/clinical/,/clinical/cckm-tools/,/clinical/cckm-tools/content/,/clinical/cckm-tools/content/questionnaires/,/clinical/cckm-tools/content/questionnaires/related/,

/clinical/cckm-tools/content/questionnaires/related/name-97209-en.cckm

201606168

page

100

UWHC,UWMF,

Clinical Hub,UW Health Clinical Tool Search,UW Health Clinical Tool Search,Questionnaires,Related

Validation of the Critical-Care Pain Observation Tool In Adult Patients

Validation of the Critical-Care Pain Observation Tool In Adult Patients - Clinical Hub, UW Health Clinical Tool Search, UW Health Clinical Tool Search, Questionnaires, Related


420 AMERICAN JOURNAL OF CRITICAL CARE, July 2006, Volume 15, No. 4 http://ajcc.aacnjournals.org
• BACKGROUND Little research has been conducted to validate pain assessment tools in critical care,
especially for patients who cannot communicate verbally.
• OBJECTIVE To validate the Critical-Care Pain Observation Tool.
• METHODS A total of 105 cardiac surgery patients in the intensive care unit, recruited in a cardiology
health center in Quebec, Canada, participated in the study. Following surgery, 33 of the 105 were evalu-
ated while unconscious and intubated and 99 while conscious and intubated; all 105 were evaluated
after extubation. For each of the 3 testing periods, patients were evaluated by using the Critical-Care
Pain Observation Tool at rest, during a nociceptive procedure (positioning), and 20 minutes after the
procedure, for a total of 9 assessments. Each patient’s self-report of pain was obtained while the patient
was conscious and intubated and after extubation.
• RESULTS The reliability and validity of the Critical-Care Pain Observation Tool were acceptable.
Interrater reliability was supported by moderate to high weighted κcoefficients. For criterion validity,
significant associations were found between the patients’ self-reports of pain and the scores on the Criti-
cal-Care Pain Observation Tool. Discriminant validity was supported by higher scores during position-
ing (a nociceptive procedure) versus at rest.
• CONCLUSIONS The Critical-Care Pain Observation Tool showed that no matter their level of con-
sciousness, critically ill adult patients react to a noxious stimulus by expressing different behaviors that
may be associated with pain. Therefore, the tool could be used to assess the effect of various measures
for the management of pain. (American Journal of Critical Care. 2006;15:420-427)
VALIDATION OF THE CRITICAL-CARE PAIN
OBSERVATION TOOL IN ADULT PATIENTS
To purchase electronic or print reprints, contact The InnoVision Group, 101
Columbia, Aliso Viejo, CA 92656. Phone, (800) 809-2273 or (949) 362-2050
(ext 532); fax, (949) 362-2049; e-mail, reprints@aacn.org.
By Céline Gélinas, RN, PhD, Lise Fillion, RN, PhD, Kathleen A. Puntillo, RN, DNSc, Chantal Viens, RN, PhD, and
Martine Fortier, MPs. From School of Nursing, McGill University, Montreal, Quebec (CG), Faculty of Nursing,
Laval University, Quebec City, Quebec (LF, CV, MF), and Department of Physiological Nursing, University of
California, San Francisco, Calif (KP).
P
ain is an important stressor for many patients in
critical care,
1-3
and it is not unusual for the inten-
sity of the pain to be described as moderate to
severe.
4-12
Pain assessment is the first step in proper pain
relief, an important goal in patients’ care. Although
critical care clinicians strive to obtain each patient’s
self-report of pain, many factors compromise patients’
ability to communicate verbally, including the use of
sedative agents, mechanical ventilation, and changes in
the level of consciousness.
13,14
Several pain scales have
been used to document self-reporting of pain in intubated
patients.
7-9,12
In the absence of a patient’s self-report,
observable behavioral and physiological indicators
become important indices for the assessment of pain.
13,15-17
Preliminary research
18-20
has been conducted to
validate instruments that include behavioral and/or
physiological indicators. Use of these instruments in
critical care practice is restricted because of the limi-
tations of the studies. Limitations include small sample
sizes (<40 patients),
18-20
lack of validation in intubated
at University of Wisconsin-Madison on February 19, 2015ajcc.aacnjournals.orgDownloaded from

