American Journal of Emergency Medicine xxx (2018) xxx-xxx

Contents lists available at ScienceDirect

American Journal of Emergency Medicine

F
journal homepage: www.elsevier.com

OO
A systematic review of the pain scales in adults: Which to use?
Ozgur Karcioglua, ⁎, Hakan Topacoglub, Ozgur Dikmea, Ozlem Dikmec
a
University of Health Sciences, Istanbul Training and Research Hospital, Emergency Department, Fatih, Istanbul, Turkey
b
Duzce University, School of Medicine, Emergency Department, Duzce, Turkey
c

PR
Koc University, School of Medicine, Emergency Department, Istanbul, Turkey

ARTICLE INFO ABSTRACT

Article history: Objective

Received 26 December 2017 The study analysed the Visual Analogue Scale (VAS), the Verbal Rating Scale (VRS) and the Numerical
Accepted 3 January 2018 Rating Scale (NRS) to determine: 1. Were the compliance and usability different among scales? 2. Were any
Available online xxx

ED
of the scales superior over the other(s) for clinical use?
Methods
Keywords: A systematic review of currently published studies was performed following standard guidelines. Online
Acute pain database searches were performed for clinical trials published before November 2017, on the comparison of
Pain score the pain scores in adults and preferences of the specific patient groups. A literature search via electronic data-
Pain scale bases was carried out for the last fifteen years on English Language papers. The search terms initially included
Pain intensity pain rating scales, pain measurement, pain intensity, VAS, VRS, and NRS. Papers were examined for method-
Pain management ological soundness before being included. Data were independently extracted by two blinded reviewers. Stud-
CT
Emergency department
ies were also assessed for bias using the Cochrane criteria.
Results
The initial data search yielded 872 potentially relevant studies; of these, 853 were excluded for some rea-
son. The main reason for exclusion (33.7%) was that irrelevance to comparison of pain scales and scores,
followed by pediatric studies (32.1%). Finally, 19 underwent full-text review, and were analysed for the study
purposes. Studies were of moderate (n = 12, 63%) to low (n = 7, 37%) quality.
RE

Conclusions
All three scales are valid, reliable and appropriate for use in clinical practice, although the VAS is more
difficulties than the others. For general purposes the NRS has good sensitivity and generates data that can be
analysed for audit purposes.
© 2017.

1. Pain in the emergency department Pain estimations need to be elicited and recorded to highlight both
R

the presence of pain and the efficacy of pain treatment. The patients'
Acute pain is one of the most common chief complaints reported perception of pain should be documented during the initial assessment
by most patients admitted to the ED, while its perception and expres- of a patient. Current evidence provides a general recommendation that
CO

sion have great variations between countries [1]. The definition of pain
by International Association for the Study of Pain (IASP) as ‘an un-
pleasant sensory and emotional experience associated with actual or
potential tissue damage, or described in terms of such damage’ is ac-
cepted worldwide [2].
Subjective and multidimensional nature of the pain experience ren-
der pain assessment really challenging. In the Joint Commission on
UN
pain needs to be evaluated and managed within 20–25 min of initial
healthcare provider assessment in the ED [4]. Pain treatment should be
targeted to a goal of reducing the pain score (e.g., by 50%, below 4/10,
or referred to as mild/moderate or severe) rather than a specific anal-
gesic dose [5].
2. Pain scores and documentation of pain

Accreditation of Healthcare Organizations (JCAHO) guidelines, im-

plementation of this standard in clinical practice comprised the addi-
tion of pain as the “fifth” vital sign to be noted in the context of initial
assessment; the use of pain intensity ratings; and posting of a state-
ment on pain management in all patient care areas. Supplemented with
regular pain reassessments, the schedule of pain reassessment should
be driven by patients' pain severity [3].
⁎
Corresponding author at: Dept. of Emergency Medicine, Univ. of Health Sciences,
Istanbul Education and Research Hospital, PK 34098, Fatih, Istanbul, Turkey.
Email address: okarcioglu@gmail.com (O. Karcioglu)
The patient's self-report is the most accurate and reliable evidence
of the existence of pain and its intensity, and this holds true for pa-
tients of all ages, regardless of communication or cognitive deficits
[6].
In the absence of objective measures, the clinician must depend on
the patient to supply key information on the localization, quality and
severity of the pain. Although physicians commonly question the re-
ported severity and rely on their own estimates, the value of the pa-
tients' description of the location and nature of the discomfort has been
proved on the theoretical basis and routine practice [7].

