Assignment Writing Service http://StudyHub.vip/A-Comparison-Of-Faces-Scales-For-The-Me 👈

1. A comparison of faces scales for the measurement of pediatric pain:
children's and parents' ratings
Christine T. Chambersa,*, Kelly Giesbrechta
, Kenneth D. Craiga
, Susan M. Bennettb
,
Elizabeth Huntsmanb
a
Department of Psychology, University of British Columbia, 2136 West Mall, Vancouver, BC, V6T 1Z4, Canada
b
Department of Psychology, British Columbia's Children's Hospital Vancouver, BC, Canada
Received 16 September 1998; received in revised form 10 February 1999; accepted 19 March 1999
Abstract
Faces scales have become the most popular approach to eliciting children's self-reports of pain, although different formats are available.
The present study examined: (a) the potential for bias in children's self-reported ratings of clinical pain when using scales with smiling rather
than neutral `no pain' faces; (b) levels of agreement between child and parent reports of pain using different faces scales; and (c) preferences
for scales by children and parents. Participants were 75 children between the ages of 5 and 12 years undergoing venepuncture, and their
parents. Following venepuncture, children and parents independently rated the child's pain using ®ve different randomly presented faces
scales and indicated which of the scales they preferred and why. Children's ratings across scales were very highly correlated; however, they
rated signi®cantly more pain when using scales with a smiling rather than a neutral `no pain' face. Girls reported signi®cantly greater levels
of pain than boys, regardless of scale type. There were no age differences in children's pain reports. Parents' ratings across scales were also
highly correlated; however, parents also had higher pain ratings using scales with smiling `no pain' faces. The level of agreement between
child and parent reports of pain was low and did not vary as a function of the scale type used; parents overestimated their children's pain using
all ®ve scales. Children and parents preferred scales that they perceived to be happy and cartoon-like. The results of this study indicate that
subtle variations in the format of faces scales do in¯uence children's and parents' ratings of pain in clinical settings. q 1999 International
Association for the Study of Pain. Published by Elsevier Science B.V.
Keywords: Faces scales; Pediatric pain; Venepuncture
1. Introduction
Pain assessment is one of the most dif®cult yet imperative
challenges facing health professionals and researchers who
work with children. Accurate assessment is necessary not
only to ensure the proper management of pediatric pain, but
also to facilitate the scienti®c investigation of pain. Pediatric
pain assessment has made important advances in the past
decade, and there now exist a myriad of assessment tools
developed for use with children, including self-report, beha-
vioral and physiological measures (McGrath, 1996; Cham-
bers and McGrath, 1998; Finley and McGrath, 1998). Pain
is a highly individualized and subjective event. Therefore, a
child's self-report (i.e. what a child says) has generally been
considered to be the `gold standard' for pain assessment
(Merskey and Bogduk, 1994), despite its limitations
(Jensen, 1997). Notwithstanding debate as to the validity
of children's self-report when not used in conjunction
with behavioral and/or physiological measures (Craig,
1992), self-report alone has become the most common
measure of pain obtained from pediatric patients. Not
surprisingly, a variety of measures have been developed to
elicit self-reports of pain from children (Champion et al.,
1998), including the Poker Chip Tool (Hester et al., 1979),
visual analogue scales (VAS; Huskisson, 1974), pain ther-
mometers (Jay et al., 1983), and color scales (Eland, 1981).
In recent years considerable attention has been devoted to
what are referred to as `faces scales'. Faces scales show a
series of faces, typically hand-drawn, with the faces graded
in increasing intensity between `no pain' and `worst pain
possible' (Chambers and Craig, 1998). When presented with
a faces scale, children are asked to point to the face that best
shows how much pain they are currently experiencing.
Faces scales, unlike other self-report measures, are thought
to be easily understood by children in that they do not
require the child to translate their pain experience into a
numerical value. Several studies have shown that faces
scales are preferred by children, parents and nurses, when
Pain 83 (1999) 25±35
0304-3959/99/$20.00 q 1999 International Association for the Study of Pain. Published by Elsevier Science B.V.
PII: S0304-3959(99)00086-X
www.elsevier.nl/locate/pain
* Corresponding author. Tel.: 11-604-822-5280, fax: 11-604-822-
6923.
E-mail address: cchamber@interchange.ubc.ca (C.T. Chambers)

2. C.T. Chambers et al. / Pain 83 (1999) 25±35
26
Table 1
Descriptions of the various faces scales available for measuring children's paina
Reference Population for use Description of Scale Scoring Reliability and validityb
Beyer (1984) Three to 12-year-olds; non-
speci®c, but versions
available for use with either
Black and Hispanic children
Consists of both a
photographic scale (six color
photographs of children in
pain) ranging from a neutral to
a pained expression, and a
numerical scale ranging from
`no hurt at all' to `the biggest
hurt you could ever have'
The six faces are assigned
numerical values from 0
to 100 (i.e. 0, 20, 40, 60,
80, 100) which
correspond to the
accompanying numerical
scale
Photographic and numerical
scales are highly correlated
(r ˆ 0.82), good agreement
among children in sequencing
the photographs, ratings highly
correlated with ratings on other
measures (r ˆ 0.88±0.98),
measure shows sensitivity to
analgesic intervention
Bieri et al. (1990) Three to 12-year-olds; have
been reports of successful
use with children as young
as 2
Seven hand-drawn faces,
showing gradual increases in
pain expression (neutral to
pain); developed from
children's own drawings of
facial expressions of pain
The ordered faces are
scored from 0 to 6
Good overall agreement
between children on rank
ordering of the faces; faces
appear to be perceived by
children as equally spaced;
scores obtained are relatively
consistent over time
Douthit (1990); Tyler et al.
(1993)
Three to 12-year-olds Five hand-drawn faces
ranging from smiling to
crying; faces based on the
scale by Beyer (1984)
The ordered faces are
scored from 0 to 4
Scale highly correlated with
other self-report and
behavioral measures (r ˆ
0.74±0.79)
Frank et al. (1982) Adults Eight hand-drawn faces
ranging from smiling (`no
pain') to crying (`very severe
pain'); no data provided on
how the faces were developed
The ordered faces are
scored from 0 to 7
The faces were correctly
ranked by 50 subjects and were
evaluated by the experimenter
as representing `a reasonable
spectrum of visual expressions
between the two poles of pain
experience'
Goddard and Pickup (1996) Two to 16-year-olds Consists of ®ve hand-drawn
faces ranging from smiling
(`no pain') to crying (`most
severe') along with a
numerical descriptive scale
ranging from 0 to 4; scale
developed based on Wong and
Baker (1988) scale
The ordered faces are
scored from 0 to 4
Authors report a high percent
agreement (87%) between
child and nurse ratings of pain
using this scale
Kuttner and LePage (1989) Four to 12-year-olds One set of ®ve drawings of
children's faces exhibiting
increasing levels of pain
(ranging from neutral to
severe) and a second set of ®ve
drawings indicating increasing
levels of anxiety
The ordered faces are
scored from 0 to 4 on each
scale
Children were able to choose
from all possible paired
comparisons the face they felt
was the most scared or hurt;
faces shown to possess interval
scale properties; good
agreement among children
with respect to what the scales
were designed to measure
LeBaron and Zeltzer (1984) Six to 10-year-olds Five hand-drawn faces
ranging from neutral to sad;
same scale used to assess both
pain and anxiety; no data
provided on how faces were
developed
The ordered faces are
scored from 0 to 4
Moderate correlations (from
r ˆ 0.11 to 0.50) between
patient and observer rating of
pain using the scale
Lehmann et al. (1990) Three to 8-year-olds Scale consists of ®ve sets of
faces beginning with a single
neutral face and each
successive item consisting of
an increasing number of
distressed faces, with the
number ranging from two to
®ve
The ordered faces are
scored from 0 to 4
Children's ordering of a series
of painful experiences (e.g.