patients,
18
use of a subjective scale (eg, absence,
slight, moderate, and extreme intensity of behaviors),
18
confusion in the definition of behaviors (eg, body
movements and muscle rigidity), and use of dependent
observations (ie, statistical analysis of the observa-
tions rather than of the sample of patients).
19
The aim
of our study was to examine the reliability and validity
of a newly developed instrument for pain assessment:
the Critical-Care Pain Observation Tool (CPOT).
Method
Design, Sample, and Ethics
A repeated measures design was chosen for this
quantitative study. A convenience sample of 105 cardiac
surgery patients in the intensive care unit (ICU) at a car-
diology health center in Quebec, Canada, was recruited
for the study. Patients were considered for inclusion if
they were 18 years or older, had been admitted for car-
diac surgery, understood French, were in the ICU after
surgery, and were able to hear and to see. Patients
were excluded if they had been admitted for a heart
transplant or thoracic aortal aneurysm repair, received
medical treatment for chronic pain, had an ejection
fraction less than 0.25, had preexisting psychiatric or
neurological problems, had a dependence on alcohol
or drugs, received neuromuscular blockers following
surgery, or had complications after surgery (eg, hem-
orrhage, delirium, death).
This study was approved by the human research
committee of the health center. Recruitment was done
the day before the surgery; the study was explained to
eligible patients, and informed consent was obtained.
At this time, patients were taught how to use the pain
intensity descriptive scale.
Instruments
Critical-Care Pain Observation Tool. The CPOT,
developed in French, has 4 sections, each with differ-
ent behavioral categories: facial expression, body
movements, muscle tension, and compliance with the
ventilator for intubated patients or vocalization for
extubated patients (Table 1). Items in each section are
http://ajcc.aacnjournals.org AMERICAN JOURNAL OF CRITICAL CARE, July 2006, Volume 15, No. 4 421
Indicator
Facial expression
Body movements
Muscle tension
Evaluation by passive flexion and
extension of upper extremities
Compliance with the ventilator
(intubated patients)
OR
Vocalization (extubated patients)
Total, range
Description
No muscular tension observed
Presence of frowning, brow lowering, orbit tightening,
and levator contraction
All of the above facial movements plus eyelid tightly
closed
Does not move at all (does not necessarily mean
absence of pain)
Slow, cautious movements, touching or rubbing the
pain site, seeking attention through movements
Pulling tube, attempting to sit up, moving limbs/
thrashing, not following commands, striking at staff,
trying to climb out of bed
No resistance to passive movements
Resistance to passive movements
Strong resistance to passive movements, inability to
complete them
Alarms not activated, easy ventilation
Alarms stop spontaneously
Asynchrony: blocking ventilation, alarms frequently
activated
Talking in normal tone or no sound
Sighing, moaning
Crying out, sobbing
Score
Relaxed, neutral 0
Tense 1
Grimacing 2
Absence of movements 0
Protection 1
Restlessness 2
Relaxed 0
Tense, rigid 1
Very tense or rigid 2
Tolerating ventilator or 0
movement
Coughing but tolerating 1
Fighting ventilator 2
Talking in normal tone
or no sound 0
Sighing, moaning 1
Crying out, sobbing 2
0-8
Table 1 Description of the Critical-Care Pain Observation Tool
Behavioral and physiological indicators
are important indices for assessment of
pain in patients unable to self-report.
at University of Wisconsin-Madison on February 19, 2015ajcc.aacnjournals.orgDownloaded from