https://doi.org/10.1016/j.ajem.2018.01.008
0735-6757/© 2017.
2 American Journal of Emergency Medicine xxx (2018) xxx-xxx
Pain scores have gained acceptance as the most accurate and reli-
able measure of assessing a patient's pain and response to pain treat-
ment [5]. Scales devised to estimate and/or express the patient's pain
can be evaluated in two groups: Unidimensional and multidimensional
measures. It should be noted that unidimensional scales measure only
intensity, they cannot be viewed as a comprehensive pain assessment.
Comprehensive pain assessment is expected to encompass both the
unidimensional measurement of pain intensity and the multidimen-
sional evaluation of the pain perception. The unidimensional pain in-
tensity scales commonly used bedsides are:
• Numeric Rating Scale (NRS),
• Visual Analog Scale (VAS),
• Verbal Rating/Descriptor Scale (VRS/VDS).
2.1. B.1
The VAS is the most widely used tool for estimating both sever-
ities of pain and to judge the extent of pain relief [8]. Healthcare
ED
CT
R RE
CO
UN
Fig. 1. A. Visual Analog Scale (VAS). B. The Numerical Rating Scale (NRS) C. Verbal Rating Scale (VRS) or Verbal DescriptorScale (VRS). D. FACES Pain Rating Scale.
worker asks the patient to select a point on a line drawn between
two ends to express how intense he/she perceives pain (Fig. 1). The
VAS is a continuous scale comprised of a horizontal (HVAS) or ver-
tical (VVAS) line, usually 100 mm long, anchored by two verbal de-
scriptors (i.e., “no pain” and “worst imaginable pain”) [9, 10]. Patients
are asked to rate “current” pain intensity or pain intensity “in the last
F
24 h”.
The VAS is an easy-to-use instrument which does not warrant us-
OO
ing a sophisticated device. It is also highly sensitive in detecting treat-
ment effects, and its results can be analysed by parametric tests [11].
Minimal translation difficulties have led to an unknown number of
cross-cultural adaptations [10]. Although this tool is suitable for use
with older children and adults, the need for a marking and for being
able to visualize and mark the line, can make the VAS impractical to
use in the emergency situation. On the other hand, most experts be-
lieve that the VAS offers little practical advantage over verbal reports
PR
in the clinical practice [5].
 American Journal of Emergency Medicine xxx (2018) xxx-xxx
2.2. B.2
The numeric rating scale (NRS) is a single 11-point numeric scale
broadly validated across myriad patient types. Data obtained via NRS
are easily documented, intuitively interpretable, and meet regulatory
requirements for pain assessment and documentation [12]. To date,
findings demonstrated that even in the chaotic prehospital phase most
acute care patients allow evaluation via a simple “zero-to-10 scale”
or NRS reliably, respecting their pain levels [13]. Like the pain VAS,
minimal language translation difficulties support the use of the NRS
across cultures and languages [14].
Evidence indicated that patients really want to give a pain num-
ber, rather than simply relate whether they want analgesia. Strengths
of this measure over the pain VAS are the ability to be administered
both verbally (therefore by telephone) and in writing, as well as its
simplicity of scoring. However, similar to the pain VAS, the pain
ED
CT
R RE
CO
UN
Fig. 2. Flow diagram of study selection for systematic review to compare the clinical use of three commonly used pain rating scales, namely the Visual Analogue Scale (VAS), the
Verbal Rating Scale (VRS) and the Numerical Rating Scale (NRS).
3
NRS evaluates only 1 component of the pain experience, pain inten-
sity, and therefore does not capture the complexity and idiosyncratic
nature of the pain experience or improvements due to symptom fluc-
tuations [10].
NRS is a commonly used tool necessitating the patient rate his pain
on a scale from 0 to 10, with 0 indicating no pain and 10 reflecting the
F
worst possible pain (Fig. 2). NRS are often conducted as a scale from
1 to 10 which does not give the patient a solution to indicate no pain at
OO
all. It can be used with children who are able to understand numbers.
The pain scores are interpreted as:
• 0 = no pain
• 1–3 = mild pain
• 4–6 = moderate pain
• 7–10 = severe pain
NRS can be used with most children older than 8 years of age, and
PR
behavioral observation scales are required for those unable to provide
4 American Journal of Emergency Medicine xxx (2018) xxx-xxx
a self-report [15]. For patients with cancer-related pain, the NRS is the
most frequently used instrument to measure pain intensity [16]. Goulet
et al. examined the agreement and correlation of electronic medical
record-based ratings of NRS and self-administered NRS in 1643 adult
patients [17]. The correlation was high, but the mean electronic med-
ical record-based NRS score was significantly lower than the survey
score (1.72 vs. 2.79; p < 0.0001).
2.3. B.3
Verbal Pain Scores (VPSs), Verbal Rating Scales (VRS) or Ver-
bal Descriptor Scales: These tools may discern those patients who are
truly in pain but who may not express their discomfort, as well as in-
fluence the physician to inquire about the patient's pain.
VRS consist of a number of statements describing increasing pain
intensities (Fig. 1). Patients are told to choose the word which best de-
scribes their pain intensity. The number of descriptors used has ranged
from four (none, mild, moderate, severe) to 15 [18]. For patients who
have limited literacy or cognitive impairment, use of these scales may
be difficult, and they do not provide the number of choices available
