Journal of Critical Care 36 (2016) 207211

Contents lists available at ScienceDirect

Journal of Critical Care

journal homepage: www.jccjournal.org

Results of implementing a pain management algorithm in intensive care

unit patients: The impact on pain assessment, length of stay, and
duration of ventilation
Brita F. Olsen, RN, MNSc a,b,i,, Tone Rusten, RN, PhD b,c, Leiv Sandvik, MD, PhD d,
Morten Jacobsen, MD, PhD e,f,g, Berit T. Valeberg, RN, PhD h
a

Intensive and Post Operative Unit, stfold Hospital Trust, Postbox 300, 1714 Grlum, Norway
Division of Emergencies and Critical Care, Oslo University Hospital, Postbox 4950, Nydalen, 0424 Oslo, Norway
c
Institute of Health and Society, Department of Nursing Science, Faculty of Medicine, University of Oslo, Postbox 0316, Oslo, Norway
d
Oslo Center for Biostatistics and Epidemiology, Oslo University Hospital, Postbox 1122, Blindern, 0317 Oslo, Norway
e
Norwegian University of Life Sciences, Postbox 5003, 1432 s, Norway
f
Faculty of Medicine, University of Oslo, Postbox 0316, Oslo, Norway
g
Department of Medicine, stfold Hospital Trust, Postbox 300, 1714 Grlum, Norway
h
Oslo and Akershus University College of Applied Sciences, Postbox 4, St Olavs plass, 0130 Oslo, Norway
i
Institute of Clinical Medicine, Faculty of Medicine, University of Oslo, Postbox 1078, 0316 Oslo, Norway
b

a r t i c l e

i n f o

Available online xxxx

Keywords:
Pain
Acute pain
Pain management
Critical care
Intensive care units
Algorithms

a b s t r a c t
Purpose: This study aimed to measure the impact of implementing a pain management algorithm in adult
intensive care unit (ICU) patients able to express pain. No controlled study has previously evaluated the impact
of a pain management algorithm both at rest and during procedures, including both patients able to self-report
and express pain behavior, intubated and nonintubated patients, throughout their ICU stay.
Materials and methods: The algorithm instructed nurses to assess pain, guided them in pain treatment, and was
implemented in 3 units. A time period after implementing the algorithm (intervention group) was compared
with a time period the previous year (control group) on the outcome variables: pain assessments, duration of
ventilation, length of ICU stay, length of hospital stay, use of analgesic and sedative medications, and the
incidence of agitation events.
Results: Totally, 650 patients were included. The number of pain assessments was higher in the intervention
group compared with the control group. In addition, duration of ventilation and length of ICU stay decreased
signicantly in the intervention group compared with the control group. This difference remained signicant
after adjusting for patient characteristics.
Conclusion: Several outcome variables were signicantly improved after implementation of the algorithm
compared with the control group.
2016 Elsevier Inc. All rights reserved.

1. Introduction
Many patients in intensive care units (ICUs) experience pain [1,2].
Pain should be assessed routinely and repetitively [3] but is not always
done [4]. Valid pain assessment tools are available and recommended
[3], but a substantial proportion of ICU nurses do not use them [5].
When implementing these tools in clinical practice, knowledge decits,
resistance, and barriers against changing practice have been documented
among clinicians [6-9].
Disclosure of funding: the authors do not have any ethical conicts or nancial interests to disclose. South-Eastern Norway Regional Health Authority, stfold Hospital Trust,
and Oslo University Hospital funded this study.
Corresponding author at: Intensive and Post Operative Unit, stfold Hospital Trust,
Postbox 300, 1714 Grlum, Norway.
E-mail address: [email protected] (B.F. Olsen).
http://dx.doi.org/10.1016/j.jcrc.2016.07.011
0883-9441/ 2016 Elsevier Inc. All rights reserved.

Clinical evidence-based algorithms are suitable for implementing pain

management in clinical practice [10]. However, because an appropriate
algorithm for adult ICU patients that included both pain assessment and
pain management was not available, a comprehensive new algorithm
was developed [11]. This algorithm was implemented in 3 units [12]. To
our knowledge, no controlled studies have previously evaluated the impact of a pain management algorithm both at rest and during procedures
[13], including both patients able to self-report and express pain behavior,
intubated and nonintubated patients, throughout their ICU stay.
However, the implementation of a single pain assessment tool has
been evaluated in several studies [14-17]. Of note, not all ICU patients
able to express pain were included in these studies. Other studies
have evaluated the implementation of several assessment tools,
including tools to assess pain, agitation, and delirium [18-21]. When
introducing several tools targeting different variables, it is difcult to

