Critical Care Evaluation

EHAVIORAL

PAIN

ASSESSMENT TOOL
FOR CRITICALLY ILL
ADULTS UNABLE TO
SELF-REPORT PAIN
By Louise Rose, RN, BN, MN, PhD, ICU Cert, Lynn Haslam, RN, BScN(Hons), MN,
NP(Hons), Craig Dale, RN, BScN(Hons), MN, CNCC(C), Leasa Knechtel, RN,
BScN(Hons), MN, CNCC(C), and Michael McGillion, RN, BScN, PhD

CNE

1.0 Hour

Notice to CNE enrollees:

A closed-book, multiple-choice examination
following this article tests your understanding of
the following objectives:
1. Describe key elements of the behavioral assessment Critical-Care Pain Observation Tool
(CPOT).
2. Evaluate effects of CPOT with change in practice for documentation and administration of
analgesics and sedatives.
3. Compare results among studies regarding compliance with pain assessment documentation
and practice recommendations.
To read this article and take the CNE test online,
visit www.ajcconline.org and click CNE Articles
in This Issue. No CNE test fee for AACN members.
2013 American Association of Critical-Care Nurses
doi: http://dx.doi.org/10.4037/ajcc2013200

246

Background Critically ill adults often cannot self-report pain.

Objective To determine the effect of the Critical-Care Pain
Observation Tool on frequency of documentation of pain
assessment and administration of analgesics and sedatives in
critically ill patients unable to self-report pain.
Methods Data on patients in 2 intensive care units of a
university-affiliated hospital were collected before and after
implementation of the tool. Patients were prospectively screened
for eligibility; data were extracted retrospectively.
Results Data were recorded for a maximum of 72 hours before
and after implementation of the tool in the cardiovascular intensive care unit (130 patients before and 132 after) and in the
medical/surgical/trauma unit (59 patients before and 52 after).
Proportion of pain assessment intervals with pain assessment
documented increased from 15% to 64% (P < .001) in the cardiovascular unit and from 22% to 80% (P < .001) in the other
unit. Median total dose of opioid analgesics decreased from 5
mg to 4 mg in the cardiovascular unit (P = .02) and increased
from 27 mg to 75 mg (P = .002) in the other unit. Median total
dose of benzodiazepines decreased from 12 mg to 2 mg (P <
.001) in the cardiovascular unit and remained unchanged in
the other unit. Increased documentation of pain assessment
was associated with increased age in the cardiovascular unit
and with decreased maximum scores on the Sequential Organ
Failure Assessment in the other unit.
Conclusion Implementation of the tool increased frequency
of pain assessment and appeared to influence administration
of analgesics in both units. (American Journal of Critical Care.
2013;22:246-255)

AJCC AMERICAN JOURNAL OF CRITICAL CARE, May 2013, Volume 22, No. 3

www.ajcconline.org

tudies1-5 on recall of moderate to severe pain challenge the effectiveness of pain

management during an intensive care unit (ICU) stay. Patients inability to selfreport pain is a marked barrier to effective assessment and management of pain.
Because impaired communication is common among critically ill patients as a
result of sedation, altered level of consciousness, and endotracheal intubation,6
clinicians cannot use self-report tools and must rely on alternative methods to determine if a
patient has pain.

Several behavioral pain assessment tools7-11 are

now available that facilitate detection of pain experienced by critically ill patients unable to communicate. Systematic pain assessment, with either patient
self-reporting or use of behavioral pain assessment
tools as appropriate, can improve patients outcomes.
In a large multicenter observational study,12 pain
assessment was associated with reductions in the duration of mechanical ventilation and ICU stay. In a
small study13 involving patients in a neurotrauma
ICU, introduction of the Nonverbal Pain Scale9
increased documentation of pain assessments and
decreased recalled severity of the pain patients
experienced. More recently, Glinas et al14 reported
increased pain documentation and decreased administration of analgesic and sedative agents after introduction of the Critical-Care Pain Observation Tool
(CPOT) in a small mixed ICU population (30 patients
before, 30 at 3 months, and 30 at 12 months after
implementation). However, few studies have evaluated the effect of these tools on pain assessment and
management practices; most published studies7,8,14
have been conducted by investigators involved in
the development and or validation of the tools.
Our goal was to determine the effect of implementing the CPOT7 in 2 ICUs of a university-affiliated
hospital that provide services to a mixed population

About the Authors

Louise Rose is a Lawrence S. Bloomberg limited-term
professor in critical care, Lawrence S. Bloomberg Faculty
of Nursing, University of Toronto, Toronto, Ontario,
Lynn Haslam is an advanced practice nurse, Department
of Anaesthesia, and Leasa Knechtel is the director of
nursing education, Sunnybrook Health Sciences Centre,
Toronto, Ontario, Craig Dale is an advanced practice
nurse, Department of Trauma, Emergency and Critical
Care, Sunnybrook Health Sciences Centre, and a PhD
candidate, Lawrence S. Bloomberg Faculty of Nursing,
and Michael McGillion is an assistant professor, Lawrence
S. Bloomberg Faculty of Nursing, and a member of the
board of directors of the Canadian Pain Society.
Corresponding author: Louise Rose, Lawrence S. Bloomberg
Faculty of Nursing, University of Toronto, 155 College
St, Toronto, Ontario, Canada, M5T IP8 (e-mail: louise.rose
@utoronto.ca).

www.ajcconline.org

of patients, including trauma and cardiothoracic

surgery patients. We hypothesized that implementation of the CPOT would increase documentation of
pain assessment and influence administration of
analgesics and sedatives. Our primary objective was
to determine the effects on the frequency of documentation of pain assessment (pain scores and narrative) and on the administration of analgesics and
sedatives in patients unable to self-report pain. Our
secondary objectives were to determine patient factors associated with documented pain assessment
and opioid administration and to
examine the impact of CPOT implementation on ICU length of stay and
the duration of mechanical ventilation.

