ARTICLE IN PRESS

THE JOURNAL OF PEDIATRICS www.jpeds.com ORIGINAL

ARTICLES
Time Perception during Neonatal Resuscitation
Daniele Trevisanuto, MD1, Giuseppe De Bernardo, MD2, Giulia Res, MD1, Desiree Sordino, MD2, Nicoletta Doglioni, MD1,
Gary Weiner, MD3, and Francesco Cavallin, MS4

Objective To assess the accuracy of time perception during a simulated complex neonatal resuscitation.
Study design Participants in 5 neonatal resuscitation program courses were directly involved in a complex simu-
lation scenario. They were asked to assume the role of team leader, assistant 1, or assistant 2. At the end of the
scenario, each participant completed a questionnaire on perceived time intervals for key resuscitation interven-
tions. During the scenario, actual times were documented by an external observer and video recorded for later
review. In addition, participants were asked to evaluate their self-perceived level of stress and preparation.
Results Health care providers (68 physicians and 40 nurses) were involved in 36 scenarios. Perceived time in-
tervals for the initiation of key resuscitation interventions were shorter than the actual time intervals, regardless of
the participants role in the scenario. Self-assessed levels of stress and preparation did not influence time perception.
Conclusions Health care providers underestimate the passage of time, irrespective of their role in a simulated
complex neonatal resuscitation. Participants self-assessed levels of stress and preparation were not related to the
accuracy of their time perception. These findings highlight the importance of assigning a dedicated individual to
document interventions and the passage of time during a neonatal resuscitation. (J Pediatr 2016;:-).

D
uring neonatal resuscitation, specific time intervals for each intervention are recommended in the neonatal resuscita-
tion program (NRP) and European Resuscitation Council algorithms.1,2 In addition, time intervals are used to guide
the duration of resuscitative efforts and eligibility for therapies such as postresuscitation therapeutic hypothermia. However,
the accuracy of time perception by health care providers during neonatal resuscitation is not known. Studies in cognitive psy-
chology demonstrate that the accuracy of time perception varies depending on several factors such as the complexity of the
event, emotional status, stress, personality, previous experience, and evaluation methods.3-6
Although accurate perception of time is important for responders to medical emergencies, previous studies have shown that
clinicians experience time distortion. Emergency medical technicians in the field and responders to simulated in-hospital adult
cardiac arrest events underestimate the passage of time in some situations and overestimate time in others.7-10 As a result, the
initiation of interventions and the accuracy of documentation reported by the health care providers at the end of an acute event
could be affected. Neonatal resuscitation is unique because most often there is a well-defined start time and an electronic timer
clearly visible; however, the time perceived by health care providers in a neonatal resuscitation setting is unknown.
The aim of the present study was to investigate the time perceived by the health care providers during a simulated complex
neonatal resuscitation in relation to the participants role.

Methods
This prospective observational study was conducted at 5 Italian hospitals. The study protocol was approved by the Ethics Com-
mittee of Azienda Ospedaliera, University of Padova, Padova, Italy.
Consent to record the scenario and to use the data was obtained by all participants. A 2-day NRP course was conducted in
5 Italian hospitals by a team of 6 instructors from February to November 2014. The course consisted of didactic sessions fol-
lowed by hands-on skill stations and practice scenarios. Participants included physicians and nurses who were routinely in-
volved in the care of newborns in the delivery room. At the end of the course, participants were involved in a high-fidelity simulation
using a neonatal simulator (SimNewB, Laerdal, Stavanger, Norway). The scenario consisted of an asphyxiated term infant needing
a complex resuscitation including positive pressure ventilation, endotracheal in-
tubation (ETT), chest compressions, and emergency medications. Heart rate, re-
spiratory rate, and breath sounds were controlled remotely and could be assessed
by auscultation of the thorax, observation of chest movements, and From the 1Department of Womens and Childrens Health,
Maternal-Fetal Medicine Unit, University of Padua School
of Medicine, Padua, Italy; 2AOP Santobono Pausilipon,
Napoli, Italy; 3Department of Pediatrics, Neonatal-
Perinatal Medicine, University of Michigan, Ann Arbor, MI;
and 4Independent Statistician, Padova, Italy
A1 Assistant 1
The authors declare no conflicts of interest.
A2 Assistant 2
ETT Endotracheal intubation 0022-3476/$ - see front matter. 2016 Elsevier Inc. All rights
NRP Neonatal resuscitation program reserved.
http://dx.doi.org10.1016/j.jpeds.2016.07.003