scored from 0 to 2, with a possible total score ranging
from 0 to 8. The CPOT was developed as follows.
Some items and their operational definitions were
derived from previously described instruments for
pain assessment.
18-21
In addition, pain indicators were
described by using findings from a chart review of the
medical files of 52 critically ill patients
22
and from 9
focus groups with 48 critical care nurses and inter-
views of 12 physicians.
23
Content validity of the CPOT was established with
4 physicians and 13 critical care nurses. The physicians
and nurses completed a questionnaire on the relevance
of the inclusion of these indicators in the CPOT by
using a Likert scale (1=not at all, 2=a little, 3=moder-
ately, and 4 = very much). Content validity indices,
which are the proportion of participants who answered
3 or 4 on the Likert scale, were calculated. All indica-
tors had indices of 0.88 to 1.00. Content validity indices
greater than 0.80 were sufficiently satisfactory
24
to con-
sider including all these indicators in the CPOT.
Pain Intensity Descriptive Scale. A previously vali-
dated pain intensity descriptive scale (0=none, 1=mild,
2 = moderate, 3 = severe, 4 = unbearable) was used.
This scale has been used in previous studies
18,25
in
acute and critical care.
Confusion Assessment Method for the Intensive
Care Unit. The Confusion Assessment Method for the
Intensive Care Unit (CAM-ICU) was used to assess
delirium. The instrument has good sensitivity and speci-
ficity for assessing delirium in critically ill patients.
26,27
Two modifications were made in the CAM-ICU to
adapt it to the sedation scale used in our study and to
facilitate assessment of patients’ inattention. First, the
Ramsay Scale
28
was used to assess the level of sedation.
Second, patients’ inattention was verified by assessing
their capacity to concentrate on the pain intensity
descriptive scale used in our study.
Procedure
Three testing periods, each including 3 assess-
ments for a total of 9 pain assessments (T1-T9) with
the CPOT, were completed during each patient’s early
postoperative course (Figure 1). For each patient, the
first 3 assessments (T1-T3) were done while the
patient was intubated and still unconscious (ie, with a
sedation score of 5 or 6 on the Ramsay Scale). T1 was
done with the patient at rest, approximately 2 hours
after the end of surgery. T2 was completed a few min-
utes after T1 during positioning of the patient. Posi-
tioning represented a previously confirmed nociceptive
procedure.
9
On the basis of the patient’s needs, endo-
tracheal suctioning often was performed at the same
time as positioning. Finally, T3 was done at recovery,
20 minutes after the positioning procedure.
The second testing period (assessments T4-T6)
was 3 hours after the first testing period. During this
time, the patient was still intubated but conscious.
Patients were considered conscious if they had a score
of 2, 3, or 4 on the Ramsay Scale.
Finally, the third testing period (assessments T7-
T9) was after the patient was extubated, approximately
5 hours after the second testing period. The position-
ing procedure at T8 sometimes occurred with ambula-
tion and/or respiratory exercises, which were part of
the postoperative care protocol.
For each of the 3 testing periods, patients were
evaluated with the CPOT for 1 minute at rest both
before and after positioning and for the duration of the
positioning procedure. This standardization of proce-
dures was based on the work of Puntillo et al.
9
One of
us (C.G.) and a critical care nurse (G.N.) evaluated the
patients. Upon completion of the CPOT during the
second testing period (ie, assessments T4-T6), intu-
bated patients communicated the presence or absence
of pain by nodding their heads (yes or no) to the
422 AMERICAN JOURNAL OF CRITICAL CARE, July 2006, Volume 15, No. 4 http://ajcc.aacnjournals.org
Figure 1 Illustration of the time of consent and the 3 testing periods.
Abbreviations: P, positioning procedure; R, rest; REC, recovery (20 minutes after the positioning procedure).
Intensive care unit
T1 T2 T3 T4 T5 T6 T7 T8 T9
R P REC R P REC R P REC
First testing period
Unconscious
intubated patients
Second testing period
Conscious intubated
patients
Third testing period
Conscious extubated
patients
Operating
room
Consent
day before
surgery
at University of Wisconsin-Madison on February 19, 2015ajcc.aacnjournals.orgDownloaded from

question, Do you have pain? This procedure was
selected because many intubated patients during this
phase of their recovery were unable to use the pain
intensity descriptive scale. Before the third testing
period, at T7, patients were evaluated by using the
CAM-ICU to determine the presence of delirium.
Three patients were excluded because of delirium.
During the third testing period (ie, assessments T7-
T9) after completion of the CPOT, the extubated
patients used the pain intensity descriptive scale to
grade their pain.
Data Analysis
Statistical analyses were completed by using ver-
sion 11.5 of SPSS for Windows (SPSS Inc, Chicago,
Ill). Descriptive statistics were computed for all vari-
ables. Interrater reliability was examined. Weighted κ
coefficients were calculated for all assessments (T1-
T9).
29
To test validity of the CPOT, we determined cri-
terion and discriminant validity (Table 2). Criterion
validity was examined by measuring the relationship
between the CPOT scores and the patients’ self-
reports, the gold standard measure of pain. Analysis
of variance was used to examine the differences between
the intubated patients’ self-reports of pain (yes or no)
and the CPOT scores (assessments T4-T6). Also,
Spearman correlations
31
were calculated between the
extubated patients’ self-reports of pain intensity (ordi-
nal descriptive scale) and the CPOT scores (assess-
ments T7-T9). Finally, discriminant validity
31
was
examined by performing paired t tests between assess-
ments with the CPOT taken at rest and during posi-
tioning (T1 with T2, T4 with T5, and T7 with T8).
Results
Characteristics of the Sample
A total of 131 patients were approached for consent
the day before surgery, and 117 (89%) agreed to partici-
pate in the study. Reasons for refusal were as follows:
anxious about the surgery (n = 9 patients), not interested
(n = 3), undecided (n = 1), and bad experience with
research (n = 1). During the course of the study, 8
patients were excluded because of postoperative compli-
cations (hemorrhage, delirium, death), 3 because their
surgery was canceled, and 1 because of extubation right
after surgery. The final sample size was 105 patients
enrolled during a 3-month period. Table 3 gives the
demographic characteristics of the patients.
Anesthesia was similar for all patients, and all
were receiving continuous infusions of propofol after
surgery (mean dose 85.4 mg/h, SD 39.7 mg/h). For
each patient, this medication was tapered off and
stopped 1 to 3 hours after the patient’s arrival in the
ICU. Thus, all patients were receiving propofol during
http://ajcc.aacnjournals.org AMERICAN JOURNAL OF CRITICAL CARE, July 2006, Volume 15, No. 4 423
Table 2 Description of reliability and validity methods examined in this study
Psychometric
property
Interrater
reliability
Criterion
validity
Discriminant
validity
Description
Interrater reliability is the consistency
with which 2 raters agree on their
measurement/observation (ie, the CPOT)
of a phenomenon (ie, pain)
Two raters assessed the patients in this
study: the principal investigator and 1
critical care nurse
Criterion validity refers to the relationship
between the instrument (ie, the CPOT)
and the gold standard measure of pain
(ie, the patient’s self-report)
In this study, yes/no and pain intensity
were used as the gold standard self-report
measures
Discriminant validity refers to evidence
that instruments measuring 2 different
constructs should not correlate
In this study, we examined whether the
CPOT could be used to discriminate
between pain during positioning and
lack of pain at rest
Coefficient or analysis
κcoefficient (proportion
of responses in which the
2 raters agreed)
Intubated patients (T4-T6):
analysis of variance
Extubated patients (T7-T9):
Spearman correlation
Rest time compared with
positioning for all three
testing periods: paired t test
Level of acceptability*
<0 Poor
0-0.20 Slight
0.21-0.40 Fair
0.41-0.60 Moderate
0.61-0.80 Substantial
0.81-1.00 Almost perfect
P ≤.01
P ≤.01
Abbreviation: CPOT, Critical-Care Pain Observation Tool.
* Levels of acceptability for interrater reliability scores from Landis and Koch.
30
at University of Wisconsin-Madison on February 19, 2015ajcc.aacnjournals.orgDownloaded from