getting a shot, a cut) using the
scale were in the order
hypothesized by the authors
Maunuksela et al. (1987) Four to 12-year-olds Five hand-drawn faces
ranging from smiling to
crying; no data provided on
how the faces were developed
The ordered faces are
scored from 0 to 4
High correlations between the
faces scale and other self-
report and behavioral
measures (r ˆ 0.59 to 0.83)

3. compared with other assessment tools, including visual
analogue scales and word descriptor scales (Wong and
Baker, 1988; West et al., 1994; Fogel-Keck et al., 1996).
Acceptance of faces scales has also likely been facilitated by
the importance of facial expression in the social communi-
cation of pain (Craig et al., 1996). However, there has been
minimal effort to empirically validate the relative merits of
different faces scales (Johnston, 1998).
The ®rst faces scale, a series of seven line-drawn faces,
was developed by Katz, 1979. Since then, a number of faces
scales have been developed (e.g. Frank et al., 1982; Beyer,
1984; LeBaron and Zeltzer, 1984; McGrath et al., 1985;
Smith and Covino, 1985; Maunuksela et al., 1987; Wong
and Baker, 1988; Kuttner and LePage, 1989; Bieri et al.,
1990; Douthit, 1990; Lehmann et al., 1990; Pothmann,
1990; Tyler et al., 1993; Goddard and Pickup, 1996).
Table 1 reviews currently available faces scales, lists ages
for which they were designed, and brie¯y summarizes their
psychometric properties, including scoring. Despite a simi-
lar underlying conceptual basis, there are considerable
differences between the various faces scales. The scales
differ in format, ranging from simple line drawings
(Maunuksela et al., 1987), through cartoon-like representa-
tions (Wong and Baker, 1988), to more detailed depictions
of facial expressions (Kuttner and LePage, 1989) and actual
photographs of children's faces (Beyer, 1984). Further, the
scales vary with respect to the number of faces included in
the array. Some include only ®ve faces (LeBaron and Zelt-
zer, 1984) while others include six (Wong and Baker, 1988)
or seven (Bieri et al., 1990) faces. The scales also vary
depending on whether tears are present in the `worst pain'
face (Kuttner and LePage, 1989) or not (Bieri et al., 1990),
and whether the `no pain' face is a neutral face (Bieri et al.,
1990) or a smiling face (Wong and Baker, 1988).
Despite these often marked differences, little attention has
been paid to how variations in scale format affect children's
self-reported pain ratings, with the apparent assumption that
the various scales are equivalent and interchangeable.
Evidence is emerging to the contrary. There has been criti-
cism that faces scales beginning with a smiling `no pain' face
confound non-noxious affective states with pain; for exam-
ple, children who are not in pain are not necessarily happy
(Champion et al., 1998). Chambers and Craig (1998) found,
using modi®ed versions of existing faces scales so that
aspects other than mouth expression were comparable, that
children's ratings of the severity of pain in hypothetical vign-
ettes depended on whether the faces scale used began with a
smiling `no pain' face or a neutral `no pain' face. Speci®-
cally, the use of a scale with a smiling `no pain' face resulted
in signi®cantly higher pain ratings (i.e. `false positives' for
pain or overestimations of pain severity) in situations that
involved negative emotions but not pain (e.g. waking up in
a thunder and lightning storm) and in situations with both
negative emotions and pain (e.g. getting a needle injection at
the doctor's), when compared with ratings using a neutral
scale. Similarly, the use of a scale with a smiling `no pain'
face in situations that involved positive emotions and pain
(e.g. `®nally' getting an ear pierced) resulted in signi®cantly
C.T. Chambers et al. / Pain 83 (1999) 25±35 27
Table 1 (continued)
Reference Population for use Description of Scale Scoring Reliability and validityb
McGrath et al. (1985) Three to 15-year-olds Nine hand-drawn faces
ranging from smiling to
crying; no data provided on
how faces were developed;
developed as a measure of
pain affect and not pain
intensity
The faces are scored on a
0±1 scale, where the
maximum negative
affective value equals 1
(i.e. 0.04, 0.17, 0.37, 0.47,
0.59, 0.75, 0.79, 0.85,
0.97)
Showed good consistency
between the scale and a VAS
during different medical
procedures; moderate
correlations between faces
scale and behavioral measures;
used cross-modality matching
to generate a ratio scale for the
faces
Pothmann (1990) Three to 18-year-olds Five hand-drawn faces
ranging from neutral to sad; no
data provided on how faces
were developed
The ordered faces are
scored from 0 to 4
High correlation between the
faces scale and other self-
report measures (r ˆ 0.87)
Smith and Covino (1985) Not stated Eight hand-drawn faces
ranging from happy to sad;
developed based on scale by
Frank et al. (1982)
The ordered faces are
scored from 0 to 7
None provided
Wong and Baker (1988) Three to 18-year-olds Six hand-drawn faces ranging
from smiling to crying; faces
were developed based on
analyses of children's
drawings of faces representing
different degrees of `hurt'
The ordered faces are
scored from 0 to 5
No signi®cant differences in
test-retest scores, preferred by
children in comparison to other
self-report measures
a
These scales were located through literature reviews using Medline and Psychlit computer databases.
b
There can be high correlations between measures of pain that indicate similar rank ordering, even though there are substantial mean differences in the
distributions being correlated (Chambers et al., 1998).

4. lower pain ratings (i.e. `false negatives' or underestimations
of children's pain), in comparison to ratings with a neutral
face scale. These ®ndings were consistent across the sample
of 100 healthy children ranging in age from 5 to 12 years.
This study demonstrated systematic biases in children's pain
ratings using hypothetical vignettes. The question of general-
izability of ®ndings to children actually experiencing pain in
clinical settings still needs to be tested. In addition, since the
previous work in this area used modi®ed versions of faces
scales speci®cally designed for the investigation, research is
needed to examine whether the biasing effect occurs when
using the original faces scales most commonly used in
research and clinical practice.
