For debate

Health services administration
Volume 200 - Issue 7

Safer hours for doctors and improved safety for patients

Authors:  Dev A S Kevat, Peter A Cameron, Andrew R Davies, Chris P Landrigan and Shantha W Rajaratnam

Med J Aust 2014; 200 (7): 396-398. || doi: 10.5694/mja13.10412
Published online: 21 April 2014
Safety concerns about sleep deprivation in a growing junior doctor workforce present a chance for roster reforms
Download PDF (120 KB)
Share: Share on Blue Sky Share on Facebook Share on LinkedIn Share by email

Abstract

  • An increasing weight of evidence is demonstrating that sleep deprivation and circadian rhythm disruption in doctors are associated with human error and harm to both patients and doctors.

  • The increasing junior doctor workforce entering the hospital system in Australia provides a rare opportunity for workplace and roster reforms.

  • There are cultural, educational and industrial challenges to reforming working hours.

  • Any changes should be evidence-based and monitored to ensure that training for junior doctors and patient care are not compromised.

An increasing weight of evidence is demonstrating that sleep deprivation and circadian (24-hour) rhythm disruption in doctors are associated with human error and harm to both patients and doctors. Sleep deprivation has been linked to a greater risk of surgical complications1 and unintercepted medical errors in intensive care units,2 as well as an increased rate of needlestick injuries3 and car accidents after shifts.4

In Australia, an increasing junior doctor workforce is entering the hospital system, with growth from 1335 domestic medical graduates in 2006 to a projected graduating cohort of more than 3200 in 2014.5 This unprecedented increase in domestic production of doctors has been a result of government policy to overcome workforce shortages. All medical training pathways involve accruing at least 2 years of hospital experience, with many requiring at least 5 years.

This combination of heightened safety concerns and surge in medical labour supply provides a rare opportunity to implement best-practice workplace and roster reforms. Any changes made should be evidence-based and monitored to ensure that training for junior doctors and continuity of patient care are not compromised. It is important to ensure that wider industrial agendas and cost-saving measures are not masked as improved patient safety initiatives.

Systematic reviews on specific work hour restrictions for doctors have been published,6,7 and the topic is the focus of international debate.8 Overseas measures, such as the European Union’s Working Time Directive, which seeks to limit all workers to an average of 48 working hours a week,9 may not be appropriate for doctors in the Australian setting. Here, we present recent evidence relating to doctor fatigue and patient harm in a relevant cultural and organisational context for Australia.

Sleep propensity and performance impairment

The closely overlapping states of sleepiness, drowsiness, fatigue, sleep deprivation and resultant increase in sleep propensity have been the subject of detailed research, which has quantified associated cognitive and physical impairments including decreased working memory, attention lapses, diminished motor control and microsleeps.10-12 Sleep propensity is the likelihood of falling asleep, often measured as the speed of transition from wakefulness to sleep. It is dependent on the time spent awake, recent sleep gained (particularly over the preceding 48 hours) and the time of day or, more specifically, the timing of the endogenous circadian pacemaker.13

Sleep propensity in doctors has been associated with an increased risk of many types of error, including diagnostic, medication and procedural mistakes.1,2 The impact of increasing sleep propensity on performance can be quantified using simulation technology, including colonoscopy, laparoscopy and other simulation scenarios.14 Standardised patient scenarios can also be used to examine doctors’ performance. Actual errors can be discerned through chart review or real-time observation.2

What is the minimum duration of sleep required to function well?

