particular levodopa (a medication used to manage Parkinsons). The programme will run until 2026, focusing on different types of time critical medication each year (Royal College of Emergency Medicine, 2023d).

1.7.2 Participation in the QIP is voluntary. Organisations were invited to submit their data and use the information to make changes to their local practice. In addition, the data platform allowed organisations to compare themselves against the national mean. At the time of writing this report, the QIP was still in its first year.

1.7.3 The QIP sets out a process map for organisations for time critical medications. It also sets out standards, process measures and outcome measures that organisations can assess themselves against. The supporting organisational checklist allows organisations to benchmark their preparedness for time critical medications in the ED.

3.1.1 The patient spent a total of 52 hours in the ED. During this time he was due to receive 18 doses of co-beneldopa. A dose refers to each capsule to be taken, for example at 20:00 hours two doses are due. Table 1 sets out if and how these doses were provided.

Table 1 Recorded doses of the patient’s Parkinson’s medication

Roles and responsibilities in the ED

3.2.1 The investigation heard that the focus of the ED was to assess patients and decide whether they could be treated immediately and discharged home or would require admission to hospital.

3.2.2 The patient was seen by several staff members when he attended the ED. The investigation found these staff each had a different role within the ED. However, importantly, there was a lack of clarity about which staff may be responsible for identifying that patients needed time critical medication and which staff were then required to act on this information. The roles and responsibilities for time critical medication are described in table 2.

Table 2 ED roles and responsibilities

Corridor care/temporary escalation spaces

3.3.1 At the time of the patient safety event, the Trust had been struggling to meet demand for beds for a number of months and routinely used ED corridors as bed spaces for patients who needed to be admitted to the hospital. The patient was placed on a corridor bed space and spent the duration of his ED stay there. The corridor bed spaces were staffed on a ratio of one nurse to four patients. Staff recognised that patients received a sub-optimal experience when on a bed in the corridor. Due to the lack of privacy on the corridor, any assessments or interventions were undertaken in a designated bed space.

3.3.2 Nurses who spoke to the investigation explained that although they thought having patients on the corridor was unsafe, they did their utmost to ensure they received standard nursing care including regular medications. The nurses did not have access to a medication trolley and therefore it took them longer to give medications as they had to leave the area to obtain the medications required. Nurses told the investigation that frequently required medications were not readily available or accessible for various reasons, including the medication running out and not having yet been replaced, or the medication not being stocked in the ED. Therefore, they would have to source them away from the ward, leaving their patients.

3.3.3 In addition, because the patient was being cared for on a corridor it was more difficult for his medicine to be kept with him. The patient’s son told the investigation that at some point during day 2, a doctor told him to take his father’s medication home as it would be prescribed and given by the hospital. The investigation has not been able to clarify when this conversation took place or who it was with.

3.3.4 The problem of corridor care is not unique to the Trust. In June 2024 the Royal College of Nursing (2024) published a report into corridor care where one third of nurses stated they had cared for a patient in a corridor or temporary escalation space. NHS England (2024a) states that corridor care ‘is not acceptable and should not be considered as standard’.

3.3.5 The investigation observed that the Trust involved in the patient safety event had accepted that corridor care occurred and had taken steps to make it as safe as possible in the circumstances. Although NHS England does not support the use of corridors and temporary bed spaces, it has recently acknowledged that current demand exceeds capacity, with the expectation being that demand will continue to increase and organisations need to build in more capacity to meet the demand (NHS England, 2024b).

Self-administration of medication

3.3.6 The patient had brought his medication from home with him to the ED. He had been self-administering throughout the day while attending an outpatient clinic appointment. He then proceeded to self-administer two doses while in the ED, with the support of his son. The nurse documented in a free-text field in the EPR that he had self-administered his morning dose, as it had not been prescribed on the ePMA system. The Trust’s EPR system did not interact with its ePMA system and no information about self-administration was entered separately onto the ePMA system.

3.3.7 On speaking with members of the medical team the strong consensus was a preference for doctor prescribing and nurse administration of medication rather than patient self-administration of medication.

3.3.8 The hospital had a self-administration policy that had been approved in September 2021 but could not be implemented. A Trust representative told the investigation that the key issue was a lack of lockers and keys for patients to access their medication if it was left with them in the hospital. There were also the additional challenges posed by corridor care (see 3.3.3). At the time of the patient’s ED attendance the policy was not in use.