FLA 5.4.0 DTD YMPD8513_proof August 4, 2016

THE JOURNAL OF PEDIATRICS www.jpeds.com Volume

palpation of the umbilical cord. The vital signs that are typi-
cally available in the delivery room (heart rate and hemoglo-
bin oxygen saturation via pulse oximetry [SpO2]) were displayed
on the bedside monitor about 40 seconds after the position-
ing of the oximeter probe on the right hand of the manikin.
The SpO2 was not shown on the monitor when the heart rate
was less than 60 beats per minute. The external observer pro-
vided verbal feedbacks during the scenario only if specifi-
cally required by the team and not provided by the manikin
(ie, the presence of secretions). A bedside Apgar timer ringing
at 1, 5, and 10 minutes was available for the team.
Participants were divided into groups of 3 and were asked
to assume the roles of team leader responsible for coordinat-
ing the team and managing the airway, assistant 1 (A1) re-
sponsible for chest compressions, and assistant 2 (A2)
responsible for umbilical catheter insertion and medication ad-
ministration. All the other tasks, such as time recording during
the scenario, were left to the decision of the team. During each
simulation, an external observer documented the actual time
of each intervention and the duration of the entire scenario
for calculation of accurate time intervals. All scenarios were
video-recorded, stored, and reviewed by the same observer to
confirm the documented intervention times. We chose this ap-
proach to have a double evaluation of time intervals; however,
video tape was used as the gold standard method of ascer-
taining timing.
At the end of the scenario, each participant completed a
13-item questionnaire. Participants were asked to estimate the
time elapsed from birth to the following events: beginning
positive pressure ventilation, ETT, beginning chest compres-
sions, administration of the first dose of adrenaline, first spon-
taneous breath, and duration of the entire scenario. The
questionnaire did not include the assignment of the Apgar
score. Finally, each participant was asked to describe how stress-
ful he/she found the scenario by using the following Likert
scale: 0 = not stressful; 1 = mildly stressful; 2 = moderately stress-
ful; and 3 = very stressful. A similar scale was used to measure
each participants assessment of their preparation for the re-
suscitation scenario: 0 = very well prepared; 1 = moderately
well prepared; 2 = unprepared; and 3 = very unprepared.
Participants were informed only at the end of the scenario
that they were required to estimate the time (retrospective
timing).
The primary outcome of the study was the difference
between the actual time and perceived time according to the
participants role in the scenario (team leader, A1, A2). In ad-
dition, we assessed the relationships between self-perceived stress
and preparation with time perception.
Statistical analysis was performed using R 2.12 software (R
Foundation for Statistical Computing, Vienna, Austria).11 Given
the lack of information about the time perceived by health care
providers during neonatal resuscitation, the sample size could
not be estimated using mathematical methods. Therefore, all
the participants in the NRP courses were included in the study
sample. Time data were expressed as median and IQR and rep-
resented the difference (seconds) between perceived time and
actual time measured by the observer. These time differences
were compared using the Wilcoxon signed rank test. Self-
perceived stress level and preparation were expressed as a
median score (IQR) and were compared among the 3 roles
using the Kruskal-Wallis test. The correlation between stress
level and preparation was evaluated using the Spearman rank
correlation coefficient.
Two multiple regression models were performed to iden-
tify the effects of role, stress level, and preparation on the per-
ception of time from birth to the first spontaneous breath and
the duration of the entire scenario. A P value of less than .05
was considered statistically significant.
Results
Health care providers (68 physicians and 40 nurses) attended
the courses and were involved in a total of 36 complex sce-
narios. The role of team leader was most frequently assumed
by physicians (97.2%), and the roles of A1 and A2 were dis-
tributed between physicians (A1: 55.6%; A2: 36.1%) and nurses
(A1: 44.4%; A2: 63.9%) (P < .001).
Study participants perceived that the time interval between
birth and the initiation of key resuscitation interventions was
significantly shorter than the actual time interval (Table