Methods

Systematic pain
assessment
improves patients
outcomes, reducing mechanical
ventilation time
and length of stay.

Study Design, Participants, and

Setting
A before-and-after design was
used to examine the effect of CPOT
implementation in 2 ICUs at Sunnybrook Health Sciences Centre, a 600bed university-affiliated hospital in
Toronto, Ontario. The ICUs were a 20-bed mixed
medical/surgical/trauma ICU (CRCU) that admits
more than 1100 patients annually and a 14-bed cardiovascular ICU (CVICU) that admits 1150 patients
each year. Both ICUs functioned as closed intensivistled units. Each week, the 20-bed CRCU was overseen
by 2 intensivists; the 14-bed CVICU was supervised
by 1 intensivist. A team of medical trainees, including fellows and residents, supported each intensivist
to provide 24-hour care. These ICUs employed more
than 100 (CRCU) and 65 registered nurses (CVICU)
in full- and part-time positions.
The CPOT consists of 4 domains: facial
expression, body movement, muscle tension, and
compliance with the ventilator (or vocalization for
nonintubated patients). Each domain is scored from
0 to 2, with a maximum score of 8. The tool has
content validity, moderate to high interrater reliability, discriminate validity, and moderate criterion

AJCC AMERICAN JOURNAL OF CRITICAL CARE, May 2013, Volume 22, No. 3

247

validity.7,16,17 Before implementation of the CPOT,

the pain management policies of the ICUs did not
include use of a behavioral pain assessment tool,
although individual nurses self-reported that they
used various tools, including the CPOT.18 The recommended frequency of pain assessment, or pain
assessment intervals (PAIs), for surgical patients was
hourly for 6 hours postoperatively and then a minimum of every 4 hours. Nonsurgical
patients were expected to have pain
assessment documented a minimum
of every 4 hours.
For the baseline phase of the
study, patients were recruited from
September 2008 to January 2009.
After a 4-month implementation
phase, patients were recruited from
June to October 2009. Patients were
eligible if they were unable to communicate verbally or via other means, as determined
by documented failure to follow verbal commands
or a motor score of 5 or less on the Glasgow Coma
Scale. Each patients inability to communicate was
confirmed with the patients bedside nurse. Patients
were excluded if they were receiving neuromuscular
blockers at the time of screening, were readmitted
to the ICU and had previously been enrolled in the
study, or were in the ICU during both study phases.

Two intensive
care units were
used, a 20-bed
mixed unit and a
14-bed cardiovascular unit.

Data Collection
Research staff prospectively screened the eligibility of all consecutive patients admitted to the 2
ICUs. In order to minimize the impact of data collection on critical care nurses practices of documenting pain assessment, the relevant data were extracted
from each study patients record retrospectively after
the patient had been discharged from the ICU.
Demographic data included age,
sex, admission type, primary reason
for ICU admission, and number of
invasive catheters or tubes. Additional information collected included
frequency and type of documentation (either pain score or narrative
description) of pain assessment
from the time of inclusion in the
study until the patient regained the
ability to communicate (indicated
by a motor score of 6 on the Glasgow Coma Scale or nursing documentation) or a
maximum of 72 hours; type, delivery method, and
dose of analgesic and sedative medications administered; and daily scores on the Sequential Organ
Failure Assessment (SOFA), duration of mechanical

For extraction of
pain assessment
descriptors, potential behavioral
descriptors were
compiled from
published tools.

248

ventilation, and length of ICU stay. In order to

guide extraction of narrative descriptions of pain
assessment, a reference compendium of potential
behavioral descriptors was compiled from published
behavioral pain assessment tools.7-9,11,19,20 Data abstractors were instructed to record verbatim all documentation potentially related to pain assessment
and management, including ambiguous documentation. The abstractors excluded any reference to
pain behaviors elicited during routine neurological
assessment. Random audits of data extraction were
done to ensure consistency of nurses narratives of
pain documentation recorded by research staff.
Tool Implementation
Before use of the CPOT was implemented, all
nurses attended educational sessions that included
video demonstration of pain behaviors and instruction on application of CPOT. Videos were provided
by Dr Glinas, who developed the CPOT and who
used the videos in the study of CPOT implementation.14 Existing unit protocols and ICU flow sheets
were modified to incorporate the CPOT. Point-ofcare CPOT scoring guides were available at every
bedside, posters were displayed in prominent locations, and educational materials were posted on the
ICUs Web portal and published in newsletters. The
senior nursing team provided focused 1-on-1 bedside education during implementation and monitored compliance via monthly random chart audits.
Results of monthly audits were e-mailed to staff,
posted on notice boards, and discussed at staff
meetings. Auditing for compliance with the pain
assessment policy was incorporated into individual
performance reviews. Senior nurses and physicians
were involved in tool selection and championed
implementation through existing quality and education forums.
Statistical Methods
Assessment of the primary outcome, the frequency of documentation of pain assessment,
required a sample size of 524 PAIs (524 before and
524 after) in each participating ICU to detect a 10%
difference in the frequency of the documentation
with 90% power and a = .05.
Descriptive statistics were used to summarize
demographic characteristics and doses of medications.
Continuous variables were described by using measures of central tendency and spread (means and
standard deviations or medians and interquartile
ranges, depending on data distribution). Frequencies,
proportions, and 95% confidence intervals were
used to describe categorical variables. The overall

