Overnight staffing in Canadian neonatal and pediatric intensive care units.

Aim: Infants and children who require specialized medical attention are admitted to neonatal and pediatric intensive care units (ICUs) for continuous and closely supervised care. Overnight in-house physician coverage is frequently considered the ideal staffing model. It remains unclear how often this is achieved in both pediatric and neonatal ICUs in Canada. The aim of this study is to describe overnight in-house physician staffing in Canadian pediatric and level-3 neonatal ICUs (NICUs) in the pre-COVID-19 era. Methods: A national cross-sectional survey was conducted in 34 NICUs and 19 pediatric ICUs (PICUs). ICU directors or their delegates completed a 29-question survey describing overnight staffing by resident physicians, fellow physicians, nurse practitioners, and attending physicians. A comparative analysis was conducted between ICUs with and without in-house physicians. Results: We obtained responses from all 34 NICUs and 19 PICUs included in this study. A total of 44 ICUs (83%) with in-house overnight physician coverage provided advanced technologies, such as extracorporeal life support, and included all ICUs that catered to patients with cardiac, transplant, or trauma conditions. Residents provided the majority of overnight coverage, followed by the Critical Care Medicine fellows. An attending physician was in-house overnight in eight (15%) out of the 53 ICUs, seven of which were NICUs. Residents participating in rotations in the ICU would often have rotation durations of less than 6 weeks and were often responsible for providing care during shifts lasting 20-24 h. Conclusion: Most PICUs and level-3 NICUs in Canada have a dedicated in-house physician overnight. These physicians are mainly residents or fellows, but a notable variation exists in this arrangement. The potential effects on patient outcomes, resident learning, and physician satisfaction remain unclear and warrant further investigation.

Evaluating the Validity of the Responses to Illness Severity Quantification Score to Discriminate Illness Severity and Level of Care Transitions in Hospitalized Children with Severe Acute Malnutrition.

OBJECTIVE: To evaluate the validity of the Responses to Illness Severity Quantification (RISQ) score to discriminate illness severity and transitions between levels of care during hospitalization. STUDY DESIGN: A prospective observational study conducted in Maiduguri, Nigeria, enrolled inpatients aged 1-59 months with severe acute malnutrition. The primary outcome was the RISQ score associated with the patient state. Heart and respiratory rate, oxygen saturation, respiratory effort, oxygen use, temperature, and level of consciousness are summed to calculate the RISQ score. Five states were defined by levels of care and hospital discharge outcome. The states were classified hierarchically, reflecting illness severity: hospital mortality was the most severe state, then intensive care unit (ICU), care in the stabilization phase (SP), care in the rehabilitation phase (RP), and lowest severity, survival at hospital discharge. A multistate statistical model examined performance of the RISQ score in predicting clinical states and transitions. RESULTS: Of 903 children enrolled (mean age, 14.6 months), 63 (7%) died. Mean RISQ scores during care in each phase were 3.5 (n = 2265) in the ICU, 1.7 (n = 6301) in the SP, and 1.5 (n = 2377) in the RP. Mean scores and HRs for a 3-point change in score at transitions: ICU to death, 6.9 (HR, 1.80); SP to ICU, 2.8 (HR, 2.00); ICU to SP, 2.0 (HR, 0.5); and RP to discharge, 1.4 (HR, 0.91). CONCLUSIONS: The RISQ score can discriminate between points of escalation or de-escalation of care and reflects illness severity in hospitalized children with severe acute malnutrition. Evaluation of clinical implementation and demonstration of benefit will be important before widespread adoption.

A Retrospective Study of Complications of Enteral Feeding in Critically Ill Children on Noninvasive Ventilation.

The utilization of noninvasive ventilation (NIV) in pediatric intensive care units (PICUs), to support children with respiratory failure and avoid endotracheal intubation, has increased. Current guidelines recommend initiating enteral nutrition (EN) within the first 24-48 h post admission. This practice remains variable among PICUs due to perceptions of a lack of safety data and the potential increase in respiratory and gastric complications. The objective of this retrospective study was to evaluate the association between EN and development of extraintestinal complications in children 0-18 years of age on NIV for acute respiratory failure. Of 332 patients supported with NIV, 249 (75%) were enterally fed within the first 48 h of admission. Respiratory complications occurred in 132 (40%) of the total cohort and predominantly in non-enterally fed patients (60/83, 72% vs. 72/249, 29%; p < 0.01), and they occurred earlier during ICU admission (0 vs. 2 days; p < 0.01). The majority of complications were changes in the fraction of inspired oxygen (220/290, 76%). In the multivariate evaluation, children on bilevel positive airway pressure (BiPAP) (23/132, 17% vs. 96/200, 48%; odds ratio [OR] = 5.3; p < 0.01), receiving a higher fraction of inspired oxygen (FiO2) (0.42 vs. 0.35; OR = 6; p = 0.03), and with lower oxygen saturation (SpO2) (91% vs. 97%; OR = 0.8; p < 0.01) were more likely to develop a complication. Time to discharge from the intensive care unit (ICU) was longer for patients with complications (11 vs. 3 days; OR = 1.12; p < 0.01). The large majority of patients requiring NIV can be enterally fed without an increase in respiratory complications after an initial period of ICU stabilization.

