Analysis of the validation of existing behavioral pain and distress scales for use in the procedural setting

Review article

Analysis of the validation of existing behavioralpain and distress scales for use in the proceduralsetting

DIANNE CRELLIN R N B N M N N PR N B N M N N P*‡, THOMAS P. SULLIVAN

B .B . MMeded SS c ic i†, FRANZ E. BABL M D M P H F R A C P F A A PM D M P H F R A C P F A A P*†‡,

RONAN O’SULLIVAN M D F R C S I F C E RM D F R C S I F C E R* AND ADRIAN

HUTCHINSON R N B NR N B N*

*Department of Emergency, Royal Children’s Hospital, †Melbourne University and ‡MurdochChildren’s Research Institute, Melbourne, Victoria, Australia

SummaryBackground: Assessing procedural pain and distress in young children

is difficult. A number of behavior-based pain and distress scales exist

which can be used in preverbal and early-verbal children, and these

are validated in particular settings and to variable degrees.

Methods: We identified validated preverbal and early-verbal behavioral

pain and distress scales and critically analysed the validation and

reliability testing of these scales as well as their use in procedural pain

and distress research. We analysed in detail six behavioral pain and

distress scales: Children’s Hospital of Eastern Ontario Pain Scale

(CHEOPS), Faces Legs Activity Cry Consolability Pain Scale (FLACC),

Toddler Preschooler Postoperative Pain Scale (TPPPS), Preverbal Early

Verbal Pediatric Pain Scale (PEPPS), the observer Visual Analog Scale

(VASobs) and the Observation Scale of Behavioral Distress (OSBD).

Results: Despite their use in procedural pain studies none of the

behavioral pain scales reviewed had been adequately validated in the

procedural setting and validation of the single distress scale was limited.

Conclusions: There is a need to validate behavioral pain and distress

scales for procedural use in preverbal or early-verbal children.

Keywords: pain; validation studies; pain measurement; procedure;

child

Introduction

Children presenting to emergency departments (ED)

frequently undergo diagnostic and therapeutic pro-

cedures during their admission. Tools to accurately

assess the level of pain and distress experienced are

essential to improve the management of children

undergoing painful procedures in clinical practice

and through research. It is widely accepted that

assessing pain and distress in preverbal and early-

verbal children in particular is inherently difficult.

They are developmentally unable to self-report pain

Correspondence to: Franz E Babl, Clinical Associate Professor, TheUniversity of Melbourne Consultant Emergency Paediatrician,ED, Royal Children’s Hospital, Parkville, Victoria 3052, Australia(email: [email protected]).

Pediatric Anesthesia 2007 17: 720–733 doi:10.1111/j.1460-9592.2007.02218.x

� 2007 The Authors720 Journal compilation � 2007 Blackwell Publishing Ltd

and distress and assessment is primarily based on

observation of behavior. A number of pain and

distress scales exist which have been used in

preverbal and early-verbal children and these are

validated in particular settings and to variable

degrees.

Recommendations and clinical practice guidelines

detailing the importance of minimizing pain and

distress associated with diagnostic and therapeutic

procedures abound in the emergency literature

(1–4). In line with this focus on procedural pain,

increasing numbers of clinical studies are conducted

in EDs aimed at evaluating the value of interven-

tions designed to minimize the level of pain and

distress experienced during painful procedures

(5–24).

Despite the attention procedural pain has received

in the pain and emergency literature, pain assess-

ment in pre- and early-verbal children is largely

dependent on observational tools developed for use

in nonprocedural settings (e.g., postoperative pain).

The aim of a recent systematic review of observa-

tional behavior measures of pain for children and

adolescents aged 3–18 years by von Baeyer and

Spagrud (25) was to identify scales that could be

recommended for use in clinical trials for a variety of

clinical settings. For ‘medical procedures and brief

painful events’ they recommend Faces Legs Activity

Cry Consolability Pain Scale (FLACC) and Chil-

dren’s Hospital of Eastern Ontario Pain Scale (CHE-

OPS) for pain assessment in children aged 3–

18 years. We set out to determine whether scales

frequently used in clinical trials and those recom-

mended in the literature (e.g., the scales recommen-

ded by von Baeyer and Spagrud) have indeed been

validated for the assessment of procedural pain in

pre- and early-verbal children. Our aim was to make

recommendations about which scales may be used

to measure procedural pain in pre- and early-verbal

children for clinical research in the ED.

A review of the existing literature was conducted

to identify validated preverbal and early-verbal

behavioral pain and distress scales. We selected

scales that had been either validated in the proce-

dural setting as pain or distress tools, or had been

validated in the nonprocedural setting and either

used in published research or could be potentially

used to assess procedural pain and distress in the

preverbal and early-verbal population. A critical

analysis of the validity and reliability testing of these

scales as well as their use in procedural pain and

distress research was then conducted.

Elements of validity and reliability

The clinical and research utility of a pain and

distress scale depends primarily on the adequacy

of its validity and reliability testing (26). An exam-

ination of the validity and reliability of an instru-

ment requires an understanding of the various

components of each and the methods by which they

may be tested.

An instrument is valid if it measures the construct

it intends to measure (27,28). The various aspects of

validity commonly assessed include content, criter-

ion, and construct validity.

Content validity reflects the extent to which the

instrument is representative or inclusive of all

features of the construct. Reviewing the literature

and canvassing experts informs the development of

the instrument, which is usually assessed by expert

opinion to establish content validity (26,28,29). Con-

tent validity is important for the development of an

instrument, but does not provide acceptable evi-

dence of validity beyond the development phase of

the instrument (27).

Criterion validity indicates the extent that a

subject’s performance on the instrument is related

to actual behavior (28). Concurrent validity is a

component of criterion validity (28) and describes

the level of correlation between two instruments

considered to measure the same construct when

applied at the same time. High correlation indicates

a high level of agreement between the two measures

and therefore increased likelihood that they are a

valid measure of the construct (28). In pain mea-

surement, this standard is usually a self-report.

Observer ratings, on a Visual Analog Scale (VAS),

have often been used as a proxy for self-report in

preverbal and early-verbal children (30).

Construct validity provides the strongest evi-

dence for validity (29) and is determined by the

extent to which the measurement reflects the

actual construct (28). Construct validity will exam-

ine whether the tool detects theoretically predict-

able changes in the construct, e.g., whether pain

scores change with administration of analgesia

(26,28). This assessment of validity may also be

BEHAVIORAL PAIN AND DISTRESS SCALES 721

� 2007 The AuthorsJournal compilation � 2007 Blackwell Publishing Ltd, Pediatric Anesthesia, 17, 720–733

established by comparison with a previously vali-

dated instrument (28,29). Finally, establishing

construct validity requires demonstration of the

capacity of the instrument to discriminate between

the construct and other similar concepts, i.e., a

pain scale should have the capacity to discriminate

between pain and fear. Establishing construct

validity may require several studies using a

number of different approaches.

Reliability is a measure of the consistency of a

tool. There are three major components of reli-

ability – intrarater reliability, interrater reliability

and internal consistency. Intrarater reliability

examines the stability of the instrument across time

(27,28). Interrater reliability is a measure of equiv-

alence and assesses the agreement among users of

the tool (27,28). Internal consistency assesses the

extent that the components of the tool are measuring

the same construct (27,28). Beyer and Wells (26)

suggest that interrater reliability is important when

observation is the method of data collection, and

internal consistency is important when several items

are summed to provide a global score.