208

B.F. Olsen et al. / Journal of Critical Care 36 (2016) 207211

evaluate the effect of implementing the pain assessment tools. Despite

these limitations, these studies found a decrease in pain and agitation
[17,20], decreased duration of ventilation [15,18-20], decreased length
of ICU stay [14,15,18,19], decreased length of hospital stay [18],
a decrease in complications [15], nosocomial infections [20], decreased
mortality [18,19], more frequently charted pain assessments in the
medical records [14-17,20], and better and more dedicated analgesia
[14-16,18,20,21].
Based on earlier research, the objective of the present study was to
evaluate the use of a pain assessment and pain management algorithm
in all groups of ICU patients able to express pain on pain assessments,
duration of ventilation, length of ICU stay, length of hospital stay,
use of analgesic and sedative medications, and the incidence of
agitation events.
2. Materials and methods
2.1. Development of the algorithm
A short, evidence-based algorithm was developed [11]. The
algorithm instructed ICU nurses to assess patients' pain at least once a
shift, both at rest and during turning [22,23]. A numeric rating scale
(NRS) was used when patients could self-report pain [24]. The
behavioral pain scale (BPS) was used when patients were receiving
mechanical ventilation and unable to self-report pain [25]. Finally,
the BPS-NonIntubated (BPS-NI) was used when patients were not
intubated but unable to self-report pain [26]. Pain treatment
actions were chosen based on cutoff points that dened a pain
event. Pain events were dened as NRS scores N3 [20,27], or BPS and
BPS-NI scores N5 [20,25,26]. If the pain score was higher than the
prespecied cutoff, the nurses were guided to consider increasing pain
treatment. If the pain score was less than the cutoff, the nurses were
guided to consider either decreasing or continuing the present pain
treatment. Pain treatments included analgesics within each patient's
prescription or nonpharmacological interventions such as changing
the patient's position.
2.2. Implementation and evaluation of the algorithm
The algorithm was implemented in 3 units (1 medical/surgical ICU, 1
surgical ICU, and 1 postanesthesia care unit) at 2 Norwegian hospitals.
Before the implementation, these units had no protocols or guidelines
for pain assessment or management, but the nurses were able to titrate
doses of prescribed analgesics. Nurses at the 3 units were educated in
pain assessment and management for a 3-week period [12].
The algorithm was used for a 22-week period for all ICU patients 18
years old admitted to the 3 units [12]. Patients were included in the
study if they were able to self-report pain or express pain behaviors
and were excluded if they could not self-report pain or express pain
behaviors (having quadriplegia, receiving neuromuscular blockade or
paralyzing drugs, and being investigated for brain death). During
these 22 weeks, the nurses' level of adherence with the algorithm was
74.6% [12].
To evaluate the impact of the algorithm in ICU patients, we used
a pre-post intervention design. To cover the same time period
of the year, we compared a period (from May to November 2012)
after implementing the algorithm (intervention group) with a
similar time period (from May to November 2011) the previous year
(control group).
2.3. Ethics approval
The Regional Ethics Committee (2011/2582D) and the leadership
at the hospitals that participated in the study approved this study.
The study was registered at ClinicalTrials.gov (NCT01599663).

2.4. Data collection

Demographic and clinical data including sex, age, ventilation status
(yes/no), diagnosis using the International Classication of Diseases10 codes, total ventilation time during ICU stay (for patients receiving
mechanical ventilation or noninvasive ventilation), length of ICU stay,
length of hospital stay, severity of disease, and nurses' workload were
collected from the medical records. In addition, the use of analgesic
and sedative medications (yes/no), daily dosages of analgesic or sedative medications, sedation level (day/evening/night), and pain assessments (day/evening/night, at rest and during turning) during the rst
6 days of each patient's ICU stay were collected from medical records.
Sedation level was measured by the Motor Activity Assessment Scale
(MAAS) [28] or the Richmond Agitation-Sedation Scale (RASS) [29], depending on which scale the units used. MAAS ranges from 0 (unresponsive) to 6 (dangerous agitation), and a score of 3 denotes a calm and
cooperative patient. RASS ranges from 5 (unarousable) to 4 (combative), and 0 denotes an alert and calm patient. Severity of disease was
measured with the Simplied Acute Physiology Score (SAPS II) [30].
SAPS is calculated during the rst 24 hours of ICU stay, ranges from 0
to 163, with higher scores indicating high disease severity and high
risk of hospital mortality. Nurses' workload was measured by the
Nine Equivalents of Nursing Manpower Score (NEMS) [31]. The NEMS
includes 9 variables. Scoring range is from 0 (low workload) to 66
(high workload).
Pain assessments were measured using 3 different tools. The NRS is a
scale that ranges from 0 (no pain) to 10 (worst possible pain). The BPS
contains 3 domains (facial expressions, movements of upper limbs, and
compliance with ventilation). In the BPS-NI, the domain compliance
with ventilation is replaced with vocalization. Each domain contains descriptors rated on a 1-to-4 scale. The ratings for each domain are
summed, with a total score from 3 (no pain) to 12 (worst possible pain).