Table. Difference between perceived time and actual time in relation to the participants role (team leader, A1, A2) in
the scenario
Difference Difference Difference
Team leader-EO A1-EO A2-EO Actual time
Beginning PPV 13 (40 to 5)* 6 (44 to 5) 25 (43 to 0)* 57 (43-70)
Insertion of endotracheal tube 81 (226 to 9.5) 239 (420 to 90) 136 (340 to 54) 246 (174-451)
Beginning chest compressions 65 (119 to 15) 45 (87 to 0) 50 (90 to 0) 128 (100-211)
Administration first dose of adrenaline 93 (160 to 30) 109 (215 to 55) 96 (199 to 41) 304 (243-395)
Duration of chest compressions 25 (86 to 70) 15 (76 to 48) 30 (88 to 100) 148 (108-215)
First spontaneous breath 60 (135 to 0) 60 (170 to 0) 60 (300 to 0) 600 (360-600)
Duration of the entire scenario 60 (120 to 5) 60 (174 to 20) 38 (120 to 0) 600 (447-630)

EO, external observer; PPV, positive pressure ventilation.

Data (seconds) expressed as median (IQR).
*P < .05.
P < .01.
P < .001.

2 Trevisanuto et al

FLA 5.4.0 DTD YMPD8513_proof August 4, 2016

 2016
Figure. Difference between A, perceived time and B, actual time in relation to the participants role (TL, A1, A2) in the sce-
nario. Data are expressed as medians (IQR). *P < .05; P < .01. PPV, positive pressure ventilation; TL, team leader.
and Figure, A) regardless of the participants role in the sce-
nario. Participants also underestimated the time of the first
spontaneous breath and the duration of the whole scenario.
There was no significant difference between the perceived and
actual duration of chest compressions (Table and Figure, B).
Participants needed a median of 2 (IQR 1-3) attempts for suc-
cessful intubation. The median duration of ETT was 22 (IQR
15-40) seconds.
The self-assessed stress level was significantly greater
(P < .001) among the team leaders (team leader median 3; IQR
2-3) compared with either assistant (A1 median 2; IQR 1.5-
3, A2 median 2; IQR 2-2); however, the level of self-assessed
preparation was not significantly different (P = .80) (team leader
median 2; IQR 2-2, A1 median 2; IQR 2-2, A2 median 2; IQR
2-2). There was no statistically significant association between
self-perceived levels of stress and self-perceived preparation
among either team leaders or assistants (team leader rho -0.23;
P = .18, A1 rho 0.17; P = .31, A2 rho 0.01; P = .99).
Using multivariable analyses, perception of time elapsed from
birth to the first spontaneous breath was not associated with
role (P = .72), self-perceived stress level (P = .34), or self-
perceived preparation (P = .82). Similarly, perception of the du-
ration of the whole scenario was not associated with role
(P = .77), self-perceived stress level (P = .71), or self-perceived
preparation (P = .14).
Time Perception during Neonatal Resuscitation
FLA 5.4.0 DTD YMPD8513_proof August 4, 2016
ORIGINAL ARTICLES
Discussion
In this study, we assessed time perception by the health care
providers during a simulated complex neonatal resuscitation
in relation to the participants role. Our results showed that
participants consistently underestimated the time interval from
birth until key resuscitation interventions were performed, and
their time perception was not influenced by their role in the
scenario.
During medical emergencies, time plays a crucial role. Pub-
lished guidelines for cardiopulmonary resuscitation recom-
mend specific time intervals for interventions, duration of
intubation attempts, and total duration of resuscitative efforts.1,2
An altered perception of time might, therefore, affect the
efficacy of interventions, the patients outcome, and the ac-
curacy of clinical documentation. Furthermore, inaccurate
documentation caused by distorted time perception may alter
medical decisions concerning postresuscitation care, such as
eligibility for therapeutic hypothermia, and have medical-
legal consequences. For example, in infants with an Apgar score
of 0 after 10 minutes of resuscitation, if the heart rate remains
undetectable, it may be reasonable to stop assisted ventilation.2
However, it remains to be established if these timing errors
could have serious consequences to the infant or if they are
merely recall biases. Studies using aids, such as voice prompts
3
THE JOURNAL OF PEDIATRICS www.jpeds.com
and monitor alarms, suggest that these will improve timeli-
ness and intervention in emergency situations.12 A recent study
showed that health care professionals using a decision support
tool exhibit significantly fewer deviations from the NRP al-
gorithm compared with those working from memory alone
during simulated neonatal resuscitation,13 but implications of
such recall on the real patient remain to be evaluated.
Our results suggest that, in addition to the 3 persons re-
quired for a complex resuscitation, such as described in this