with the VAS or NRS, thus potentially limiting precision [19].
This article reviews the current literature to provide systematic data
regarding the results from comparative studies on unidimensional as-
ED
sessment of pain intensity using the NRS, VRS, or VAS. The follow-
ing points were investigated to determine evidence-based recommen-
dations:
- Were the compliance and usability different among scales?
- Were any of the scales superior over the other(s) for clinical use?
CT
3. Methods
A systematic review of currently published studies was performed
following standard guidelines. Online database searches were per-
formed for randomized controlled trials published before November
RE
2017, on the comparison of the pain scores in adults and preferences
of the specific patient groups. A literature search via the Cochrane
Central Register of Controlled Trials, PubMed/Medline, Clinical Key,
EMBASE, the Cumulative Index to Nursing and Allied Health Lit-
erature (CINAHL), and BIOSIS was carried out for the last fifteen
years on English Language papers. Published studies evaluating the
R
patients' preferences and usability of the pain intensity scales were
targeted. The reference lists of retrieved articles were used to gen-
erate more papers and search terms. Data were independently ex-
tracted by two blinded reviewers. The discrepancies, on the other
CO
hand, were resolved by the primary author. The research protocol to
answer these questions was registered in PROSPERO, the Interna-
tional Prospective Register of Systematic Reviews (registration num-
ber is: CRD42017080974).
3.1. Search methodology
UN
A comprehensive literature search was carried out using the fol-
lowing strategy:
Online searches were performed using the following search key-
words and terms: (‘pain assessment’ OR ‘pain intensity’ OR ‘pain
score’ OR ‘pain comparison’ OR ‘pain scale’ OR ‘acute pain’ OR
‘pain rating’) AND (‘emergency’ AND ‘acute’ AND ‘score’). The
search was limited to human studies (clinical trials) conducted on
adults and published in English.
3.2. Study selection, data screening and critical appraisal
The study included all comparative trials conducted to assess the
use of commonly used scales measuring acute pain intensity and
to compare them with each other on specific patient groups, exclu-
sively in adults. All RCTs of any duration that investigated pain
F
scores in comparison to each other were identified. All potentially
eligible papers were critically appraised, with the emphasis on evi-
OO
dence from randomized trials and international guidelines rather than
smaller studies, case series and case reports. Reference lists of rele-
vant systematic reviews and all included studies were checked to iden-
tify additional eligible articles. Conference abstracts and proceedings
were not deemed eligible for inclusion in the review. Citation titles
and abstracts were independently screened and assessed regarding the
methodological quality by two reviewers (H.T. and O.D.). Any dis-
agreements between the two reviewers were then resolved by consen-
PR
sus or in consultation with a third reviewer (O.K.) if needed.
3.3. Quality assessment and risk of bias
Eligible clinical studies were rated regarding the quality of evi-
dence as per GRADE guidelines and assigned to one of four groups:
High (A), moderate (B), low (C) and very low (D) quality [20].
Studies that met the inclusion criteria for the review were assessed
for bias using the risk of bias criteria developed by Cochrane's EPOC
group [21] which is based upon Cochrane's Risk of Bias Tool [22].
Studies were assessed with regard to selection bias, performance bias,
detection bias, attrition bias, reporting bias, and other sources of bias.
Studies were rated as “low risk of bias (L)”, “high risk of bias (H),” or
“unclear risk of bias (U)” on a general impression after evaluating all
criteria (Table 1).
4. Results
The initial electronic data search yielded a total of 872 potentially
relevant studies; of these, 853 were excluded for some reason, and fi-
nally 19 trials fully met the selection criteria based on inclusion of in-
formation regarding comparative data on the pain scales, and specific
populations' preferences on the scales (Fig. 2). The main reason for ex-
clusion (33.7%, 288/853) was that irrelevance to comparison of pain
scales and scores, followed by pediatric studies (32.1%, 274/853).
Data collected for the review of the 19 clinical studies included
in the analysis of the pain scales used in the acute setting were tabu-
lated and summarized (Table 1). With respect to quality of evidence
per GRADE guidelines, there were 12 (63%) moderate quality (B) and
7 (37%) low quality (C) evidence derived from the studies.
5. Discussion
In order to use pain-rating scales well clinicians need to appreci-
ate the potential for error within the tools, and the potential they have
to provide the required information. Interpretation of the data from
a pain-rating scale is not as straightforward as it might first appear.
Leigheb M, et al. pointed out that there is substantial discordance be-
tween NRS and VAS scores which is suggestive of a need for clinical
judgment to be incorporated into assessment of actual pain intensity
and concluded that leaning on pain scale data alone is not a compre-
hensive approach [23].
In the present study, most of the studies in the analysis indicated a
good correlation between VAS, VRS/VDS and NRS, although some
 American Journal of Emergency Medicine xxx (2018) xxx-xxx 5

