Increased Occurrence of Tracheal

IntubationAssociated Events During Nights

and Weekends in the PICU*
Kyle J. Rehder, MD, FCCP1; John S. Giuliano Jr, MD, FAAP2; Natalie Napolitano, MPH, RRT-NPS, FAARC3;
David A. Turner, MD, FCCM, FCCP1; Gabrielle Nuthall, MBChB, FRACP, CICM4;
Vinay M. Nadkarni, MD, FCCM5; Akira Nishisaki, MD, MSCE5; for the National Emergency Airway
Registry for Children and Pediatric Acute Lung Injury and Sepsis Investigators
*See also p. 2698.
Department of Pediatrics, Division of Pediatric Critical Care Medicine,
Duke Childrens Hospital, Durham, NC.

Department of Pediatrics, Division of Critical Care Medicine, Yale University School of Medicine, New Haven, CT.

Department of Nursing, Respiratory Care, and Neurodiagnostics, The

Childrens Hospital of Philadelphia, Philadelphia, PA.

Division of Pediatric Intensive Care, Starship Childrens Health Center,

Grafton, Auckland, New Zealand.

grant support from Nihon Kohden America (Evaluation of mainstream capnography). The remaining authors have disclosed that they do not have any
potential conflicts of interest.
Address requests for reprints to: Kyle Rehder, MD, Division of Pediatric
Critical Care, Duke Childrens Hospital, DUMC Box 3046, Durham, NC
27710. E-mail: kyle.rehder@duke.edu

Department of Anesthesiology and Critical Care Medicine, The Childrens Hospital of Philadelphia, Philadelphia, PA.

Drs. Rehder and Giuliano Jr shared first authorship.

This study was performed at 20 centers in the National Emergency Airway
for Children Registry, and data analysis was performed at Duke University Medical Center (Durham, NC) and Childrens Hospital of Philadelphia
(Philadelphia, PA).
Supplemental digital content is available for this article. Direct URL citations
appear in the printed text and are provided in the HTML and PDF versions
of this article on the journals website (http://journals.lww.com/ccmjournal).
For a full author list of National Emergency Airway Registry for Children (NEAR4KIDS) and Pediatric Acute Lung Injury and Sepsis Investigators (PALISI),
see Supplemental Digital Content 3 (http://links.lww.com/CCM/B502).
Supported, in part, by Endowed Chair, Critical Care Medicine, The Childrens
Hospital of Philadelphia, unrestricted research fund from Laerdal Foundation Acute Care Medicine, and Agency for Healthcare Research and Quality
(1R03HS021583-01 and 1 R18 HS022464-01).
Ms. Napolitano served as a board member for American Association for
Respiratory Care (AARC) and Allergy and Asthma Network (AAN) (compensation received only for travel to meetings), consulted for AAN (qualitative
project on decisions of families for asthma care), lectured for Draeger Medical, and received support for travel from AARC (for AARC Conference). Her
institution has a pending patent for noninvasive ventilation mask and received
grant support from the Agency for Healthcare Research and Quality (AHRQ),
CVS Health, Aerogen, and Nihon Kohden. Ms. Nuthall is employed by Auckland District Health Board. Her institution received other support from A+
Trust (contributed to research nurse's salary to collect ongoing data for the
project). Dr. Nadkarnis institution received grant support from AHRQ (R18
Grant). Dr. Nishisaki received support for article research from the AHRQ. His
institution received grant support from AHRQ (Improving the safety and quality of tracheal intubation in PICUs, R18HS022464; and Evaluating safety and
quality of tracheal intubation in pediatric ICUs, R03HS021583) and received
Copyright 2015 by the Society of Critical Care Medicine and Wolters
Kluwer Health, Inc. All Rights Reserved.
DOI: 10.1097/CCM.0000000000001313