The primary purpose of the present study was to examine
whether the previous ®ndings of bias in children's self-
reported ratings of pain (Chambers and Craig, 1998) would
be observed in children experiencing clinical pain (i.e. vene-
puncture) when applying ®ve commonly used faces scales.
These commonly used scales vary in whether the `no pain'
face is a smiling face or a neutral face. Two of the faces scales
used in the current study had smiling `no pain' faces (Maun-
uksela et al., 1987; Wong and Baker, 1988), while the others
had neutral `no pain' faces (LeBaron and Zeltzer, 1984; Kutt-
ner and LePage, 1989; Bieri et al., 1990). These ®ve faces
scales were chosen from the many available faces scales, not
only because they varied with respect to whether the `no
pain' face was smiling or neutral, but also because systematic
literature reviews con®rmed that they are commonly cited in
empirical research and are widely used in clinical practice. It
was predicted that the scales with smiling `no pain' faces
would elicit higher pain ratings, in comparison to ratings
using scales with neutral `no pain' faces. As well, the study
examined age and gender differences in children's pain
ratings. Younger children consistently have reported more
pain than older children (Fradet et al., 1990; Lander and
Fowler-Kerry, 1991; Arts et al., 1994; Goodenough et al.,
1997a; Chambers and Craig, 1998), and females have
reported more pain than males (Lautenbacher and Rollman,
1993; Unruh, 1996; Goodenough et al., 1997a).
Parents are often called upon to provide pain ratings when
their children cannot do so or there are questions concerning
the credibility of the child's ratings (McGrath et al., 1994;
Chambers et al., 1998). Consequently, another purpose of
the current study was to examine the appropriateness of the
use of faces scales to elicit parent reports of their child's
pain and to examine whether the presence of a smiling `no
pain' face would produce biased ratings among parents as
well as children. Further, there have been con¯icting
accounts regarding the accuracy of parent reports of pain,
with some studies indicating good agreement with their
children's ratings (O'Hara et al., 1987; Bennett-Branson
and Craig, 1993; Miller, 1996), and other studies showing
poor agreement (Manne et al., 1992; Bellman and Paley,
1993; Chambers et al., 1998). It has been suggested that
the differences across studies might be related to the use
of different rating scales (Chambers et al., 1998). Therefore,
we also compared levels of agreement between child and
parent reports of pain using the ®ve different faces scales.
Since the actual choice of scales used clinically may be
affected by factors other than research data (e.g. attractive-
ness of the scale), the ®nal purposes of this study were to
examine which of the ®ve faces scales children and parents
would prefer and to explore what characteristics of faces
scales make them appealing to both children and parents.
2. Method
2.1. Participants
Participants were 75 children (39 girls, 36 boys) between
the ages of 5 and 12 years (mean ˆ 8:73 years, SD ˆ 2:38),
C.T. Chambers et al. / Pain 83 (1999) 25±35
28
Fig. 1. The ®ve faces scales. (A) Bieri et al. scale: Reprinted from Bieri D,
Reeve RA, Champion GD, Addicoat L, Ziegler JB. The Faces Pain Scale
for the self-assessment of the severity of pain experience by children:
development, initial validation, and preliminary investigation for ratio
scale properties. Pain 1990;41:139±150, with permission from Elsevier
Science; (B) Wong & Baker FACES Pain Rating Scale: From Wong D.
Whaley and Wong's essentials of pediatric nursing 5th ed. 1997 p. 1215.
Copyrighted by Mosby-Year Book, Inc. Reprinted by permission. The
instructions that typically accompany the scale are: Face 0 ˆ no hurt;
Face 1 ˆ hurts a little bit; Face 2 ˆ hurts little more; Face 3 ˆ hurts even
more; Face 4 ˆ hurts whole lot; Face 5: hurts worst; (C) Maunuksela et al.
scale: From Maunuksela EL, Olkkola KT, Korpela, R. Measurement of pain
in children with self-reporting and behavioral assessment. Clinical Pharma-
cology Therapeutics 1987;42:137±141, with permission; (D) LeBaron &
Zeltzer scale: LeBaron S, Zeltzer L. Assessment of acute pain and anxiety
in children and adolescents by self-reports, observer reports, and a beha-
viour checklist. Journal of Consulting and Clinical Psychology 1984;52:
729±738. Copyright q 1984 by the American Psychological Association.
Reprinted with permission; (E) Kuttner & LePage scale: From Kuttner L,
LePage T. Faces scales for the assessment of pediatric pain: a critical
review. Canadian Journal of Behavioral Science 1989;21:198±209, with
permission.

5. recruited from the endocrine (n ˆ 37) and diabetic (n ˆ 38)
units at British Columbia's Children's Hospital (BCCH),
who were scheduled for venepuncture either in the central
lab or metabolic investigation unit. Sixty-seven percent of
the sample had had ten or more previous venepunctures and
49% of the children opted to use EMLA (Eutectic Mixture
of Local Anaesthetics) for their current procedure, with
approximately equal numbers of boys and girls choosing
to use EMLA or not. Parents (62 mothers, 13 fathers) ranged
in age from 25 to 56 years (mean ˆ 38:24, SD ˆ 6:00) and
were of middle social class (mean ˆ 38:13, SD ˆ 14:56;
Hollingshead Index).
2.2. Measures
Five faces scales were chosen for use in this study
(LeBaron and Zeltzer, 1984; Maunuksela et al., 1987;
Wong and Baker, 1988; Kuttner and LePage, 1989; Bieri
et al., 1990). A more detailed description of each of the ®ve
scales is provided in Table 1, and they are shown in Fig. 1.
With respect to the scales developed by Kuttner and LePage
(1989), only the pain scale, and not the anxiety scale, was
used in the current study. Faces scales were shown in
succession on laminated white paper 10 cm in height by
28 cm in width to keep size and background parameters
consistent. No visual written descriptions or numbers
accompanied the faces scales. Each scale was prepared so
that the diameter of each of the faces was 3 cm. Depending
on the number of faces presented in each of the ®ve arrays,
children's and parents' ratings were scored from 0 to 4, 0 to
5, or 0 to 6. Scores were later standardized to ratings on a 0±
10 scale (Goodenough et al., 1997b). Verbal instructions
(see below) were also modi®ed to be consistent across all
scales used.
C.T. Chambers et al. / Pain 83 (1999) 25±35 29
Table 2
Descriptive statistics of children's and parents' ratings for each of the ®ve faces scales
Bieri et al.
(0±6)
Wong and Baker
(0±5)
Maunuksela et al.