Furosemide in pediatric intensive care: a retrospective cohort analysis.

Introduction: Furosemide is the most commonly used medication in pediatric intensive care. Growing data indicates improved hemodynamic stability and efficacy of furosemide infusions compared to intermittent injections, thereby suggesting furosemide infusions might be considered as first line therapy in critically ill, paediatric patients. The objective of this study is to examine furosemide treatment as either continuous infusions or intermittent injections and subsequent patient outcomes. Methods: This is a retrospective cohort analysis of patients treated in a pediatric intensive care unit (ICU) over a nine year period (July 31st 2006 and July 31, 2015). Eligible patients were admitted to either the general pediatric or cardiac specific ICU for a duration of at least 6 hours and who received intravenous furosemide treatment. Results: A total of 7,478 patients were identified who received a total of 118,438 furosemide administrations for a total of 113,951 (96%) intermittent doses and 4,487 (4%) infusions running for a total of 1,588,750 hours. A total of 5,996 (80%) patients received exclusively furosemide injections and 1,482 (20%) patients received at least one furosemide infusion. A total of 193 patients died during ICU admission, amounting to 87 (6%) of the 1,482 patients who received an infusion and 106 (2%) of the 5,996 who received intermittent injections. Multivariable regression analysis showed no statistically significant decrease in adjusted mortality for patients who received furosemide injections compared to furosemide infusions (aOR 1.20, CI 0.76-1.89). Discussion: This retrospective study observed similar mortality for patients who received furosemide infusions compared to furosemide injections. More research on furosemide in the ICU could provide insights on fluid management, drug effectiveness, and pharmacologic stewardship for critically ill children.

Implementing a complex hospital innovation: conceptual underpinnings, program design and implementation of a complex innovation in an international multi-site hospital trial.

BACKGROUND: Designing implementation programs that effectively integrate complex healthcare innovations into complex settings is a fundamental aspect of knowledge translation. We describe the development of a conceptually grounded implementation program for a complex healthcare innovation and its subsequent application in pediatric hospital settings. METHODS: We conducted multiple case observations of the application of the Phased Reciprocal Implementation Synergy Model (PRISM) framework in the design and operationalization of an implementation program for a complex hospital wide innovation in pediatric hospital settings. RESULTS: PRISM informed the design and delivery of 10 international hospital wide implementations of the complex innovation, BedsidePEWS. Implementation and innovation specific goals, overarching implementation program design principles, and a phased-based, customizable, and context responsive implementation program including innovation specific tools and evaluation plans emerged from the experience. CONCLUSION: Theoretically grounded implementation approaches customized for organizational contexts are feasible for the adoption and integration of this complex hospital-wide innovation. Attention to the fitting of the innovation to local practices, setting, organizational culture and end-user preferences can be achieved while maintaining the integrity of the innovation.

Risk factors for unplanned paediatric intensive care unit admission after anaesthesia-an international multicentre study.

Objectives: Unplanned intensive care unit (ICU) admissions are associated with near-miss events, morbidity, and mortality. We describe the rate, resource utilization, and outcomes of paediatric patients urgently admitted directly to ICU post-anaesthesia compared to other sources of unplanned ICU admissions. Methods: We performed a secondary analysis of data from specialist paediatric hospitals in 7 countries. Patients urgently admitted to the ICU post-anaesthesia were combined and matched with 1 to 3 unique controls from unplanned ICU admissions from other locations by age and hospital. Demographic, clinical, and outcome variables were compared using the Wilcoxon rank-sum test for continuous variables and chi-square or Fisher's exact test for categorical variables. The effect of admission sources on binary outcomes was estimated using univariable conditional logistic regression models with stratification by matched set of anaesthesia and non-anaesthesia admission sources. Results: Most admissions were <1 year of age and for respiratory reasons. Admissions post-anaesthesia were shorter, occurred later in the day, and were more likely to be mechanically ventilated. Admissions post-anaesthesia were less likely to have had a previous ICU admission (4.8% compared to 11%, P=0.032) or PIM 'high-risk diagnosis' (9.5% versus 17.2%, P=0.035) but there was no difference in the number of subsequent ICU admissions. There was no difference in the PIM severity of illness score and no mortality difference between the groups. Conclusions: Young children and respiratory indications dominated unplanned ICU admissions post-anaesthesia, which was more likely later in the day and with mechanical ventilation.