2.5. Statistical analysis

A sample size calculation was performed, using ventilation time as
the variable. In a study by Chanques and colleagues [20], median
ventilation times were 120 and 65 hours for the control group and the
intervention group, respectively, and the interquartile ranges were
264 and 168 hours, respectively. In normally distributed data, the standard deviation (SD) is 0.694*interquartile range; the SDs of the ventilation times were 183 and 117 hours, respectively. Based on these data,
we assumed that, in the present study, the difference in mean ventilation time between the groups would be at least 55 hours and that the
SD would be 150 hours for each group. To achieve 80% test power, we
needed to include at least 117 mechanically ventilated patients in
each group. Because we included both mechanically ventilated and
nonmechanically ventilated patients, data from all patients enrolled
were collected until at least 117 of the included patients in each group
were mechanically ventilated. In that way, we could also include ventilation time as an outcome measure in the group of ICU patients that
were mechanically ventilated.
Diagnostic groups including b5% of the patients were merged into 1
category named other diagnoses. Mean daily NEMS was calculated for
each patient. Ventilator time was calculated for patients receiving mechanical ventilation or noninvasive ventilation. Analgesic and sedative
medications used in N5% of patients are analyzed [32]. Median daily dosages of each medication were calculated for each patient. Doses of
ketobemidone [33], morphine [33], oxycodone [33], fentanyl [34],
remifentanil [34], and alfentanil [34] were converted into intravenous
morphine equianalgesic dosages. Sedation level was divided into no agitation events (RASS 1 or MAAS 4) and agitation events (RASS N1 or
MAAS N4) [20,28,29]. The total number of pain assessments that were
recorded during the rst 6 days of the patients' ICU stay was divided
by the total number of pain assessments that should have been recorded

B.F. Olsen et al. / Journal of Critical Care 36 (2016) 207211

based on the number of shifts that they were cared for in the ICU,
multiplied by 100.
When comparing patients in the intervention group with patients in
the control group, the 2 test was used for categorical variables and the t
test or Mann-Whitney U test for continuous variables. A P value of b.05
was considered signicant. Outcome variables that differed signicantly
between the 2 groups were included in a linear regression analysis to
control for confounding variables. Linear regression analysis was performed with the outcome variables as dependent variables and with
group and confounders (baseline variables that differed signicantly
between the control group and the intervention group) as independent
variables. Nonnormally distributed dependent variables were log transformed before regression analysis. Statistical analyses were performed
using Statistical Package for the Social Sciences (IBM SPSS Statistics for
Windows, version 21.0; IBM Corp., Armonk, NY).

209

Table 2
Impact of using the pain management algorithm
Outcome measures

Control group
(n = 252)

Intervention group
(n = 398)

Ventilation timeb, (h)

Length of ICU stay, (d)
Length of hospital stay, (d)

79 (26-205)
3.0 (1.7-6.9)
13 (7-24)
(n = 252)

46 (17-153)
2.6 (1.7-5.4)
13 (7-24)
(n = 285)

.01
.04
.79

Sedation level
MAAS
RASS
Agitation eventa, n (%)

2 (0-3)
1 (3 to 0)
14 (6)

2 (1-3)
0 (2 to 0)
9 (3)

.28
.09
.02

Quantitative data are expressed as median (25th-75th percentiles).

a
An agitation event was dened as MAAS N4 or RASS N1.
b
Total ventilation time during ICU stay (for patients receiving mechanical ventilation or
noninvasive ventilation).