2668

www.ccmjournal.org

Objectives: Adverse tracheal intubationassociated events are

common in PICUs. Prior studies suggest provider and practice factors are important contributors to tracheal intubation
associated events. Little is known about how the incidence of tracheal intubationassociated events is affected by the time of day,
day of the week, or presence of in-hospital attending-level intensivists. We hypothesize that tracheal intubations occurring during
nights and weekends are associated with a higher frequency of
tracheal intubationassociated events.
Design: Retrospective observational cohort study.
Setting: Twenty international PICUs.
Subjects: Critically ill children requiring tracheal intubation.
Interventions: None.
Measurements and Main Results: We analyzed 5,096 tracheal intubation courses from July 2010 to March 2014 from the prospective multicenter National Emergency Airway Registry for Children.
Frequency of a prioridefined tracheal intubationassociated events
was the primary outcome. Occurrence of any tracheal intubation
associated events and severe tracheal intubationassociated events
were more common during nights (19:00 to 06:59) and weekends
compared with weekdays (19% vs 16%, p = 0.01; 7% vs 6%,
p = 0.05, respectively). This difference was significant in emergent
intubations after adjusting for site-level clustering and patient factors:
for any tracheal intubationassociated events: adjusted odds ratio,
1.20; 95% CI, 1.021.41; p = 0.03; but not significant in nonemergent intubations: adjusted odds ratio, 0.94; 95% CI, 0.631.40;
p = 0.75. For emergent intubations, PICUs with home-call attending
coverage had a significantly higher frequency of tracheal intubation
associated events during nights and weekends (adjusted odds ratio,
1.29; 95% CI, 1.011.66; p = 0.04), and this difference was attenuDecember 2015 Volume 43 Number 12

Copyright 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

Pediatric Critical Care

ated in PICUs with in-hospital attending coverage (adjusted odds
ratio, 1.12; 95% CI, 0.911.39; p = 0.28).
Conclusions: Higher occurrence of tracheal intubationassociated
events was observed during nights and weekends. This difference
was primarily attributed to emergent intubations. In-hospital attending
physician coverage attenuated this discrepancy between weekdays
versus nights and weekends but was not fully protective for tracheal
intubationassociated events. (Crit Care Med 2015; 43:26682674)
Key Words: airway; in-house; intubation; patient safety; pediatric;
personnel staffing

ediatric advanced airway management is challenging and

can result in significant morbidity or mortality (14).
Patients requiring emergent tracheal intubation (henceforth referred to as intubation) rather than nonemergent
intubation are at the greatest risk for complications (1, 5). Some
data also suggest worse outcomes for critically ill patients during
nights and weekends compared to traditional weekday staffing
models (1, 69). Furthermore, intubation attempts by inexperienced providers have been associated with an increased incidence
of adverse events (3). Considering these factors and others, many
childrens hospitals have employed 24-hour in-hospital intensive
care attending coverage with the intent to mediate these risks.
However, the presence of in-hospital coverage has not consistently demonstrated improved patient outcomes (1012).
The aim of this study was to evaluate the effect of time of
day and 24/7 in-hospital coverage on tracheal intubation
associated events (TIAEs) and outcomes in the PICU using
the National Emergency Airway Registry for Children
(NEAR4KIDS) database (5). We hypothesized that intubations
during nighttime and weekends are associated with higher
occurrences of adverse TIAEs. Since night and weekend intubations are more likely emergent and at higher risk for TIAEs,
we further evaluated our hypothesis among the intubations for
emergent indications. We then evaluated the impact of in-hospital coverage on TIAEs during nights and weekends.

MATERIALS AND METHODS

The previously described NEAR4KIDS registry is a multicenter
quality improvement collaborative to prospectively collect safety
and quality data on intubations in PICUs (5). Data collection was
either approved or declared exempt by the institutional review
board at each participating center. Data collection included basic
patient demographics, provider presence, intubating provider
specialty and level of training, equipment used, and details on the
intubation encounter. Centers also reported if the intubation was
elective (nonemergent) or emergent. An encounter was defined
as one completed episode of advanced airway management
intervention, typically ending with successful intubation. Within
each intubation encounter, a course encompassed one method
or approach to secure an artificial airway (e.g., switching from
direct laryngoscopy to fiber optic laryngoscopy was defined as
two courses), while an attempt was defined as a single distinct
advanced airway maneuver (e.g., insertion of a device such as
Critical Care Medicine