Escalation of care in children at high risk of clinical deterioration in a tertiary care children's hospital using the Bedside Pediatric Early Warning System.

BACKGROUND: Escalation and de-escalation are a routine part of high-quality care that should be matched with clinical needs. The aim of this study was to describe escalation of care in relation to the occurrence and timing of Pediatric Intensive Care Unit (PICU) admission in a cohort of pediatric inpatients with acute worsening of their clinical condition. METHODS: A monocentric, observational cohort study was performed from January to December 2018. Eligible patients were children: 1) admitted to one of the inpatient wards other than ICU; 2) under the age of 18 years at the time of admission; 3) with two or more Bedside-Paediatric-Early-Warning-System (BedsidePEWS) scores ≥ 7 recorded at a distance of at least one hour and for a period of 4 h during admission. The main outcome -the 24-h disposition - was defined as admission to PICU within 24-h of enrolment or staying in the inpatient ward. Escalation of care was measured using an eight-point scale-the Escalation Index (EI), developed by the authors. The EI was calculated every 6 h, starting from the moment the patient was considered eligible. Analyses used multivariate quantile and logistic regression models. RESULTS: The 228 episodes included 574 EI calculated scores. The 24-h disposition was the ward in 129 (57%) and the PICU in 99 (43%) episodes. Patients who were admitted to PICU within 24-h had higher top EI scores [median (IQR) 6 (5-7) vs 4 (3-5), p < 0.001]; higher initial BedsidePEWS scores [median (IQR) 10(8-13) vs. 9 (8-11), p = 0.02], were less likely to have a chronic disease [n = 62 (63%) vs. n = 127 (98%), p < 0.0001], and were rated by physicians as being at a higher risk of having a cardiac arrest (p = 0.01) than patients remaining on the ward. The EI increased over 24 h before urgent admission to PICU or cardiac arrest by 0.53 every 6-h interval (CI 0.37-0.70, p < 0.001), while it decreased by 0.25 every 6-h interval (CI -0.36-0.15, p < 0.001) in patients who stayed on the wards. CONCLUSION: Escalation of care was related to temporal changes in severity of illness, patient background and environmental factors. The EI index can improve responses to evolving critical illness.

Development and an initial validation of the Responses to Illness Severity Quantification (RISQ) score for severely malnourished children.

AIM: To develop and perform an initial validation of a score to measure the severity of illness in hospitalised children with severe acute malnutrition (SAM). METHODS: A prospective study enrolled SAM children aged 6-59 months hospitalised in Borno State, Nigeria. Candidate items associated with inpatient mortality were combined and evaluated as candidate scores. Clinical and statistical methods were used to identify a preferred score. RESULTS: The 513 children enrolled had a mean age of 15.6 months of whom 48 (9%) died. Seven of the 10 evaluated items were significantly associated with mortality. Five different candidate scores were tested. The final score, Responses to Illness Severity Quantification (RISQ), included seven items: heart rate, respiratory rate, respiratory effort, oxygen saturation, oxygen delivery, temperature and level of consciousness. The mean RISQ score on admission was 2.6 in hospital survivors and 7.3 for children dying <48 h. RISQ scores <24 h before death had an area under the receiver operating characteristic curve (AUROC) of 0.93. The RISQ score performed similarly across differing clinical conditions with AUROCs 0.77-0.98 for all conditions except oedema. CONCLUSION: The RISQ score can identify high-risk malnourished children at and during hospital admission. Clinical application may help prioritise care and potentially improve survival.

Dosing of enteral acetaminophen in critically ill children: a cohort study.