Table 1
Main characteristics of the outstanding human studies that were explained and reviewed in the present study.

Investigator(s), Quality of Risk

title and date, Sample size and evidence of
Ref. # population (GRADE)⁎ bias⁎⁎ Objectives Findings Notes, conclusions

F
McLean, et al. 1227 prehospital C H To determine the feasibility Prehospital pain assessment using a VRS and NRS An 11-point scale is preferable for
2004. [13] patients of prehospital pain was feasible in this patient population. Further prehospital practice and could also
> 13 yrs old measurement among studies are needed to confirm this in other settings. be useful for research applications.

patients 13 yrs of age or
older using a VRS and
NRS.
Leigheb M, et 137 adult ED B H To evaluate the intensity
al. 2017. [23] patients and location of pain
experienced by patients in
the ED, the time to
analgesic therapy in the
ED, and the patient's
OO
The magnitude of NRS pain measurements were
higher than VAS measurements.
Pain assessment using a VRS and
NRS can be implemented with
minimal paramedic training.
The discrepancy between NRS and
VAS scores suggests that
painintensity cannot be determined
accurately according
topainscaledata alone but should
also incorporate clinical judgment.

satisfaction so to identify
potential interventions to
improve management
Pereira et al. 101 C U Correlating two
2015. [24] institutionalized unidimensional scales for
elderly measurement of self-
reported pain intensity for
elderly and identifying a
preference for one of the
scales.
PR
There were moderate to strong, positive and Pain measurements among
statistically significant associations between the institutionalized elderly can be
scores of NRS and VDS: overall assessment made by NRS and VDS; however,
(r = 0.75), the rest (r = 0.92) and movement the preferred scale for the elderly
(r = 0.87). Higher mean scores were associated in was the VDS, regardless of
NRS to higher categories of pain intensity in VDS. gender.
The association between the mean scores of NRS

ED
with the categories of VDS was significant,
indicating convergent validity and a similar metric
between the scales.
Herr KA, et al. 175 B U To determine: (1) the All 5 pain scales (VAS, FPS, VDS, 21-point NRS, Although all 5 of the pain intensity
2004. [25] 86 younger psychometric properties 11-point VNS) were effective in discriminating rating scales were
(age 25-55) and and utility of 5 types of different levels of pain sensation; however the psychometrically sound when used
89 older (> 65) commonly used pain rating VDS was most sensitive and reliable. Failure rates with either age group, failures,
adults scaleswhen used with for pain scale completion were minimal, except internal consistency reliability,
CT
younger and older adults, for the VAS. construct validity, scale
(2) factors related to failure The scale most preferred to represent pain sensitivity, and preference suggest
to successfully use a pain intensity in both cohorts of subjects was the NRS, that the VDS is the scale of choice
rating scale, (3) pain rating followed by the VDS. for assessing pain intensity among
scale preference, and (4) older adults, including those with
factors impacting scale mild to moderate cognitive
preference. impairment.
RE

