Working Hours in Anaesthesiology and their Effect on Patient Safety? A Review of “Effect on Patient Safety of a Resident Physician Schedule without 24-Hour Shifts”. N Engl J Med, 2020.

Since the introduction of mandatory compliance with the European Working Time Directive in 2009, healthcare organisations have had greater responsibility for overseeing physician working hours. Despite this, long working hours have remained a reality for many Irish trainees. The 2017 ‘Your Training Counts’ report from the Irish Medical Council reported that 33.5% of trainee doctors worked for 60 hours or more in a working week. The authors of the report identified a clear association between hours worked per week and self- rated quality of life. The College of Anaesthesiologists of Ireland has deemed physician wellbeing as a “crucial element to ensure good professional practice and delivery of the highest possible standard of care”, and have adopted a wellbeing programme as part of the College Strategy 2019-2024. And, many of our anaesthesiology departments have now transitioned to shift work for trainees, be it day/night shifts or on call shifts of 16 hours, as opposed to more traditional 24-hour shifts. Whilst much of the impetus for this appears to be physician wellbeing, concerns about a potential relationship between clinician fatigue and patient safety might also be regarded as a reasonable motivating factor for reducing shift hours. Proponents of reduced working hours cite the effects of fatigue upon decision-making and increased chance for medical error. Data from other industries have highlighted the adverse effects fatigue can have on human performance and decision making, with one systematic review (Wagstaff and Sigstad Lie, 2011) noting an increased risk of workplace accidents after 8 hours, and a doubling after 12 hours. However, this is not a straightforward narrative. There are those who argue the contrary, who suggest that longer shifts provide improved continuity in patient care and enhanced safety as a consequence, through reducing the chance for error because of multiple inputs or the risks associated with handover. Thus, in reality, the relationship between trainee shift patterns and their effect on patient safety is a complex matter. The ROSTERS study group (Landrigan et al., 2020) at Boston Children’s Hospital attempted to investigate the issue, and published their findings in NEJM in June, 2020. They conducted a 4 year multicentre cluster- randomised crossover trial in six paediatric intensive care units over four years, with paediatric residents randomised to either shifts of 24 hours or more (control) or 16 hours or less (intervention). Contrary to the investigator’s purported hypothesis, the primary outcome of ‘serious medical error’ per admission was reported as being significantly higher in the intervention group i.e. those working shorter shifts (181.3 vs. 131.5 per 1000 patient-days; relative risk, 1.56; 95% CI, 1.43 to 1.71). This was despite the intervention group recording lower mean weekly working hours and, higher mean weekly sleep hours than those working on the control schedule. At first glance, those in favour of reduced working hour shifts will be disappointed at this finding. On closer inspection, however, this trial has many flaws that make interpretation of its results difficult. For example, the authors acknowledge that physicians in the intervention arm were responsible for a higher mean number of patients, indicating that the workload of both groups was not comparable in the first instance. Indeed, when a secondary analysis included adjustment for the number of patients per resident physician as a potential confounder, there was no longer a difference in errors. Other issues with the study included a non-blinding of the data collectors, and a high variability in the primary outcome across hospital sites. It is likely that a much larger sample size is required to draw upon any significant conclusions. In essence, the title of the study could equally have stated that physicians are more likely to make a medical error when they are responsible for a higher number of patients, as opposed to being on a shorter shift. Medical error often results from the highly complex interactions of multiple processes, with failure at many levels of an organisation. In general, the relationship of medical error to duration of physician shifts requires the accounting for the multitude of other confounding factors. Although of questionable methodology, the publication of this “headlined as negative trial” further serves to highlight that this is an important and topical issue. The concept of reducing physician working hours in the name of patient safety is a relatively new one. Prior to this study, few investigators (Bilimoria et al., 2016) (Silber et al., 2019) have looked at the impact of trainee shift duration upon patient safety, and these have shown no conclusive benefit with elimination of extended shifts. Thus, whilst there remains no established causal relationship between physician working hours and patient safety, this topic is particularly relevant to Irish anaesthesiology trainees. One cannot ignore the potential serious consequences of medical errors and, if the incidence of these could be reduced by changing physician working patterns, then it is certainly worth further examination. and many important questions arise from this study: Does an increased frequency of handover associated with shorter shifts lead to a disruption in continuity of patient care? Does this necessarily compromise patient safety? Or is it the quality of hand over which is important? Do we simply (or not so simply, as it may be) need more doctors on call, thus generating a more manageable workload? Or are extended shifts the correct way to practise all along? In the future, hospitals will undoubtedly face further pressure from trainees, other organisations and the general public to comply with working hour directives. For now, however, with inconclusive evidence on patient safety and a below European average doctor to patient ratio, it appears long working shifts will persist for many Irish trainees. Whilst it is easy to get lost in the debate over what the optimal shift pattern might look like, we should not lose sight of the inextricable relationship between physician wellbeing and patient safety. Looking after ourselves and each other seems like a good place to start.