AJCC AMERICAN JOURNAL OF CRITICAL CARE, May 2013, Volume 22, No. 3

www.ajcconline.org

Table 1
Patients demographics
No. (%) of patients
CVICU

CRCU

Before
(n = 130)

After
(n = 132)

70 (61-76)
91 (70)

Admission category
Surgical
Trauma
Medical
Reason for ICU admission
Cardiovascular
Respiratory
Gastrointestinal
Neurologic
Trauma
Metabolic
Genitourinary
Hematologic
Sepsis

Age, median (IQR), y

Male sex

Before
(n = 59)

After
(n = 52)

67 (60-75)

57 (38-75)

54 (43-63)

93 (70)

42 (71)

40 (77)

123 (95)
0 (0)
7 (5)

124 (94)
2 (2)
6 (4)

13 (22)
28 (48)
18 (30)

6 (12)
23 (44)
23 (44)

123 (95)
3 (2)
0 (0)
2 (2)
0 (0)
2 (2)
0 (0)
0 (0)
0 (0)

125 (95)
3 (2)
1 (1)
0 (0)
2 (2)
0 (0)
0 (0)
3 (2)
1 (1)

1 (2)
14 (24)
2 (3)
9 (15)
28 (48)
1 (2)
2 (3)
0 (0)
2 (3)

2 (4)
5 (10)
7 (14)
6 (12)
23 (44)
0 (0)
0 (0)
3 (6)
6 (12)

Number of invasive catheters/tubes, median (IQR)

9 (8-9)

9 (8-10)

8 (7-9)

6 (6-7)a

Maximum SOFA score,b median (IQR)

6 (3-8)

4 (3-6)a

8 (5-11)

8 (6-11)

Abbreviations: CRCU, medical/surgical/trauma unit; CVICU, cardiovascular intensive care unit; ICU, intensive care unit; IQR, interquartile range.
a
b

P < .05 before and after study phases.

The maximum Sequential Organ Failure Assessment (SOFA) score22 was calculated by summing the worst (greatest) scores for all 6 components of the
SOFA score recorded daily during inclusion in the study. The 6 components of the SOFA score are respiratory (ratio of PaO2 to fraction of inspired oxygen),
coagulation (platelets), liver (bilirubin), cardiovascular (hypotension), central nervous system (score on Glasgow Coma Scale), and renal (creatinine). Each
is scored from 0 (no organ failure) to 4, with a maximum score of 24.

proportion of PAIs in which pain assessment was

documented before and after CPOT implementation was determined for each ICU and the results
were compared by using c2 tests. Median pain
assessments per patient and medication doses
before and after CPOT implementation were
compared by using Wilcoxon rank sum tests.
Duration of mechanical ventilation and ICU stay
were determined by using time-to-event methods,
which account for censoring due to death, and
were compared by using log-rank tests. Multiple
Poisson regression was used to examine prospectively chosen patient factors associated with pain
assessment in each ICU. In order to examine
patient characteristics associated with median
opioid equivalent total dose, multivariable models were constructed according to ICU and study
phase. Models were assessed for collinearity and
goodness of fit. All tests were 2-tailed, and P = .05
was considered significant. Analyses were performed by an independent statistician who used
SAS 9.1 software (SAS Institute). Analysis of the
narrative descriptions of pain has been described
in detail elsewhere.21

www.ajcconline.org

Results
A total of 189 patients were recruited before
implementation of the CPOT and 184 patients after
implementation. Demographic characteristics for
both study phases according to ICU are shown in
Table 1. Patient characteristics in the 2 study phases
were similar except for median maximum SOFA
scores in the CVICU and number of catheters in
the CRCU cohort. In both units, the number of
PAIs did not differ significantly during the study
phases. In the CVICU, the proportion of PAIs with
pain assessment documented increased from 15%
to 64% (P < .001) and from 22% to 80% (P < .001)
in the CRCU. The median number of PAIs for each
patient with documented pain assessment increased
after CPOT implementation in both ICUs (Table 2).
Because an increase in documentation of a
behavioral pain score after implementation of the
CPOT was anticipated, the frequency of documentation of narrative assessments of behavioral and physiological indicators of pain was determined. The
number of narrative assessments increased in the
CVICU and were unchanged in the CRCU (Table 2).
For the CVICU patients, the median maximum

AJCC AMERICAN JOURNAL OF CRITICAL CARE, May 2013, Volume 22, No. 3

249

Table 2
Documentation of pain assessment
No. of patients, median (IQR)
CVICU

CRCU

Documentation

Before
(n = 130)

After
(n = 132)

Eligible hoursa

1321, 3 (2-6)

893, 3 (2-6)

.20

2390, 37 (14-72)

2627, 68 (26-72)

.09

633, 3 (2-6)

519, 3 (2-5)

.18

580, 9 (3-18)

646, 17 (7-18)

.08

180, 1 (0-2)

341, 2 (1-3)

<.001

213, 2 (1-5)

516, 11 (5-16)

<.001

96, 0 (0-1)

333, 2 (1-3)

<.001

127, 1 (1-4)

515, 11 (5-16)

<.001

Narrative episodes (all)

254, 1 (0-2)

172, 1 (1-2)

.001

336, 3 (2-8)

310, 2 (2-8)

.56

Narrative episodes (eligible)

130, 0 (0-1)

147, 1 (1-2)

<.001

262, 2 (1-7)

218, 3 (1-6)

.93

PAIs
PAIs with pain assessment (all)
PAIs with pain assessment

(eligible)b

Before
(n = 59)

After
(n = 52)

Abbreviations: CRCU, medical/surgical/trauma unit; CVICU, cardiovascular intensive care unit; IQR, interquartile range; PAIs, pain assessment intervals.
a
b

Refers to the number of hours of data collection during which patients remained eligible for the study.
Excludes PAIs coded as 2 (ambiguous as to whether documented assessment referred to pain, agitation, or delirium because interventions for management of agitation and delirium coincided with documentation) and 4 (documentation of analgesia given, though no assessment data recorded).