OBJECTIVE: Acetaminophen is the most common medication prescribed in children's hospitals. The aim of the study was to estimate the frequency and risk factors for acetaminophen underdosing and overdosing in the paediatric intensive care unit (PICU). DESIGN: Retrospective cohort of drug administrations in a large tertiary care PICU. PATIENTS: All PICU admissions, less than 18 years of age, admitted between 1 January 2008 and 1 January 2018, having received at least one dose of enteral acetaminophen. METHODS: The primary outcome was acetaminophen underdosing and overdosing, defined as doses exceeding the 10% upper and lower limits of the standard reference range (10-15 mg/kg) and 10% above daily maximum dose (75 mg/kg). A generalised estimating equation regression assessed patient risk factors for single underdosing, single overdosing and cumulative daily overdosing of acetaminophen. RESULTS: Of the 147 485 doses of enteral acetaminophen administered, 7814 (5.3%) were single underdoses (1 in every 19 doses) and 4640 (3.1%) were single overdoses (1 in every 32 doses). There were 6813 cumulative overdose days (1 in every 9 patient-days). Risk factors for both underdosing and overdosing included older age and cardiac admission, whereas risk factors for cumulative overdosing were young age and cardiac admission. Electronic prescribing increased the risk of underdosing and overdosing, but decreased cumulative acetaminophen overdosing (relative risk 0.51, p=0.001). CONCLUSION: Acetaminophen underdosing and overdosing are common in the PICU and can be detected with pharmacoepidemiology. Electronic prescribing increased the risk of single underdosing and overdosing, although it reduced the risk of cumulative overdosing.

The role of hospital characteristics in patient safety: a protocol for a national cohort study.

BACKGROUND: Substantial expenditures on health care safety programs have been justified by their goal of reducing health care associated-harm (adverse events), but adverse event rates have not changed over the past 4 decades. The objective of this study is to describe hospital-level factors that are relevant to safety in Canadian hospitals and the impact of these factors on hospital adverse events. METHODS: This is a protocol for a national cohort study to describe the association between hospital-level factors and adverse events. We will survey at least 90 (35%) Canadian hospitals to describe 4 safety-relevant domains, chosen based on the literature and expert consultation, namely patient safety culture, safety strategies, staffing, and volume and capacity. We will retrospectively identify hospital adverse events from a national data source. We will evaluate organization-level factors using established scales and a survey, codesigned by the study team and hospital leaders. Hospital leaders, clinical unit leaders and front-line staff will complete the surveys once a year for 3 years, with an anticipated start date of winter 2022. We will use national health administrative data to estimate the rate and type of hospital adverse events corresponding to each 1-year survey period. INTERPRETATION: Analysis of data from this project will describe hospital organizational factors that are relevant to safety and help identify organizational initiatives that improve hospital patient safety. In addition to biyearly reports to the leaders of the participating hospitals, we have a multifaceted and tailored dissemination strategy that includes integrating the knowledge users into the study team to increase the likelihood that our study will lead to improved hospital patient safety.

Intensive Care Unit Built Environments: A Comprehensive Literature Review (2005-2020).

BACKGROUND: The intensive care environment in hospitals has been the subject of significant empirical and qualitative research in the 2005-2020 period. Particular attention has been devoted to the role of infection control, family engagement, staff performance, and the built environment ramifications of the recent COVID-19 global pandemic. A comprehensive review of this literature is reported summarizing recent advancements in this rapidly expanding body of knowledge. PURPOSE AND AIM: This comprehensive review conceptually structures the recent medical intensive care literature to provide conceptual clarity and identify current priorities and future evidence-based research and design priorities. METHOD AND RESULT: Each source reviewed was classified as one of the five types-opinion pieces/essays, cross-sectional empirical investigations, nonrandomized comparative investigations, randomized studies, and policy review essays-and into nine content categories: nature engagement and outdoor views; family accommodations; intensive care unit (ICU), neonatal ICU, and pediatric ICU spatial configuration and amenity; noise considerations; artificial and natural lighting; patient safety and infection control; portable critical care field hospitals and disaster mitigation facilities including COVID-19; ecological sustainability; and recent planning and design trends and prognostications. CONCLUSIONS: Among the findings embodied in the 135 literature sources reviewed, single-bed ICU rooms have increasingly become the norm; family engagement in the ICU experience has increased; acknowledgment of the therapeutic role of staff amenities; exposure to nature, view, and natural daylight has increased; the importance of ecological sustainability; and pandemic concerns have increased significantly in the wake of the coronavirus pandemic. Discussion of the results of this comprehensive review includes topics noticeably overlooked or underinvestigated in the 2005-2020 period and priorities for future research.