The amount of sleep required to function optimally is relatively consistent across the population. Increased sleep debt (due to chronic sleep deficiency) has been observed in individuals who claim to require less sleep than that recommended for the broader population.15 Studies in healthy volunteers have demonstrated that reducing the opportunity to sleep from 8 hours to 6 hours per night resulted in significant impairments in attention and cognitive performance; restriction to 4 hours resulted in even greater deficits.16 Participants were inaccurate in estimating their sleepiness levels, particularly when the deficits accumulated over several days — a concern in the hospital setting. A randomised prospective intervention study that eliminated extended work and on-call shifts of 24 hours or more resulted in increased sleep duration for junior doctors, along with reduced sleepiness and attention failures during wake time10 — 22% more serious errors were made by the control group compared with the intervention group.2

Recent evidence indicates that senior doctors also experience the effects of sleep deprivation. A study in a United States hospital assessed complication rates for consultant surgeons performing day-time procedures after spending a night on call. Surgeons who had received 6 or more hours of sleep opportunity recorded a complication rate of 3.4%, compared with a rate of 6.2% for the same surgeons when they had less than 6 hours’ sleep opportunity (odds ratio, 1.72).1

What should the maximum shift length be?

Neurobehavioural performance shows a sharp decline after about 16 hours of wakefulness,17 due to the interactive effect of the two fundamental processes that drive the sleep–wake cycle: homoeostatic sleep pressure and circadian rhythmicity. In the industrial setting, longer shift lengths of 10 or 12 hours are associated with a “dose-dependent” increase in the risk of accidents.18 Limited work has been conducted on the effect of medical shifts of 12 hours,19 although the deleterious effect of doctors working shifts longer than 16 hours has been more extensively studied.20

While most Australian jurisdictions do not roster doctors for shifts longer than 16 hours, long shifts followed by an on-call period are common. On-call periods may be a legal loophole to continued work and are a patient safety blind spot.21,22 Across weekends in particular, many trainee doctors are likely to be working at levels of sleep deprivation that place them at high risk of making errors that cause patient harm.21,22 At these times, in our opinion, doctors-in-training are also likely to be under the least direct supervision, be susceptible to self-censorship of seeking advice, and have poorer access to diagnostic services.

How well do doctors function on night shifts?

The alternative to having doctors on call is to have doctors rostered on night shifts. This practice usually requires more doctors, who are now being provided by increased graduate numbers. However, less than 3% of permanent night-shift workers demonstrate full circadian adjustment to the schedule.23 Research has found a 21% reduction in night-time performance compared with day-time performance in tertiary hospital emergency registrars presented with Fellowship examination standard scenarios.24 Some local and international authorities have argued against working seven consecutive night shifts.25,26 Due to the strong circadian drive for wakefulness during the daytime, sleep between night shifts is of lower quality and duration, effectively leading to an accumulating sleep debt over successive shifts. A meta-analysis has identified an increased risk of industrial accidents on the fourth night of four consecutive night shifts.18 To our knowledge, no studies have systematically examined the effect of successive night shifts on performance in the medical sphere.

Challenges and potential solutions

In considering roster reforms, real challenges exist in ensuring adequate training exposure and continuity of patient care. In the current Australian work paradigm, residents and registrars provide around-the-clock coverage of patients in public hospitals, including admitting many new patients who present out of hours. Consultants in most specialties work closer to core daylight hours (eg, 0700–1800) on weekdays and provide input outside these hours as needed. Registrars perform most out-of-hours surgical procedures (eg, appendicectomies). Specialties such as emergency medicine and intensive care now often provide evening consultant cover as the norm. As times of consultant cover provide the highest-value training time for junior doctors, a reduction in working hours can result in a lesser proportion of hours spent under direct consultant supervision — realistically, junior doctors will still have to provide most of the care around the clock.27 Careful planning is required to ensure sufficient high-value training hours. Workplace agreements, such as the Queensland Medical Officers’ Certified Agreement,28 are allowing hospitals to roster consultants to work evenings and weekend hours. It is likely this power will be exercised in coming years, expanding the number of high-value training hours. We believe that consultants will simultaneously need to be wary of “oversupervision” and of focusing solely on reaching service targets, rather than on training opportunities.