laryngoscope, endotracheal tube, or laryngeal mask into patients

mouth or nose) (2). Pediatric Index of Mortality (PIM)-2 was
captured at the time of PICU admission (13).
Adverse TIAEs were defined a priori. Severe TIAEs were
defined as cardiac arrest, esophageal intubation with delayed
recognition, emesis with witnessed aspiration, hypotension
requiring intervention (fluid and/or pressors), laryngospasm,
pneumothorax or pneumomediastinum, or direct airway
injury. Nonsevere TIAEs included mainstem bronchial intubation (confirmed by chest radiograph), esophageal intubation with immediate recognition, emesis without aspiration,
hypertension requiring therapy, epistaxis, dental or lip trauma,
medication error, arrhythmia, or pain and/or agitation requiring additional medication with delay in intubation.
Intubation time was recorded on the data collection form
for each intubation encounter. Weekdays were defined as
07:00 to 18:59, Monday through Friday (i.e., excluding nights
and weekends). Nights and weekends included 1) weeknights
(defined as 19:00 to 06:59 the next morning) and 2) weekends
(defined as Friday at 19:00 to 06:59 Monday morning). For the
data collection period, centers identified themselves as having
24/7 in-hospital pediatric intensivist coverage or home coverage
and provided the start date for in-hospital coverage (to properly
identify intubations that occurred under in-hospital coverage).
Statistical Analysis
The primary outcome was occurrence of any TIAE during an
intubation course. The primary exposure variable was weekday
versus nights and weekends. Nonemergent intubation status
was considered as an effect modifier, and analyses were stratified by emergent status when the association of TIAEs and
weekday status was evaluated. For the analysis of in-hospital
coverage, only emergent intubations were included to focus on
those intubations where in-hospital coverage was most likely
to affect outcomes. Summary statistics are provided as percentages for categorical variables or median and interquartile range
(IQR) for continuous variables. Dichotomous categorical variables were analyzed using chi-square test, whereas continuous
variables were compared using Wilcoxon rank-sum test. Statistical analysis was performed with JMP 11 (SAS, Cary, NC) and
STATA 11.2 (Stata Corp, College Station, TX). A random effect
multivariate logistic regression model was developed with the
site as a group variable (20 sites) and occurrence of TIAE as a
dichotomous outcome. Factors associated with weekday versus
night and weekend status in the univariate analysis (p < 0.1)
and also known to be associated with TIAEs in previous studies
were included as covariates in this model. We also conducted
sensitivity analyses using two alternative definitions for nights
(nighttime, 16:00 to 06:59 and nighttime, 18:00 to 06:59). p
values of less than 0.05 were considered statistically significant.

RESULTS
Patient Demographics and Provider Characteristics
Five thousand ninety-six intubation courses (representing
4,775 encounters) were reported at 20 participating institutions
www.ccmjournal.org

2669

Copyright 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

Rehder et al

during the study period (July 2010 to March 2014). A median of

162 courses (IQR, 81347) were reported from each site. Patient
demographics and indications for intubation are reported in
Table 1.

Table 1. Respiratory failure, shock status, and emergent intubations were associated with night and weekend intubations.
Of note, admission PIM-2 scores and history of difficult airway

Patient Demographics and Indications for Tracheal Intubation

All Intubations
(n = 5,096)

Patient Characteristics

Patient age (yr; median, IQR)

Weekday Intubations
(n = 2,394)

1 (06)

1 (06)

Night and Weekend

Intubations (n = 2,702)

1 (06)

0.54
0.22

Diagnostic category (%)

Respiratorylower airway/pulmonary

1,716 (34)

773 (32)

943 (35)

Neurologic

906 (18)

445 (18)

461 (17)

Respiratoryupper airway

545 (11)

254 (11)

291 (11)

Sepsis/shock

345 (7)

153 (6)

192 (7)

Cardiacsurgical

292 (6)

143 (6)

149 (5)

Cardiacmedical

209 (4)

114 (5)

95 (4)

Trauma (including traumatic brain injury)

131 (2)

63 (3)

68 (3)

Other

774 (15)

363 (15)

411 (15)

Missing

178 (3)

86 (4)

92 (3)

Oxygenation failure

1,977 (39)

854 (36)

1,123 (42)

< 0.01

Ventilation failure

1,911 (38)

867 (36)

1,044 (39)

0.08

Upper airway obstruction

662 (13)

324 (14)

338 (13)

0.28

Unstable hemodynamics

562 (11)

239 (10)

323 (12)

0.03

Impaired airway reflexes

360 (7)

149 (6)

211 (8)

0.03

Pulmonary toilet

268 (5)

117 (5)

151 (6)

0.26

Neuromuscular weakness

172 (3)

84 (4)

88 (3)

0.62

Therapeutic hyperventilation

117 (2)

42 (2)