3.3.9 One member of the medical team told the investigation they wanted to make sure they knew exactly what had been prescribed and taken for each patient and for this to be documented in the medication section of the electronic record. The ePMA system at the Trust was part of the EPR but fields were not replicated from one to the other. Patients were not able to access their medical records on the EPR and ePMA system to update whether or not they had taken their own medication.

3.3.10 The patient involved in the safety event was in a bed in the corridor. This meant there was no privacy, or secure storage for any of the patient’s belongings, including any medication he had brought with him. The lack of space meant that it was not possible to provide a place for patients to keep their belongings safely and staff urged them to only keep essentials with them. The use of corridor bed spaces made the safe storage of patients’ medication impossible.

3.3.11 One member of staff supported that if a patient was well enough to take their own medication, then as the expert in their own condition they should be empowered to continue to self-administer while in hospital. The paper medication chart that was in use for a short period during the patient’s care had a designated section for self-administration.

3.3.12 At some point, the Trust considered self-administration of medication for patients as it was included on the paper prescription chart and there was a draft policy. In the patient’s case, he was allowed to self-administer two doses. However, there was no process in place to support the decision making and documentation of self-administration. Once the patient’s son was advised to take the medication home, self-administration was not possible for the patient.

3.3.13 One doctor reported that the patient was a ‘poor historian’ and they phoned the patient’s son to obtain a medical history and information about current medications. This may have demonstrated that the patient was unwell as his Parkinson’s specialist nurse had previously described the patient as being very involved in his own care and that he knew what worked well for him. The deterioration of the patient’s health from his baseline, coupled with the general reluctance to allow patients to self-administer medication, likely meant that self-administration was not a consideration for the patient as he became more unwell.

Interface between GP and hospital IT systems

Neurology clinic

3.3.14 The patient was under the care of a specialist neurology team for his Parkinson’s. The neurology team was employed by another NHS trust (Trust B) to provide care for patients from the ED Trust’s catchment area (Trust A). As a result, all the patient’s information, including medical records and clinic letters, was stored on Trust B’s EPR system. This situation would be similar for patients who attend a tertiary (specialist) centre for care of specific medical conditions.

3.3.15 There had been a previous agreement for all clinic letters from neurology to be downloaded from Trust B’s EPR system and uploaded to Trust A’s system. The investigation was told that there were issues in maintaining this and as a result there was a known backlog. The investigation requested from Trust A the most recent neurology letter on its system relating to the patient’s care. The letter was dated 13 months prior to the patient’s attendance at the ED, demonstrating the scale of the backlog.

3.3.16 The patient was seen by the ED assessment resident doctor out of hours. This meant the doctor could not contact the neurology team directly when they were unable to access the patient’s neurology letters in the EPR. During working hours (Monday to Friday, 09:00 to 17:00 hours), the neurology team are available and provide support to the ED when requested.

3.3.17 The investigation was told that both Trust A and Trust B used the same EPR and ePMA systems, but that they were configured differently. At Trust B the neurology team were able to receive an alert when a patient was admitted to the Trust and they were coded to have Parkinson’s within its EPR. The neurology team at Trust A requested the same functionality and were told by the IT team that it was not possible.

3.3.18 All clinic letters are were sent to the patients’ GPs, so there was a record of the patient’s care within primary care and Trust B. However, no record of his care was available to Trust A.

Primary care

3.3.19 The patient’s GP’s computer system was a standalone system. Data from certain fields was transferred to the national patient record system (also referred to as the GP summary care record or the Spine). The hospital system was also a separate standalone system. The hospital system was in use in other trusts across the country and is a system which originated in the USA. Although a number of other local trusts use the same system, the version, configuration and settings are separate which means there is no interoperability.

3.3.20 The Trust was not able to access information in the GP system and is only able to obtain information that has been sent from key fields to the national system. The investigation observed that the limited sharing of information contributed to the under dosing of the patient’s morning co-beneldopa. Key information relating to the dose was contained in a field that is not shared with the GP summary care record.

3.3.21 As explained above, the patient was seen out of hours when the GP surgery was closed and could not be contacted to confirm the patient’s medication dose. The clinic letters which contained information relating to the patient’s co-beneldopa were also not part of the GP summary care record. Therefore, the letters were only available to staff who work for the GP surgery or Trust B, where the letters originated.