Alterations in coordinated EEG activity precede the development of seizures in comatose children.

OBJECTIVE: We aimed to test the hypothesis that computational features of the first several minutes of EEG recording can be used to estimate the risk for development of acute seizures in comatose critically-ill children. METHODS: In a prospective cohort of 118 comatose children, we computed features of the first five minutes of artifact-free EEG recording (spectral power, inter-regional synchronization and cross-frequency coupling) and tested if these features could help identify the 25 children who went on to develop acute symptomatic seizures during the subsequent 48 hours of cEEG monitoring. RESULTS: Children who developed acute seizures demonstrated higher average spectral power, particularly in the theta frequency range, and distinct patterns of inter-regional connectivity, characterized by greater connectivity at delta and theta frequencies, but weaker connectivity at beta and low gamma frequencies. Subgroup analyses among the 97 children with the same baseline EEG background pattern (generalized slowing) yielded qualitatively and quantitatively similar results. CONCLUSIONS: These computational features could be applied to baseline EEG recordings to identify critically-ill children at high risk for acute symptomatic seizures. SIGNIFICANCE: If confirmed in independent prospective cohorts, these features would merit incorporation into a decision support system in order to optimize diagnostic and therapeutic management of seizures among comatose children.

Objective Sleep Characteristics and Factors Associated With Sleep Duration and Waking During Pediatric Hospitalization.

Importance: Children's sleep may be affected by hospitalization, yet few objective determinations of sleep patterns are reported for children in intensive care or general medicine units. There is limited research on relationships between sleep in hospital and child (eg, age, pain), treatment (eg, medications, nurse presence), or environmental (eg, noise, light, type of unit) factors. Objective: To determine sleep quantity and patterns in hospitalized children and determine factors associated with sleep quantity and nighttime waking for children in hospital. Design, Setting, and Participants: This was a prospective cross-sectional study of children admitted to a general pediatric unit or a pediatric intensive care unit at a pediatric quaternary teaching hospital in Toronto, Ontario, Canada, from October 2007 to July 2008. Participants included children aged 1 to 18 years who were expected to stay in hospital for at least 2 nights. Demographic data, information about the hospital stay and illness, and usual sleep habits were collected. Children wore an actigraph for 1 to 3 consecutive days and nights and completed a sleep diary. Sound and light meters were placed at the bedside. Data analyses took place in April 2009. Main Outcomes and Measures: The primary outcome was the mean number of minutes of child nighttime sleep from 7:30 pm to 7:29 am. Sleep variables were averaged over days and nights recorded (mean [SD] days and nights of wear, 2.54 [0.71]) and examined for associations with sleep quantity and patterns, as well as hazard of waking in the night. Results: Of 124 eligible children approached for inclusion, 69 children consented (35 [51%] female; 20 [29%] aged 1-3 years, 10 [14%] aged 4-7 years, 17 [24%] aged 8-12 years, and 22 [32%] aged 13-18 years; 58 [84%] in the general pediatric unit). Children aged 1 to 3, 4 to 7, 8 to 12, and 13 to 18 years obtained a mean (SD) of 444 (132), 475 (86), 436 (114), and 384 (83) minutes of nighttime sleep, respectively; mean (SD) number of night awakenings was 14 (3), 18 (3), 14 (8), and 12 (6), respectively. Children on general pediatric units slept 258 minutes more per night than children sleeping in the pediatric intensive care unit (95% CI, 165.16-350.56 minutes; P < .001), children admitted for planned surgery slept 123 minutes more than children admitted for exacerbations of chronic illness (95% CI, 49.23-196.01 minutes; P < .01), and children sleeping in rooms with other patients slept 141 minutes fewer than children in private rooms (95% CI, -253.51 to -28.35 minutes; P = .01). Sound events greater than 80 dB were significantly associated with increased risk of instantaneous waking (hazard ratio [HR], 1.35; 95% CI, 1.02-1.80; P = .04), as were light events greater than 150 lux (HR, 1.17; 95% CI, 1.01-1.36; P = .03), receiving a medication that promoted sleep (HR, 1.04; 95% CI, 1.00-1.08; P = .03), and having a nurse in the room for most or all of the night (HR, 1.08; 95% CI, 1.03-1.13; P = .003). Sleeping on the general pediatrics unit was significantly associated with decreased risk of instantaneous waking (HR, 0.81; 95% CI, 0.77-0.85; P < .001), as was being admitted for planned surgery (HR, 0.95; 95% CI, 0.91-0.99; P = .04), receiving a medication that promoted wakefulness (HR, 0.96; 95% CI, 0.93-0.995; P = .02), and sharing a room with another patient (HR, 0.78; 95% CI, 0.72-0.84; P < .001). Conclusions and Relevance: In this cross-sectional study of hospitalized children, children experienced considerable nighttime waking and sleep restriction to levels below national clinical recommendations at a time when they most needed the benefits of sleep. Given light and noise were the greatest contributors to nighttime waking in hospital, clinicians, administrators and hospital design experts should work together for solutions.