The difficulty in establishing the validity and

reliability of a scale means that a single study is

insufficient to establish either validity or reliability

(26). Rather, multiple studies by researchers other

than the original authors of a scale are necessary

before a scale is deemed adequate for either research

or clinical use (26). Further, the assessment of

validity and reliability of a scale should be conduc-

ted prior to using the scale in either of these settings

(26).

Methods

Search strategy and scale selection

A search of the literature was undertaken to identify

published work relevant to the aims of this review,

which were to identify valid behavioral scales for

assessment of procedural pain in pre- and early-

verbal children. An online search using MEDLINE

(1966–December 2006) was conducted for seminal

works detailing scale development and validation

studies using the search terms ‘pain assessment’,

‘infant’ and ‘child’ and ‘pediatrics’.

A second online search was conducted using

MEDLINE (January 1996–December 2006) to identify

the clinical studies evaluating pain management

strategies for painful procedures conducted in the

ED and current clinical practice guidelines devel-

oped for ED pain management. The terms ‘emer-

gency services,’ ‘pain assessment,’ ‘infant,’ ‘child’

and ‘pediatrics’ were used for this search. Finally, a

MEDLINE (January 1966–December 2006) search

was conducted using the name of the selected scales

as the search term.

Searches were supplemented by reviewing the

reference lists of relevant papers identified in the

MEDLINE searches. Websites for relevant profes-

sional bodies, in particular, Initiative on Methods,

Measurements and Pain Assessment in Clinical

Trials (IMMPACT http://www.impact.org) and

the International Association for the Study of Pain

(IASP http://www.iasp-pain.org) were also

reviewed for material that may have contributed to

this review.

Studies were considered relevant if they

addressed validation of a pain and distress scale

for use in pre- or early-verbal children (6 months–

4 years) or use of a pain and distress scale in

procedural pain in children of any age.

Review of the published material sourced identi-

fied a number of scales which had been validated in

previous studies and ⁄ or were used as outcome

measures in clinical trials.

Scales recommended in the emergency literature,

which had not been validated for pre- and early-

verbal children or for procedural use were excluded

unless they had also been used extensively in clinical

trials in the procedural setting. Additionally, scales

that had been designed and validated for neo-

nates and sedated patients were excluded from

consideration.

A number of scales identified in the literature

deemed similar to other scales were excluded. The

Procedure Behavioral Rating Scale (PBRS) (31) has

given rise to several derivatives, i.e., the Procedural

Behavior Checklist (PBC) (32) and the Observation

Scale for Behavioral Distress (OSBD) (33). The

OSBD has been used and validated most exten-

sively in the literature. Therefore, the similar PBRS

and PBC were excluded from the review. The

Modified Behavioral Pain Scale (34) a derivative of

CHEOPS was also rejected as the original scale

appears more frequently in validation studies and

clinical trials.

722 D. CRELLIN ET AL.


Three scales arising from studies attempting to

validate scales for use in the ED were initially

considered but closer scrutiny resulted in their

exclusion. Bulloch and Tenenbein (35) conducted a

study to validate two pain scales for use in the ED.

However, the scales used in this study ‘Colour

Analog Scale’ and a ‘7-point Faces Pain Scale’ are

both self report scales for children aged 5–16 years

of age and focused on acute pain assessment

associated with the presenting complaint, therefore

adding little to the search for a tool to assess

procedural pain in pre- and early-verbal children.

Another recent study conducted in the ED attemp-

ted to validate the Alder Hey Triage Pain Scale in

similar circumstances for children too young to self-

report (36). However, once again this scale was not

validated for procedural use and has not been used

for clinical studies in procedural pain.

The scales critically analysed (Table 1) were the

Children’s Hospital of Eastern Ontario Pain Scale

(CHEOPS) (37), Faces Legs Activity Cry Consolabi-

lity (FLACC) (38), Toddler Preschooler Postopera-

tive Pain Scale (TPPPS) (39), Preverbal, Early Verbal

Pediatric Pain Scale (PEPPS), Observer Visual Ana-

log Scales (VASobs) (40,41), and the Observation

Scale of Behavioral Distress (OSBD) (33).

Results

Analysis of selected behavioral pain anddistress scales

Children’s Hospital of Eastern Ontario Pain Scale. The

Children’s Hospital of Eastern Ontario Pain Scale

(CHEOPS) is an observational pain scale consisting

of six behavioral components that provide a global

score ranging from four to thirteen (Table 1) (37).

The scale was developed following consultation

with experienced postanesthesia care unit (PACU)

nurses and underwent piloting before further refine-

ment. CHEOPS is scored in time periods, each

consisting of 5 s to observe the child’s behavior

followed by 25 s to record the score.

The initial validation study was conducted by

McGrath et al. (1985) (37) postoperatively and con-

sisted of four components. CHEOPS was shown to

be a reliable scale, with average item correlation

between two nurses ranging from 0.9 to 0.99.

CHEOPS showed good internal correlation, with

good item-to-total-score correlations and moderate

to good item-to-item correlation.

Concurrent validity was demonstrated by com-

paring simultaneous CHEOPS scores, rated by a

researcher, with observer VAS pain scores, rated

independently by two PACU nurses. However, the

scale was a culmination of the behavioral markers

used by PACU nurses to indicate pain, which are

likely to be similar to those used to inform a global

assessment of pain (VASobs) resulting in predictably

high correlation between the two scores. The authors

attempted to demonstrate construct validity by

showing a significant drop in CHEOPS scores

following analgesia administration. However, the

nurse applying CHEOPS was the treating nurse and

so not blinded to the analgesia, which weakens

support for the establishment of construct validity.

McGrath and colleagues also undertook ‘social

validation,’ a method of validity testing the authors

acknowledge as unproven in pain assessment.

Eighty-eight teachers viewed 14 five-second seg-

ments of footage of children in PACU, and then

independently rated the child’s pain using a VAS.

The teacher scores correlated well with CHEOPS.

However, this method is unproven and cannot be

considered a determinant of validity.

The postoperative setting was used in a second

study aimed at validating CHEOPS for pain assess-

ment (42). CHEOPS and the Objective Pain Scale

(OPS) were applied to the youngest age group

(6 months to 3 years) and compared with establish

concurrent validity. The Faces Scale was added for

older children (3–6.5 years) and CHEOPS, the Faces

Scale and self report VAS were used for comparison

in the oldest age group (6.5–12 years). Pearson’s

correlation demonstrated high levels of correlation

between the various scales (r = 0.743–0.921).

Construct validity was established by analysing

the trend in pain scores collected prior to surgery,

before and after analgesia postoperatively and

scores collected once oral analgesia was commenced.

Significant trends existed for all scales.

The authors’ of this methodologically sound study

conclude that all scales can be considered valid for

postoperative pain assessment in children aged

6 months to 12 years.

