TL;DR
This randomized trial investigated whether shorter resident duty hours in intensive care units (ICUs) improve patient safety and resident well-being. Contrary to expectations, the study found no significant differences in adverse events or resident sleepiness across 24-, 16-, and 12-hour shifts, suggesting that the purported benefits of shorter hours may not materialize and that trade-offs exist.
What Problem Does This Paper Address?
The medical field has been grappling with the issue of physician fatigue, particularly among residents, due to long working hours. This fatigue is linked to increased medical errors and compromised patient safety. In response, regulatory bodies have increasingly mandated shorter resident duty hours, with the aim of improving both patient care and resident well-being. However, a key concern is that more frequent handovers associated with shorter shifts could lead to information loss and continuity-related errors, potentially negating the benefits of reduced fatigue. This study aimed to directly evaluate the net effect of different resident duty schedules (24, 16, and 12 hours) on patient safety, resident well-being, and continuity of care in the high-acuity environment of the ICU.
Methods
The researchers conducted a randomized trial in two academic medical-surgical ICUs in Toronto, Ontario. Residents were randomly assigned to one of three in-house overnight duty schedules: 24-hour, 16-hour, or 12-hour shifts. These assignments were made in 2-month rotation blocks. The primary patient outcome measured was adverse events, defined as any unplanned injury arising from medical care that led to morbidity, required treatment, prolonged hospital stay, or resulted in disability. The primary resident outcome was sleepiness, assessed using the 7-point Stanford Sleepiness Scale. Secondary outcomes included patient deaths, preventable adverse events, residents' physical symptoms, burnout, continuity of care, and ICU staff perceptions of resident performance. Adverse events were identified through a multimodal screening process involving nurse observation, voluntary reporting, and medical chart review, followed by physician verification. Sleepiness was measured every four hours using the Stanford Sleepiness Scale. Somatic symptoms and burnout were assessed using validated questionnaires.
Datasets
Not discussed in the paper.
Key Insights and Ideas
- Shorter hours, same safety? The study's most striking finding is the lack of a significant difference in overall adverse event rates across the different duty schedules. This challenges the assumption that simply reducing hours automatically translates to improved patient safety.
- Trade-offs are real: While overall adverse events didn't change, the study did reveal potential trade-offs. Residents on the 24-hour schedule reported more physical symptoms, and ICU staff perceived residents on the 16-hour schedule as less familiar with patient details and making lower-quality decisions.
- Preventable errors lean towards shorter shifts: A concerning signal emerged with preventable adverse events, with seven out of eight occurring during the 12-hour schedule. While not statistically significant due to the low number of events, this warrants further investigation.
- Sleepiness is complex: Resident sleepiness didn't significantly differ between schedules, and it increased at night, consistent with circadian rhythms. This suggests that time of day might be a more dominant factor in fatigue than shift duration alone.
- Burnout is schedule-independent: The study found no significant effect of duty schedule on resident burnout, as measured by the Maslach Burnout Inventory. However, a decrease in the personal accomplishment domain was observed over the course of the ICU rotation, suggesting that burnout's impact might be felt over time rather than being acutely influenced by daily shift length.
Key Results
- Adverse Events: No significant difference in adverse event rates per 1000 patient-days across the 24-hour (81.3), 16-hour (76.3), and 12-hour (78.2) schedules (p = 0.7).
- Resident Sleepiness: No significant differences in daytime (mean rating 2.33, 2.61, 2.30; p = 0.3) or nighttime (mean rating 3.06, 2.73, 2.42; p = 0.2) sleepiness ratings between schedules.
- Preventable Adverse Events: Seven of eight preventable adverse events occurred with the 12-hour schedule (p = 0.1).
- Physical Symptoms: Residents working the 24-hour schedule reported more severe and frequent somatic symptoms (p = 0.04).
- ICU Staff Perceptions: Staff rated residents' knowledge and decision-making worst with the 16-hour schedule (p = 0.002 for clinical details, p = 0.03 for social details, p = 0.02 for suboptimal decisions).
- Mortality: ICU mortality rates were similar across all three schedules (18.3% for 24-hour, 17.1% for 16-hour, 14.4% for 12-hour; p = 0.2).
- Burnout: Burnout levels were similar across all groups (p > 0.5).
Limitations and Future Work
The authors acknowledge several limitations:
- Lack of Blinding: The study design could not conceal the schedule from residents, potentially introducing bias in subjective responses. However, physician reviewers of adverse events were blinded to schedule allocation.
- Underpowered for Rare Events: The study may have been underpowered to detect smaller effects, particularly for rare outcomes like preventable adverse events. Larger studies are needed for more robust estimates.
- Continuity Measurement: Measuring continuity of care proved challenging, leading to its prioritization as a secondary outcome. Future work should address this limitation.
- Adherence Monitoring: Formal monitoring of resident adherence to schedules was not conducted, though anecdotal reports suggested many disliked the 12-hour overnight schedule due to social disconnection.
- Generalizability: The study was conducted in Canadian ICUs. While the findings are relevant to other regions, direct applicability may vary.
The authors suggest that more precise quantification of the effects of duty schedules will require larger randomized studies. They also highlight the need to further delineate the emerging signals regarding trade-offs between resident well-being and patient safety measures, particularly concerning preventable adverse events and staff perceptions of resident performance. Future research could also explore the qualitative aspects of resident experience, such as social contact and personal well-being, which were not fully captured by the quantitative measures.
Summary updated