75 (3)

0.02

Indications for intubation (%)b

4,156 (82)

1,844 (77)

2,312 (86)

774 (15)

364 (15)

410 (15)

0.98

Limited neck extension

317 (6)

126 (5)

191 (7)

0.008

Midface hypoplasia

139 (3)

64 (3)

75 (3)

0.81

Upper airway obstruction

642 (13)

292 (12)

350 (13)

0.39

Thyromental space < 3 fingerbreadths

932 (18)

416 (17)

516 (19)

0.09

1,032 (20)

464 (19)

568 (21)

0.12

Emergent intubations

History of difficult airway (%)

< 0.001

Airway examination suggesting difficult airway (%)

Widest mouth opening < 3 fingerbreadths

Pediatric Index of Mortality-2,e

median (IQR)

2.2 (0.96.4)

2.3 (0.96.7)

2.1 (0.96.1)

0.25

IQR = interquartile range.

a
A single category was selected for each tracheal intubation encounter.
b
More than one indication may be selected for each tracheal intubation course.
c
Emergent intubations are defined as tracheal intubations for nonprocedural indication.
d
History of difficult airway was reported by providers.
e
Available in 4,054 tracheal intubation courses.
Analysis based on Course. Please refer to Materials and Methods section for more details.

2670

www.ccmjournal.org

December 2015 Volume 43 Number 12

Copyright 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

Pediatric Critical Care

were not associated with time of intubation. First attempt intubation providers for the course are reported in Table 2.
Weekday Versus Night and Weekend Intubations
and Outcomes
Overall, 2,702 intubation courses (53%) occurred during nights
and weekends. Adverse TIAEs were reported in 898 (18%) of all
intubation courses, and severe TIAEs were reported in 319 (6%).
Tracheal intubations during nights and weekends were
associated with a higher occurrence of TIAEs, as shown in
Table 3. This was primarily attributed to emergent intubations.
This difference was significant in emergent intubations after
adjusting for site-level clustering and patient factors: adjusted
odds ratio (aOR) for any TIAEs: aOR, 1.20; 95% CI, 1.021.41;
p = 0.03; but not significant in nonemergent intubations: aOR,
0.94; 95% CI, 0.631.40; p = 0.75.
Attending-level providers were more likely to be present at
weekday intubations. Pediatric residents were more likely to be
involved as first intubators during nights and weekends. This
latter finding was consistent for both nonemergent and emergent intubations.
24/7 In-Hospital Intensivist Coverage and Attending
Physician Presence at Emergent Intubations
Nine of 20 PICUs (45%) reported having in-hospital attending
physician coverage during the study period: seven PICUs had inhospital coverage throughout and two PICUs transitioned from
home coverage to an in-hospital coverage system during the study
period. Overall, 56% of emergent night and weekend intubations
occurred in PICUs with in-hospital coverage. Table 4 displays the
occurrence of TIAEs and severe TIAEs, as well as provider characteristics for emergent intubations, based on the presence of
in-hospital coverage. Emergent intubations at the PICUs with
Table 2.

in-hospital attending coverage were associated with a higher occurrence of TIAEs during weekdays as well as nights and weekends.
For emergent intubations, the odds of having a TIAE were
significantly higher in night and weekend intubations versus
weekday intubations (odds ratio [OR], 1.29; p = 0.04) (Table 5)
in PICUs with home-call attending coverage but not for PICUs
with in-hospital coverage (OR, 1.12; p = 0.28) after adjusted
for covariates associated with time of intubation (respiratory
failure and shock status) and age.
Attending physicians in in-hospital coverage were more
likely than those in home coverage to be present for both night
and weekend and weekday intubations, although home coverage attending physicians were commonly present at night and
weekend intubations (68%). Attending physician presence for
emergent night and weekend intubations was not significantly
associated with occurrence of any TIAEs (attending presence:
393/1,923 [20%] vs attending not present: 70/319 [18%];
p = 0.27). This was also the case for the occurrence of severe
TIAEs (146/1,923 [7%] vs 29/389 [7%]; p = 0.93). Attending
physicians were more likely and pediatric residents were less
likely to be the first intubating providers for night and weekend
intubations in PICUs with in-hospital coverage (Table 4).
Sensitivity Analysis With Alternative Definitions for
Nighttime Hours
Alternative definitions of nighttime (18:00 to 06:59 and 16:00
to 06:59) were also used to evaluate the association of TIAEs
with nights and weekends. For the first definition (nighttime:
18:00 to 06:59), 2,871 intubation courses were classified as
intubations during nights and weekends. The night and weekend hours were significantly associated with any TIAEs (weekday 16% vs night and weekend 19%; p = 0.01) but not with
severe TIAEs (6% vs 7%; p = 0.11).