In a more recent effort, Suraseranivongse et al. (29)

conducted a cross validation study to evaluate

CHEOPS, FLACC, TPPS, and OPS in 167 children



Tab

le1

Val

idat

ion

of

exis

tin

gb

ehav

iora

lp

ain

and

dis

tres

ssc

ales

Sca

leE

lem

ents

and

scor

eV

alid

atio

nst

udi

esS

etti

ng

and

age

ran

geV

alid

ity

and

reli

abil

ity

asse

ssed

CH

EO

PS

Cry

(1,2

or

3)F

acia

l(1

or

2)C

hil

dV

erb

al(0

,1o

r2)

To

rso

(1o

r2)

To

uch

(1o

r2)

Leg

s(1

or

2)S

core

ran

ge:

4–13

McG

rath

etal

.(3

7)P

ost

op

erat

ive

(i)

27ch

ild

ren

aged

1–5

yea

rs(i

i)30

chil

dre

nag

ed1–

7y

ears

(iii

)20

chil

dre

nag

en

ot

rep

ort

ed

(i)

Inte

rrat

erre

liab

ilit

y(i

i)In

tern

alco

rrel

atio

n,

crit

erio

nv

alid

ity

–co

ncu

rren

t(i

ii)

Co

nst

ruct

val

idit

y–

con

tras

ted

gro

up

sS

ura

sera

niv

on

gse

etal

.(2

9)*

Po

sto

per

ativ

e16

7ch

ild

ren

aged

1–5

yea

rsC

on

stru

ctv

alid

ity

Cri

teri

on

val

idit

y-

con

curr

ent

Co

nte

nt

val

idit

yIn

ter-

and

intr

arat

erre

liab

ilit

yT

yle

ret

al.

(42)

Po

sto

per

ativ

e43

chil

dre

nag

ed6

mo

nth

s–12

yea

rsC

on

stru

ctv

alid

ity

Cri

teri

on

val

idit

y-

con

curr

ent

FL

AC

CF

ace

(0,1

or

2)L

egs

(0,1

or

2)A

ctiv

ity

(0,1

or

2)C

ry(0

,1o

r2)

Co

nso

lab

ilit

y(0

,1o

r2)

Sco

rera

ng

e:0–

10

Mer

kel

etal

.(3

8)P

ost

op

erat

ive

89h

ealt

hy

chil

dre

nag

ed2

mo

nth

sto

7y

ears

(i)

30(i

i)29

(iii

)30

(i)

Inte

rrat

erre

liab

ilit

y(i

i)C

on

stru

ctv

alid

ity

(iii

)C

ross

val

idat

ion

Wil

lis

etal

.(4

5)P

ost

op

erat

ive

147

chil

dre

nag

ed3–

7y

ears

Cri

teri

on

val

idit

yIn

terr

ater

reli

abil

ity

Vo

epel

-Lew

iset

al.

(46)

Po

sto

per

ativ

e.79

chil

dre

nag

ed4–

18y

ears

wit

hC

og

nit

ive

Imp

airm

ent

Co

nst

ruct

val

idit

yC

rite

rio

nv

alid

ity

Inte

r-an

din

trar

ater

reli

abil

ity

Man

wo

rren

and

Hy

nan

(44)

Mu

ltip

leu

nit

s,al

lre

qu

irin

gan

alg

esia

for

pai

n.

147

pat

ien

tsag

edle

ssth

an3

yea

rs

Co

nst

ruct

val

idit

yC

on

ten

tv

alid

ity

Su

rase

ran

ivo

ng

seet

al.

(29)

*as

abo

ve

asab

ov

e

TP

PP

SV

oca

lex

pre

ssio

n(0

,1,2

or

3)F

acia

lex

pre

ssio

n(0

,1,2

or

3)B

od

ym

ov

emen

t(0

or

1)S

core

ran

ge:

0–7

Tar

bel

let

al.

(39)

Po

sto

per

ativ

e74

pat

ien

tsag

ed1–

5y

ears

Co

nst

ruct

val

idit

yC

rite

rio

nv

alid

ity

–co

ncu

rren

tIn

terr

ater

reli

abil

ity

Inte

rnal

corr

elat

ion

Har

tric

kan

dK

ov

an(4

8)P

ost

op

erat

ive

and

no

np

ain

ful

rad

iolo

gic

alp

roce

du

res

51ch

ild

ren

aged

1–5

yea

rs

Co

nst

ruct

val

idit

y-

Co

nv

erg

ent

and

dis

crim

inan

tIn

terr

ater

reli

abil

ity

McC

arth

yet

al.

(49)

Acu

tep

ain

100

chil

dre

nag

ed1–

5y

ears

Cri

teri

on

val

idit

y

Su

rase

ran

ivo

ng

seet

al.

(29)

*as

abo

ve

asab

ov

e



Ta

ble

1(C

onti

nu

ed)

Sca

leE

lem

ents

and

scor

eV

alid

atio

nst

udi

esS

etti

ng

and

age

ran

geV

alid

ity

and

reli

abil

ity

asse

ssed

PE

PP

SH

eart

Rat

e(0

,1,2

,3

or

4)F

acia

l(0

,1,2

,3

or

4)C

ry(0

,1,2

,3

or

4)C

on

sola

bil

ity

(0,1

,2,

3o

r4)

Bo

dy

po

stu

re(0

,1,2

,3

or

4)S

oci

abil

ity

(0,1

,2,

3o

r4)

Su

ckin

g⁄f

eed

ing

(0,1

,2,

3o

r4)

Sco

rera

ng

e:0–

28

Sch

ult

zet

al.

(15)

Po

sto

per

ativ

e40

chil

dre

n,

aged

12–2

4m

on

ths.

Co

nst

ruct

val

idit

yIn

terr

ater

reli

abil

ity

Intr

arat

erre

liab

ilit

yIn

tern

alco

nsi

sten

cy

VA

S(o

bs)

100

mm

ho

rizo

nta

lli

ne

–m

ark

edat

each

end

wit

h‘n

op

ain

’an

d‘m

ost

pai

n’.

Ob

serv

erm

ark

so

nth

eli

ne

the

lev

elo

fp

erce

ived

pai

nex

per

ien

ced

by

the

chil

d

Sin

ger

etal

.(5

9)P

roce

du

ral

pai

n1,

171

adu

ltp

roce

du

res

Sin

ger

etal

.(4

1)A

cute

pai

nan

dp

roce

du

ral

pai

n63

chil

dre

nag

ed4–

7y

ears

Co

nst

ruct

val

idit

y–

Co

nv

erg

ent

Cri

teri

on

val

idit

yK

elly

etal

.(4

0)A

cute

pai

n78

chil

dre

nag

ed8–

15y

ears

OS

BD

Sco

red

for

pre

sen

ce(0

for

abse

nce

).B

rack

etin

dic

ates

wei

gh

tin

gIn

form

atio

nse

ekin

g(1

.5)

Cry

(1.5

)S

crea

m(4

.0)

Ph

ysi

cal

rest

rain

t(4

.0)

Ver

bal

resi

stan

ce(2

.5)

See

ks

emo

tio

nal

sup

po

rt(2

.0)

Ver

bal

pai

n(2

.5)

Fla

il(4

.0)

Ver

bal

fear

(2.5

)**

Mu

scu

lar

rig

idit

y(2

.5)*

*N

erv

ou

sb

ehav

ior

(1.0

)**

Sco

rera

ng

e:0–

28

Jay

etal

.(3

3)P

roce

du

ral

(BM

A)