3. Results
To ensure that there were at least 117 mechanically ventilated
patients in each group, we included 650 patients overall. The intervention group (n = 398) and the control group (n = 252) were similar
regarding sex, age, diagnoses, and use of ventilation. Patients in the
intervention group had signicantly lower disease severity (SAPS, 36
vs 40; P = .02) and lower nursing workload (NEMS, 31 vs 36;
P b .001) compared with those in the control group (Table 1).
In the intervention group, 4223 pain assessments using NRS, BPS, or
BPS-NI were recorded for 2832 shifts (that should have equated to 5664
assessments, 74.6% adherence [12]). In the control group, 370 pain
assessments using NRS were recorded for 2542 shifts (that should
have equated to 5084 assessments, 7.3% adherence).
Compared with those in the control group, patients in the intervention group had shorter ventilation times (median 46 vs 79 hours, P =
.01), shorter length of ICU stay (median 2.6 vs 3.0 days, P = .04), and
fewer agitation events (3% vs 6%, P = .02) (Table 2). When adjusting
for confounding variables, disease severity (SAPS) was included as an
independent variable in the regression analyses. Because nursing workload (NEMS) could not be excluded as an effect outcome, NEMS was not
included as independent variable. After adjusting for SAPS, we found
that the differences between the 2 groups remained signicant for
ventilation time (P = .01) and length of ICU stay (P = .03).
The use of analgesic and sedative medications was similar in the 2
groups (Table 3): 96% of the intervention group and 97% of the control
group received analgesics, and 78% of the intervention group and 81%

Table 1
Patient and unit characteristics.
Characteristics

Age (y)
Severity of disease (SAPS)
Nursing workload (NEMS) per day
Sex
Male, n (%)
Female, n (%)
Diagnoses (International Classication of
Diseases-10 codes), n (%)
Diseases of the respiratory system, n (%)
Neoplasms, n (%)
Diseases of the circulatory system, n (%)
Diseases of the digestive system, n (%)
Other diagnoses, n (%)
Injury, poisoning, and certain other
consequences of external cause, n (%)
Mechanical ventilation
Yes, n (%)
No, n (%)
Quantitative data are expressed as mean (SD).

Control
group
(n = 252)
57 (20)
40 (20)
36 (13)

Intervention P
group
(n = 398)
60 (18)
36 (19)
31 (11)

157 (62)
95 (38)

265 (67)
133 (33)

19 (8)
24 (10)
29 (11)
36 (14)
57 (23)
87 (34)

38 (9)
47 (12)
47 (12)
73 (18)
66 (17)
127 (32)

139 (55)
113 (45)

209 (52)
189 (48)

.11
.02
b.001
.26

.28

.51

of the control group received sedatives. With regard to medication

dosages, those for epidural fentanyl were higher (median 310 vs 223 g,
P = .01) and those for intravenous midazolam (77 vs 105 mg, P = .03)
were lower in the intervention group compared with the control group.
4. Discussion
To our knowledge, no controlled studies have previously evaluated
the impact of a comprehensive pain management algorithm both at
Table 3
Impact on use and dosage of medication when using the pain management algorithm
Medications

Use of analgesic medications

Ketobemidone, n (%)
Fentanyl, n (%)
Oxycodone, n (%)
Morphine, n (%)
Remifentanil, n (%)
Alfentanil, n (%)
Morphine equianalgesic
intravenous dosagea, n (%)
Fentanyl epidural, n (%)
Paracetamol, n (%)
Use of sedative medications
Propofol, n (%)
Midazolam, n (%)
Zopiclone, n (%)
Diazepam, n (%)
Oxazepam, n (%)
Ketamine, n (%)
Daily dosage of analgesic
medications
Ketobemidone, (mg)
Fentanyl, (g)
Oxycodone, (mg)
Morphine, (mg)
Remifentanil, (mg)
Alfentanil, (mg)
Morphine equianalgesic
intravenous dosagea, (mg)
Fentanyl epidural, (g)
Paracetamol, (g)
Daily dosages of sedative medication
Propofol, (mg)
Midazolam, (mg)
Zopiclone, (mg)
Diazepam, (mg)
Oxazepam, (mg)
Ketamine, (mg)

Control
group
(n = 252)

Intervention
group
(n = 285)

245 (97)
141 (56)
130 (52)
66 (26)
19 (8)
53 (21)
42 (17)
233 (93)

274 (96)
167 (59)
126 (44)
75 (26)
18 (6)
80 (28)
56 (20)
262 (92)

.48
.59
.19
1.00
.69
.07
.43
.94

60 (24)
210 (83)
204 (81)
132 (52)
85 (34)
70 (28)
49 (19)
21 (8)
17 (7)

60 (21)
235 (83)
222 (78)
146 (51)
95 (33)
73 (26)
60 (21)
39 (14)
16 (6)