CPOT score was 0 (interquartile range [IQR], 0-2),

and 28 of the 274 documented scores (10%) were 3
or greater, indicating the presence of pain.23 For 11 of
the 28 scores (39%), patients medical records had
no documentation of administration of an analgesic.
For CRCU patients, the median score was 4 (IQR,
1-5), and 104 of the 693 scores (15%) were 3 or
greater; for 43 scores (41%), medical records had no
documentation of administration of an analgesic.
Table 3 presents median total and hourly doses
of analgesic and sedative agents administered before
and after CPOT implementation. Median total opioid equivalent doses decreased by 1 mg in the
CVICU (P < .001) and increased by 48 mg in the
CRCU (P < .001). The total dose of benzodiazepines
decreased in the CVICU by 10 mg (P < .001), but
remained unchanged in the CRCU. In the CVICU,
both before and after CPOT implementation,
higher total opioid doses were received by sicker
patients (higher maximum SOFA scores) and by
patients admitted for medical indications. In the
CRCU, after CPOT implementation, patients admitted for medical indications received less total opioid
compared with patients admitted after surgery or
trauma (P = .001), when adjustments were made for
age, sex, number of invasive devices, maximum SOFA
score, and duration of ICU stay; no differences were
noted before CPOT implementation.
Accounting for the number of PAIs, admission
category, sex, and maximum SOFA score, increased
age was associated with increased documentation
of pain assessment in the CVICU. In the CRCU,
decreased maximum SOFA scores were associated
with increased documentation (Table 4). In the
CVICU, the estimated median duration of ICU stay,

250

determined by using the Kaplan-Meier method,

decreased from 2.0 (IQR, 1.0-5.0) days to 1.8 (IQR,
1.0-3.0) days (P = .007); no difference was found in
the CRCU (median, 5.9; IQR, 2.9-13.6 days before
and median, 7.0; IQR, 5.0-14.7 days after). No difference was found in the duration of mechanical
ventilation before and after CPOT implementation
in either ICU. The CVICU median was 0.6 (IQR,
0.3-0.8) days before CPOT and 0.5 (IQR, 0.3-0.8)
days after implementation. The CRCU median was
3.9 (IQR, 1.4-10.7) days before implementation and
5.9 (IQR, 3.0-9.1) days after use of the CPOT began.

Discussion
Our findings indicate successful implementation of the CPOT in terms of improved compliance
with regular documentation of pain assessment in
2 ICUs with no previous formal use of this, or any
other, behavioral pain assessment tool. Although
we showed improved compliance with the policyrecommended frequency of documented pain
assessment, we did not achieve compliance rates
higher than 80%. Although 100% is obviously
preferable, 80% compliance may be acceptable in
ICUs with large numbers of nurses, staff turnover,
and acutely ill patients. The rate was also a marked
improvement from compliance at baseline. A similar compliance rate for documented pain assessment
was reported by Topolovec-Vranic et al,13 whereas
Glinas et al14 reported a median of 12 documented
assessments (every 2 hours) per 24 hours of ICU
stay 12 months after CPOT implementation,
although compliance cannot be calculated because
institutional recommendations for frequency of
documentation were not described.

AJCC AMERICAN JOURNAL OF CRITICAL CARE, May 2013, Volume 22, No. 3

www.ajcconline.org

Table 3
Drug dosesa
CVICU
Drug
Fentanyl, total, g

No.
21

94

107

104

2
(2-2)

5
(3-9)

112

1.6
(1.6-1.6)

24
(24-24)

Ketamine, per hour

Gabapentin, total

0.6
(0.4-1.0)

1.4
(0.8-3.0)

Opioid equivalents, per hour

Ketamine, total, g

11

0.0
(0.0-0.0)

Morphine, per hour

Opioid equivalents, total

800
(100-2650)

0.2
(0.1-0.3)

Hydromorphone, per hour

Morphine, total

No.

47
(15-82)

Fentanyl, per hour, g

Hydromorphone, total

Before
(n = 130)

CRCU

16

100
(100-225)

14
(9-18)

Gabapentin, per hour

After
(n = 132)

100
(50-670)

.17

14
(4-50)

.09

0.6
(0.4-0.8)

.26

0.2
(0.1-0.3)

.84

4
(3-7)

.02

1.2
(0.6-2.0)

.04

1.0
(1.0-1.0)

31
(31-31)

100
(100-200)

.85

17
(11-21)

.50

No.
38

Before
(n = 59)
300
(100-725)

No.
44

After
(n = 52)
775
(313-2712)

.003

20
(8-83)

.006

2.2
(1.9 -2.4)

.85

0.1
(0.0-0.2)

>.99

16
(11-20)

.79

0.2
(0.2-0.3)

.79

75
(30-258)

.002

2
(0.6-8.0)

.008

10
(4-21)
5

1.6
(0.7-3.0)

0.1
(0.0-0.3)
3

15
(2-17)

0.2
(0.0-0.3)
42

27
(11-73)