(0±4)
LeBaron and Zeltzer
(0±4)
Kuttner and LePage
(0±4)
Children's ratings
Mean (raw) 1.49 1.51 1.19 0.99 1.00
Mean (standardized/10)a
2.49* 3.01** 2.97** 2.47* 2.50*
Median (raw) 1.00 1.00 1.00 1.00 1.00
Median (/10) 1.67 2.00 2.50 2.50 2.50
Standard deviation (/10) 3.15 3.18 3.33 3.14 2.99
Skew 1.30 0.99 0.85 1.33 1.26
Kurtosis 0.68 0.10 20.45 0.79 0.89
Parents' ratings
Mean (raw) 2.09 2.15 1.47 1.21 1.24
Mean (standardized/10)a
3.49* 4.29** 3.67* 3.03*** 3.10***
Median (raw) 2.00 2.00 1.00 1.00 1.00
Median (/10) 3.33 4.00 2.50 2.50 2.50
Standard deviation (/10) 3.04 2.82 2.97 2.89 2.90
Skew 0.83 0.50 0.90 1.09 0.90
Kurtosis 20.31 20.38 20.01 0.61 0.14
a
Standardized means in the same row with different numbers of asterisks differ at P , 0:05 or better. Raw means are reported using the scales of the original
measure, but, for purpose of comparison, the raw scores were standardized as proportions of a scale of 10.
Table 3
Spearman correlation matrix for all ®ve faces scales (parents and children's ratings)a
Bieri et al. Wong and Baker Maunuksela et al. LeBaron and Zeltzer
Children's ratings
Wong and Baker 0.91 ± ± ±
Maunuksela et al. 0.89 0.86 ± ±
LeBaron and Zeltzer 0.90 0.82 0.81 ±
Kuttner and LePage 0.91 0.85 0.81 0.93
Parents' ratings
Wong and Baker 0.88 ± ± ±
Maunuksela et al. 0.81 0.83 ± ±
LeBaron and Zeltzer 0.85 0.84 0.78 ±
Kuttner and LePage 0.84 0.83 0.78 0.84
a
Note. All correlations are signi®cant at P , 0:001.

6. 2.3. Procedure
Parents of children scheduled for venepuncture were
informed of the study by a clinic nurse. Those parents
who indicated an interest in participating (n ˆ 83) were
approached by a researcher and provided with additional
information (e.g. nature of the tasks, time required). Three
parents did not speak English suf®ciently well to participate
and ®ve parents elected not to take part. The others (n ˆ 75)
provided written informed consent. Written consent was
obtained from children over the age of 7 years. Verbal
consent was obtained from the younger children.
Prior to the procedure, parents provided basic demo-
graphic information (e.g. parent education, occupation)
and answered brief questions regarding their child's medical
history. Immediately following the blood sampling proce-
dure, ®rst the children and then their parents were indepen-
dently shown the ®ve faces scales in random order. Upon
presentation of the ®rst faces scale, they were provided with
the following instructions: `Each of these faces is of a
person who has no hurt or pain, or some, or a lot of hurt
or pain. This ®rst face doesn't have any hurt or pain at all.
The next face has just a little bit of hurt or pain. Each of the
next faces has a little more hurt or pain, until you get to this
last face, who has the worse hurt or pain you can imagine'
(Chambers and Craig, 1998). They were asked to point to
the face that showed how much hurt or pain the child felt
during the blood test. Parents were not allowed to watch
while their children provided ratings. Upon presentation
of each of the remaining four faces scales, children were
given the following reminder, `Remember, this ®rst face
means no hurt or pain and each of these faces means a little
bit more hurt or pain until you get to this last face which
means the worst hurt or pain you can imagine.' When
presented with the faces scale by Maunuksela et al.
(1987), their attention was directed to the fact that this
scale is oriented from right to left.
After providing their pain intensity ratings, children, and
then parents, were independently shown all ®ve faces scales
and were asked to indicate which scale they `liked' the most,
as well as asked an open-ended question about why they
preferred the scale that they chose. Their responses were
later classi®ed into broader categories, which were judged
by the authors (K.G. and C.T.C.) to be representative of the
qualitatively distinct reasons why children and parents
preferred scales. This study was approved by both the
University of British Columbia (UBC) and BCCH Research
Ethics Boards.
2.4. Statistical analyses
Descriptive measures of central tendency (i.e. mean,
median), variability, skew and kurtosis were used to exam-
ine the distributions of children's and parents' responses.
Skew values provide a description of the degree of asym-
metry or skewness in a distribution, while kurtosis values
provide a description of whether there are more or fewer
extreme scores than expected in a normal distribution.
Spearman correlation coef®cients (for skewed data) were
used to examine the relations between each of the ®ve scales
for both children's and parents' ratings. As correlational
analyses do not disclose mean differences between
measures, analyses of variance (ANOVA) were also under-
taken. Since the ®ve faces scales had different ranges, the
scores on each were standardized (i.e. converted to ratings
out of 10) to enable meaningful comparisons between scales
and for the purposes of statistical analyses using ANOVA
techniques (Goodenough et al., 1997b). Bartlett Box's tests
of homogeneity of covariance matrices and Mauchley's
tests of sphericity were used to examine any violations of
the multi-sample sphericity assumption underlying the
mixed ANOVA analyses for parents' and children's ratings.
To examine age differences in children's use of the faces
scales, children were divided into three age groups: 5±7 year
olds (n ˆ 24), 8±10 year olds (n ˆ 29), and 11±13 year olds
(n ˆ 22). A set of 2 gender† £ 3 (age group† £ 5 (faces
scale) mixed ANOVAs were used to examine differences
in children's self-reported pain ratings, as well as parents'
ratings of their children's pain (as a function of the chil-
dren's age and gender). Given the limitations of correlations
for examining parent±child agreement (Chambers et al.,
1998), Kappa statistics were used to describe the level of
agreement between parents' and children's ratings and a
series of paired t-tests examined mean differences between
parents' and children's ratings using each of the ®ve scales.
Percentages were used to describe the scales preferred by
children and by parents, as well as the reasons for their
preference.
C.T. Chambers et al. / Pain 83 (1999) 25±35
30
Table 5
Children's and parents' preferences (%) for faces scales
% Preferred by children % Preferred by parents
Bieri et al. 8.2 25.0
Wong and Baker 64.4 40.3
Maunuksela et al. 13.7 12.5
LeBaron and Zeltzer 2.7 4.2
Kuttner and LePage 11.0 18.1
Table 4
Standardized means and standard deviations of boys' and girls' pain ratings
using the ®ve faces scalesa
Boys' ratings (0±10) Girls' ratings (0±10)
Mean (SD) Mean (SD)
Bieri et al. 1.94 (2.72) 2.99 (3.47)
Wong and Baker 2.17 (2.36) 3.79 (3.64)
Maunuksela et al. 2.43 (3.13) 3.46 (3.47)
LeBaron and Zeltzer 1.74 (2.45) 3.14 (3.57)
Kuttner and LePage 1.94 (2.61) 3.01 (3.25)
a
Note. Boys' and girls' pain ratings using all ®ve scales are signi®cantly
different at P ˆ 0:05.