.50
.87
.38
.85
.99
.64
.72
.06
.71

10 (5-15)
9 (5-15)
1644 (547-3020) 1490 (500-3081)
20 (14-26)
20 (10-30)
10 (5-14)
11 (8-21)
4 (2-7)
4 (1-6)
1 (1-4)
2 (1-3)
115 (18-349)
97 (16-378)

.32
.47
.36
.38
.45
.98
.66

223 (159-331)
3 (2-4)

310 (221-415)
3 (2-4)

.01
.31

550 (166-1551)
105 (47-257)
8 (8-8)
5 (3-10)
20 (10-33)
282 (150-541)

566 (200-1150)
77 (33-135)
8 (6-8)
5 (3-12)
20 (13-30)
192 (100-250)

.99
.03
.28
.43
.68
.12

Quantitative data are expressed as median (25th-75th percentiles).

a
Morphine equianalgesic intravenous dosages were calculated using ketobemidone,
fentanyl, oxycodone, morphine, remifentanil, and alfentanil.

210

B.F. Olsen et al. / Journal of Critical Care 36 (2016) 207211

rest and during procedures, including both patients able to self-report

and express pain behavior, intubated and nonintubated patients,
throughout their ICU stay. The patient mix and the staff were the
same for the 2 study periods, and the included units did not implement
new medications or new procedures for weaning from mechanical
ventilation between these 2 study periods.
The number of pain assessments was higher in the intervention
group compared with the control group. This result is consistent with
several studies [14-17,20] showing that pain assessments and identication of pain episodes are more frequently documented after the implementation of pain assessment tools. Guidelines [3] recommend a
regular pain assessment of ICU patients using pain assessment tools.
Patients in the intervention group were pain assessed with different
tools according to their ability to report pain. In the control group,
only NRS was used. These ndings are not surprising, because BPS and
BPS-NI were not available for Norwegian ICU patients during data collection
for the control group. However, it is important to use different tools for different groups of ICU patients. Therefore, we presume that implementing a
pain management algorithm may improve pain assessment in ICU patients.
The median decrease in ventilation time was 33 hours. This decrease
is consistent with previous research [15,18-20], which found a decrease
in ventilation time ranging from 21 to 70 hours. Despite different designs
and interventions, all these studies included a regular and systematic assessment of ICU patients, as did the present study. It is a bit puzzling that
the overall use of analgesic and sedative medications did not change
after implementing the algorithm. However, decreased daily dosages of
midazolam could be one of the explanations for the decreased ventilation time in the intervention group. Because increased ventilation time
is associated with a higher incidence of ventilator-associated pneumonia
[35], ventilation-induced lung damage [36], and higher treatment cost
[37], we consider our nding to be of clinical importance.
Length of ICU stay decreased signicantly in the intervention group
compared with the control group (0.4 days) and is consistent with prior
research [14,15,18,19] reporting a decrease in the length of stay from
0.2 to 5.2 days. As treatment of ICU patients is expensive [38,39], a
reduced ICU stay could have socioeconomic consequences. Despite the
decreased length of ICU stay, we found no impact on the length of
hospital stay. This could indicate that the algorithm has an impact
during the stay in ICU but without a long-term impact.
It is worth noting that we could not evaluate inherent changes in
clinical practices in the ICU over time regardless any specic action to
improve pain management. In addition, it is not possible to evaluate
that the presence of resource persons also could have an impact on
the result. We do not know whether using pain assessment tools,
protocols, or an algorithm is the best. However, all these studies have
1 thing in common. Pain in ICU patients is assessed using valid pain
assessment tools, and pain assessment is performed regularly and
systematically, as clinical guidelines recommend [3]. This may be
the most important explanation for the positive outcomes across all
these studies.
The incidence of agitation events decreased signicantly in the intervention group compared with the control group, even though only
small numbers of patients were found to be agitated. A decrease in agitation events is consistent with a study [20], which found that the incidence of agitation decreased after a systematic evaluation of pain and
agitation. The authors suggested that this might be explained by a better
match between supply of analgesic and sedative medications and each
patient's needs. This explanation is supported by Payen and colleagues
[40,41], who observed that patients assessed for pain were more likely
to be treated with nonopioids, to receive multimodal analgesia, and to
have dedicated pain treatment during procedural pain events compared
with patients who were not pain assessed. Because data on specic pain
treatment actions after the pain assessments were not available in our
study, we could not identify therapeutic changes. Because of the small
number of agitation events, this difference was not adjusted in the
regression analysis.