45

1.0
(0.4-2.3)

350
(200-500)

0
(0-0)

0.2
(0.1-0.8)

13
(8-18)

Acetaminophen, total, g

36

1
(1-3)

16

1
(1-4)

.95

32

3
(1-6)

33

5
(2-9)

.17

Benzodiazepine, total

99

12
(4-60)

63

2
(2-5)

<.001

21

2
(1-3)

25

2
(2-2)

.34

1
(0.3-2)

<.001

0.0
(0.0-0.2)

.93

418
(118-1513)

.46

14
(3-30)

.28

8
(5-13)

>.99

3
(1-19)

Benzodiazepine, per hour

Propofol, total

11

Haloperidol, per hour

20

15
(10-50)

Propofol, per hour

Haloperidol, total

150
(90-400)

5
(5-5)

0.1
(0.1-0.1)

103
(30-225)

.33

15
(9-27)

.43

5
(3-7)

>.99

0.5
(0.2-0.7)

.44

0.1
(0.0-0.2)
20

603
(378-1375)

28

23
(11-34)
2

10
(2-80)
0.3
(0.0-1.4)

0.1
(0.1-0.2)

.68

Abbreviations: CRCU, medical/surgical/trauma unit; CVICU, cardiovascular intensive care unit.

a

Data are median (interquartile range) for hourly dose. All doses are in milligrams unless otherwise indicated. Dashes indicate no data.

Numerous observational studies24-26 have indicated poor compliance with various evidence-based
practice recommendations, indicating that implementation of the recommendations remains a
challenge in the ICU. Assessment of the need for

www.ajcconline.org

medication and of the effect of the medication is a

central tenet of medication administration that
guides initiation, escalation, and discontinuation of
therapy. However, systematic assessment of pain in
critically ill patients remains infrequent and is done

AJCC AMERICAN JOURNAL OF CRITICAL CARE, May 2013, Volume 22, No. 3

251

Table 4
Multivariable analysis of patients characteristics
associated with documented pain assessment
CVICU
Before CPOT

CRCU
After CPOT

IRR (95% CI)

IRR (95% CI)

PAIs

1.15 (1.11-1.19)

<.001

1.16 (1.13-1.19)

Age

Before CPOT
P

<.001

After CPOT

IRR (95% CI)

1.19 (1.14-1.25)

<.001

IRR (95% CI)

1.12 (1.10-1.14)

<.001

0.99 (0.98-1.00)

.15

1.01 (1.00-1.02)

.003

0.99 (0.98-1.00)

.14

1.00 (0.99-1.00

.16

Maximum SOFA score 0.98 (0.92-1.05)

.53

1.03 (1.00-1.07

.09

0.99 (0.93-1.06)

.76

0.98 (0.96-1.00)

.04

Sex

0.86 (0.52-1.44)

.58

1.17 (0.96-1.41

.11

1.13 (0.73-1.76)

.59

0.94 (0.78-1.13)

.50

Admission category
Surgical
Trauma
Medical

1
1.28 (0.66-2.50)
1.0 (1.00-1.00)

.46
>.99

1
2.48 (1.27-4.81)
1.26 (0.95-1.68)

.007
.10

1
0.89 (0.48 -1.64)
1.41 (0.79-2.51)

.71
.25

1
0.95 (0.75-1.20)
0.91 (0.73-1.14)

.66
.43

Abbreviations: CRCU, medical/surgical/trauma unit; CVICU, cardiovascular intensive care unit; IRR, incidence rate ratio; PAIs, pain assessment intervals;
SOFA, Sequential Organ Failure Assessment.

inconsistently. For example, Payen et al27 found that

only half of the patients treated with opioids on day
2 of the patients ICU stay had documented assessments of pain. Other studies13,14 on implementation
of behavioral pain assessment tools had low baseline rates of documented pain assessment. In a recent
multicenter 1-day point prevalence study13 of 10
routine care processes involving 50 ICUs and more
than 650 patients, variability was detected in compliance in all care practices, but documentation of
pain scores was one of the practices with the lowest
compliance (35%; IQR, 17%-62%). We noted
increased pain documentation for older patients in
the CVICU and for patients with decreased organ
failure in the CRCU. Possibly nurses thought that
older adults need more frequent pain assessments
than do younger patients. Competing priorities for
nurses time may influence the nurses ability to
complete and document pain assessment in patients
with higher severity of illness.
We found that implementation of CPOT had
different effects on opioid and benzodiazepine
administration in the 2 study ICUs. In the CVICU,
a small but significant decrease occurred in use of
opioid analgesics and a large decrease in the
administration of benzodiazepines. In the CRCU,
the amount of opioid analgesics administered
increased dramatically, and benzodiazepine usage
was unchanged. The decreases in the CVICU may
have been due to both implementation of sedation
strategies targeting minimal sedation and provision
of the CPOT to guide pain assessment. Before
implementation of the CPOT, and in the absence
of a systematic pain assessment tool, nurses in the

252

CVICU may have thought that most patients had

pain related to procedures and incision sites28 and
routinely administered preemptive analgesics. Possibly CPOT implementation enabled nurses to better detect the presence or absence of pain and to
titrate administration of analgesics accordingly.
Because of the minimal differences in patient
characteristics across the 2 ICUs after CPOT implementation, increased administration of opioids in
the CRCU might be due to improved detection and
management of pain. However, in other studies,
implementation of a behavioral pain tool had the
opposite effect on opioid administration. Glinas
et al14 detected a reduction in opioid administration
and postulated that this effect was due to improved
guidance in pain management decisions and ability
to discriminate pain from other symptoms as well
as a reduction in the number of trauma patients
after implementation. Similarly, decreased use of
opiates and better pain control were achieved in a
large before-and-after study29 in which protocols for
management of analgesia, sedation, and delirium
were implemented that included targeting of analgesia to pain scores. Divergent findings such as
these on opioid usage are context dependent and
expected. Targeting either increased or decreased
administration of analgesics as a desirable outcome
has limited clinical usefulness. Rather, administration of analgesics must be based on the intensity
and nature of the pain problem for individual
patients so that the patients receive appropriate
doses of the medications.
Although implementation of CPOT increased
documentation of pain assessment and potentially