7. 3. Results1
Descriptive statistics for children's and parents' pain
ratings on each of the ®ve faces scales, collapsing across
age and gender, appear in Table 2. Overall, mean ratings
were in the low to moderate range (means ranged from 2.57
to 3.01 on the transformed 0±10 scale). Children's and
parents' ratings on all ®ve faces scales tended to be skewed
toward the `no pain' end of the measures (as evidenced by
the positive skew values, 0.50±1.33). Kurtosis values
revealed that the distributions of children's and parents'
ratings varied with some approximating normal distribu-
tions (i.e. kurtosis values close to 0) and others having
greater (i.e. positive kurtosis values) or fewer (i.e. negative
kurtosis values) extreme scores than what would be
expected from a normal distribution.
Correlations among ratings on all of the ®ve faces scales
are shown in Table 3, separately for both children and
parents. Correlations among children's ratings were gener-
ally very high, ranging from 0.81 (Maunuksela et al., 1987
scale with both the LeBaron and Zeltzer, 1984 and the
Kuttner and LePage, 1989 scales) to 0.93 (LeBaron and
Zeltzer scale with the Kuttner and LePage scale). Correla-
tions among parents' ratings were also very high, ranging
from 0.78 (Maunuksela et al. scale with both the LeBaron
and Zeltzer and the Kuttner and LePage scale) to 0.88 (Bieri
et al. scale with the Wong and Baker scale).
For both parents' and children's ratings, the Bartlett Box
tests were signi®cant (P , 0:05), indicating that the
observed covariance matrices of the dependent variables
were not equal across groups. However, when the correla-
tions between the cell sample sizes and the cell variances for
parent and child ratings were examined, neither of these sets
of correlations were signi®cantly negatively correlated,
which would have indicated a liberal bias in the signi®cance
of the results. Therefore, the analysis was appropriate with-
out adjustment for the lack of homogeneity (Glass and
Hopkins, 1996). Further, the Mauchley's tests of sphericity
were also signi®cant (P , 0:05) (i.e. the pooled matrices
were not spherical) for parents' and children's ratings.
Therefore, as recommended when a data set is not spherical
(Glass and Hopkins, 1996) a Greenhouse±Geisser adjust-
ment, a conservative approach which slightly adjusts the
degrees of freedom to account for the lack of sphericity,
was used. The ANOVA results reported below represent
the Greenhouse±Geisser adjusted values for the within-
subjects effects.
The 2 gender† £ 3 age† £ 5 (faces scale) ANOVA on
children's ratings showed no signi®cant interaction effects.
The main effect of age was also not signi®cant, F(2,
69† ˆ 1:45, n.s. There was a signi®cant main effect for
difference between faces scales, F(3.27, 225:45† ˆ 5:14,
p ˆ 0:001. Student±Newman±Keuls post-hoc tests revealed
that children's ratings using the scales by Wong and Baker
(1988) and Maunuksela et al. (1987) were signi®cantly
higher than were their ratings using the scales by Bieri et
al. (1990); LeBaron and Zeltzer (1984) and Kuttner and
LePage (1989) (see Table 2). There was also a signi®cant
main effect for gender, F(1, 69† ˆ 3:70, P ˆ 0:05, with girls
scoring higher on all ®ve faces scales in comparison to boys
(the standardized means and standard deviations for boys'
and girls' ratings using the ®ve scales are shown in Table 4).
The ANOVA on parents' ratings showed no signi®cant
interaction effects, and the main effects of child's age, F(2,
69† ˆ 1:85, n.s., and gender, F(1, 69† ˆ 0:92, n.s. were also
not signi®cant. There was a signi®cant main effect for faces
scales, F(3.51, 241:99† ˆ 14:35, P , 0:001. Student±
Newman±Keuls post-hoc tests revealed that parents' ratings
were highest when using the Wong and Baker scale,
followed by the Maunuksela et al. and Bieri et al. scales,
and then by the LeBaron and Zeltzer and Kuttner and
LePage scales (see Table 2).
Kappa statistics describing the relationship between
parents' and children's ratings for each of the ®ve scales
were as follows: Bieri et al. (1990), k ˆ 0:21; Wong and
Baker (1988), k ˆ 0:21; Maunuksela et al. (1987),
k ˆ 0:21; LeBaron and Zeltzer (1984), k ˆ 0:36; and Kutt-
ner and LePage (1989), k ˆ 0:28. All of the Kappas repre-
sented agreement beyond chance but at a level usually
classi®ed as ``poor'' (Fleiss, 1981). The paired t-tests
revealed that parents had signi®cantly higher pain ratings
in comparison to their children's ratings using all ®ve of the
faces scales (Bieri et al., t 74† ˆ 23:45, P , 0:001; Wong
C.T. Chambers et al. / Pain 83 (1999) 25±35 31
1
Copies of the complete ANOVA tables and post-hoc testing for the
analyses described are available from the authors upon request.
Table 6
Children's and parents' reasons (%) for scale preferencea
% Endorsed by children % Endorsed by parents
Simple and easy to use 6.8 15.3
Happy or cartoon-like faces 52.1 27.8
More options/choices 2.7 6.9
Realistic/life-like faces 4.1 23.6
Descriptive/expressive faces 2.7 13.9
Other or not sure 31.5 12.5
a
Note. Children and parents provided answers in response to an open-ended question about why they chose their preferred scale. Responses were later
classi®ed into the above categories, which were judged to capture the qualitatively distinct reasons why children and parents preferred scales.

8. and Baker, t 74† ˆ 24:65, P , 0:001; Maunuksela et al.,
t 74† ˆ 22:07, P , 0:05; LeBaron and Zeltzer, t 74† ˆ
22.27, P , 0:05; Kuttner and LePage, t 74† ˆ 22:28,
P , 0:05). The means and standard deviations of children's
and parents' ratings are found in Table 2.
The percentages of children and parents preferring each
of the scales, and reasons for choosing their preferred scale,
are shown in Tables 5 and 6. The majority of children
(64.4%) and parents (40.3%) reported a preference for the
scale by Wong and Baker (1988); however, a large portion
of parents also preferred the Bieri et al. (1990) scale
(25.0%). The least preferred for both children (2.7%) and
parents (4.2%) was the LeBaron and Zeltzer (1984) scale.
Over half of the children (52.1%) and 27.8% of the parents
reported that what they liked about their preferred scale was
that it was `happy' or `cartoon-like'. A number of parents
also reported liking scales that were `realistic' or `life-like'
(23.6%) and `simple' or `easy to use' (15.3%).
4. Discussion
The primary purpose of the current study was to examine
whether children's self-reported ratings of clinical pain
using faces scales would be biased by the use of scales
with `no pain' faces that depicted a positive, non-noxious
affective state (i.e. smiling faces) rather than an affectively
neutral `no pain' display. Our ®ndings indicated that,
despite high to very high correlations among children's
ratings across the ®ve faces scales, there were signi®cant
mean differences between ratings. Children had signi®-
cantly higher pain ratings when using two scales that
commence with smiling `no pain' faces (Maunuksela et
al., (1987) and Wong and Baker (1988)), in comparison to
ratings using scales with neutral `no pain' faces (Kuttner
and LePage (1989); LeBaron and Zeltzer (1984) and Bieri
et al. (1990)).