the first testing period (ie, assessments T1-T3). All
patients also were receiving continuous infusions of
fentanyl when they were admitted to the ICU from
surgery. The mean dosage of fentanyl decreased from
73.7 µg/h (SD 21.8) at the first testing period to 50.7
µg/h (SD 31.1) at the third testing period. Rarely,
patients (n = 4) received an intravenous bolus of fen-
tanyl before positioning.
Sample at the 3 Testing Periods
First Testing Period. For assessments T1 to T3, data
were collected on 33 of the 105 intubated patients who
were unconscious, a criterion for testing during this
period. CPOT scores were higher during the positioning
procedure (T2) than during rest (T1) or recovery (T3;
Figure 2).
Second Testing Period. For assessments T4 to T6,
data were collected on 99 of the awake 105 intubated
patients. The remaining 6 patients were extubated before
the completion of this testing period. Again, CPOT
scores were higher during the positioning procedure
(T5) than during rest (T4) or recovery (T6). Moreover,
in this testing period, patients had the highest scores
on the CPOT (Figure 2).
Third Testing Period. Finally, for assessments T7 to
T9, all 105 patients were assessed after they were extu-
bated. The CPOT scores were similar to those of the 2
previous testing periods (Figure 2).
Interrater Reliability
Together, the principal investigator and the critical
care nurse (C.G. and G.N.) completed the CPOT at all
9 assessments and were blinded to each other’s scores.
The sample sizes for interrater reliability differed for
each time, reflecting the times when both were present.
424 AMERICAN JOURNAL OF CRITICAL CARE, July 2006, Volume 15, No. 4 http://ajcc.aacnjournals.org
Table 3 Description of the study sample (n = 105)*
Variable
Age, mean (SD)
Sex, No. (%) of patients
Male
Female
Type of cardiac surgery, No. of patients (%)
Coronary artery bypass graft
Valvular repair or replacement
Coronary artery bypass graft and valvular
surgery
Interauricular/interventricular
communication repair
Value
60 (8)
83 (79)
22 (21)
83 (79)
11 (10)
9 (9)
2 (2)
*All patients had sternal incisions.
Conscious
intubated
Conscious
extubated
Unconscious
intubated
Figure 2 Mean scores and standard deviations of the Critical-
Care Pain Observation Tool for the 3 testing periods (N = 105
patients). The scores can range from 0 to 8.
Testing period
1 Unconscious
intubated patients
(n = 33)
2 Conscious
intubated patients
(n = 99)
3 Conscious
extubated patients
(n = 105)
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
M
e
a
n

s
c
o
r
e
Assessment
T1 Rest
T2 Procedure
T3 Recovery
T4 Rest
T5 Procedure
T6 Recovery
T7 Rest
T8 Procedure
T9 Recovery
Time
Rest Procedure
T5
T8
T2
T4
T7
T1
T6
T9
T3
Recovery
Mean
0.55
2.70
0.67
1.21
3.38
1.35
0.69
2.79
0.87
SD
1.03
1.36
0.89
1.23
1.38
1.42
0.87
1.31
1.04
Table 4 Weighted κcoefficients for each assessment from
T1 to T9*
Assessment
T1
T2
T3
T4
T5
T6
T7
T8
T9
12
12
14
29
33
33
34
33
34
Weighted κκcoefficient
0.83
0.63
0.85
0.52
0.85
0.88
0.62
0.77
0.71
*Assessments made by using the Critical-Care Pain Observation
Tool were independently completed by the principal investigator
and the critical care nurse when both were present.
No. of
patients
at University of Wisconsin-Madison on February 19, 2015ajcc.aacnjournals.orgDownloaded from