First Attempt Providers for Tracheal Intubation Course

All Intubations
(n = 5,096) (%)

Weekday Intubations
(n = 2,394) (%)

1,024 (20)

432 (18)

592 (22)

Emergency medicine resident

209 (4)

122 (5)

87 (3)

Anesthesiology resident

115 (2)

50 (2)

65 (2)

1,959 (38)

945 (39)

First Attempt Provider

Pediatric resident

PCCM fellow
Anesthesiology fellow

Night and Weekend

Intubations (n = 2,702) (%)

1,014 (38)

70 (1)

31 (1)

39 (1)

576 (11)

260 (11)

316 (12)

Anesthesiology attending

66 (1)

22 (1)

44 (2)

ENTa

49 (1)

24 (1)

25 (1)

Hospitalist

17 (< 1)

14 (1)

PCCM attending

Nurse practitioner
Respiratory therapist
Other

3 (< 1)

405 (8)

212 (9)

193 (7)

66 (1)

29 (1)

37 (1)

540 (11)

253 (11)

287 (11)

PCCM = pediatric critical care medicine.

a
ENT includes otolaryngologists at all training levels (resident, fellow, and attending).

Critical Care Medicine

www.ccmjournal.org

2671

Copyright 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

Rehder et al

Table 3.

Status

Tracheal Intubation Outcomes and Providers Stratified by Emergent Intubation

Nonemergent Intubations
(n = 940)

All Intubations (n = 5,096)

Emergent Intubations (n = 4,156)

Weekday
(n = 2,394)
(%)

Night and
Weekend
(n = 2,702)
(%)

Any TIAEa

387 (16)

511 (19)

0.01

73 (13)

48 (12)

0.66

314 (17)

463 (20)

0.01

Severe TIAE

133 (6)

186 (7)

0.05

12 (2)

11 (3)

0.53

121 (7)

175 (8)

0.21

Nonsevere TIAE

285 (12)

367 (14)

0.07

66 (12)

40 (10)

0.41

219 (12)

327 (14)

0.03

2,279 (84) < 0.001 515 (94)

356 (91)

0.17 1,682 (91)

1,923 (83)

Outcome or
Provider Condition

Weekday
Night and
(n = 550)
Weekend
(%)
(n = 390) (%)

Weekday
(n = 1,844)
(%)

Night and
Weekend
(n = 2,312) (%)

Outcomes

Provider
Attending
present

2,197 (92)

Attending first
attemptb

260 (11)

316 (12)

0.35

27 (5)

Resident first
attemptc

432 (18)

592 (22)

0.001 110 (20)

< 0.001

20 (5)

0.88

233 (13)

296 (13)

0.87

103 (26)

0.02

322 (17)

489 (21)

0.003

TIAE = tracheal intubationassociated events.

a
Note each tracheal intubation course may experience more than one TIAE.
b
Pediatric critical care attending.
c
Pediatric residents.
Boldface values indicate statistically significant association (p < 0.05).

For the second definition (16:00 to 06:59), 3,337 intubation courses were classified as intubations during nights and
weekends. The night and weekend hours were significantly
associated with both any TIAEs (weekday 16% vs night and
weekend 19%; p = 0.004) and severe TIAEs (weekday 5% vs
night and weekend 7%; p = 0.03). Multivariate analyses with
these two alternative definitions revealed that in-hospital
attending coverage and the occurrence of TIAEs were significantly associated for weekday intubations but not for night
and weekend intubations, which was consistent with the original definition of nighttime (19:00 to 06:59) (Supplemental
Table 1, Supplemental Digital Content 1, http://links.lww.
com/CCM/B440; and Supplemental Table 2, Supplemental
Digital Content 2, http://links.lww.com/CCM/B441).