Ware LJ, et al. 68 cognitively C H To determine the reliability Concurrent validity was supported with correlations In terms of pain scale preference,
2006. [26] impaired and validity of selected ranging from 0.56 to 0.90. The lowest correlations the NRS (33%) was the preferred
minority pain intensity scales were found between the FPS-R and the other scale in the cognitively intact
sample including the FPS-R, VDS, scales, suggesting that the FPS-R may be group and the FPS-R (54%) was
NRS, and Iowa Pain measuring a broader construct incorporating pain the preferred scale in the
Thermometer (IPT) with a cognitively impaired group.
cognitively impaired African-Americans and Hispanics
minority sample. preferred the FPS-R. Severely,
R

moderately, and mildly impaired

participants also preferred the
FPS-R.
Yazici et al. 621 B L To determine patient pain Patient preference for pain scales were as follows: The NRS, TPS, FPS, and VDS
CO

2014. [27] postoperative scale preferences and 97.4% FPS, 88.6% NRS, 84.1% VDS, 78.1% were appropriate pain rating scales
adult patients compare the level of TPS, 60.1% SFMPQ, 37.0% BPI, 11.4% VAS, for the participants in this study,
agreement among pain and 10.5% MPQ. Education was an important and that the VAS should be used
scales commonly used factor in the preferences for all scales (p < .000). in combination with one of these
during postoperative pain The level of pain determined by the VAS did not scales.
assessment. correlate with the level of pain identified by the
NRS, TPS, FPS, and VDS (p < .05).
Bahreini M, et 150 adult ED C H To assess the agreement The three pain scales were strongly correlated at all Spearman correlation coefficients
al. 2015. [30] patients between VAS, Color time periods. The findings suggest that NRS, between NRS and CAS, NRS and
UN

Analog Scale (CAS), and
NRS in the emergency
setting.
CAS, and VAS can be interchangeably applied for VAS, and CAS and VAS were
acute pain measurement in adult patients. 0.95, 0.94, and 0.94, respectively
(p < 0.001). On a scale of 0 to 10,
the 95% limits of agreement
between the paired NRS and VAS,
VAS and CAS, and CAS and NRS
ranged from − 2.0 to 2.6, from
− 2.7 to 2.0, and from − 2.1 to 2.0,
respectively.
6 American Journal of Emergency Medicine xxx (2018) xxx-xxx

Table 1 (Continued)

Investigator(s), Quality of Risk

title and date, Sample size and evidence of
Ref. # population (GRADE)⁎ bias⁎⁎ Objectives Findings Notes, conclusions

Aziato L, et al. 150 post- B L To select, develop, and (Color-Circle Pain Scale–[CCPS]) had higher scale Using a valid tool for pain

F
2015. [31] operative validate context- preference than NRS and FPS. Convergent assessment gives the clinician an
patients appropriate unidimensional validity was very good and significant (0.70 objective criterion for pain
pain scales for pain –0.75). Inter-rater reliability was high management.

Göransson KE, 217 adult ED
et al. 2015. patients
[32].
OO
assessment among adult (0.923–0.928) and all the scales were sensitive to Due to the subjective nature of
post-operative patients. change in the intensity or level of pain pain, consideration of socio-
experienced before and after analgesia. cultural factors for the particular
context ensures that the
appropriate tool is used.
B L To compare correlations The pain scores generated from the NRS and the A majority reported that the NRS
between values on the VAS were found to strongly correlate (mean reflected/described their pain
VAS and the NRS in difference, 0.41). Most patients found the NRS better than the VAS (53% and
patients in the ED and to easier to use than the VAS (61% and 22%, 26%, respectively; p < .01). NRS

Edelen MO, et 1960 elderly
al. 2010. [33]. residents from
71 nursing
homes
assess the patients'
preference of scale
B L To compare VDS and NRS
using item response theory
(IRT) methods to identify
the
correspondencebetween
the scales response options
by estimating item
parameters for these and
PR
respectively; p < .001).
The sample reported moderate amounts of pain on
average. Examination of the IRT location
parameters for the pain intensity items indicated
the following approximate correspondence: VDS
mild ≈ NRS 1–4, VDS moderate ≈ NRS 5–7, VDS
severe ≈ NRS 8–9, and VDS very severe,
horrible ≈ NRS 10.
might be more appropriate to use
in the ED
Either scale (VDS and NRS) can be
used in practice depending on the
preference of the clinician and
respondent.