Weighted κcoefficients were moderate to high at all
assessments (Table 4).
Criterion Validity
Mean CPOT scores according to patients’ self-
reports of the presence or absence of pain during the
second testing period (ie, assessments T4-T6) and the
analysis of variance are presented in Table 5. At each
assessment in this testing period, CPOT scores were
significantly higher for intubated patients reporting
pain than for those who had no pain.
During the third testing period (ie, assessments
T7-T9), mean pain intensity scores were significantly
higher during the positioning procedure at T8 (2.01)
than during rest at T7 (1.71) and recovery at T9
(1.40). Spearman correlations of 0.49, 0.59, and 0.40
(P ≤.001) at T7 to T9 showed that the patients’ self-
reported pain intensity scores were moderately corre-
lated with the CPOT scores.
Discriminant Validity
At the 3 testing periods, CPOT scores were signif-
icantly higher during positioning than during the rest
periods. Table 6 gives the results of the paired t tests.
Discussion
Our findings validated the CPOT, which was
developed specifically to assess pain in ICU patients.
Interrater reliability was high for most assessments
and moderate at T4. Payen et al
19
obtained a weighted
κcoefficient of 0.74 when they compared behavioral
pain scores between pairs of evaluators. A total of 46
nurses and nurse’s aides, 1 physical therapist, and 1
physician participated in that study.
19
In our study,
only 2 evaluators used the instrument, which is a limi-
tation to the examination of interrater reliability, and
results cannot be generalized to other ICU nurses.
When patients were intubated during the second
testing period, CPOT scores differed significantly
between those who reported pain and those who did
not. Moreover, when patients were extubated during
the third testing period, the higher a patient’s self-
report of pain was, the higher was the patient’s score
on the CPOT. These results are consistent with those
of previous studies
18,32
in which self-reports of pain of
patients in a postanesthesia care unit were moderately
related to pain behaviors. Our results support the crite-
rion validity of the CPOT because the indicators were
tested against the most valid measurement of pain;
that is, the patients’ self-reports.
Discriminant validity was supported by the find-
ing that CPOT scores were higher during positioning
than at rest in the 3 testing periods. Payen et al
19
also
found higher behavioral scores during positioning
than at rest in unconscious critically ill patients. Such
results emphasize that pain behaviors are observable
even if a patient cannot report pain.
Our study, however, is the first to document differ-
ences in pain behavior scores according to levels of
activity during different states of consciousness and
intubation: unconscious and intubated, conscious and
intubated, and then awake and extubated. These results
http://ajcc.aacnjournals.org AMERICAN JOURNAL OF CRITICAL CARE, July 2006, Volume 15, No. 4 425
Table 5 Differences in scores on the Critical-Care Pain Observation Tool according to patients’ self-reports of pain in the
second testing period (T4, T5, and T6)
Assessment
T4
T5
T6
Patients’ self-reports of pain:
pain present or absent*
Yes, present (n = 53)
No, absent (n = 41)
Yes, present (n = 79)
No, absent (n = 18)
Yes, present (n = 54)
No, absent (n = 39)
Mean
1.62
0.78
3.65
2.11
2.07
0.49
SD
1.38
0.85
1.31
0.90
1.40
0.88
F
11.83

22.18

38.89

df
(1, 92)
(1, 95)
(1, 91)
ANOVAScores
Abbreviation: ANOVA, analysis of variance.
*Because of intermittent drowsiness postoperatively, 5 patients at T4, 2 at T5, and 6 at T6 were unable to give their self-reports of pain by
nodding their heads.