While it is the case that the ®ve scales used varied in
several ways (e.g. cartoon-like versus realistic face draw-
ings, presence of tears in the `worst pain' face), the common
element that distinguishes the two scales on which scores
were signi®cantly higher from the other three is the presence
of a smiling rather than a neutral `no pain' face. Therefore,
the ®ndings in this clinical setting were consistent with the
®ndings of Chambers and Craig (1998), who also found
signi®cantly higher pain ratings when a smiling scale was
used to rate hypothetical situations involving negative
emotions and pain (e.g. getting an injection). It appears
that the presence of a smiling face as the `no pain' face
biases toward shifting children's ratings away from the
`no pain' end of the spectrum, resulting in signi®cantly
higher pain ratings in comparison to scales with neutral
`no pain' faces. Again, similar to the ®ndings of Chambers
and Craig (1998), this effect was constant across age (i.e. 5±
12 years) and gender. Perhaps most important was the
evidence of bias in children with important medical
problems experiencing clinical pain using scales commonly
used for health service delivery, rather than healthy children
responding to hypothetical vignettes, as in the earlier study.
Assuming there is validity to the proposition that the deliv-
ery of pain control should be predicated on the presence,
nature, and severity of the pain, it is important that pain
reports should be valid (i.e. disclose the subjective experi-
ence of pain), rather than be in¯uenced by non-nociceptive
events and experience.
Similarly, parents' ratings of their children's pain varied
with the ®ve faces scales. Parents had higher ratings using
the scale by Wong and Baker, followed by the scales by
Maunuksela et al. and Bieri et al., and then by the LeBaron
and Zeltzer and Kuttner and LePage scales. Thus, parents
also appear to have been biased toward providing higher
ratings with the smiling `no pain' face. In contrast to their
children, parents also had higher pain ratings when using the
Bieri et al. scale. The pattern of results from the parents is
not as clear as with the children, and it is dif®cult to spec-
ulate post-hoc what other factors might account for between
scale differences. One possible reason for the relatively
higher pain ratings using the Bieri et al. scale might be
that there were relatively more response options to select
from, a characteristic shared with the Wong and Baker
scale. The Bieri et al. and Wong and Baker scales offer
seven and six alternatives, respectively, whereas the remain-
ing scales provide only ®ve options. Consequently, it would
appear that scales with smiling `no pain' faces, as well as
scales with a greater number of response options, produce
relatively higher pain ratings of children's clinical pain by
their parents. This combination could account for the higher
parent ratings using the Wong and Baker scale compared
with their ratings using any other scale.
However, this interpretation assumes that faces scales are
appropriate for use with adults, when they were developed
to elicit self-reports of pain from children. Anecdotally,
parents appeared to experience dif®culty understanding
how to apply the scales to judge pain in their children,
even when given the same verbal instructions their children
received. For example, some parents would report, `Well, he
had a lot of pain, but his face never looked like that,' and
they would choose a face toward the lower end of the array.
On the other hand, some parents seemed unable to separate
their children's more general behavioral distress during the
procedure from the amount of pain the child experienced.
They would report to the experimenter, `Oh, I didn't think it
hurt her too much. She was just scared,' but then would
proceed to point to a face at the `worst pain' end of the
array. None of the children in our study appeared to experi-
ence dif®culties understanding the instructions that were
provided with the faces scales. It may be that adults are
capable of providing more sophisticated and differentiated
ratings and descriptions of pain (e.g. they are able discrimi-
nate sensory and affective qualities of pain), and providing
ratings on a single continuum, such as a faces scale, is too
simple or global a format for them to communicate their
C.T. Chambers et al. / Pain 83 (1999) 25±35
32

9. more complex judgements. The appropriateness of faces
scales for eliciting parent reports of their children's pain is
an issue that deserves further attention.
There was a gender difference in children's reports of
pain, with girls reporting signi®cantly more pain than
boys using all ®ve faces scales. This is consistent with
previous research reporting similar gender differences in
pain ratings (Lautenbacher and Rollman, 1993; Unruh,
1996; Goodenough et al., 1997a). Parents' ratings, however,
did not vary as a function of the child's gender. We remain
uncertain as to how differences in pain expression relate to
variations in pain experience, and the basis for these differ-
ences (i.e. whether the differences are due to biological
factors or differential socialization) remains to be explored.
Interestingly, in contrast to previous research (Fradet et al.,
1990; Lander and Fowler-Kerry, 1991; Arts et al., 1994;
Goodenough et al., 1997a; Chambers and Craig, 1998) no
age differences in either children's pain ratings or parents'
ratings of their children's pain were found. However, a large
proportion (57%) of younger children in our sample used
EMLA to manage their pain from the venepuncture proce-
dure, while only 31% of older children used EMLA. Conse-
quently, the greater use of EMLA among the younger
children may have served to suppress otherwise higher
ratings and bring them into a similar range as the ratings
of older children.
Another purpose of the current study was to shed addi-
tional light on the unresolved issue of agreement between
parent and child reports of pain. There were non-chance
levels of agreement across the ®ve faces scales, with some
variability (Kappa statistics ranged from 0.21 to 0.36), but
all of these values represented poor levels of agreement
(Fleiss, 1981). Indeed, parents consistently overestimated
the level of their children's pain using each of the faces
scales.
The ®nding of overestimation was inconsistent with
previous research investigating postoperative pain in chil-
dren, which typically reveals parental underestimation of
children's pain (Bellman and Paley, 1993; Chambers et al.,
1998). There has been considerable speculation as to why
adults misperceive children's pain. The inconsistencies
may represent a different perspective on the child's pain
experience (Manne et al., 1992). The context and type of
pain (i.e. short, sharp pains vs. longer term pains) being
rated may affect parental accuracy. As parents were present
to observe their child's painful procedure during venepunc-
ture, they may have based their pain ratings on behavioral
distress as well as their perception of how painful the vene-
puncture was. Alternatively, the low levels of parent±child
agreement may be due in part to the methodological dif®cul-
ties parents experienced with rating pain using the faces
scales as discussed above. Future research examining
parent±child agreement in pain reports should include an
examination of agreement when using not only faces scales,
but other types of self-report measures as well (e.g. color
analogue scales). As parents' ratings are also amenable to
bias, caution must be exercised in using them to disentangle
the `truth' in children's pain experiences. The most fruitful
research strategy would appear to compare multiple meas-
urement approaches, including self-report (when available),
adult judgements, behavioral measures (e.g. facial expres-
sion) and biological measures (e.g. heart rate, sweating).