ED
five additional pain items.
Pratici E, et al. 97 women in C H To determine the level of There was moderate agreement between calculated The concordance correlation
2017. [34] labor agreement between percentage pain reduction from a VAS or NRS coefficient with patient-reported
calculated percentage pain and patient-reported % pain reduction in epidural percentage pain reduction was
reduction, derived from analgesia. The difference could range up to 30%. 0.76 and 0.77 for the VAS and
VAS or NRS, and patient- NRS, respectively.
reported % pain reduction
in patients having epidural
CT
analgesia.
Gagliese L, et 504 B L To compare the feasibility Psychometric analyses suggested that the NRS was Difficulties with VAS use among
al. 2005. [35] postoperative and validity of the NRS, the preferred pain intensity scale. The VDS also the elderly were identified,
adults. VDS, and VAS (horizontal had a favourable profile with low error rates and including high rates of unscorable
and vertical line good face, convergent and criterion validity. data and low face validity
orientation) for the
assessment of pain
RE

intensity in younger and

older surgical patients.
Bijur PE, et al. 108 adult ED C H To assess the comparability NRS scores were strongly correlated to VAS scores The verbally administered NRS can
2003. [36] patients of the NRS and VAS as at all time periods (r = 0.94). The slope of the be substituted for the VAS in acute
measures of acute pain, regression line was 1.01 and the y-intercept was pain measurement.
and to identify the − 0.34.
minimum clinically
significant difference in
R

pain that could be detected

on the NRS.
Herr K, et al. 97 adults with B L To compare the sensitivity The IPT showed the lowest failure rate of all pain All five pain scales were sensitive
2007. [37] chronic joint and utility of the new IPT scales evaluated. Other scale failure rates were in detecting changes in pain
CO

pain with four other pain scales: relatively low except for the VNS and the VAS.
NRS, VNS, FPS, and
VAS, using a naturally
occurring pain condition
and a controlled treatment
with rheumatology
patients.
UN
intensity pre and post joint
No significant difference was noted in scale failure injection. All correlations between
by age, gender or education level, but cognitive the scales were strong and
impairment was significantly related to failure on significant; however, the
the VAS and the NRS. intercorrelations for the older
cohort were weaker. The scale
most preferred in both cohorts of
patients was the IPT, followed by
the FPS.

Taylor LJ, et 66 cognitively B L To determine the reliability Concurrent validity of the VDS, NRS, and IPT was When asked about scale preference,
al. 2005. [38] impaired and validity of selected supported with Spearman rank correlation both the cognitively impaired and
elderly painintensity scales such as coefficients ranging from .78 to .86 in the the intact groups preferred the IPT
the FPS, VDS, NRS, and cognitively impaired group. The FPS, however, and the VDS. This study revealed
the Iowa Pain demonstrated weak correlations with other that cognitive impairment did not
Thermometer (IPT) to scaleswhen used with the impaired group, ranging inhibit participants' ability to use a
assess pain in cognitively from.48 to .53. In the cognitively intact group, variety of pain intensity scales, but
impaired older adults. strong correlations ranging from.96 to .97 were the stability issue must be
found among all of thescales. considered.
 American Journal of Emergency Medicine xxx (2018) xxx-xxx 7

Table 1 (Continued)

Investigator(s), Quality of Risk

title and date, Sample size and evidence of
Ref. # population (GRADE)⁎ bias⁎⁎ Objectives Findings Notes, conclusions

Li L, et al. 173 B U To determine the All four pain intensity scales had good reliability Although all four scales can be

F
2007. [39] postoperative psychometric properties and validity when used with Chinese adults. The options for Chinese adults to
adults. and applicability of four ICCs of the four scales across current, worst, least, report pain intensity, the FPS-R
pain scales in Chinese and average pain on each postoperative day were appears to be the best one. No