study, a further member of the team is required for accurate
timing. For those units that are less well-resourced, it could
be a problem because there may not be sufficient personnel
to perform the recording role in addition to the resuscitation
interventions. We found that times recorded by the external
observer and those captured by the video were nearly identi-
cal, suggesting that a member of the team exclusively dedi-
cated to the time documentation is very accurate. On the other
hand, our results suggest that video recording should be a part
of training simulations during NRP to support the review and
discussion of the team performance during the debriefing phase.
We believe that the use of video during actual resuscitations
is also important for evaluation of neonatal resuscitation per-
formance of health care providers.14,15
Emotional state and stress have been shown to cause
distorted time perception. Cognitive psychology studies in-
volving adult volunteers have shown that participants tend
to overestimate the perceived time duration when exposed
to high-arousal conditions and negative emotional states.16
These situations seem to fit medical emergencies, supporting
the hypothesis that medical emergencies will seem to last
longer than they actually do.2,3,16 Small studies evaluating
health care providers time perception during adult emergen-
cies, however, have shown conflicting results.8-10 In a simulation
study, Eisen et al10 showed that internal medicine residents
overestimated the total duration of an adult cardiac arrest
scenario by an average of 42.6 seconds; however, the sub-
group of first year residents tended to underestimate the
duration. This variation in the experience of time suggests
that previous experience, preparation, stress, or the level of
responsibility during the event may effect time perception.
Hosbond et al8 showed an overall underestimation of the
elapsed time (median 22.5 seconds shorter) during a simu-
lated cardiac arrest when the participants were queried during
the event but an overestimation of the total elapsed time
(median 36.5 seconds longer) when participants were queried
after the event was completed. In another clinical study, an
independent observer was placed with paramedic teams to
determine the paramedics perception of elapsed field time.
Team-member perception of elapsed time for the entire run
varied from the actual time by an average absolute value of
20%, with the greatest errors occurring in their estimates of
on scene time. Paramedics tended to overestimate elapsed
time during short runs and underestimate elapsed time
during long runs.9
As the accuracy of time perceived by health care providers
involved in the context of neonatal resuscitation is unknown,
we aimed to measure it in a scenario of advanced neonatal re-
4 Trevisanuto et al
FLA 5.4.0 DTD YMPD8513_proof August 4, 2016
Volume
suscitation. Overall, we found an underestimation of time
elapsed from birth to the initiation of all resuscitative inter-
ventions. In addition, our data show that time perception was
comparable between participants assuming the role of team
leader or assistant, suggesting that the level of responsibility
plays a limited role in time perception during neonatal resus-
citation. Importantly, distorted time perception may be a factor
explaining the frequent inaccuracies found in the medical
records documenting neonatal resuscitations.17 Based on pre-
vious experimental psychology studies showing the influ-
ence of stress and experience on the perception of the time,3,5,16
we evaluated the participants self- perceived levels of stress
and preparation using a Likert scale. We found that the team
leader reported a higher level of stress than either of the as-
sistants, even though the self-perceived preparation was similar
between the 2 roles. Neither stress level, nor self-perceived
preparation, were associated with the perception of time elapsed
from birth to the first spontaneous breath or the time of the
entire scenario.
Participants were not informed before the experiment that
they would have to estimate time after the scenario. This choice
could have influenced the results because prospective and ret-
rospective time estimation involves different mechanisms. In
situations in which participants are asked to judge time ex-
plicitly, cognitive psychologists make a distinction between 2
paradigms: one in which participants are informed before they
perform the task that they are required to make a time-
related judgment (prospective timing), and the other in which
they receive no prior warning (retrospective timing). Retro-
spective timing is mainly associated with memory processes,
but the structure of events is a critical determinant of remem-
bered duration.6