DISCUSSION
Our study using a large multicenter intubation quality
improvement database (NEAR4KIDS) demonstrated an
increased occurrence of TIAEs for tracheal intubations performed during nights and weekends. This remained significant
after adjusting for patient factors and excluding nonemergent
intubations. When in-hospital attending coverage was compared against home-call coverage, the increased risk of TIAEs
during nights and weekends persisted in home coverage models but not in in-hospital coverage, suggesting a protective
effect of 24/7 in-hospital coverage. Because in-hospital coverage does not assure attending presence at each intubation, we
separately analyzed documented attending presence regardless
2672

www.ccmjournal.org

of the hospital model. Attending physician presence at the bedside was not associated with occurrence of TIAEs among night
and weekend emergent intubations.
This is the largest study to date evaluating the association
between time of day and TIAEs in multiple PICUs. Our study
adds to the growing body of literature evaluating patient care
in ICUs outside of traditional weekday work hours. Similar
to much of the other published literature, our study demonstrates inconsistency in outcomes by time and day of week.
However, this variability is likely multifactorial. In evaluating
this finding, we must consider the night and weekend differences in patient population and disease processes, indications
for the procedure, unit staffing, and provider experience level.
Several studies have demonstrated worse outcomes for
patients admitted during nights and weekends (7, 1416);
however, this discrepancy may be secondary to confounders
such as severity of illness or admitting diagnoses (17, 18). Less
data exist regarding procedure outcomes by time of day. In a
single-center pediatric study, Carroll et al (1) demonstrated a
three-fold risk of complications for intubations occurring during nights and weekends. Our data support this finding in a
larger multicenter cohort.
Concerns over differences in outcomes during off-hours
have led to guidelines recommending 24/7 in-hospital intensivist coverage of all level I and level II adult ICUs (19).
However, the published data regarding in-hospital coverage are
unclear regarding its effect on patient outcomes (11, 2023).
Furthermore, aggregate outcomes may not directly assess the
impact of in-hospital attending coverage, as these outcomes
December 2015 Volume 43 Number 12

Copyright 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

Pediatric Critical Care

Table 4. Tracheal Intubation Outcomes and Providers by Attending Coverage Model

Stratified by Weekday and Night and Weekend for Emergent Intubations, Unadjusted
Analysis (n = 4,156)
Weekday
Outcome or
Provider Condition

Night and Weekend

In-Hospital
(n = 929) (%)

Home Coverage
(n = 915) (%)

187 (20)

127 (14)

< 0.001

283 (22)

180 (18)

66 (7)

55 (6)

0.34

99 (8)

76 (7)

0.78

Nonsevere TIAE

139 (15)

80 (9)

< 0.001

203 (16)

124 (12)

0.01

Attending present

893 (96)

789 (86)

< 0.001

1,221 (95)

702 (68)

< 0.001

Attending first
attempt

165 (18)

68 (7)

< 0.001

231 (18)

65 (6)

< 0.001

Resident first attempt

158 (17)

164 (18)

0.60

211 (16)

278 (27)

< 0.001

Any TIAEa
Severe TIAE

In-Hospital
(n = 1,285) (%)

Home Coverage
(n = 1,027) (%)

0.007

TIAE = tracheal intubationassociated events.

a
Each tracheal intubation course may experience one or more TIAEs.
Boldface values indicate statistically significant association (p < 0.05).

are also affected by care provided during weekday hours.

Investigating procedural outcomes during nights and weekends provides the unique opportunity to measure the immediate effect of in-hospital attending coverage. In our study, the
association between nights and weekends and occurrence of
TIAEs was somewhat attenuated, but not eliminated, in the
presence of in-hospital attending coverage. This result suggests
that the higher intubation risks associated with nights and
weekends may be mainly from patient factors.
While we may have been unable to fully adjust for patientlevel factors in our multivariate model, it is also possible that
the occurrence of TIAEs was more susceptible to nonattending
physician ICU staffing (physician trainees, nurses, and respiratory therapists), which may have been quite different during nights and weekends. We may have also overestimated the
clinical impact of having in-hospital attending physician coverage in our hypothesis. Our recent publication demonstrated
similar intubation skills among pediatric critical care medicine
fellows and attending physicians when analyzed with TIAEs as

an outcome (3), suggesting that attending presence may not be

the most important factor in minimizing TIAEs. It is important to note that even when present at intubations, attending
physicians were rarely the first attempt providers (Table4).
Our study also demonstrated increased occurrence of TIAEs
among intubations with 24/7 in-hospital attending coverage
during the weekday hours (OR, 1.53), with a lesser, nonsignificant increase during nights and weekends (OR, 1.33). It is likely
that hospitals choosing to provide in-hospital coverage represent
a higher acuity population, which could explain this discrepancy. Although there may be other inherent differences in the
patient populations or provider characteristics as confounders
for the increased TIAE frequency with in-hospital coverage, it
is improbable that differences attributed to the coverage model
would lead to a larger discrepancy during weekday hours.
This study has several limitations. It is an observational study
relying on self-reported data; however, reporting bias is minimized by prospective data collection and careful monitoring of
center compliance (5). We are unable to account for unmeasured