OO
postoperative adults. consistently high (0.673-0.825), and all scales at significant differences were noted
each rating were strongly correlated in terms of gender, age, and
(r = 0.71–0.99). educational level.
Both the VDS and the FPS-R had low error rates.
Nearly half of the participants (48.1%) preferred
the FPS-R, followed by the NRS (24.4%), the
VDS (23.1%), and the VAS (4.4%).
Li L, et al. 180 B U To evaluate the reliability The intraclass correlation coefficients across Although all three scales show
2009. [40] postoperative and validity of the FPS-R, current, worst, and least pain on each good reliability, validity, and

PR
elderly. NRS, and the Iowa Pain postoperative day were consistently high (0.949 to sensitivity for assessing
Thermometer (IPT) for 0.965), and all scales at each rating were strongly postoperative pain intensity in
pain assessment in Chinese correlated (r = 0.833 to 0.962). The scale mostly Chinese elders, the IPT appears to
elders who have had preferred was the IPT (54.7%), followed by the be a better choice based on patient
surgery. FPS-R (28.5%) and the NRS (15.6%). No preference.
significant differences were noted in participant
preference by age and cognitive status, but
preference for the IPT and the FPS-R were
significantly related to gender and education level.
Ismail AK, et 133 ambulance B H To evaluate the agreement VAS performs as well as VNRS in assessing acute There was a strong correlation

ED
al. 2015. [41] patients between verbal NRS and pain in prehospital setting. VAS and VNRS must between VNRS and VAS at the
VAS in measuring acute not be used interchangeably to assess acute pain; scene (r = 0.865; p < 0.001), as
pain in prehospital setting either method should be used consistently. well as on arrival at the hospital
and to identify the (r = 0.933; p < 0.001). Kappa
preference among values and analysis indicates good
paramedics and patients. agreement between both scales for
measuring acute pain.
Akinpelu AO, 35 women with C H To evaluate correlation of There was no significant difference between the The three pain rating scales
CT
et al. 2002. caesarian pain scores obtained on the pain scores obtained on the 3 pain rating scales. measure the same construct, and
[42] section VAS, the Box Numerical Significant correlations existed between pain could be used for pain
Scale (BNS) and Verbal scores obtained on the VAS and VRS (r = 0.48, measurement in obstetrically
Rating Scale (VRS) was p = 0.003); VAS and BNS (r = 0.74, p = 0.000); related conditions in this
studied. BNS and VRS (r = 0.74, p = 0.000). environment.
Abbreviations: The Faces pain scale (FPS), visual analog scale (VAS), numeric rating scale (NRS), verbal descriptor scale (VDS), thermometer pain scale (TPS), McGill Pain
Questionnaire (MPQ), Short-form McGill Pain Questionnaire (SFMPQ), and Brief Pain Inventory (BPI).
RE

(GRADE system) (GGHO)

Grade A: High level of evidence (The true effect lies close to our estimate of the effect.)
Grade B: Moderate level of evidence (The true effect is likely to be close to our estimate of the effect, but there is a possibility that it is substantially different.)
Grade C: Low level of evidence (The true effect may be substantially different from our estimate of the effect.)
Grade D: Very low level of evidence (Our estimate of the effect is just a guess, and it is very likely that the true effect is substantially different from our estimate of the effect.)
⁎
Quality of evidence and definitions.
⁎⁎
Risk of bias: Studies were assessed for bias using the risk of bias criteria developed by Cochrane [21] which is based upon Cochrane's Risk of Bias Tool [22]. Studies were
R

assessed with regard to selection bias, performance bias, detection bias, attrition bias, reporting bias, and other sources of bias. Studies were rated as “low risk of bias (L)”, “high
risk of bias (H),” or “unclear risk of bias (U)” on a general impression after evaluating all criteria.
CO

pointed out there is a discrepancy in some situations. One study re-
ported a moderate agreement between calculated percentage pain re-
duction from a VAS or NRS and the difference could range up to 30%.
VDS and NRS were also found to have strong correlation and can
be used in practice depending on the preference. The elderly were
found to prefer VDS to express their pain intensity [24, 25] including
those with mild to moderate cognitive impairment. Accordingly, Ware
UN
One of the first reviews on comparison of the three pain scales
(VAS, VRS, and NRS) were published by Williamson and Hoggart in
2005 and they reported that all three scales were valid, reliable and
appropriate for use, although the VAS had more practical difficulties
than the other two scales [28]. They stressed that for general purposes
the NRS has good sensitivity and produces data that can be analysed
for audit purposes. Likewise, Hjermstad MJ, et al. performed a sys-