The overall use of analgesic and sedative medications did not change
after implementing the algorithm in the present study. However, it is
worth noting that 97% of patients in the control group received
analgesia. This high percentage could explain why the use of analgesia
did not increase after implementing the algorithm. The daily dosages
of analgesic and sedative medications did not change, except for 2 of
the medications. The median daily dosage of epidural fentanyl increased, and the median daily dosage of midazolam decreased. A decrease in the dosages of midazolam was also found in a multicenter
prospective cohort study by Payen and colleagues [40]. During recent
years, we have seen a trend toward more awake and cooperative ICU
patients [42], and analgosedation has been recommended [43]. The
units included in this study did not implement any such new treatment
between the 2 study periods, and we did not nd a change in sedation
level toward more awake and cooperative patients, as measured by
MAAS or RASS. However, we cannot exclude the possibility that the
new trend has inuenced the use of midazolam.
4.1. Strengths and limitations
A strength of the present study is that all ICU patients who could
express pain were included, and not only those patients who could
self-report pain. Moreover, the study included a large number of
patients. Furthermore, the present study is innovative, because little
research has been done in this area.
Because the data for pain assessments from patients in the control
group were inadequately documented in the medical records, and we
did not have independent observers assessing pain in the control
group [20], we could not evaluate the impact of implementing an algorithm on the occurrence of pain and pain intensity. Furthermore, it is a
limitation that for 113 patients in the intervention group, no written
documentation on pain assessment, medications, or sedation levels
was available. However, because all the nurses in the 3 units were educated and trained in using the algorithm, we included available data
from all patients admitted to the ICU during the study period. In addition, several other outcomes such as the incidence of nosocomial infections [20], complications [15], and ventilator-associated pneumonias
[40] could have been evaluated. A pre-post intervention design has
weaknesses, as confounding variables could inuence the outcome,
but differences in baseline variables between groups were controlled,
using regression analysis. In the present study, ICU patients were not
randomized regarding pain assessment, because pain assessment of
ICU patients is already highly recommended.
5. Conclusion
The implementation of a pain management algorithm in adult ICU
patients able to express pain was associated with a higher number of
pain assessments and a decreased duration of ventilation and length
of ICU stay. This algorithm detects and treats pain at rest and during
turning, based on the assessment of pain with validated tools specically dedicated to patients' condition throughout their ICU stay (intubated/
nonintubated and able/unable to self-report pain). These ndings could
have an impact on ICU patient outcomes and have economic consequences. Findings from this study support the concept that pain should
be assessed regularly and systematically with valid pain assessment
tools, but further research is needed. An evaluation is needed for the occurrence of pain and of pain intensity, for several additional outcome
variables, and for more specic pain treatment actions after the implementation of an algorithm. A comparison of the outcomes of using algorithms, protocols, and pain assessment tools is also warranted.
Acknowledgment
We would like to thank all the nurses who participated in and
contributed to the study. We acknowledge the South-Eastern Norway