Table 5. Multivariate Analysis for Tracheal IntubationAssociated Events for Emergent

Intubations Adjusted for Patient Factors and Site-Level Clustering (n = 4,156)
Condition

Comparison

OR

95% CI

Home-call only

Night and weekend

(vs weekday)

1.29

1.011.66

0.04

In-house only

Night and weekend

(vs weekday)

1.12

0.911.39

0.28

Weekday only

In-house (vs home call)

1.53

1.072.18

0.02

Night and weekend only

In-house (vs home call)

1.33

0.951.84

0.09

OR = odds ratio.
Based on a random effect model as the center as a group variable, also adjusted for the following patient factors: age, respiratory failure, and shock. Overall,
the model was significant: Wald chi-square (df = 7) = 43.49; p < 0.0001. Note that these data also demonstrate increased occurrence of tracheal intubation
associated events among intubations with 24/7 in-hospital attending coverage during the weekday hours (OR = 1.53), with a lesser, nonsignificant increase
during nights and weekends (OR = 1.33). It is likely that hospitals choosing to provide in-hospital coverage represent a higher acuity population, which could
explain this discrepancy. Boldface values indicate statistically significant association (p < 0.05).

Critical Care Medicine

www.ccmjournal.org

2673

Copyright 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

Rehder et al

patient and institutional factors that may affect the occurrence

of TIAEs, including prior institutional team training or the presence of difficult airway algorithms. The data did not specify when
the attending physician arrived at the intubation scene; therefore,
we cannot report the degree of attending physician involvement
for each intubation. Also, our analysis did not include patient
outcomes beyond the intubation encounter, including length
of ventilation or mortality. The data collection form does not
specify when the attending arrives; therefore, we cannot report
which intubations had attending presence for the entire procedure. Finally, although this study reports a large number of intubation courses from 20 academic centers, this cohort may not be
representative of all intubations that occur in PICUs, particularly
intubations that occur in community PICUs.
Although PICUs strive to provide equivalent care at any
hour of the day, this study demonstrates that there is variability
in the frequency of TIAEs between traditional weekday work
hours and those occurring during nights and weekends. This
variance remains present after adjusting for patient factors
and excluding nonemergent intubations and was somewhat
attenuated in the presence of in-hospital attending coverage.
Further investigation is needed to elucidate factors that contribute to this variance and to determine processes which may
lead to safer tracheal intubations for this high-risk population.

CONCLUSIONS
Adverse TIAEs are more common during nights and weekends. This association remained significant after adjusting for
patient factors. Presence of 24/7 in-hospital attending coverage
was partially protective for TIAEs during nights and weekends.
Further research is required to identify explanatory factors.

ACKNOWLEDGMENT
We thank Hayley Buffman for her tireless efforts as the coordinator for the multicenter NEAR4KIDS registry.

REFERENCES

1. Carroll CL, Spinella PC, Corsi JM, et al: Emergent endotracheal intubations in children: Be careful if its late when you intubate. Pediatr
Crit Care Med 2010; 11:343348
2. Nishisaki A, Ferry S, Colborn S, et al; National Emergency Airway
Registry (NEAR); National Emergency Airway Registry for kids
(NEAR4KIDS) Investigators: Characterization of tracheal intubation
process of care and safety outcomes in a tertiary pediatric intensive
care unit. Pediatr Crit Care Med 2012; 13:e510
3. Sanders RC Jr, Giuliano JS Jr, Sullivan JE, et al; National Emergency
Airway Registry for Children Investigators and Pediatric Acute Lung
Injury and Sepsis Investigators Network: Level of trainee and tracheal
intubation outcomes. Pediatrics 2013; 131:e821e828
4. Easley RB, Segeleon JE, Haun SE, et al: Prospective study of airway management of children requiring endotracheal intubation before
admission to a pediatric intensive care unit. Crit Care Med 2000;
28:20582063