As trainee numbers increase, Colleges may need to better define and ration high-value training hours between trainees. Case numbers and patient exposure for junior doctors should be prospectively tracked and compared when rosters are being altered, to ensure College requirements regarding clinical experience and patient exposure are being met. Greater use of simulation learning may help offset the effects of less patient contact. Software programs can be useful for developing safer rosters that meet trainee and hospital expectations.29

As reduced shift lengths require a greater number of handovers, an improvement in handover practices and team delegation of tasks is key to successful roster reform and patient care. Lack of effective handover is a significant patient safety risk.30 Increased doctor numbers provide an opportunity for more doctors to be rostered in the evening and night hours, rather than the present often skeletal cover at these times. This could reduce long shifts, as well as the need for on-call duty. In particular, the increased numbers of Postgraduate Year 3 doctors and unaccredited registrars may allow a reduction in shift hours and improve the sleep opportunity for more senior doctors. However, care must be taken to rotate all doctors through day shifts, when greater educational opportunities and supervision currently occur.

According to a 2012 report from Health Workforce Australia, subspecialised areas may continue to struggle in providing 24-hour care due to the shortage of appropriately qualified doctors.31 In the United Kingdom, senior nurses have been used to alleviate this shortage by screening pages and calls.32 This has resulted in improved patient outcomes, but the practice is less common in Australia.

The importance of a protected period of rest was recognised by the US Institute of Medicine in its most recent guidelines, which included a recommendation of a 5-hour protected break during extended 30-hour shifts.25 The Australian Medical Association also has a code of practice for preventing and managing sleep deprivation.33 The Royal Australasian College of Surgeons has been proactive in publishing its own guidelines,26 and surgical trainees have reported that working about 60 hours a week is sufficient for training experience.34

We recommend that rostering processes should look beyond total working hours to include reviewing the need for prolonged on-call periods, as well as auditing out-of-hours calls to identify those that have low clinical and training significance and could be safely diverted to nursing staff or generalist night-shift doctors. Protocols for managing night-time calls (particularly during the hours of 0000–0600) and adequate escalation protocols will ultimately improve care for patients. In the UK, restructuring of after-hours care into teams that focus on required competencies, rather than teams based on traditional specialist competencies, have been shown to reduce unplanned transfers to coronary care and intensive care units from wards and surgical or medical assessment units (17% v 6%; P < 0.01).30 Although it would require substantial reorganisation and clinician support, we believe similar models could be of benefit in Australia to help manage the effects of the 4-hour National Emergency Access Target, provided implementation was tailored to local conditions. Restructuring night teams would allow generalist doctors and other health staff to manage most service needs, with escalation as clinically indicated.

Without planning and reform, the opportunity to effectively utilise an increasing workforce may be lost. While it may be easy to simply add more labour in outdated roster patterns, this will not reduce working hours, nor provide effective care for growing after-hours demand. It is time to bring work practices in line with the substantial body of evidence on safer working hours for doctors, which will provide better care for patients.

Authors

  • Dev A S Kevat MB BS, LLB, MPH1,2
  • Peter A Cameron MB BS, FACEM, MD1
  • Andrew R Davies MB BS, FRACP, FJFICM1
  • Chris P Landrigan MD, MPH3
  • Shantha W Rajaratnam BSc, LLB, PhD1
  • 1. Monash University, Melbourne, VIC.
  • 2. Royal Brisbane and Women’s Hospital, Brisbane, QLD.
  • 3. Harvard University, Boston, Mass, USA.

Correspondence: dev.kevat@monash.edu

Competing interests

Chris Landrigan has received honoraria from the US Committee of Interns and Residents (CIR) for grand rounds presentations on handovers, sleep deprivation and patient safety at CIR member hospitals. Shantha Rajaratnam has had travel expenses for attendance at scientific conferences reimbursed by Vanda Pharmaceuticals; is Immediate Past President of the Australasian Sleep Association; has performed consultancy work for the National Transport Commission, EdanSafe, Philips Respironics and the Australian Workers Union; and his institution has received research grants from the ResMed Foundation, Cephalon, Philips Respironics, Takeda Pharmaceuticals, and Vanda Pharmaceuticals. No grants were made for research or writing of this article.

References

Provenance: Not commissioned; externally peer reviewed.

View this article on Wiley Online Library