B.F. Olsen et al. / Journal of Critical Care 36 (2016) 207211

Regional Health Authority, stfold Hospital Trust, and Oslo University

Hospital for funding the study.
References
[1] Rotondi AJ, Chelluri L, Sirio C, Mendelsohn A, Schulz R, Belle S, et al. Patients' recollections of
stressful experiences while receiving prolonged mechanical ventilation in an
intensive care unit. Crit Care Med 2002;30(4):74652.
[2] Gelinas C. Management of pain in cardiac surgery ICU patients: have we improved
over time? Intensive Crit Care Nurs 2007;23(5):298303.
[3] Barr J, Fraser GL, Puntillo K, Ely EW, Gelinas C, Dasta JF, et al. Clinical practice guidelines for the management of pain, agitation, and delirium in adult patients in the intensive care unit. Crit Care Med 2013;41(1):263306.
[4] Payen JF, Chanques G, Mantz J, Hercule C, Auriant I, Leguillou JL, et al. Current practices in sedation and analgesia for mechanically ventilated critically ill patients: a
prospective multicenter patient-based study. Anesthesiology 2007;106(4):68795.
[5] Rose L, Smith O, Gelinas C, Haslam L, Dale C, Luk E, et al. Critical care nurses' pain assessment and management practices: a survey in Canada. Am J Crit Care 2012;
21(4):2519.
[6] Berben SA, Meijs TH, van Grunsven PM, Schoonhoven L, van Achterberg T. Facilitators and barriers in pain management for trauma patients in the chain of emergency
care. Injury 2012;43(9):1397402.
[7] Horbury C, Henderson A, Bromley B. Inuences of patient behavior on clinical
nurses' pain assessment: implications for continuing education. J Contin Educ Nurs
2005;36(1):1824.
[8] Yildirim YK, Cicek F, Uyar M. Knowledge and attitudes of Turkish oncology nurses
about cancer pain management. Pain Manag Nurs 2008;9(1):1725.
[9] Rose L, Haslam L, Dale C, Knechtel L, Fraser M, Pinto R, et al. Survey of assessment
and management of pain for critically ill adults. Intensive Crit Care Nurs 2011;
27(3):1218.
[10] Coyne P, Lyckholm L, Bobb B, Blaney-Brouse D, Harrington S, Yanni L. Managing pain
with algorithms: an opportunity for improvement? Or: the development and utilization of algorithms to manage acute pain. Pain Manag Nurs 2013;14(4):e1858.
[11] Olsen BF, Rustoen T, Sandvik L, Miaskowski C, Jacobsen M, Valeberg BT. Development of a pain management algorithm for intensive care units. Heart Lung 2015;
44(6):5217.
[12] Olsen BF, Rustoen T, Sandvik L, Miaskowski C, Jacobsen M, Valeberg BT. Implementation of a pain management algorithm in intensive care units and evaluation of
nurses' level of adherence with the algorithm. Heart Lung 2015;44(6):52833.
[13] De Jong A, Molinari N, de Lattre S, Gniadek C, Carr J, Conseil M, et al. Decreasing severe pain and serious adverse events while moving intensive care unit patients: a
prospective interventional study (the NURSE-DO project). Crit Care 2013;17(2):R74.
[14] Rose L, Haslam L, Dale C, Knechtel L, McGillion M. Behavioral pain assessment tool
for critically ill adults unable to self-report pain. Am J Crit Care 2013;22(3):24655.
[15] Arbour CBS, Gelinas C, Michaud C. Impact of the implementation of the critical-care
pain observation tool (CPOT) on pain management and clinical outcomes in mechanically ventilated trauma intensive care unit patients: a pilot study. J Trauma
Nurs 2011;18(1):5260.
[16] Gelinas C, Arbour C, Michaud C, Vaillant F, Desjardins S. Implementation of the
critical-care pain observation tool on pain assessment/management nursing practices in an intensive care unit with nonverbal critically ill adults: a before and
after study. Int J Nurs Stud 2011;48(12):1495504.
[17] Topolovec-Vranic J, Canzian S, Innis J, Pollmann-Mudryj MA, McFarlan AW, Baker AJ.
Patient satisfaction and documentation of pain assessments and management after
implementing the adult nonverbal pain scale. Am J Crit Care 2010;19(4):34554.
[18] Skrobik Y, Ahern S, Leblanc M, Marquis F, Awissi DK, Kavanagh BP. Protocolized intensive care unit management of analgesia, sedation, and delirium improves analgesia and subsyndromal delirium rates. Anesth Analg 2010;111(2):45163.
[19] Mansouri P, Javadpour S, Zand F, Ghodsbin F, Sabetian G, Masjedi M, et al. Implementation of a protocol for integrated management of pain, agitation, and delirium
can improve clinical outcomes in the intensive care unit: a randomized clinical trial.
J Crit Care 2013;28(6):91822.
[20] Chanques G, Jaber S, Barbotte E, Violet S, Sebbane M, Perrigault PF, et al. Impact of
systematic evaluation of pain and agitation in an intensive care unit. Crit Care Med
2006;34(6):16919.

