Therefore, should a patient undergo elective 12 h surgery on the weekend, which, whilst negating the twilight period in our study, would potentially be associated with the worst outcomes in our cohort? Morton and Snow 1 caution against limiting the large number of operations admitted electively to intensive care out-of-officehours, because we can only measure 'those patients operated on'. It is our belief that this cohort does not reflect patients denied surgery solely based on the availability of in-office-hour surgical lists, nor do we believe that greater in-office-hour surgical completion would result in a zero-sum game (i.e. delaying elective surgery to in-office-hours will result in more adverse outcomes). 5 Our results suggest that using out-of-office-hour intensive-care facilities as a vehicle to reduce waiting lists for elective major surgery may not be the safest or most cost-effective means. There are certainly other options in reducing waiting lists for elective surgery, including expanding the number of operating theatres and the development of high dependency units. We hope that our findings stimulate further research and consideration, especially given that the timing and order of elective surgical lists, in conjunction with a patient's perioperative workup, are theoretically amenable to strategic health system planning. Whilst we believe the political, administrative, and workforce priorities will always mandate a need for out-ofoffice-hour elective procedures, the one-third of all planned cases requiring intensive care services are well above the 1e8% of planned surgical cases returning to ward-based care during out-of-office-hour periods. 6,7 We are grateful to Morton and Snow 1 for encouraging further debate in this important area of surgical, anaesthesia, and intensive care practice. Authors' contributions Morton & Snow wrote a letter to the BJA editors about our manuscript and we are responding. It will make sense when the letters are placed in order and the BJA provide a reference or doi for the Morton & Snow letter.

EditordDrug errors are defined as any unintended patient safety incident relating to the handling of drugs that could have led, or did lead, to harm. They are the most frequent critical incidents in anaesthesia. 1 Whilst in adults approximately one-third of errors lead to harm, life-threatening errors are even more likely in paediatric anaesthesia. This is because of variable exposure of some anaesthetists to paediatric patients, age and weight variations, and the potential for large-scale errors. 2 The Paediatric Anaesthesia Trainee Research Network conducted a tablet device survey amongst anaesthetists attending the 2017 Association of Paediatric Anaesthetists of Great Britain and Ireland Annual Scientific Meeting. This evaluated the frequency of drug errors, reporting attitudes, and potential avenues for improvement. Of 354 delegates, there were 162 respondents (46%). They ranged in grade from core trainee (n¼4) to consultant (n¼99; 61%) with an approximate 50:50 split amongst those with more or less than five years of paediatric anaesthetic experience. A total of 60% of anaesthetists experienced paediatric drug errors at least once every year, 15% of whom reported experiencing an error at least once every month. These frequencies are likely to underestimate true frequencies because of reluctance to self-report errors and failure of error recognition. Calculation and dilution errors accounted for half of all errors. This was followed by the i.v. cannula not being flushed (16%) and wrong drug administration (11%). In order of frequency, antibiotics, opioids, paracetamol, and neuromuscular blocking agents were the most commonly involved drugs. The most common causative factors were distractions and interruptions whilst handling drugs, dose or dilution miscalculations, and anaesthetist fatigue. Amongst reporting attitudes, 36% of respondents stated that they would only report errors resulting in actual patient 4. Kaufmann J, Roth B, Engelhardt T, et al. Development and prospective federal statewide evaluation of a device for height-based dose recommendations in prehospital pediatric emergencies: a simple tool to prevent most severe drug errors.

No abstract

Lack of continuing education and physician anaesthetist support are commonly cited problems amongst Ethiopian anaesthetic providers. Whilst operating at Jimma University Medical Centre (JUMC), Operation Smile volunteers identified a clear need for improvement in anaesthetic care delivery at JUMC. JUMC is a 450-bed university teaching hospital 350 km southwest of Addis Ababa. At the start of this programme it had two physician anaesthetists, with the majority of anaesthesia historically having been provided by non-physician anaesthesia providers. A visiting lecturer programme was established at JUMC in 2012 following collaboration between two consultant anaesthetists, working for Operation Smile and JUMC respectively. UK trainee anaesthetists in their final years of anaesthetic training volunteered at JUMC for periods of two to six months, providing sustainable education and consistent physician anaesthetist presence to support service provision and training. Over its six-year history, nine visiting lecturers have volunteered at JUMC. They have helped establish a postgraduate training programme in anaesthesia, assisting in the provision of a future physician anaesthetist workforce. Four different training courses designed for low- and middle-income countries (LMICs) have been delivered and visiting lecturers have trained local anaesthetists in subsequent course delivery. Patient safety and quality improvement projects have included introducing the World Health Organization Surgical Safety Checklist, Lifebox pulse oximeters, obstetric spinal anaesthesia packs, improving critical care delivery and establishing two post-anaesthetic care units. Development of partnerships on local, national and global platforms were key to the effective delivery of relevant sustainable education and support. Instilling local ownership proved fundamental to implementing change in the local safety culture at JUMC. Sound mentorship from anaesthetic consultant supervisors both in the UK and in Jimma was crucial to support the UK trainee anaesthetists working in a challenging global setting. This model of sustainable capacity building in an LMIC with a significant deficit in its physician workforce could be replicated in a similar LMIC setting.