42ca

nce

rp

atie

nts

aged

2–20

yea

rsC

rite

rio

nv

alid

ity

Inte

rrat

erre

liab

ilit

y

Ell

iott

etal

.(6

5)P

roce

du

ral

(BM

A)

55ch

ild

ren

aged

3–13

yea

rsC

on

ver

gen

tv

alid

ity

Inte

rrat

erre

liab

ilit

y

CH

EO

PS

,C

hil

dre

n’s

Ho

spit

alo

fE

aste

rnO

nta

rio

Pai

nS

cale

;F

LA

CC

,F

aces

Leg

sA

ctiv

ity

Cry

Co

nso

lab

ilit

yP

ain

Sca

le;

TP

PP

S,

To

dd

ler

Pre

sch

oo

ler

Po

sto

per

ativ

eP

ain

Sca

le;

PE

PP

S,

Pre

ver

bal

,E

arly

Ver

bal

Ped

iatr

icP

ain

Sca

le;

VA

So

bs,

Vis

ual

An

alo

gS

cale

(ob

serv

er);

OS

BD

,O

bse

rvat

ion

Sca

leo

fB

ehav

iora

lD

istr

ess;

BM

A,

Bo

ne

mar

row

asp

irat

e.*N

ote

:A

cro

ssv

alid

atio

nst

ud

yex

amin

ing

CH

EO

PS

,F

LA

CC

,T

PP

PS

and

Ob

ject

ive

Pai

nsc

ale

(OP

S).



aged 1–5 years. Children were video taped prior to

surgery and following surgery before analgesia was

given. Four observers viewed the video segments

and pain scores were recorded using each of the four

scales.

Comparison of presurgery scores with those

obtained afterwards, when children may reason-

ably expect to be experiencing pain, demonstrated

that all scales had the capacity to discriminate

(P = 0.001) between varying levels of pain. Con-

current validity was established by comparing the

scores achieved with the scales with each other

and moderate to good correlation was found

(r = 0.62–0.77). The highest level of agreement

between scales and the clinical decision to treat

pain was achieved with CHEOPS. Inter- and

intrarater reliability was also tested and found to

be high for all scales.

The authors recommend CHEOPS for pain assess-

ment in the immediate postoperative phase for

children aged 1–5 years. This decision appeared to

be based on the high level of agreement with the

clinical decision to treat as little else discriminated

the validity of the scales from one another.

In a clinical trial (20) evaluating the analgesic

effectiveness of topical local anaesthetic and nitrous

oxide (N2O) or use of the combination for venous

cannulation, the results showed that CHEOPS scores

may be moderated by the use of N2O. CHEOPS

scores measured before and during cannulation were

significantly increased during cannulation for those

treated with N2O but to a significantly lesser extent

than those treated with topical local anesthetic.

However, the self-reported pain scores for both

treatment groups were not significantly different.

These results introduce doubt about the ability of

CHEOPS to adequately reflect the pain experience.

Furthermore, Beyer et al. (43) found that CHEOPS

scores were generally low after discharge from

PACU and at that over time correlated increasingly

poorly with self-reports of pain. Thus, CHEOPS may

be valid only for the immediate postsurgery period.

Willis et al. (45) suggest CHEOPS as the gold

standard preverbal pain tool for comparative pur-

poses. However, as a consequence of the tight

observational and recording intervals and many

types of behavior to evaluate it is burdensome to

apply. Tarbell et al. (39) also note that the strong

correlation between CHEOPS and observer VAS

measures of pain may mean it is more practical to

use a VAS.

A number of studies have provided varying levels

of support for the validity of CHEOPS for the

assessment of pain in postoperative children. How-

ever, CHEOPS has not been validated in the acute

procedural setting, although a number of clinical

trials, including many conducted in the ED have

used CHEOPS to assess procedural pain (20–24).

Von Baeyer and Spagrud (25) report ‘excellent’

evidence of validity and recommend this scale’s

use in medical procedures. However, it is unclear

how validity for the use of CHEOPS to assess pain

during medical procedures was established.

Face Legs Activity Cry Consolability Pain Scale. The

Faces Legs Activity Cry Consolability (FLACC) Pain

Scale is an observational pain scale consisting of five

behavioral components that provide a global pain

score ranging from 0 to 10 (Table 1) (38,44–46).

FLACC was designed for use after a period of up to

5 min of observation. The authors suggest observing

the child with their legs uncovered, testing tone and

tenseness, and initiating consoling of the child when

applying FLACC (47).

The initial validation study of FLACC by Merkel

and colleagues (38), included 89 children aged

2 months to 7 years, was conducted postoperatively

and consisted of three parts.

Interrater reliability was reported as good, and

was assessed by correlating FLACC scores of two

nurses, who simultaneously and independently

applied FLACC to 30 children.

Construct validity was assessed by applying

FLACC to 29 children before and after analgesia

administration, and showed a significant decrease in

FLACC score following analgesia. However, the

treating nurse also applied FLACC and so was not

blinded to analgesia administered.

In part three, 30 children were simultaneously

rated by two independent nurses, one using a pain

scale identified by the authors as previously valid-

ated and the other using FLACC. The two scales

demonstrated good correlation and thus concurrent

validity.

The study supported the preliminary validity and

reliability of FLACC, although the sample size was

small and all children were English speaking with-

out cognitive impairment.



Willis et al. (45) conducted a further validation

study on 30 children aged 3–7 years, at least 18 and a

mean of 20 h after surgery. The same nurse scored

all 30 children after a period of observation. These

scores were compared with the child’s report of pain

using a Faces pain scale. To establish reliability, a

second nurse independently applied FLACC to 6 of

the 30 children.

Overall, FLACC and the Faces self-report scale

showed good agreement – supporting concurrent

validity – and interrater reliability was 100%. How-

ever, there was no agreement between the self-report

and FLACC scores of children younger than 5 years

of age, which suggests that either the younger

children failed to comprehend the Faces Scale or

that FLACC is not valid in this population.

The study was undertaken on a small sample size,

limited to English speaking children without cogni-

tive impairment, and did not assess any other form

of validity.

Voepel-Lewis et al. (46) examined the validity and

reliability of FLACC for postoperative use in chil-

dren with cognitive impairment. They enrolled 79

children, aged 4–18 years, undergoing general or

orthopedic surgery. Postsurgery but prior to anal-

gesia administration, each child was filmed for later

review. The treating nurse and a parent simulta-

neously rated the child’s pain using FLACC and a

VAS respectively. Those children who received

analgesia in PACU were filmed and scored a second

time. Two nurses, blinded to analgesia, independ-

ently viewed the videos in random order and

applied FLACC. Fifty were viewed for a second

time 2 months later to establish intrarater reliability.

The study found by deos FLACC had good inter-

and intrarater reliability, demonstrated construct

validity with a decrease in FLACC following anal-

gesia and at concurrent validity was supported by a

strong correlation between FLACC and parent VAS

scores.

All children were tested for the ability to self-

report, and although 16 children were deemed

capable of providing a self-report of pain, only eight

did. Thus, FLACC scores were unable to be reliably

compared with self-report.

