Medical advances in diagnosing and treating cancer have vastly improved patient outcomes.1 In England and Wales, in 1971–72 24% of cancer patients survived 10 or more years post diagnosis. In 2010–11 this had more than doubled to 49.8% of patients.2 Additionally, development of novel systemic anti-cancer therapy (SACT—previously referred to as ‘chemotherapy’) means more people can be treated. This has resulted in increasing pressure on oncology services to provide treatments safely and on time.

In the UK, preparing SACT is predominately carried out within, or managed by, hospital pharmacy departments. It is a high risk, multi-step process, requiring aseptic compounding.3 Multiple checks by pharmacy staff are necessary to ensure the SACT is prepared correctly and safely.4

Figure 1 shows a common SACT production workflow within a UK hospital aseptic unit. The process is time consuming and error prone. Human verification is carried out to ensure the safe production and delivery of SACT.4 5 Irrespective of current safety measures, medication errors still occur.5 Analysis of the National Aseptic Error Reporting Scheme (NAERS) internal error data showed a UK error rate of 0.49%. Errors associated with SACT made up 40% of all reported internal errors, with the majority of errors associated with labelling (34.2%) and transcription (11.1%).6

Systemic anti-cancer treatment (SACT) production workflow.

Medication errors are defined here using the National Reporting and Learning Systems definition, and include any unintended incident relating to prescribing, preparing, dispensing, administering, monitoring, or providing advice on medicines.7 In the context of compounding SACT, the NAERS categories are: labelling errors, transcription errors, incorrect expiry, incorrect final volume, calculation errors, incorrect dose/strength, incorrect diluent/infusion fluid, incorrect drug, and incorrect container.6 Errors detected before leaving the aseptic service are classed as internal errors.

Errors can be attributed to a multitude of factors including workload, work environment and stress. Additionally, errors themselves cause more work, increased stress levels and potentially further error. Healthcare settings have previously been deemed ‘hectic, demanding, time-constrained environments’. Work environments involving frequent interruptions (eg, phone calls, pagers, or other healthcare professionals (HCPs) such as doctors, nurses and pharmacists) have been attributed to HCPs making errors.8 Interruptions during drug dispensing, for example, have been found to increase internal error rates by 3.42%.9 Similarly, a UK study which interviewed pharmacy staff about errors they had been involved in reported interruptions and distractions as frequently causing errors.10 This same study found that pharmacy staff more frequently attributed errors they made to a high workload. More recently a Swiss study evaluated the impact of simulated workloads on accuracy and error rates in manual aseptic preparations.1 Three 1 hour scenarios were designed to replicate low, medium and high workloads. Each scenario was carried out by 21 pharmacists and pharmacy technicians in a randomised order and observed by a member of the research team. In total 1007 syringes were prepared. Errors, including wrong concentration, labelling errors and selection errors, were found to increase significantly from 1.8% to 5.4% as workload increased.

In the manual processing of SACT, transcription of prescriptions onto worksheets and labels is one of the most error prone elements. For example, a root cause analysis of 401 Danish community pharmacy medication errors identified 59% of errors at the transcription stage, with the most significant of these being wrong strength or wrong drug errors.11

There is a drive in the UK to evaluate and implement technology to improve workflow safety in the preparation of SACT.12 Research exists where technological advancements in compounding workflow systems and robotics have been made. However, these areas of research have focused on improvements to the final product compounding time.13–16 They do not evaluate the impact on errors from moving from paper to computerised processes, akin to the implementation of electronic prescribing. In addition, there are few studies which have looked at the views of staff relating to the adoption of new technologies, and none which use qualitative interviewing methodology to explore views in depth.

Undetected errors put patients at risk of serious harm.11 Additionally, errors can directly or indirectly have an impact on staff.17 18 Involvement in errors can result in increased stress levels, decreased job satisfaction,19 20 and subsequent increases in sick leave and staff turnover. Workplace stress is ‘a major cause of occupational ill health, poor productivity and human error’.20 The Health and Safety Executive reported an estimated 1.4 million work-related ill-health cases in 2018/19. Of these, 44% were related to stress, depression or anxiety resulting in 12.8 million lost working days. In addition, recording, reporting and rectifying errors takes time, affecting workflow and slowing down service delivery, with community pharmacists reporting that patients become frustrated when there is a delay.21 22

The cause and impact of errors can become cyclical due to a switch of focus on error correction rather than prevention. Two UK based qualitative studies explored community pharmacists’ experiences of errors, work environment, workload and stress levels.23 24 A number of the pharmacists interviewed considered the implications of these factors on increased risk of error: ‘so much pressure on the staff they stress them out, pressure and stress cause errors’.23