3.3.22 The investigation was told, based on the information in the GP system, that co-beneldopa 50/200 mg was to be taken three times a day. The administrative note accompanying this entry stated ‘will increase gradually each week till on 4 per day as per neurology letter’. The investigation was not able to identify a clinic letter with these instructions. The investigation reviewed a clinic letter from March 2019 which had similar instructions but for a different dose of co-beneldopa. It is possible this entry became attached to the 50/200 mg dose of co-beneldopa in error.

3.3.23 The process at the patient’s GP surgery (and many others) was for a member of the administration team to review all correspondence coming into the practice. Any letters which indicated that a GP needed to take action would be passed to the GP. All other correspondence which was for information only was scanned into the patients’ electronic records. This process was a response to the increased workload faced by GPs and aimed to enable them to focus on patient care and tasks that required action.

3.3.24 The dosing information entered into the GPs EPR was contained in a field that would transfer information to the summary care record. However, the administrative note that the patient was required to gradually increase medication was added into a field that does not transfer to the summary care record. This meant that this information was not available to staff in the ED when they reviewed the summary care record.

3.3.25 During an interview with the patient’s GP and while reviewing their electronic records, the GP identified this issue and on discussion could not identify how this issue might be mitigated in the future.

3.3.26 The investigation was later able to review details of the patient’s neurology clinic letters from March 2019 to August 2023. From March 2021, it was stated that the patient was taking both doses of medication four times a day. It is not clear how the prescription remained unchanged for 4 years despite multiple changes in co-beneldopa dosing.

3.3.27 HSSIB is currently undertaking an investigation titled ‘Workforce and patient safety: the digital environment – integration of systems’. While this piece of work is not yet complete, early analysis indicates that many of NHS IT systems are not fully integrated and therefore the seamless sharing of a patient’s medical information between primary and secondary care, and even between departments within some trusts, is not yet achievable.

Involving the patient and/or family in decisions

3.3.28 The investigation heard from the patient’s son and his Parkinson’s specialist nurse that the patient was very involved in the management of his condition. As Parkinson’s progresses, the brain gets used to the increased dopamine from medication but the natural dopamine continues to decrease. This means people may need to increase their dose to maintain movement, or take other medications to support the brain and body in dopamine absorption and movement. These dose adjustments would routinely be managed by an outpatient neurology team. The patient was on a higher dose than was usual for a man aged 85; however, this dose worked for his symptoms and any attempts to reduce the dose resulted in a worsening of his symptoms. During his early diagnosis the patient’s wife always went with him to appointments, and later the patient’s son went with him.

3.3.29 When the patient was assessed by the medical doctor in the ED, his son was not present. The doctor told the investigation that the patient was not able to tell them his regular dose and so the doctor phoned his son. The son told the doctor the usual frequency for both strengths of co-beneldopa was four times a day. On this occasion, the doctor cross-referenced the information with the GP summary care record and attempted to review the neurology clinic letters.

3.3.30 As set out above, the doctor was unable to review any of the latest neurology clinic letters in the patient’s electronic record, or contact the Parkinson’s team or GP as this was outside of the working hours for those services. The doctor discussed this with their consultant who advised that the information could be followed up in the morning. This meant that the information in the GP summary care record was used to inform prescribing until confirmation could be made. This was not followed up with the GP (or Parkinson’s team) the following day. The investigation could not identify whether this was handed over as a task to the day team or why it was not followed up.

3.3.31 The patient’s morning dose was under prescribed; his son reported that both strengths should be taken, whereas the GP summary care record stated only the lower strength should be taken. Staff recognised that the GP summary care record was often not updated for various reasons. These could include the clinic letter not being sent to the GP practice or being sent but not acted on, or delays in updating the electronic record system.

3.3.32 The investigation heard conflicting opinions about listening to patients and/or their family members. Some staff thought that the GP summary care record was the key piece of information for prescribing and would only go to the patient/family if information was not available.

3.3.33 The patient’s son told the investigation that the morning dose was his father’s most important dose as the effects of the evening dose had worn off, and that his father had needed the medication in the morning to help him get moving. He shared that both he and his wife had visited EDs on several occasions with elderly parents who have complex medical needs. Although they had been able to provide information about their parents medication needs, they felt they were not always listened to by staff. They worried about what the experience may be like for other patients, who did not have family or carers with them when they attended an ED.