Manworren and Hynan (44) tested FLACC in a

variety of settings, including PACU; the pediatric

intensive care unit; surgery ⁄ trauma unit; hemato-

logy ⁄ oncology unit; and the infant unit. One hundred

and forty-seven children aged less than 3 years who

had analgesia ordered on an ‘as required’ basis were

included in the study. A pilot study was conducted,

and 25 volunteer nurses assessed ten video segments

to establish interrater reliability. Nineteen nurses

showed sufficient reliability; fourteen collected data.

Children deemed to be in pain by their treating

nurse were then rated, at the bedside, by one of the

research nurses before and after analgesic adminis-

tration. Construct validity was demonstrated by a

fall in FLACC score after analgesic administration.

The nurses were not blinded to the analgesia. This

study lends preliminary support to the use of

FLACC in a variety of nonprocedural settings.

FLACC has not been validated in the acute

procedural setting, however like CHEOPS, FLACC

appears in the literature as the outcome measure in

clinical trials examining interventions to minimise

procedural pain in young children (18,19). Further-

more, it is also recommended by von Baeyer and

Spagrud (25) for use in pain associated with ‘proce-

dures and other brief painful events.’ Once again the

basis on which they established the validity of this

scale in this setting is unclear.

Toddler Preschooler Postoperative Pain Scale. The

Toddler Preschooler Postoperative Pain Scale

(TPPPS) is an observational pain scale consisting of

three categories that provide a global score ranging

from zero to seven (Table 1) (39). The scale compo-

nents were derived from preliminary studies and

observed pain behavior in other studies (39).

The initial validation of TPPPS was conducted by

Tarbell et al. (39) on 74 children, aged 1–5 years,

emerging from anesthesia following surgical repair

of a hernia or a hydrocoele. Children were

observed for six 5 min intervals following arousal

from anesthesia and the treating nurse rated the

child’s pain using TPPPS. A second independent

nurse observed 28 of the children to assess reliab-

ility and interrater reliability was reported as good

across both the scale as a whole and its individual

items.

The authors assessed internal consistency by

comparing the six interval scores for each patient,

and by aggregating the six scores and comparing the

total between patients. High item-to-total-score cor-

relations were observed in both the individual and

aggregated TPPPS scores, however, the correlation



between the individual items varied from poor to

good. This may reflect the different degrees of ‘pain’

that the various items are measuring.

Construct validity was demonstrated by a signi-

ficant fall in TPPPS scores following postoperative

analgesia, observed in the 25 children who received

further analgesia.

The study demonstrates preliminary validity and

reliability of TPPPS, but has a number of weaknes-

ses. The population consisted of predominately

Caucasian boys and the nurses were not blinded to

analgesic administration.

The authors note that although TPPPS was found

to be sensitive to analgesic regimens, they are

uncertain whether this sensitivity was because of

decreased pain experienced by the child or as a

consequence of the sedative action of some analge-

sics.

Hartrick and Kovan (48) conducted a further

validation study on 51 children aged 1–5 years.

The study was conducted in three parts and two

nurses observed each patient simultaneously and

independently at 5 min intervals.

In part one, twenty patients were observed in

PACU following ear, nose and throat (ENT) surgery

until discharge to the ward. Part one demonstrated

good interrater reliability and construct validity,

with a significant fall in TPPPS score following

analgesia administration.

Part two compared the scores obtained in part one

with the scores from two other groups: eight

children who had undergone urological surgery

and were treated for pain with caudal anesthesia;

and 11 patients emerging from anesthesia following

nonpainful radiological procedures. No statistical

difference in pain scores between the radiological

and urological groups was observed. However,

patients undergoing ENT surgery had significantly

higher TPPPS scores than the radiological group,

demonstrating discriminate validity.

Part three was a cross-validation study of TPPPS,

FLACC and a third pain scale, undertaken on 12

children emerging from anesthesia. Both of the

observing nurses applied all three scales at each

observation interval. All three scales showed a

response to analgesia, although the authors con-

clude that TPPPS may be better than FLACC for

discriminating between painful and nonpainful

situations.

The study provides further support for the valid-

ity of TPPPS, however, a small sample size and lack

of blinding limit the ability to generalize the study’s

findings. The authors also query whether TPPPS is a

measure of pain or generalized distress.

McCarthy et al. (49) conducted an exploratory

study using a modified version of TPPPS to assess

the pain of 100 children, aged 1–5 years, presenting

to the ED. This study was not an attempt to examine

its use for procedural pain. The child’s pain was

assessed by a researcher using the modified TPPPS

and by parents and nursing staff using a VAS.

TPPPS scores correlated significantly with both the

nurse and parent VAS scores, and the authors

conclude that TPPPS may be appropriate to assess

pain in the ED, but that more research is required.

This is one of only two attempts to validate a pain

assessment tool for use in the ED.

Despite one attempt to validate TPPPS in the ED,

there have been no further attempts to validate the

scale in this setting and no attempts to validate it for

the acute procedural setting. However, TPPPS has

been used in clinical research as an outcome meas-

ure in nonprocedural settings in several trials

(16,17).

Preverbal, Early Verbal Pediatric Pain Scale. The Pre-

verbal, Early Verbal Pediatric Pain Scale (PEPPS) is

an observational pain scale consisting of seven

behavioral categories that provide a global score

ranging from 0 to 26 (Table 1) (15). PEPPS was

developed in consultation with experienced pediat-

ric nurses following a review of the literature and

observation of children’s pain behavior (15).

The preliminary validity of PEPPS was assessed

by Schultz et al. (15). Forty children, aged 12 months

to 24 months, were filmed postoperatively following

abdominal or urological surgery. The footage was

then split into 3 min segments and two to four

segments of footage from each child were collated,

totalling 120 segments. These segments of footage

were randomly ordered and then presented inde-

pendently to four pediatric research nurses, all

blinded to analgesia administration, who applied

PEPPS.

Interrater and intrarater reliability were reported

as very good, while item-to-item correlations varied

from poor to good, and item-to-total score correla-

tions from moderate to good. However, the authors



suggest that eliminating those items with poorer

item-to-total score correlations would only slightly

improve internal consistency. Construct validity was

demonstrated by significant differences in pre- and

postanalgesia PEPPS scores.

The main limitations of the study are the small

sample size, male dominated population (35 patients

were boys) and limited age-range.

The Preverbal, Early Verbal Pediatric Pain Scale

has not been validated in the acute procedural

setting. However, PEPPS has been used as an

outcome measure in procedural pain research con-

ducted in the ED (15).

Observer Visual Analog Scale. Visual Analog Scales

(VAS) have been validated and repeatedly used to

quantify self reports of pain intensity in adults and

children as young as five years (50–52). Younger

children (4 years) have been shown to be capable of

reliably self-reporting pain using modified self

report scales, which often use a series of faces or

other visual representations of pain intensity (33,53–

58). However, for pre- and early-verbal children,

unable to self report, observer VAS (VASobs)

appears in the literature as a proxy for self-report

(5,11–14).

Several attempts to validate the use of VASobs for

pain assessment have been made. Singer et al. (59)

conducted a study comparing patient self report

using a 100 mm VAS with independent reports

provided by clinicians using the same tool for adults

undergoing painful procedures in the ED. Patient

and clinician scores for 1171 procedures performed

during the study period were evaluated. The mean

VAS for patients was 20.8 mm (95% CI: 19.4–

22.2 mm) and for clinicians 23.5 mm (95% CI:

22.3–24.6 mm). However, the trend for higher scores

for clinicians was not consistent across different

types of procedures and the correlation (Pearson’s

correlation coefficient) between patient and clinician

VASobs varied from r = 0.26–0.68.