P ≤ .001. Alpha is adjusted to 0.01 because 3 comparisons were made on the same subjects.
CPOT scores were associated with
patients’ self-reports of pain.
CPOT scores were higher during painful
procedures, lending support to its validity.
at University of Wisconsin-Madison on February 19, 2015ajcc.aacnjournals.orgDownloaded from

suggest that patients, whatever their levels of con-
sciousness, may demonstrate pain behaviors in response
to a nociceptive procedure. Whether a behavioral
response to a noxious procedure is accompanied by
perception of pain in an unconscious patient is
unknown. Until there is evidence to the contrary,
experts recommend that healthcare providers assume
that unconscious patients may have pain, especially if
behavioral responses to a known noxious stimulus
occur. The experts
33,34
recommend that these patients
be treated the same way as conscious patients when the
patients are exposed to sources of pain.
Indeed, in a study by Lawrence,
35
formerly uncon-
scious patients revealed that they could hear, under-
stand, and respond emotionally to what was being said
while they were unconscious. In light of this finding,
perhaps the CPOT can be used to assess pain in other
populations of critical care patients. This hypothesis
requires confirmation in future studies.
Interestingly, patients seemed to have higher CPOT
scores when they were conscious and intubated than
they did when they were unconscious or extubated.
The presence of the endotracheal tube as a potential
source of pain
6,36
for cardiac surgery patients, in addi-
tion to the sternal incision,
5,8
may help to explain these
findings. Moreover, mechanical ventilation modes
provided a positive inspiratory pressure to allow dis-
tribution of oxygen throughout the lungs.
17
This pres-
sure may cause stretching of the sternal incision,
which does not occur in nonsurgical patients, and
which can be painful to some patients.
In our study, 65 conscious intubated patients had
endotracheal suctioning while they were being posi-
tioned. However, when conscious intubated patients
were positioned during assessment T5, 18 of them
who did not report pain had high CPOT scores (mean
2.11, SD 0.90). Perhaps for these patients the position-
ing was a distressful or an uncomfortable experience
1,9,37
rather than a painful one. Also, the endotracheal tube
may have caused coughing during the positioning pro-
cedure, leading to higher CPOT scores in the absence
of reported pain. This finding suggests that behaviors
observed by using the CPOT may be an indicator of
more than pain. Further research is warranted to deter-
mine the sensitivity and specificity of the CPOT as a
measure of pain.
We also found that CPOT scores were similarly low
for both unconscious and conscious extubated patients.
This result may have occurred because the patients were
highly sedated while unconscious and may have been
experiencing the residual effects of anesthesia.
19
Once
extubated, they could have experienced less severe pain
than they did when they were intubated.
Data collection in this study was completed in the
8 hours after surgery, a period when intermittent drowsi-
ness can be expected. Previous studies
7-9,12,20
in which
intubated patients provided self-reports of pain were
conducted in periods varying from 12 to 72 hours after
the end of surgery. Those patients might have had more
time to recuperate from the residual effects of anesthe-
sia than our patients did. In the study by Ferguson et
al,
5
patients’ self-reports of pain after coronary artery
bypass graft surgery were collected more than 8 hours
after the end of surgery, and 9 (21%) of 43 patients
were unable to communicate their self-report of pain
because of drowsiness. The occurrence of the same
inability to communicate in the patients in our study is
not surprising.
Limitations
This study was not without limitations. First, data
were collected by only 2 persons. More raters should
be used in tests of interrater reliability in subsequent
evaluations of the CPOT. Second, data could be col-
lected for only 33 of the 105 patients while the patients
were unconscious. Third, postoperative drowsiness led
to missing data for some patients. Finally, cardiac
surgery patients are a relatively healthy ICU group
and may not represent most ICU patients, who are
426 AMERICAN JOURNAL OF CRITICAL CARE, July 2006, Volume 15, No. 4 http://ajcc.aacnjournals.org
Table 6 Differences in scores on the Critical-Care Pain
Observation Tool measured at rest before the procedure
(T1, T4, and T7) and during the procedure (T2, T5, and T8)
Assessments
T1-T2
T4-T5
T7-T8
No. of
patients
33
99
105
t
-9.01*
-12.07*
-15.96*
df
32
98
104
*P < .001. Alpha is adjusted to 0.01 because 3 comparisons were
made on the same subjects.
Experts recommend we assume
that unconscious patients have pain,
especially if behavioral responses to
noxious stimuli occur.
Behaviors observed by using the CPOT
may indicate more than pain.
at University of Wisconsin-Madison on February 19, 2015ajcc.aacnjournals.orgDownloaded from