3.4.1 The ePMA system at the Trust was on the same system as the EPR, as described in section 3.2. The system was not bespoke but was configured to meet the local Trust’s needs. Although other trusts used the same system, the user experience would be different due to version control, upgrades and local configuration.

Outage

3.4.2 During the patient’s time in the ED the electronic systems went down unexpectedly and were not available for use. This happened on the evening of day 2 and was rectified by midday on day 3.

3.4.3 The medical doctor who saw the patient in the evening of day 2 was able to prescribe some of the patient’s medications on the ePMA as the dosages were clear from the information available. Once they had prescribed those, they moved onto the co-beneldopa. This involved speaking with the patient’s son, checking his GP summary care record and confirming the dose with the consultant, and a plan to follow up in the morning. During these tasks, the electronic system went down and the doctor had to document the prescription on a paper chart.

3.4.4 At 12:00 hours the patient was due to take both strengths of co-beneldopa. At this point the electronic system was back in use but the co-beneldopa was available as a paper prescription only. At 12:23 hours the co-beneldopa was prescribed on the ePMA from 16:00 hours. There was no 12:00 hours co-beneldopa prescription as the time had already passed and it could not be retrospectively entered into the ePMA. However, once the ePMA was back online the paper prescription was considered redundant. Therefore, the patient did not have an electronic prescription for his lunchtime co-beneldopa and the doses were missed.

System usability

3.4.5 The investigation asked staff how easy or difficult they found the electronic system to use. Staff reported that they found it easy to use and many gave examples of working in other organisations with more difficult systems.

3.4.6 The investigation observed that if a medication was prescribed but had not been given, the medication field on the computer changed colour to flag to the nurse that it was due. Within 2 hours the medication displayed as yellow and after 2 hours it displayed as red.

3.4.7 Staff accessed the ePMA system either from static computers or using a computer on wheels. Medications were located in a medication room; the primary medication room was central in the ED, the secondary medication room was on a corridor. Staff wrote down the medication required on a piece of paper, took the paper to the medication room. This was because the computer on wheels would not fit in the medication room and there was no computer in the room. They then prepared the medication based on the transcribed information. They brought the medication back to the patient and computer on wheels and checked the medication against the electronic prescription. The medication was given and the electronic prescription was marked as given.

3.4.8 The above process was an adaptation made because it was not possible to access a computer in the medication room. Staff reported that they found it the easiest way to ensure they were getting the right medication for the patient. When asked if they were told to do this by colleagues/managers they said they had not been and “just figured it out”.

3.4.9 One observation of a nurse showed that it took an hour to give a patient their medications; this included obtaining equipment to prepare the medication. Each medication involved a separate visit to the medication room to reduce the risk of error. As a result, the process of administering the patient’s medication was started 30 minutes after it was due, but they did not receive it until 1 hour after it was due.

Training

3.4.10 All staff reported that they had received training in using the ePMA system when they started working at the Trust, or when the system was implemented. The training was delivered through online sessions at staff’s own pace. The majority of staff reported that although they felt they had training it took a few weeks to really understand how to use the system.

3.4.11 The online training taught staff how to do certain tasks on the ePMA but did not reflect the context within which they were working. Therefore, the learning only happened once in ‘real-world’ use.

3.4.12 Medical staff in training grade posts had experience of working across other organisations and reported that they were used to having training for new systems when starting at a new organisation. However, the investigation found that training for the ePMA system may not have included or prepared staff for circumstances when the ePMA system may not have been functional.

3.4.13 Staff described the difficulties working when the electronic systems were not available and said that this would undoubtedly contribute to potential errors. They would need to check various different information sources which they felt increased the likelihood of error.

3.4.14 A resident doctor told the investigation about the challenges they faced in the patient’s care after the ePMA went down. They had not used a paper medication chart before and had to seek advice and guidance from an ED nurse to help locate a chart and begin to complete it. However, once they had been shown the paper medication chart they reported there were no issues with prescribing.

3.4.15 After seeking advice from the nurse, the resident doctor prescribed the co-beneldopa in the ‘Parkinson’s medication’ section of the paper chart. This section contained three pages; the third page was used to write the prescription. It is not clear whether this affected the patient receiving their Parkinson’s medication at midday, as the first two pages would have appeared blank if the chart was reviewed.