2674

www.ccmjournal.org

5. Nishisaki A, Turner DA, Brown CA 3rd, et al; National Emergency Airway Registry for Children (NEAR4KIDS); Pediatric Acute Lung Injury
and Sepsis Investigators (PALISI) Network: A National Emergency
Airway Registry for Children: Landscape of tracheal intubation in 15
PICUs. Crit Care Med 2013; 41:874885
6. Pronovost PJ, Angus DC, Dorman T, et al: Physician staffing patterns
and clinical outcomes in critically ill patients: A systematic review.
JAMA 2002; 288:21512162
7. Cavallazzi R, Marik PE, Hirani A, et al: Association between time of
admission to the ICU and mortality: A systematic review and metaanalysis. Chest 2010; 138:6875
8. Gajic O, Afessa B, Hanson AC, et al: Effect of 24-hour mandatory versus on-demand critical care specialist presence on quality of care and
family and provider satisfaction in the intensive care unit of a teaching
hospital. Crit Care Med 2008; 36:3644
9. Wallace DJ, Angus DC, Barnato AE, et al: Nighttime intensivist staffing and mortality among critically ill patients. N Engl J Med 2012;
366:20932101
10. Gajic O, Afessa B: Physician staffing models and patient safety in the
ICU. Chest 2009; 135:10381044
11. Kerlin MP, Small DS, Cooney E, et al: A randomized trial of nighttime physician staffing in an intensive care unit. N Engl J Med 2013;
368:22012209
12. Rehder KJ, Cheifetz IM, Markovitz BP, et al; Pediatric Acute Lung
Injury and Sepsis Investigators Network: Survey of in-house coverage by pediatric intensivists: Characterization of 24/7 in-hospital
pediatric critical care faculty coverage. Pediatr Crit Care Med 2014;
15:97104
13. Slater A, Shann F, Pearson G; Paediatric Index of Mortality (PIM)
Study Group: PIM2: A revised version of the Paediatric Index of Mortality. Intensive Care Med 2003; 29:278285
14. Sorita A, Ahmed A, Starr SR, et al: Off-hour presentation and outcomes in patients with acute ischemic stroke: A systematic review
and meta-analysis. Eur J Intern Med 2014; 25:394400
15. Arias Y, Taylor DS, Marcin JP: Association between evening admissions and higher mortality rates in the pediatric intensive care unit.
Pediatrics 2004; 113:e530e534
16. Bell CM, Redelmeier DA: Mortality among patients admitted to hospitals on weekends as compared with weekdays. N Engl J Med 2001;
345:663668
17. McCrory MC, Gower EW, Simpson SL, et al: Off-hours admission
to pediatric intensive care and mortality. Pediatrics 2014; 134:
e1345e1353
18. Peeters B, Jansen NJ, Bollen CW, et al: Off-hours admission and mortality in two pediatric intensive care units without 24-h in-house senior
staff attendance. Intensive Care Med 2010; 36:19231927
19. Haupt MT, Bekes CE, Brilli RJ, et al; Task Force of the American College of Critical Care Medicine, Society of Critical Care Medicine:
Guidelines on critical care services and personnel: Recommendations based on a system of categorization of three levels of care. Crit
Care Med 2003; 31:26772683
20. Arabi Y: Pro/Con debate: Should 24/7 in-house intensivist coverage
be implemented? Crit Care 2008; 12:216
21. Nishisaki A, Pines JM, Lin R, et al: The impact of 24-hr, in-hospital
pediatric critical care attending physician presence on process of
care and patient outcomes. Crit Care Med 2012; 40:21902195
22. Rehder KJ, Cheifetz IM, Willson DF, et al; Pediatric Acute Lung Injury
and Sepsis Investigators Network: Perceptions of 24/7 in-hospital
intensivist coverage on pediatric housestaff education. Pediatrics
2014; 133:8895
23. Reriani M, Biehl M, Sloan JA, et al: Effect of 24-hour mandatory vs
on-demand critical care specialist presence on long-term survival and
quality of life of critically ill patients in the intensive care unit of a teaching hospital. J Crit Care 2012; 27:421.e1421.e7

December 2015 Volume 43 Number 12

Copyright 2015 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.