Moral Distress and Other Wellness Measures in Canadian Critical Care Physicians.

Rationale: Understanding the magnitude of moral distress and its associations may point to solutions. Objectives: To understand the magnitude of moral distress and other measures of wellness in Canadian critical care physicians, to determine any associations among these measures, and to identify potentially modifiable factors. Methods: This was an online survey of Canadian critical care physicians whose e-mail addresses were registered with either the Canadian Critical Care Society or the Canadian Critical Care Trials Group. We used validated measures of moral distress, burnout, compassion fatigue, compassion satisfaction, and resilience. We also measured selected individual, practice, and workload characteristics. Results: Of the 499 physicians surveyed, 239 (48%) responded and there were 225 usable surveys. Respondents reported moderate scores of moral distress (107 ± 59; mean ± standard deviation, maximum 432), one-third of respondents had considered leaving or had previously left a position because of moral distress, about one-third met criteria for burnout syndrome, and a similar proportion reported medium-high scores of compassion fatigue. In contrast, about one-half of respondents reported a high score of compassion satisfaction, and overall, respondents reported a moderate score of resilience. Each of the "negative" wellness measures (moral distress, burnout, and compassion fatigue) were associated directly with each of the other "negative" wellness measures, and inversely with each of the "positive" wellness measures (compassion satisfaction and resilience), but moral distress was not associated with resilience. Moral distress was lower in respondents who were married or partnered compared with those who were not, and the prevalence of burnout was lower in respondents who had been in practice for longer. There were no differences in any of the wellness measures between adult and pediatric critical care physicians. Conclusions: Canadian critical care physicians report moderate scores of moral distress, burnout, and compassionate fatigue, and moderate-high scores of compassion satisfaction and resilience. We found no modifiable factors associated with any wellness measures. Further quantitative and qualitative studies are needed to identify interventions to reduce moral distress, burnout, and compassion fatigue.

In-House, Overnight Physician Staffing: A Cross-Sectional Survey of Canadian Adult ICUs.

OBJECTIVES: Overnight physician staffing in the ICU has been recommended by the Society of Critical Care Medicine and the Leapfrog Consortium. We conducted a survey to review practice in the current era and to compare this with results from a 2006 survey. DESIGN: Cross-sectional survey. SETTING: Canadian adult ICUs. PARTICIPANTS: ICU directors. INTERVENTIONS: None. MEASUREMENT AND MAIN RESULTS: A 29-question survey was sent to ICU directors describing overnight staffing by residents, fellows, nurse practitioners, and staff physicians, as well as duty duration, clinical responsibilities, and unit characteristics. We established contact with 122 ICU directors, of whom 107 (88%) responded. Of the 107 units, 60 (56%) had overnight in-house physicians. Compared with ICUs without overnight in-house physician coverage, ICUs with in-house physicians were in larger hospitals (p < 0.0001), had more beds (p < 0.0001), had more ventilated patients (p < 0.0001), and had more admissions (p < 0.0001). Overnight in-house physicians were first year residents (R1) in 20 of 60 (33%), second to fifth year residents (R2-R5) in 46 of 60 (77%), and Critical Care Medicine trainees in 19 of 60 (32%). Advanced practice nurses provided overnight coverage in four of 107 ICUs (4%). The most senior in-house physician was a staff physician in 12 of 60 ICUs (20%), a Critical Care Medicine trainee in 14 of 60 (23%), and a resident (R2-R5) in 20 of 60 (33%). The duration of overnight duty was on average 20-24 hours in 22 of 46 units (48%) with R2-R5 residents and 14 of 19 units (74%) covered by Critical Care Medicine trainees. CONCLUSIONS: Variability of in-house overnight physician presence in Canadian adult ICUs is linked to therapeutic complexity and unit characteristics and has not changed significantly over the decade since our 2006 survey. Additional evidence about patient and resident outcomes would better inform decisions to revise physician scheduling in Canadian ICUs.