AJCC AMERICAN JOURNAL OF CRITICAL CARE, May 2013, Volume 22, No. 3

www.ajcconline.org

influenced use of analgesics and benzodiazepines,

approximately 40% of CPOT scores indicating the
presence of pain were not followed by administration of an analgesic. Our intervention included nurse
education emphasizing the need for analgesics for
patients with CPOT scores of 3 or greater but did
not include revision of orders to formalize prescription of an analgesic in response to high CPOT
scores. A recent survey30 of pain assessment practices across Canada indicated infrequent targeting
of analgesia to a pain score or other assessment
parameters by physicians. However, the potential
for improved patient outcomes as a result of implementation of systematic pain assessment linked to
an intervention strategy to treat pain and prevent
escalation of pain cannot be overlooked. Chanques
et al31 reported that such a strategy reduced the incidence of pain, pain severity, the duration of mechanical ventilation and rate of nosocomial infections.
The lack of influence on duration of ICU stay and
mechanical ventilation in our study may have
occurred because pain assessment findings were
not directly linked to prescribing of analgesics.

Limitations
Because of the limitations of the study design,
our results may have been influenced by unaccounted confounders such as other ongoing quality initiatives; in particular, a sedation score and
algorithm targeting low levels of sedation were
introduced at the same time as the CPOT. Other
potential confounders include turnover of physicians
and nurses and differences in patient characteristics,
although the cohort groups were equivalent in
terms of measured baseline characteristics except
for a lower maximum SOFA scores in the CVICU
patients and fewer invasive catheters in the CRCU
patients after implementation of the CPOT. Our
results may also be subject to performance bias despite
our attempts to avoid influencing pain assessment
practices by collecting most data from the medical
record after patients were discharged from the ICU.
Ascertainment bias may have occurred despite random audits of the accuracy of data extraction because
the persons who extracted information from nurses
narratives were not blinded to the study period or
the study purpose. Because the data collection was
retrospective, we were unable to accurately determine assessment of response to analgesics. The exact
time of assessment was not well documented, and
the documentation did not directly describe response
to treatment; narrative notation is often documented
by nurses in retrospect rather than real time. We were
5 PAIs short of our target sample size in the CVICU

www.ajcconline.org

after CPOT implementation because 5 patients were

excluded during data cleaning. Because we used an
extremely conservative estimate of effect in calculating our sample size and the observed effect size was
large, this limitation does not influence interpretation of our results. Additionally, we were unable to
present data on PAIs with documented pain assessment per day of ICU stay; we discontinued data collection as soon as patients were able to communicate,
making direct comparison with the results of other
studies problematic.

Conclusion
Implementation of the CPOT increased the frequency of pain assessment and most likely influenced administration of opioid analgesics in the 2
ICUs. Few episodes of pain were detected by using
the CPOT, although approximately 40% of pain
episodes were not addressed with administration of
analgesia, suggesting the need to link pain management to assessment findings. The effectiveness of
the CPOT for optimizing appropriate administration of an analgesic in critically ill patients unable
to self-report needs to be evaluated in randomized
controlled trials.
ACKNOWLEDGMENTS
We thank Alex Kiss for his advice and assistance with
statistical analyses.
FINANCIAL DISCLOSURES
This study was supported by a grant from the Sunnybrook
Health Sciences Centre Practice Based Research Fund.
eLetters
Now that youve read the article, create or contribute to an
online discussion on this topic. Visit www.ajcconline.org
and click Submit a response in either the full-text or
PDF view of the article.

SEE ALSO
For more about pain management, visit the Critical
Care Nurse Web site, www.ccnonline.org, and read the
article by Arbour and Glinas, Setting Goals for Pain
Management When Using a Behavioral Scale: Example With the Critical-Care Pain Observation Tool.
(December 2011).
REFERENCES
1. Rotondi A, Chelluri L, Sirio C, et al. Patients recollections
of stressful experiences while receiving prolonged mechanical ventilation in an intensive care unit. Crit Care Med.
2002;30(4):746-752.
2. Puntillo KA. Pain experiences of intensive care unit patients.
Heart Lung. 1990;19:526-533.
3. Whipple J, Lewis K, Quebbeman E, et al. Analysis of pain
management in critically ill patients. Pharmacotherapy.
1995;15:592-599.
4. Arroyo-Novoa C, Figueroa-Ramos M, Puntillo K, et al. Pain
related to tracheal suctioning in awake acutely and critically
ill adults: a descriptive study. Intensive Crit Care Nurs.
2008;24:20-27.
5. Ethier C, Burry L, Martinez-Motta C, et al; Canadian Critical
Care Trials Group. Recall of intensive care unit stay in

AJCC AMERICAN JOURNAL OF CRITICAL CARE, May 2013, Volume 22, No. 3

253

6.
7.

8.

9.

10.
11.

12.

13.

14.

15.
16.

17.

18.

19.