Singer et al. in a more recent study focused on

children aged 4–7 years presenting to the ED with

acute pain or when undergoing a painful procedure

and compared self- report of pain with the ratings

provided by clinicians and their parent (41). The

children in this study used the Smiley Analog Scale,

which was validated for use in a study by Pothman

(60). The self-report scores of 57 children were

compared with the scores recorded independently

by their parent and a clinician using a 100 mm VAS.

The correlation between the child’s and their par-

ent’s scores was 0.47 (P < 0.001) and the correlation

between the child’s and the clinician’s scores was

0.08 (P = 0.54).

In a similar Study these results were reproduced

for children presenting to the ED when the child’s

self-report is compared with the parents rating of the

child’s pain (40). Children aged 5–16 years requiring

treatment for painful conditions and their parents

independently rated their pain using the 100 mm

VAS on 4 occasions during their admission to the

ED. A total of 78 children and their parents were

enrolled in the study and provided 289 score sets.

Correlation between the child and parent scores was

0.63 (95% CI: 0.56–0.70).

A study exploring the pain associated with naso-

gastric tube insertion enrolled a total of 36 children

aged 2–18 years (61). Self-report pain scores, parent

and nurse VASobs were provided for 29 participants

and show similar trends to previous studies reveal-

ing that parents and clinicians regularly under-

report the pain experience of children. However,

formal validity testing was not undertaken in this

study.

A range of studies demonstrate similar discord-

ance between self-report and observer reports of

pain in other clinical settings, such as in postoper-

ative pain (62–64). These studies indicate that con-

current validity cannot be established for VASobs as

clinicians are poor judges of a child’s pain and their

parents are only slightly better. However, the scale is

frequently used in studies attempting to validate a

newly developed pain assessment tool (30,34,37,39).

Researchers would argue where strong correlation

exists between scores that this indirectly supports

the concurrent validity of the VASobs. However,

with poor correlation between self-report (gold

standard) and observer reports of pain the validity

of this approach is weak. Furthermore, VASobs is

used repeatedly in clinical trials evaluating the

efficacy of interventions aimed at reducing pain

associated with medical procedures, including stud-

ies conducted in the ED (5,11–14).

Observation Scale of Behavioral Distress. The Observa-

tion Scale of Behavioral Distress (OSBD) is a revised

version of the Procedural Behavior Rating Scale (31)



and consists of 11 types of behavior indicative of

anxiety or distress in children (Table 1) (33). The

OSBD is scored in 15s intervals, with five seconds to

observe the child followed by 10 s to record the

OSBD score. The total score for a period of time is

then averaged. In addition, each indicator is weigh-

ted from one to four according to its intensity,

providing a global score from 0 to 28 (33).

The initial validation of OSBD was conducted by

Jay et al. (33) on 42 cancer patients aged 2–20 years

undergoing bone marrow aspiration (BMA).

Prior to each procedure the parents of all

children and children aged 8 years and older were

asked to complete a number of questionnaires and

measurements, identified by the authors as valid,

to assess child and parent anxiety levels and

coping skills. Included in these questionnaires and

measurements was a ‘Pain thermometer,’ used by

the child prior to the procedure to rate their

anticipated pain, and after the procedure to rate

their experienced pain.

Each procedure was observed by one of the two

researchers who applied the OSBD every 15 s. For

the purpose of analysis, the procedure was divided

into four separate phases.

Seven of the procedures were reviewed by both

researchers to assess interrater reliability. Inter- and

intrarater reliability were reported as good across all

time points and in total.

Validity was assessed by correlating the various

questionnaires and pain thermometer ratings with

OSBD scores. The total OSBD scores correlated

significantly with child’s Trait Anxiety scores and

anticipated and experienced pain ratings. Good

correlation was also seen between the parent’s

ratings of their child’s anxiety, obtained prior to

the procedure, and procedural distress measured

using OSBD.

A number of other measures used showed no

correlation with OSBD. Further, children were asked

to rate their pain but not ‘distress’ or ‘anxiety.’

However, the conceptual difficulty in defining ‘dis-

tress’ or ‘anxiety’ to younger children may make use

of this scale burdensome.

In light of the different developmental levels of the

participants, with a small sample spread over a wide

age range, interpretation of the results may be difficult.

The authors report that children younger than 7 years

of age experience five times the level of distress that

children older than 7 years do. However, it is possible

that older children fail to show the behavioral cues

present in the OSBD, and so record lower distress

levels.

The same group of authors that developed OSBD

conducted a subsequent validation study in 55

children, aged 3–13 years, undergoing BMA (65).

Each procedure was observed by one of the two

researchers trained in the use of the OSBD. To

establish interrater reliability and assess agreement

between the raters, both of which were reported as

good, both researchers observed 11 of the procedures.

The authors assessed two versions of the OSBD.

The ‘original OSBD’ is the OSBD scored in 15 s

intervals with weighted indicators, as explained

above. The ‘recalculated OSBD’ is the OSBD scored

once in each phase, without item weighting, such

that the scale provides a global score ranging from

zero to 11.

Validity was assessed by correlating a range of

variables with both OSBD scores. The variables

measured included heart rate and blood pressure,

taken immediately before and after the procedure;

self-reports of anticipated and experienced pain,

using a Pain Thermometer; and nurse ratings of

child distress on a 5-point Likert Scale.

The authors cite nine of the twelve variables

measured as showing ‘significant correlations’ with

OSBD, thus providing evidence of validity. All

these variables have low P values, which indicate

that the statistics obtained are likely to be a true

reflection of the population studied. However, only

two of these nine variables (nurse ratings of distress

and child heart rate during the procedure) show

moderate to strong correlation with OSBD. Only

these two measures, which show statistically signi-

ficant results and good correlation with OSBD,

provide evidence for the validity of OSBD. The

remaining variables show poor or no correlation

with OSBD.

The authors also found that the original and

recalculated OSBD scales are ‘valid’ and have

similar reliability. They conducted further analysis

on the scale, eliminating three items, reducing the

scale to revised eight item distress scale. This revised

scale, the OSBD-R, appears widely in clinical trials

focusing on interventions aimed at reducing proce-

dural pain in children (6–10), despite the limitations

of the validity testing.



Discussion

We analysed six major behavioral pain and dis-

tress scales with a particular focus on their

validation and reliability. The intent was to iden-

tify a scale addressing procedural pain which is

suitable for use in clinical practice and research in

the ED. Observer VAS is used frequently in

clinical research evaluating the efficacy of inter-

ventions aimed at alleviating pain even though it

has repeatedly been shown to be an inadequate

measure of pain. It is a practical tool to use,

however, its continued use in clinical research

should be discouraged as it does not correlate well

with self-reported pain. Although TPPPS and

PEPPS demonstrate greater levels of validity than

observer VAS in nonprocedural settings, these

scales have not been validated for use in proce-

dural pain. Despite their use in clinical trials, we

cannot recommend their use for this purpose in

pre- and early-verbal children undergoing painful

procedures.