much sicker. Future research on the effectiveness of
the CPOT as a nonverbal measure of pain in other
sicker ICU patients is warranted.
Despite these limitations, this study was innovative
in several aspects. First, development of the CPOT was
based on previous research of others as well as on
descriptive data from 2 preliminary studies that led to
selection of the behavioral indicators. Second, the
relationship between intubated patients’ self-reports of
pain and behavioral indicators was explored for the
second time. Finally, data were obtained from patients
at different levels of consciousness.
Future research should be conducted to determine
if CPOT scores can be used to differentiate pain from
other conditions. Also, receiver operating characteristic
curve analysis could be performed to examine the speci-
ficity and the sensitivity of the CPOT as a measure of
pain. This further testing could substantiate the CPOT as
a valid, reliable, and useful tool for measuring pain in
critically ill patients who are unable to self-report.
Conclusions
The CPOT had acceptable reliability and validity
in this sample of cardiac surgery ICU patients. How-
ever, the tool needs to be further validated in different
populations of critically ill patients. Appropriate pain
assessment is an important part of quality care for
critically ill patients, and use of validated measures of
pain could aid in the evaluation of multidisciplinary
pain management techniques for nonverbal critically
ill patients.
ACKNOWLEDGMENTS
Special thanks to Gaëlle Napert, the critical care nurse who contributed to data col-
lection, and to François Harel, the statistician who reviewed the analyses. Thanks to
all intensive care unit nurses and physicians for their support and collaboration in the
performance of this study. Special thanks to Drs Jean Bussières and Mathieu Simon
for their valuable collaboration in the conception and the conduction of this study.
Also, thanks to Celeste Johnston, RN, DEd, and Denise Li, RN, for their helpful comments
on this article. Finally, thanks to all the patients who participated in this study.
The research was performed at the Faculty of Nursing, Laval University, Quebec
City, Quebec, Canada. This study was supported by a nursing fellowship research
grant from the Heart and Stroke Foundation of Canada and a doctorate research
grant from the Fondation de Recherche en Sciences Infirmières du Québec.
REFERENCES
1. Hallenberg B, Bergbom-Engberg I, Haljamäe H. Patients’ experiences of
postoperative respirator treatment: influence of anaesthetic and pain treat-
ment regimens. Acta Anaesthesiol Scand. 1990;34:557-562.
2. Turner JS, Briggs SJ, Springhorn HE, Potgieter PD. Patients’ recollection
of intensive care unit experience. Crit Care Med. 1990;18:966-968.
3. Wilson VS. Identification of stressors related to patients’ psychologic
responses to the surgical intensive care unit. Heart Lung. 1987;16:267-273.
4. Desbiens NA, Wu AW, Broste SK, et al. Pain and satisfaction with pain
control in seriously ill hospitalized adults: findings from the SUPPORT
research investigations. Crit Care Med. 1996;24:1953-1961.
5. Ferguson J, Gilroy D, Puntillo KA. Dimensions of pain and analgesic
administration associated with coronary artery bypass grafting in an Aus-
tralian intensive care unit. J Adv Nurs. 1997;26:1065-1072.
6. Puntillo KA. Pain experiences of intensive care unit patients. Heart Lung.
1990;19:526-533.
7. Puntillo KA. Dimensions of procedural pain and its analgesic management
in critically ill surgical patients. Am J Crit Care. 1994;3:116-122.
8. Puntillo KA, Weiss JS. Pain: its mediators and associated morbidity in crit-
ically ill cardiovascular surgical patients. Nurs Res. 1994;43:31-36.
9. Puntillo KA, White C, Morris AB, et al. Patients’ perceptions and responses
to procedural pain: results from Thunder Project II. Am J Crit Care.
2001;10:238-251.
10. Stanik-Hutt J, Soeken KL, Fontaine DK, Gift AG. Pain experiences of
traumatically injured patients in a critical care setting. Am J Crit Care.
2001;10:252-259.
11. Valdix SW, Puntillo KA. Pain, pain relief and accuracy of their recall after
cardiac surgery. Prog Cardiovasc Nurs. Summer 1995;10:3-11.
12. Whipple JK, Lewis KS, Quebbeman EJ, et al. Analysis of pain manage-
ment in critically ill patients. Pharmacotherapy. 1995;15:592-599.
13. Hamill-Ruth RJ, Marohn L. Evaluation of pain in the critically ill patient.
Crit Care Clin. 1999;15:35-53.
14. Shannon K, Bucknall T. Pain assessment in critical care: what have we
learnt from research? Intensive Crit Care Nurs. 2003;19:154-162.
15. Christoph SB. The problem of pain in the critically ill patient. Crit Care