254

patients managed with a sedation protocol or a sedation

protocol with daily sedative interruption: a pilot study. J Crit
Care. 2011;26:127-132.
Kwekkeboom K, Herr K. Assessment of pain in the critically
ill. Crit Care Nurs Clin North Am. 2001;13:181-194.
Glinas C, Fillion L, Puntillo KA, Viens C, Fortier M. Validation of the Critical-Care Pain Observation Tool in adult
patients. Am J Crit Care. 2006;15:420-427.
Payen J-F, Bru O, Bosson J-L, et al. Assessing pain in critically ill sedated patients by using a behavioral pain scale.
Crit Care Med. 2001;29:2258-2263.
Odhner M, Wegman D, Freeland N, Steinmetz A, Ingersoll
G. Assessing pain control in nonverbal critically ill adults.
Dimens Crit Care Nurs. 2003;22(6):260-267.
Feldt K. The Checklist of Nonverbal Pain Indicators (CNPI).
Pain Manage Nurs. 2000;1:13-21.
Puntillo KA, Morris AB, Thompson CL, Stanik-Hutt J, White
CA, Wild LR. Pain behaviors observed during six common
procedures: results from Thunder Project II. Crit Care Med.
2004;32:421-427.
Payen J, Bosson J, Chanques G, Mantz J, Labarere J;
DOLOREA Investigators. 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):1308-1316.
Topolovec-Vranic J, Canzian S, Innis J, Pollmann-Mudryj
M, White McFarlan A, Baker A. Patient satisfaction and
documentation of pain assessments and management after
implementing the Adult Nonverbal Pain Scale. Am J Crit
Care. 2010;19:345-354.
Glinas 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):1495-1504.
Rubenfeld G, Angus D. Are intensivists safe? Ann Intern
Med. 2008;148:877-879.
Glinas C, Johnston C. Pain assessment in the critically ill
ventilated adult: validation of the Critical-Care Pain Observation Tool and physiologic indicators. Clin J Pain. 2007;
23:497-505.
Glinas C, Arbour C. Behavioral and physiologic indicators
during a nociceptive procedure in conscious and unconscious mechanically ventilated adults: similar or different?
J Crit Care. 2009;24:628.e627-628.e617.
Rose L, Haslam L, Dale C, et al. Survey of assessment and
management of pain for critically ill adults. Intensive Crit
Care Nurs. 2011;7:121-128.
Aissaoui Y, Zeggwagh A, Zekraoui A, Abidi K, Abouqal R.
Validation of a behavioural pain scale in critically ill,
sedated, and mechanically ventilated patients. Anesth
Analg. 2005;101:1470-1476.

20. Puntillo KA, Stannard D, Miaskowski C, Kehrle K, Gleeson

S. Use of a pain assessment and intervention notation
(PAIN) tool in critical care nursing practice: nurses evaluations. Heart Lung. 2002;31:303-314.
21. Haslam L, Rose L, Dale C, Knechtel L. Pain descriptors for
critically ill patients unable to self-report. J Adv Nurs. 2012;
68(5):1082-1089.
22. Moreno R, Vincent J, Matos R, et al. The use of maximum
SOFA score to quantify organ dysfunction/failure in intensive care: results of a prospective, multicentre study. Working Group on Sepsis-Related Problems of the ESICM.
Intensive Care Med. 1999;25(7):686-696.
23. Glinas C, Harel F, Fillion L, Puntillo K, Johnston C. Sensitivity and specificity of the Critical-Care Pain Observation
Tool for the detection of pain in intubated adults after cardiac surgery. J Pain Symptom Manage. 2009;37(1):58-67.
24. Rose L, Baldwin I, Crawford T, Parke R. Semirecumbent
positioning in ventilator-dependent patients: a multicenter,
observational study. Am J Crit Care. 2010;19(6):e100-e108.
25. Umoh N, Fan E, Mendez-Tellez P, et al. Patient and intensive care unit organizational factors associated with low
tidal volume ventilation in acute lung injury. Crit Care Med.
2008;36:1463-1468.
26. Ilan R, Fowler R, Geerts R, Pinto R, Sibbald W, Martin C.
Knowledge translation in critical care: factors associated
with prescription of commonly recommended best practices
for critically ill patients. Crit Care Med. 2007;35:1696-1702.
27. Payen J-F, Chanques G, Mantz J, et al. Current practices in
sedation and analgesia for mechanically ventilated critically ill patients: a prospective multicenter patient-based
study. Anesthesiology. 2007;106(4):687-695.
28. Chanques G, Sebbane M, Barbotte E, Viel E, Eledjam J-J,
Jaber S. A prospective study of pain at rest: incidence and
characteristics of an unrecognized symptom in surgical
and trauma versus medical intensive care unit patients.
Anesthesiology. 2007;107:858-860.
29. Skrobik Y, Ahern S, Leblanc M, Marquis F, Awissi D,
Kavanagh B. Protocolized intensive care unit management of
analgesia, sedation, and delirium improves analgesia and
subsyndromal delirium rates. Anesth Analg. 2010;111:451-463.
30. Rose L, Smith O, Glinas C, et al. Critical care nurses pain
assessment and management practices: a survey in Canada.
Am J Crit Care. 2012;21(4):251-259.
31. Chanques G, Jaber S, Barbotte E, et al. Impact of systematic
evaluation of pain and agitation in an intensive care unit.
Crit Care Med. 2006;34:1691-1699.

To purchase electronic or print reprints, contact The

InnoVision Group, 101 Columbia, Aliso Viejo, CA 92656.
Phone, (800) 899-1712 or (949) 362-2050 (ext 532); fax,
(949) 362-2049; e-mail, reprints@aacn.org.