The OSBD has been subject to limited validation

in procedural pain and distress using children aged

3–18 years suffering with cancer. These children are

older than those for which we are seeking a valid

scale and may be expected to have experienced

multiple procedures. This further limits the capacity

for assuming validity in otherwise well pre- and

early-verbal children experiencing acute procedural

pain. Additional testing is warranted if we are to

recommend this scale for use in clinical trials in

procedural pain. Furthermore, as noted by von

Baeyer and Spagrud (25) this scale attempts to

assess distress and anxiety associated with medical

procedures. Scales focusing more specifically on

pain behavior may be more appropriate tools for

clinical trials evaluating the efficacy of pain man-

agement. Finally, this scale relies heavily on verbal

markers, which will be absent in preverbal children.

Two of the scales reviewed (CHEOPS and

FLACC) have been validated for use in settings

such as postoperative pain. Additionally, they have

been used extensively in clinical trials examining

procedural pain and von Baeyer and Spagrud (25) in

their recent systematic review recommend both as

valid measures of pain in the procedural setting.

While showing potential for this purpose, CHEOPS

and FLACC have not been validated for use in

pre- and early-verbal children undergoing painful

procedures. However, until additional validation

studies are conducted we suggest their continued

use in clinical trials. A more important recommen-

dation to arise from this review is the need for

validation studies for behavioral pain and distress

tools for pre- and early-verbal children undergoing

painful procedures.

Acknowledgment

We acknowledge grant support from the Victor

Smorgon Charitable Fund, Melbourne, Australia.

References1 American Academy of Pediatrics Committee on Psychosocial

Aspects of Child Family Health and American Pain SocietyTask Force on Pain in Infants Children Adolescents. Theassessment and management of acute pain in infants, children,and adolescents. Pediatrics 2001; 108: 793–797.

2 Kelly AM. A process approach to improving pain manage-ment in the emergency department: development andevaluation. [see comment]. J Accident Emerg Med 2000; 17:185–187.

3 Royal Australasian College of Physicians, Paediatric and ChildHealth Division. Guideline Statement: Management of Procedure-related Pain in Children and Adolescents. Sydney: Royal Austra-lasian College of Physicians, 2005.

4 Zempsky WT, Cravero JP. Relief of pain and anxiety in pedi-atric patients in emergency medical systems. Pediatrics 2004;114: 1348–1356.

5 Green DA, Linares MYR, Garcia Pena BM et al. Randomizedcomparison of pain perception during radial head subluxationreduction using supination-flexion or forced pronation. PediatrEmerg Care 2006; 22: 235–238.

6 Kleiber C, Craft-Rosenberg M, Harper DC. Parents as dis-traction coaches during I.V. insertion: a randomized study. JPain Symptom Manage 2001; 22: 851–861.

7 Kanagasundaram SA, Lane LJ, Cavalletto BP et al. Efficacy andsafety of nitrous oxide in alleviating pain and anxiety duringpainful procedures. Arch Dis Childhood 2001; 84: 492–495.

8 Luhmann JD, Kennedy RM, Porter FL et al. A randomizedclinical trial of continuous-flow nitrous oxide and midazolamfor sedation of young children during laceration repair. AnnEmerg Med 2001; 37: 20–27.

9 Stevenson MD, Bivins CM, O’Brien K et al. Child lifeintervention during angiocatheter insertion in the PediatricEmergency Department. Pediatr Emerg Care 2005; 21: 712–718.

10 Tobias JD. Weak analgesics and nonsteroidal anti-inflamma-tory agents in the management of children with acute pain.Pediatr Clinics North America 2000; 47: 527–543.

11 Lewindon PJ, Harkness L, Lewindon N. Randomised con-trolled trial of sucrose by mouth for the relief of infant cryingafter immunisation. Arch Dis Childhood 1998; 78: 453–456.

12 Lindh V, Wiklund U, Blomquist HK et al. EMLA cream andoral glucose for immunization pain in 3-month-old infants.Pain 2003; 104: 381–388.



13 Schultz AA, Strout TD, Jordan P et al. Safety, tolerability, andefficacy of iontophoresis with lidocaine for dermal anesthesiain ED pediatric patients. J Emerg Nurs 2002; 28: 289–296.

14 McErlean M, Bartfield JM, Karunakar TA et al. Midazolamsyrup as a premedication to reduce the discomfort associatedwith pediatric intravenous catheter insertion. J Pediatr 2003;142: 429–430.

15 Schultz AA, Murphy E, Morton J et al. Preverbal, Early VerbalPediatric Pain Scale (PEPPS): development and early psycho-metric testing. J Pediatr Nurs 1999; 14: 19–27.

16 Memi D, Turan A, Karamanliolu B et al. Caudal neostigminefor postoperative analgesia in paediatric surgery. PaediatrAnaesth 2003; 13: 324–328.

17 Prosser DP, Davis A, Booker PD et al. Caudal tramadol forpostoperative analgesia in pediatric hypospadias surgery. Br JAnaesth 1997; 79: 293–296.

18 Bar-Meir E, Zaslansky R, Regev E et al. Nitrous oxide admin-istered by the plastic surgeon for repair of facial lacerations inchildren in the emergency room. Plast Reconstr Surg 2006; 117:

1571–1575.19 Vaughan M, Paton EA, Bush A et al. Does lidocaine gel alle-

viate the pain of bladder catheterization in young children? Arandomized, controlled trial. Pediatrics 2005; 116: 917–920.

20 Hee HI, Goy RW, Ng AS. Effective reduction of anxiety andpain during venous cannulation in children: a comparison ofanalgesic efficacy conferred by nitrous oxide, EMLA andcombination. Paediatr Anaesth 2003; 13: 210–216.

21 Burton JH, Auble TE, Fuchs SM. Effectiveness of 50% nitrousoxide ⁄ 50% oxygen during laceration repair in children. AcadEmerg Med 1998; 5: 112–117.

22 Klein EJ, Diekema DS, Paris CA et al. A randomized, clinicaltrial of oral midazolam plus placebo versus oral midazolamplus oral transmucosal fentanyl for sedation during lacerationrepair. Pediatrics 2002; 109: 894–897.

23 Pierce MC, Fuchs S. Evaluation of ketorolac in children withforearm fractures. Acad Emerg Med 1997; 4: 22–26.

24 Schutzman SA, Liebelt E, Wisk M et al. Comparison of oraltransmucosal fentanyl citrate and intramuscular meperidine,promethazine, and chlorpromazine for conscious sedation ofchildren undergoing laceration repair. Ann Emerg Med 1996;28: 385–390.

25 von Baeyer CL, Spagrud LJ. Systematic review of observational(behavioral) measures of pain for children and adolescentsaged 3 to 18 years. Pain 2007; 127: 140–150.

26 Beyer JE, Wells N. The Assessment of Pain in Children. PediatrClin North America 1989; 36: 837–854.

27 Burns N, Grove SK. The Concepts of Measurement. In: BurnsN, Grove SK, eds. The Practice of Nursing Research, 3rd edn.Philadelphia: W.B. Saunders Company, 1997: 319–341.

28 LoBiondo-Wood G, Haber J. Reliability and Validity. In:LoBiondo-Wood G, Haber J, eds. Nursing Research: Methods,Clinical Appraisal and Utilization, 4th edn. St. Louis: Mosby,1998: 327–350.