The strength of the present study consists of the measure-
ment of times perceived by a large group of health care
providers involved in a scenario of advanced neonatal resus-
citation. Some limitations should be considered when
interpreting the results. The study was undertaken immedi-
ately after participation in an NRP course, which could have
influenced the time perception of participants. The subjects
were studied during simulated neonatal resuscitation sce-
narios, and these findings should be replicated during actual
resuscitations. In addition, the sample size could not be
estimated using mathematical methods because of the lack
of information about the topic. For this reason we used a
convenience sample.
Our results show that providers underestimate the time in-
terval of key interventions irrespective of their role in the re-
suscitation scenario. The participants level of stress and
preparation did not affect their time perception. Considered
together, studies assessing the perception of time during medical
emergencies show conflicting results and suggest that more re-
search is needed to identify factors associated with time per-
ception and quantify the role of time distortion. Furthermore,
these findings highlight the importance of assigning a team
member to document interventions and the passage of time
during a neonatal resuscitation to ensure accurate documen-
tation and decision making.
 2016
Submitted for publication Jan 7, 2016; last revision received Jun 3, 2016;
accepted Jul 5, 2016
Reprint requests: Daniele Trevisanuto, MD, Department of Womens and
Childrens Health Medical School, University of Padua, Azienda Ospedaliera
Padova Via Giustiniani, 3, 35128 Padua, Italy. E-mail: trevo@pediatria.unipd.it
References
1. American Heart Association. American Academy of Pediatrics. In: Neo-
natal resuscitation textbook. 6th ed. Dallas (TX): American Academy of
Pediatrics and American Heart Association; 2011.
2. Wyllie J, Bruinenberg J, Roehr CC, Rdiger M, Trevisanuto D, Urlesberger
B. European resuscitation council guidelines for resuscitation 2015: section
7. Resuscitation and support of transition of babies at birth. Resuscita-
tion 2015;95:249-63.
3. Angrilli A, Cherubini P, Pavese A, Mantredini S. The influence of affec-
tive factors on time perception. Percept Psychophys 1997;59:972-
82.
4. Loftus EF, Schooler JW, Boone SM, Kline D. Time went by so slowly: over-
estimation of event duration by males and females. Appl Cogn Psychol
1987;1:3-13.
5. Droit-Volet S. Time perception, emotions and mood disorders. J Physiol
(Paris) 2013;107:255-64.
6. Grondin S. Timing and time perception: a review of recent behavioral
and neuroscience findings and theoretical directions. Atten Percept
Psychophys 2010;72:561-82.
7. Eisen LA. Time perception is distorted during responses to medical emer-
gencies. Med Hypotheses 2009;72:626-8.
Time Perception during Neonatal Resuscitation
FLA 5.4.0 DTD YMPD8513_proof August 4, 2016
ORIGINAL ARTICLES
8. Hosbond S, Folkestad L, Brabrand M. Does cardio-pulmonary resusci-
tation influence perception of time? Scand J Trauma Resusc Emerg Med
2009;17(Suppl 2):P8.
9. Jurkovich GJ, Campbell D, Padrta J, Luterman A. Paramedic perception
of elapsed field time. J Trauma 1987;27:892-7.
10. Eisen L, Pellechia C, Mayo P. Assessment of residents time perception
of in-hospital cardiopulmonary resuscitation (IHCPR) scenarios using
a computerized patient simulator (CPS). Chest 2007;132:446S.
11. R Development Core Team. R: A language and environment for statis-
tical computing. Vienna, Austria: R Foundation for Statistical Comput-
ing; 2011 http://www.R-project.org/. Accessed July 28, 2016.
12. Field LC, McEvoy MD, Smalley JC, Clark CA, McEvoy MB, Rieke H, et al.
Use of an electronic decision support tool improves management of simu-
lated in- hospital cardiac arrest. Resuscitation 2014;85:138-42.
13. Fuerch JH, Yamada NK, Coelho PR, Lee HC, Halamek LP. Impact of a
novel decision support tool on adherence to Neonatal Resuscitation
Program algorithm. Resuscitation 2015;88:52-6.
14. ODonnell CP, Kamlin CO, Davis PG, Morley CJ. Ethical and legal aspects
of video recording neonatal resuscitation. Arch Dis Child Fetal Neona-
tal Ed 2008;93:F82-4.
15. Trevisanuto D, Bertuola F, Lanzoni P, Cavallin F, Matediana E, Manzungu
OW, et al. Effect of a neonatal resuscitation course on healthcare pro-
viders performances assessed by video recording in a low-resource setting.
PLoS ONE 2015;10:e0144443.
16. Droit-Volet S, Meck WH. How emotions colour our perception of time.
Trends Cogn Sci 2007;11:504-13.
17. Schilleman K, Witlox RS, van Vonderen JJ, Roegholt E, Walther FJ, te Pas
AB. Auditing documentation on delivery room management using video
and physiological recordings. Arch Dis Child Fetal Neonatal Ed
2014;99:F485-90.
5