AJCC AMERICAN JOURNAL OF CRITICAL CARE, May 2013, Volume 22, No. 3

www.ajcconline.org

CNE Test

Test ID A132232: Behavioral Pain Assessment Tool for Critically Ill Adults Unable To Self-Report Pain
Learning objectives: 1. Describe key elements of the behavioral assessment Critical-Care Pain Observation Tool (CPOT). 2. Evaluate effects of CPOT with change
in practice for documentation and administration of analgesics and sedatives. 3. Compare results among studies regarding compliance with pain assessment
documentation and practice recommendations.
1. Which of the following has been identif ied as the most signif icant barrier
to effective assessment and management of pain in the intensive care unit?
a. Patient inability to self report
b. Nurses perception of pain management
c. Family interventions
d. Documentation

7. Which of the following admissions categories provided the most patients for
the study?
a. Surgical
b. Trauma
c. Medical
d. Respiratory

2. Those conducting the Critical-Care Pain Observation Tool (CPOT) study

hypothesized that implementation of CPOT would do which of the following?
a. Not affect documentation of pain assessment and would increase administration of
analgesics
b. Negatively affect documentation and decrease administration of analgesics
c. Increase documentation of pain assessment and influence administration of analgesics
d. Decrease documentation of pain assessment and increase administration of sedatives

8. As a result of CPOT implementation, the median number of pain assessment

intervals for each patient:
a. Increased
b. Decreased
c. Did not change

3. Which of the following is the recommended pain assessment interval for

postsurgical patients?
a. Every 15 minutes for 1 hour then a minimum of every 2 hours
b. Hourly for 6 hours then a minimum of every 4 hours
c. Hourly for 2 hours then every 2 hours for 12 hours
d. Hourly for the entire shift and as needed
4. Patients were eligible for the study if they met 1 of 2 criteria. The f irst
criterion refers to the patients inability to communicate as determined by
not being able to follow verbal commands. What was the second criterion?
a. A Mini-Mental Status Exam score less than 24
b. A motor score of 5 or less on the Glasgow Coma Scale
c. A vital capacity greater than 15mL/kg body weight
d. Ability to maintain a mean arterial blood pressure between 70 and 110 mmHg
5. Data abstractors excluded reference to pain behaviors during which of the
following interventions?
a. Routine neurological assessments
b. Pain assessment after narcotic administration
c. Administration of analgesia for mild pain
d. Patient mobility
6. Which of the following was used to examine prospectively chosen patient
factors associated with pain assessment in each intensive care unit (ICU)?
a. Kaplan-Meier method
b. Nonlinear regression
c. Multiple Poisson regression
d. Discriminant analysis

1. a
b
c
d

2. a
b
c
d

3. a
b
c
d

4. a
b
c
d

5. a
b
c
d

6. a
b
c
d

9. Study f indings indicated successful implementation of the CPOT was

associated with which of the following?
a. Increased patient assessment intervals
b. Minimized analgesia administration
c. Better detection for the presence of pain
d. Compliance with pain assessment documentation.
10. Which of the following supported the assumption that CPOT implementation
resulted in improved detection and use of opioids in the medical/surgical/
trauma ICU?
a. Frequency of analgesic use was targeted and met.
b. Protocols created provided guidelines for pain discrimination.
c. There was improved benchmarking and compliance.
d. There were minimal differences in patient characteristics.
11. Which of the following outcomes is anticipated with implementation of
a systematic pain assessment to treat and prevent escalation of pain?
a. Decreased length of stay
b. Improved patient satisfaction
c. Improved patient outcomes
d. Decreased postsurgical infection rates
12. Which of the following measures did the authors employ to minimize
performance bias relating to pain assessment practices?
a. Data was collected from the medical record following discharge from the ICU.
b. Participant criteria were focused on the patients ability to communicate.
c. Nursing staff was required to complete education prior to the study.
d. Data was recorded for 2 weeks before and after CPOT implementation.

7. a
b
c
d

8. a
b
c

9. a
b
c
d

10. a
b
c
d

Test ID: A132232 Contact hours: 1.0; pharma 0.0 Form expires: May 1, 2016. Test Answers: Mark only one box for your answer to each question.

11. a
b
c
d

12. a
b
c
d

Fee: AACN members, $0; nonmembers, $10 Passing score: 9 correct (75%) Category: CERP A Test writer: Jean Shinners, PhD, RN-BC
Name

Program evaluation

For faster processing, take

this CNE test online at
www.ajcconline.org
(CNE Articles in This Issue)
or mail this entire page to:
AACN, 101 Columbia,
Aliso Viejo, CA 92656.

Yes

Objective 1 was met

Objective 2 was met
Objective 3 was met
Content was relevant to my
nursing practice

My expectations were met

This method of CE is effective

for this content

The level of difficulty of this test was:

easy medium difficult
To complete this program,
it took me
hours/minutes.

No

Address
City
Country

State
AACN Customer ID#

Phone
Payment by:
Card #

ZIP

E-mail address*
Visa

M/C

AMEX

Check
Expiration Date

Signature
*E-mail address required to receive notification of completion, access to your test results, and
certificate for passing scores.

The American Association of Critical-Care Nurses is accredited as a provider of continuing nursing education by the American Nurses Credentialing Centers Commission on Accreditation.
AACN has been approved as a provider of continuing education in nursing by the State Boards of Nursing of Alabama (#ABNP0062), California (#01036), and Louisiana (#ABN12).
AACN programming meets the standards for most other states requiring mandatory continuing education credit for relicensure.