29 Suraseranivongse S, Santawat U, Kraiprasit K et al.Cross-validation of a composite pain scale for preschoolchildren within 24 hours of surgery. Br J Anaesth 2001; 87: 400–405.

30 Lawrence J, Alcock D, McGrath P et al. The developmentof a tool to assess neonatal pain. Neonatal Network 1993; 12: 59–66.

31 Katz ER, Kellerman J, Siegel SE. Behavioral distress in childrenwith cancer undergoing medical procedures: developmentalconsiderations. J Consult Clin Psychol 1980; 48: 356–365.

32 LeBaron S, Zeltzer L. Assessment of acute pain and anxiety inchildren and adolescents by self-reports, observer reports, anda behavior checklist. J Consult Clin Psychol 1984; 52: 729–738.

33 Jay SM, Ozolins M, Elliot CH et al. Assessment of Children’sDistress During Medical Procedures. Health Psychol 1983; 2:

133–147.34 Taddio A, Nulman I, Koren BS et al. A revised measure of

acute pain in infants. J Pain Symptom Manage 1995; 10: 456–463.35 Bulloch B, Tenenbein M. Validation of 2 pain scales for use in

the pediatric emergency department. Pediatrics 2002; 110:

e33–e38.36 Stewart B, Lancaster G, Lawson J et al. Validation of the Alder

Hey Triage Pain Score. Arch Dis Child 2004; 89: 625–630.37 McGrath PJ, Johnson G, Goodman JT et al. CHEOPS: A

Behavioral Scale for Rating Postoperative Pain in Children. In:Fields HL, Dubner R, Cervero F, eds. Advances in Pain Researchand Therapy. New York: Raven Press, 1985: 395–402.

38 Merkel SI, Voepel-Lewis T, Shayevitz JR et al. The FLACC: abehavioral scale for scoring postoperative pain in young chil-dren. Pediatr Nurs 1997; 23: 293–297.

39 Tarbell SE, Cohen IT, Marsh JL. The Toddler-PreschoolerPostoperative Pain Scale: an observational scale for measuringpostoperative pain in children aged 1–5. Preliminary report.Pain 1992; 50: 273–280.

40 Kelly AM, Powell C, Williams A. Parent visual analogue scaleratings of children’s pain do not reliably reflect pain reportedby child. Pediat Emerg Care 2002; 18: 159–162.

41 Singer AJ, Gulla J, Thode HC Jr. Parents and practitioners arepoor judges of young children’s pain severity. Acad Emerg Med2002; 9: 609–612.

42 Tyler DC, Tu A, Douthit J et al. Toward validation of painmeasurement tools for children: a pilot study. Pain 1993; 52:301–309.

43 Beyer JE, McGrath PJ, Berde CB. Discordance between self-report and behavioral pain measures in children aged 3–7 years after surgery. J Pain Symptom Manag 1990; 5: 350–356.

44 Manworren RC, Hynan LS. Clinical validation of FLACC:preverbal patient pain scale. Pediatr Nurs 2003; 29: 140–146.

45 Willis MHW, Merkel SI, Voepel-Lewis T et al. FLACC beha-vioral pain assessment scale: A comparison with a child’sself-report. Pediatr Nurs 2003; 29: 195–198.

46 Voepel-Lewis T, Merkel S, Tait AR et al. The reliability andvalidity of the Face, Legs, Activity, Cry, Consolability obser-vational tool as a measure of pain in children with cognitiveimpairment. Anesth Analg 2002; 95: 1224–1229.

47 Merkel S, Voepel-Lewis T, Malviya S. Pain assessment ininfants and young children: the FLACC scale. Am J Nurs 2002;102: 55–58.

48 Hartrick CT, Kovan JP. Pain assessment following generalanesthesia using the Toddler Preschooler Postoperative PainScale: a comparative study. J Clin Anesth 2002; 14: 411–415.

49 McCarthy C, Hewitt S, Choonara I. Pain in young childrenattending an accident and emergency department. J AccidentEmerg Med 2000; 17: 265–267.

50 Abu-Saad H. Assessing children’s responses to pain. Pain 1984;19: 163–171.

51 Goodenough B, Champion GD, Laubreaux L et al. Needle painseverity in children: Does the relationship between self-reportand observed behavior vary as a function of age? Austral JPsychol 1998; 50: 1–9.

52 Goodenough B, Addicoat L, Champion GD et al. Pain in 4- to6-year-old children receiving intramuscular injections: a



comparison of the Faces Pain Scale with other self-report andbehavioral measures. Clin J Pain 1997; 13: 60–73.

53 Bieri D, Reeve RA, Champion GD et al. The Faces Pain Scalefor the self-assessment of the severity of pain experienced bychildren: development, initial validation, and preliminaryinvestigation for ratio scale properties. Pain 1990; 41: 139–150.

54 Hicks CL, von Baeyer CL, Spafford PA et al. The Faces PainScale-Revised: toward a common metric in pediatric painmeasurement. Pain 2001; 93: 173–183.

55 Keck JF, Gerkensmeyer JE, Joyce BA et al. Reliability andvalidity of the Faces and Word Descriptor Scales to measureprocedural pain. J Pediatr Nurs 1996; 11: 368–374.

56 Whaley LF, Wong DL. Nursing Care of Infants and Children.St Louis, 5th edn: Mosby Year Book, 1991.

57 Beyer JE, Aradine CR. Content validity of an instrument tomeasure young children’s perceptions of the intensity of theirpain. J Pediatr Nurs 1986; 1: 386–395.

58 Benini F, Griffith P, Lago P et al. Evaluating pain in children:experience with a ‘‘do-it-yourself’’ visual analogue scale. ActaPaediatr 1996; 85: 762.

59 Singer AJ, Richman PB, Kowalska A et al. Comparison of pa-tient and practitioner assessments of pain from commonlyperformed emergency department procedures. Ann Emerg Med1999; 33: 652–658.

60 Pothman R. Comparison of the visual analogue scale (VAS)and the smiley analogue scale (SAS) for the evaluation of painin children. Adv Pain Res Ther 2000; 15: 95–99.

61 Crellin L, Johnston L. A preliminary exploration of pain asso-ciated with nasogastric tube insertion in a paediatric popula-tion. Neonatal, Paediatric Child Health Nurs 2005; 8: 19–23.

62 Hendrickson M, Myre L, Johnson DG et al. Postoperativeanalgesia in children: a prospective study in intermittentintramuscular injection versus continuous intravenous infu-sion of morphine. J Pediatr Surg 1990; 25: 185–190.

63 LaMontagne LL, Johnson BD, Hepworth JT. Children’s ratingsof postoperative pain compared to ratings by nurses andphysicians. Issues in Comprehensive Pediatr Nurs 1991; 14: 241–247.

64 Miller D. Comparisons of pain ratings from postoperativechildren, their mothers, and their nurses. Pediatr Nurs 1996; 22:

145–149.65 Elliott CH, Jay SM, Woody P. An observation scale for mea-

suring children’s distress during medical procedures. J PediatrPsychol 1987; 12: 543–551.

Accepted 30 January 2007



Analysis of the validation of existing behavioral pain and distress scales for use in the procedural setting

Documents