The largest proportion of children and parents reported a
preference for the scale by Wong and Baker (1988). Some
parents preferred the scale by Bieri et al., (1990). The
prominent characteristics that made the preferred scales
appealing to children referred to whether the scale had a
`happy' or cartoon-like format. These characteristics were
also appealing to parents; however, some parents reported
that they preferred scales that were `realistic' or `easy to
use'. Interestingly, the scale preferred by both parents and
children (i.e. the Wong and Baker scale) was the scale that
produced signi®cantly higher independent child and parent
ratings in comparison to the other scales. This indicates that
simply choosing a faces scale by asking children which one
they prefer may not result in the choice of a faces scale that
produces the more valid ratings in comparison to other
scales. While it is agreed that scales should be as child-
friendly as possible, this should not be done with a sacri®ce
of validity.
An issue not addressed in the current study concerns the
psychometric scaling and numerical scoring of the faces
scales (Johnston, 1998). Following conventional practice,
the scales used in the current study were scored as interval
data, with each successive face on the ascending scale being
assigned an incremental value of one. This method of scor-
ing assumes that the faces in the same ascending order on
each of the scales should be assigned the same numerical
values. However, present ®ndings indicate that the various
faces scales do not yield equivalent pain ratings. Conse-
quently it may be more appropriate to assign the faces
different numerical values (e.g. using a cross-modality
matching procedure; McGrath et al., 1985). If scale values
were assigned appropriately, the various faces scales might
be rendered equivalent. However, this study is the ®rst to
directly compare faces scales; future research is needed to
further explore the issue of psychometric scaling.
In addition, it is possible that smiling faces scales are
more appropriate as measures of pain affect (i.e. the
emotional aspect of pain) rather than pain intensity as they
are currently used. For example, some recent research by
Goodenough et al. (1999) examined the relationship
between children's ratings on two different faces scales
and a paired visual analogue method for measuring the
intensity and unpleasantness of pain during blood sampling.
Results showed that that the Bieri et al. faces scale (a neutral
scale) was more highly correlated with VAS ratings of
intensity than unpleasantness, whereas the McGrath et al.
(1985) Facial Affective Scale (a smiling scale) was more
highly correlated with VAS ratings of unpleasantness than
intensity (Goodenough et al., 1999). The present study
shows that smiling scales and neutral scales do not yield
C.T. Chambers et al. / Pain 83 (1999) 25±35 33

10. equivalent values when used as measures of pain intensity,
as recommended by their creators. The possibility that these
smiling scales may be more appropriate as measures of pain
affect, rather than pain intensity, needs to be explored.
A strength of the current study was its use of a clinical
sample, rather than the healthy children used in the earlier
study. However, we noted that the pain ratings provided by
both children and their parents were in the low±moderate
range. Venepuncture is a standard pain-producing procedure
that has been frequently used as a paradigm to study acute
pain in children; however, the ratings obtained in the current
study were somewhat lower than those obtained in earlier
research using venepuncture pain (Manne et al., 1992). This
could be attributed to the considerable experience that the
children surveyed in the current study had with painful
medical procedures (i.e. 67% of children had had ten or
more previous venepunctures). They may have become
less anxious about the procedure, or developed effective
coping strategies to reduce their pain. In addition, a large
number of children used EMLA in an effort to decrease their
pain. Consequently, it was not surprising that the pain
ratings from children in this study were relatively low in
comparison to previous research. Future research needs to
be conducted in other pain situations (e.g. postoperative
pain) before the ®ndings of this current study can be con®-
dently extrapolated to different pain contexts.
In summary, this study appears to con®rm that faces
scales with smiling `no pain' faces result in higher pain
ratings relative to scales with neutral `no pain' faces when
used by children. While the pattern is less clear for parents,
it appears that parents' ratings are also impacted by the
presence of smiling `no pain' faces. The appropriateness
of faces scales for eliciting parent reports of pain in their
children needs to be established. It would appear that choice
of scales for children's pain assessment is often dictated by
face validity (i.e. what the scale appears to measure) or
appeal (e.g. a `cute' faces scale), rather than thorough
consideration of what it actually measures (i.e. empirical
validity) (Johnston, 1998). The data provided here argues
that the appeal of `happy' or cartoon-like faces scales that
include smiling `no pain' faces serves to distract the user of
the scale, whether parent or child. It would seem better to
provide children with a faces scale that allows them to focus
upon the pain experience alone, rather than in combination
with other non-noxious, affective qualities.
Acknowledgements
Portions of this paper have been accepted for presentation
at the 9th World Congress on Pain, Vienna, Austria. This
project was supported by a grant from the Social Sciences
and Humanities Research Council of Canada (SSHRC)
awarded to Dr Kenneth Craig. C.T. Chambers was
supported by a University Graduate Fellowship (UGF)
from the University of British Columbia and a Doctoral
Award from the Medical Research Council of Canada
(MRC). We would like to thank Deanna Braaksma for her
assistance with data collection and data entry, and the Endo-
crine and Diabetic Unit (EDU), Metabolic Investigations
Unit, and Central Lab at British Columbia's Children's
Hospital, particularly Dr Jean-Pierre Chanoine, Sharon
Connaughty, Janet Preston, Sheila Kelton, Christina Pepe,
Fran Power, Claudette Hildebrand, Dr Bonnie Massing, and
Debbie Sheldon, for their cooperation and assistance with
this project. Finally, we would like to acknowledge the
parents and children who participated in this study.
References
Arts SE, Abu-Saad HH, Champion GD, Crawford MR, Fisher RJ, Juniper
KH, Ziegler JB. Age related responses to lidocaine prilocaine (EMLA)
emulsion and effects of music distraction on the pain of intravenous
cannulation. Pediatrics 1994;93:797±801.
Bellman MH, Paley CE. Parents underestimate children's pain. Br Med J
1993;307:1563.
Bennett-Branson SM, Craig KD. Postoperative pain in children: develop-
mental and family in¯uences on spontaneous coping strategies. Can J
Behav Sci 1993;25:355±383.
Beyer JE. The oucher: a user's manual and technical report. Evanston, IL:
The Hospital Play Equipment Company, 1984.
Bieri D, Reeve RA, Champion GD, Addicoat L, Ziegler JB. The Faces Pain
Scale for the self-assessment of the severity of pain experience by
children: development, initial validation, and preliminary investigation
for ratio scale properties. Pain 1990;41:139±150.
Chambers CT, Craig KD. An intrusive impact of anchors in children's faces
pain scales. Pain 1998;78:27±37.
Chambers CT, McGrath PJ. Pain measurement in children. In: Ashburn
MA, Rice LJ, editors. The management of pain. New York: Churchill
Livingstone, 1998. pp. 625±634.
Chambers CT, Reid GJ, Craig KD, McGrath PJ, Finley GA. Agreement
between child and parent reports of pain. Clin J Pain 1998;14:336±342.
Champion GD, Goodenough B, von Baeyer CL, Thomas W. Measurement
of pain by self-report. In: Finley GA, McGrath PJ, editors. Measure-
ment of pain in infants and children, progress in pain research and
management, Seattle, WA: IASP Press, 1998. pp. 123±160.