Nurs Clin North Am. 1991;3:11-16.
16. Kwekkeboom KL, Herr KA. Assessment of pain in the critically ill. Crit
Care Nurs Clin North Am. 2001;13:181-194.
17. Kaiser KS. Assessment and management of pain in the critically ill trauma
patient. Crit Care Nurs Q. 1992;15:14-34.
18. Mateo OM, Krenzischek DA. A pilot study to assess the relationship
between behavioral manifestations and self-report of pain in postanaesthe-
sia care unit patients. J Post Anesth Nurs. 1992;7:15-21.
19. Payen JF, Bru O, Bosson JL, et al. Assessing pain in critically ill sedated
patients by using a behavioral pain scale. Crit Care Med. 2001;29:2258-2263.
20. Puntillo KA, Miaskowski C, Kehrle K, Stannard D, Gleeson S, Nye P.
Relationship between behavioral and physiological indicators of pain, criti-
cal care patients’ self-reports of pain, and opioid administration. Crit Care
Med. 1997;25:1159-1166.
21. Ambuel B, Hamlett KW, Marx CM, Blumer JL. Assessing distress in pedi-
atric intensive care environments: the COMFORT Scale. J Pediatr Psy-
chol. 1992;17:95-109.
22. Gélinas C, Fortier M, Viens C, Fillion L, Puntillo KA. Pain assessment and
management in critically ill intubated patients: a retrospective study. Am J
Crit Care. 2004;13:126-135.
23. Gélinas C, Viens C, Fortier M, Fillion L. Les indicateurs de douleur en
soins critiques [Pain indicators in critical care]. Perspect Infirm. 2005;2:12-
14, 16-20, 22.
24. Waltz CF, Strickland OL, Lenz ER. Measurement in Nursing Research.
2nd ed. Philadelphia, Pa: FA Davis Co; 1991.
25. Choinière M, Melzack R, Girard N, Rondeau J, Paquin M-J. Comparisons
between patients’ and nurses’ assessment of pain and medication efficacy
in severe burn injuries. Pain. 1990;40:143-152.
26. Ely EW, Inouye SK, Bernard GR, et al. Delirium in mechanically venti-
lated patients: validity and reliability of the Confusion Assessment Method
for the Intensive Care Unit (CAM-ICU). JAMA. 2001;286:2703-2710.
27. Ely EW, Margolin R, Francis J, et al. Evaluation of delirium in critically ill
patients: validation of the Confusion Assessment Method for the Intensive
Care Unit (CAM-ICU). Crit Care Med. 2001;29:1370-1379.
28. Ramsay M, Savege T, Simpson B, Goodwin R. Controlled sedation with
alphaxalone-alphadolone. Br Med J. 1974;2:656-659.
29. Streiner DL, Norman GR. Health Measurement Scales: A Practical Guide
to Their Development and Use. 3rd ed. New York, NY: Oxford University
Press Inc; 2003.
30. Landis RJ, Koch GG. The measurement of observer agreement for categor-
ical data. Biometrics. 1977;33:159-174.
31. Polit DF, Hungler BP. Nursing Research: Principles and Methods. 6th ed.
Philadelphia, Pa: Lippincott; 1999.
32. Webb MR, Kennedy MG. Behavior responses and self-reported pain in
postoperative patients. J Post Anesth Nurs. 1994;9:91-95.
33. Bushnell MC. Perception of pain in the persistent vegetative state [letter]?
Eur J Pain. 1997;1:167.
34. Klein M. Perception of pain in the persistent vegetative state [letter]? Eur J
Pain. 1997;1:165-167.
35. Lawrence M. The unconscious experience. Am J Crit Care. 1995;4:227-232.
36. Gries ML, Fernsler J. Patient perceptions of the mechanical ventilation
experience. Focus Crit Care. 1988;15:52-59.
37. Bergbom-Engberg I, Haljamäe H. Assessment of patients’ experience of
discomforts during respirator therapy. Crit Care Med. 1989;17:1068-1072.
http://ajcc.aacnjournals.org AMERICAN JOURNAL OF CRITICAL CARE, July 2006, Volume 15, No. 4 427
at University of Wisconsin-Madison on February 19, 2015ajcc.aacnjournals.orgDownloaded from

http://ajcc.aacnjournals.org/cgi/external_ref?link_type=PERMISSIONDIRECT
Personal use only. For copyright permission information:

Published online http://www.ajcconline.org
' 2006 American Association of Critical-Care Nurses
2006;15:420-427Am J Crit Care

CØline GØlinas, Lise Fillion, Kathleen A. Puntillo, Chantal Viens and Martine Fortier
Validation of the Critical-Care Pain Observation Tool in Adult Patients


http://ajcc.aacnjournals.org/subscriptions/
Subscription Information
http://ajcc.aacnjournals.org/misc/ifora.xhtml
Information for authors
http://www.editorialmanager.com/ajcc
Submit a manuscript
http://ajcc.aacnjournals.org/subscriptions/etoc.xhtml
Email alerts
by AACN. All rights reserved. ' 2006 Copyright
Telephone: (800) 899-1712, (949) 362-2050, ext. 532. Fax: (949) 362-2049.
bimonthly by The InnoVision Group, 101 Columbia, Aliso Viejo, CA 92656.
journal of the American Association of Critical-Care Nurses (AACN), published
AJCC, the American Journal of Critical Care, is the official peer-reviewed research
at University of Wisconsin-Madison on February 19, 2015ajcc.aacnjournals.orgDownloaded from