211

[21] Williams TA, Martin S, Leslie G, Thomas L, Leen T, Tamaliunas S, et al. Duration of
mechanical ventilation in an adult intensive care unit after introduction of sedation
and pain scales. Am J Crit Care 2008;17(4):34956.
[22] Puntillo KA, White C, Morris AB, Perdue ST, Stanik-Hutt J, Thompson CL, et al.
Patients' perceptions and responses to procedural pain: results from Thunder Project II.
Am J Crit Care 2001;10(4):23851.
[23] Puntillo KA, Max A, Timsit JF, Vignoud L, Chanques G, Robleda G, et al. Determinants
of procedural pain intensity in the intensive care unit. The Europain(R) study. Am J
Respir Crit Care Med 2014;189(1):3947.
[24] Chanques G, Viel E, Constantin JM, Jung B, de Lattre S, Carr J, et al. The measurement
of pain in intensive care unit: comparison of 5 self-report intensity scales. Pain 2010;
151(3):71121.
[25] Payen JF, Bru O, Bosson JL, Lagrasta A, Novel E, Deschaux I, et al. Assessing pain in
critically ill sedated patients by using a behavioral pain scale. Crit Care Med 2001;
29(12):225863.
[26] Chanques G, Payen JF, Mercier G, de Lattre S, Viel E, Jung B, et al. Assessing pain in
non-intubated critically ill patients unable to self report: an adaptation of the Behavioral Pain Scale. Intensive Care Med 2009;35(12):20607.
[27] Gerbershagen HJ, Rothaug J, Kalkman CJ, Meissner W. Determination of moderateto-severe postoperative pain on the numeric rating scale: a cut-off point analysis applying four different methods. Br J Anaesth 2011;107(4):61926.
[28] Devlin JW, Boleski G, Mlynarek M, Nerenz DR, Peterson E, Jankowski M, et al. Motor
Activity Assessment Scale: a valid and reliable sedation scale for use with mechanically ventilated patients in an adult surgical intensive care unit. Crit Care Med 1999;
27(7):12715.
[29] Ely EW, Truman B, Shintani A, Thomason JW, Wheeler AP, Gordon S, et al. Monitoring sedation status over time in ICU patients: reliability and validity of the Richmond
Agitation-Sedation Scale (RASS). JAMA 2003;289(22):298391.
[30] Le Gall JR, Lemeshow S, Saulnier F. A new Simplied Acute Physiology Score (SAPS
II) based on a European/North American multicenter study. JAMA 1993;270(24):
295763.
[31] Reis Miranda D, Moreno R, Iapichino G. Nine equivalents of nursing manpower use
score (NEMS). Intensive Care Med 1997;23(7):7605.
[32] Bucknall TK, Manias E, Presneill JJ. A randomized trial of protocol-directed sedation
management for mechanical ventilation in an Australian intensive care unit. Crit
Care Med 2008;36(5):144450.
[33] Svendsen K, Borchgrevink PC, Fredheim O, Hamunen K, Mellbye A, Dale O. Choosing
the unit of measurement counts: the use of oral morphine equivalents in studies of
opioid consumption is a useful addition to dened daily doses. Palliat Med 2011;
25(7):72532.
[34] Goodman LS, Brunton LL, Chabner B, BrC K. Goodman & Gilman's pharmacological
basis of therapeutics. 12th ed. New York: McGraw-Hill; 2011.
[35] Craven DE, Lei Y, Ruthazer R, Sarwar A, Hudcova J. Incidence and outcomes of
ventilator-associated tracheobronchitis and pneumonia. Am J Med 2013;126(6):
5429.
[36] Slutsky AS, Ranieri VM. Ventilator-induced lung injury. N Engl J Med 2013;369(22):
212636.
[37] Wunsch H, Linde-Zwirble WT, Angus DC, Hartman ME, Milbrandt EB, Kahn JM. The
epidemiology of mechanical ventilation use in the United States. Crit Care Med
2010;38(10):194753.
[38] Mullins PM, Goyal M, Pines JM. National growth in intensive care unit admissions
from emergency departments in the United States from 2002 to 2009. Acad Emerg
Med 2013;20(5):47986.
[39] Lefrant JY, Garrigues B, Pribil C, Bardoulat I, Courtial F, Maurel F, et al. The daily cost
of ICU patients: a micro-costing study in 23 French intensive care units. Anaesth Crit
Care Pain Med 2015;34(3):1517.
[40] Payen JF, Bosson JL, Chanques G, Mantz J, Labarere J, Investigators D. Pain assessment
is associated with decreased duration of mechanical ventilation in the intensive care
unit: a post hoc analysis of the DOLOREA study. Anesthesiology 2009;111(6):
130816.
[41] Payen JF, Genty C, Mimoz O, Mantz J, Bosson JL, Chanques G. Prescribing nonopioids
in mechanically ventilated critically ill patients. J Crit Care 2013;28(4) [534 e712].
[42] DeBiasi EM, Akgun KM, Pisani M. Awake or sedated: trends in the evaluation and
management of agitation in the intensive care unit. Semin Respir Crit Care Med
2015;36(6):899913.
[43] Devabhakthuni S, Armahizer MJ, Dasta JF, Kane-Gill SL. Analgosedation: a paradigm
shift in intensive care unit sedation practice. Ann Pharmacother 2012;46(4):53040.