Craig KD. The facial expression of pain: better than a thousand words? Am
Pain Soc J 1992;1:153±162.
Craig KD, Gilbert CA, Lilley CM. Social barriers to optimal pain manage-
ment in infants and children. Clin J Pain 1996;12:232±242.
Douthit JL. Psychosocial assessment and management of pediatric pain. J
Emerg Nurs 1990;16:168±170.
Eland JM. Minimizing pain associated with prekindergarten intramuscular
injections. Issues Comp Pediatr Nurs 1981;5:361±372.
Finley GA, McGrath PJ, editors. Measurement of pain in infants and chil-
dren, progress in pain research and management. Seattle, WA: IASP
Press, 1998.
Fleiss JL. Statistical methods for rates and proportions. New York: Wiley,
1981.
Fogel-Keck J, Gerkensmeyer JE, Joyce BA, Schade JG. Reliability and
validity of the faces and word descriptor scales to measure procedural
pain. J Pediatr Nurs 1996;11:368±374.
Fradet C, McGrath PJ, Kay J, Adams S, Luke B. A prospective survey of
reactions to blood tests by children and adolescents. Pain 1990;40:53±
60.
Frank AJM, Moll JMH, Hort JF. A comparison of three ways of measuring
pain. Rheumatol Rehab 1982;21:211±217.
Glass GV, Hopkins KD. Statistical Methods in Education and Psychology.
Boston, MA: Allyn and Bacon, 1996.
C.T. Chambers et al. / Pain 83 (1999) 25±35
34

11. Goddard JM, Pickup SE. Postoperative pain in children: combining audit
and a clinical nurse specialist to improve management. Anaesthesia
1996;51:586±590.
Goodenough B, Kampel L, Champion GD, Laubreaux L, Nicholas MK,
Ziegler JB, McInerney M. An investigation of the placebo effect and
age-related factors in the report of needle pain from venipuncture in
children. Pain 1997a;72:383±391.
Goodenough B, Addicoat L, Champion GD, McInerney M, Young B, Juni-
per KH, Ziegler JB. Pain in 4- to 6-year-old children receiving intra-
muscular injections: a comparison of the Faces Pain Scale with other
self-report and behavioral measures. Clin J Pain 1997b;13:60±73.
Goodenough B, Champion G.D, Brouwer N, Van Dongen K, Abu-Saad H.
A comparison of the faces pain scale and the facial affective scale for
children's self-report of pain during blood sampling procedures, Paper
Accepted for Presentation at the 20th Annual Meeting of the Australian
Pain Society, 1999.
Hester NO, Foster RL, Kristensen K. Measurement of pain in children:
generalizability and validity of the pain ladder and the poker chip
tool. In: Tyler DC, Krane EJ, editors. Advances in pain research and
therapy, vol. 15, pediatric pain. New York: Raven Press, 1979. pp. 79±
84.
Huskisson E. Measurement of pain. Lancet 1974;2:1127±1131.
Jay SM, Ozolins M, Elliott CH, Caldwell S. Assessment of children's
distress during painful medical procedures. Health Psychol
1983;2:133±147.
Jensen MP. Validity of self-report and observation measures. In: Jensen TS,
Turner JA, Weisenfeld-Halleb Z, editors. Proceeding of the 8th World
Congress on Pain, Seattle, WA: IASP Press, 1997. pp. 637±661.
Johnston CC. Psychometric issues in the measurement of pain. In: Finley
GA, McGrath PJ, editors. Measurement of pain in infants and children,
progress in pain research and management, Seattle, WA: IASP Press,
1998. pp. 5-20.
Katz E. Distress behaviors in children with leukemia undergoing medical
procedures, Paper Presented at the American Psychological Association
Conference, 1979.
Kuttner L, LePage T. Faces scales for the assessment of pediatric pain: a
critical review. Can J Behav Sci 1989;21:198±209.
Lander J, Fowler-Kerry S. Age differences in children's pain. Percept Mot
Skills 1991;73:415±418.
Lautenbacher S, Rollman GB. Sex differences in responsiveness to painful
and non-painful stimuli are dependent upon the method of stimulation.
Pain 1993;53:255±264.
LeBaron S, Zeltzer L. Assessment of acute pain and anxiety in children and
adolescents by self-reports, observer reports, and a behaviour checklist.
J Consult Clin Psychol 1984;52:729±738.
Lehmann HP, Bendebba M, DeAngelis C. The consistency of young chil-
dren's assessment of remembered painful events. Dev Behav Pediatr
1990;11:128±134.
Manne SL, Jacobsen PB, Redd WH. Assessment of acute pediatric pain: do
child self-report, parent ratings, and nurse ratings measure the same
phenomenon? Pain 1992;48:45±52.
Maunuksela EL, Olkkola KT, Korpela R. Measurement of pain in children
with self-reporting and behavioral assessment. Clin Pharm Ther
1987;42:137±141.
McGrath PA, de Veber LL, Hearn MT. Multidimensional pain assessment
in children. In: Fields HL, editor. Advances in pain research and ther-
apy, 9. New York: Raven Press, 1985. pp. 387±393.
McGrath PJ. There is more to pain measurement in children than `ouch'.
Can. Psychol. 1996;37:63±75.
McGrath PJ, Finley GA, Ritchie J. Parents' roles in pain assessment and
management. Int Assoc Study Pain, March/April 1994;3±4.
Merskey H, Bogduk N. Classi®cation of chronic pain: description of
chronic pain syndromes and de®nitions of pain terms. Seattle, WA:
IASP Press, 1994.
Miller DB. Comparisons of pain ratings from postoperative children, their
mothers, and their nurses. Pediatr Nurs 1996;22:145±149.
O'Hara M, McGrath PJ, D'Astous J, Vair CA. Oral morphine versus
injected meperidine (Demerol) for pain relief in children after ortho-
paedic surgery. J Pediatr Orthop 1987;7:78±82.
Pothmann R. Comparison of the visual analogue scale (VAS) and a smiley
analogue scale (SAS) for the evaluation of pain in children. In: Tyler
DC, Krane EJ, editors. Advances in Pain Research and Therapy, 15.
New York: Raven Press, 1990. pp. 95±99.
Smith G, Covino BG. Acute pain. London: Butterworth Press, 1985.
Tyler DC, Tu A, Douthit J, Chapman CR. Toward validation of pain
measurement tools for children: a pilot study. Pain 1993;52:301±309.
Unruh AM. Gender variations in clinical pain experience. Pain
1996;65:123±167.
West N, Oakes L, Hinds PS, Sanders L, Holden R, Williams S, Fairclough
D, Bozeman P. Measuring pain in pediatric oncology ICU patients. J
Pediatr Oncol Nurs 1994;11:64±68.
Wong DL, Baker CM. Pain in children: comparison of assessment scales.
Pediatr Nurs 1988;14:9±17.
C.T. Chambers et al. / Pain 83 (1999) 25±35 35