et al. reported that NRS was the preferred scale in the cognitively in-
tact group while FPS-R was the preferred scale in the cognitively im-
paired group [26].
A number of studies cited a considerable difficulty in practical use
of VAS, especially in the elderly and populations with disadvantages.
For example, Yazici et al. noted that the NRS, TPS, FPS, and VDS
were appropriate pain rating scales for the participants in this study,
and that the VAS should be used in combination with one of these
scales [27].
tematic review of studies to culminate data on the use and perfor-
mance of unidimensional pain scales [29]. They reported that when
compared with the VAS and VRS, NRSs had better compliance in
15 of 19 studies reporting this, and were the recommended tool in 11
studies. Overall, NRS and VAS scores corresponded, with a few ex-
ceptions of systematically higher VAS scores.
Limitations of this article are similar to all review articles: the de-
pendence on previously published research and availability of refer-
ences. There is also a lack of published Level I and Level II studies
specific to this topic in the world's literature.
8 American Journal of Emergency Medicine xxx (2018) xxx-xxx
6. Summary and conclusion
“Pain cannot be treated if it cannot be assessed”. The most impor-
tant principle is that clinicians should somehow assess their patients'
pain levels, no matter which method or scale one uses to accomplish
this task. Special scales developed and validated for patients with dif-
ficulties in communication should be made available, and ED physi-
cians should have a plan for assessing pain in different case scenarios.
The bulk of evidence published to date have demonstrated a gap
for improvement to indicate pluses and minuses of each rating scale
used for acute pain. Reports focus that although all pain-rating scales
are valid, reliable and appropriate for use in emergency setting, the
VAS has somehow appeared more difficult than the others. Elderly
patients and those with cognitive impairment, communication prob-
lems and minorities have found verbal descriptor or rating scales more
practical than others in expression of their pain intensity. Ongoing re-
search in the area of ED patient pain management along with usability
of each tool should be conducted on specific patient groups and popu-
lations before firm conclusions could be drawn.
References
[1] S.F. Li, P.W. Greenwald, P. Gennis, P.E. Bijur, E.J. Gallagher, Effect of age on
acute pain perception of astandardized stimulus in the emergency department,
ED
Ann Emerg Med 38 (6) (2001) 644–647.
[2] Part III: Pain terms, a current list with definitions and notes on usage, in: H.
Merskey, N. Bogduk (Eds.), Classification of chronic pain, 2nd ed., IASP task
force on taxonomy, IASP Press, Seattle, 1994, pp. 209–214.
[3] A.W. Lozner, A. Reisner, M.L. Shear, et al., Pain severity is the key to emer-
gency department patients' preferred frequency of pain assessment, Eur J Emerg
Med 17 (1) (2010) 30–32.
[4] P.S. Grant, Analgesia delivery in the ED, Am J Emerg Med 24 (7) (2006)
CT
806–809.
[5] J.R. Miner, J.H. Burton, Pain management, in: R. Walls, R. Hockberger, M.
Gausche-Hill (Eds.), Rosen's emergency medicine - concepts and clinical prac-
tice, 9th ed., Elsevier Canada, 2018, pp. 34–51, (Copyright).
[6] Agency for Health Care Policy and Research, Acute pain management guideline
panel, In: Clinical practice guidelines, Acute pain management: Operative or
medical procedures and trauma, U.S. Department of Health and Human Services,
RE
Rockville, MD, 1992.
[7] S. Perry, G. Heidrich, Management of pain during debridement: a survey of US
burn units, Pain s13 (3) (1982) 276–280.
[8] A.M. Kelly, The minimum clinically significant difference in visual analogue
scale pain score does not differ with severity of pain, Emerg Med J 18 (3) (2001)
205–207.
[9] M.P. Jensen, P. Karoly, S. Braver, The measurement of clinical pain intensity: a
comparison of six methods, Pain 27 (1986) 117–126.
R
[10] G.A. Hawker, S. Mian, T. Kendzerska, M. French, Measures of adult pain:
visual analog scale for pain (VAS pain), numeric rating scale for pain (NRS
pain), McGill PainQuestionnaire (MPQ), short-form McGill PainQuestionnaire
(SF-MPQ), chronic PainGrade scale (CPGS), short Form-36 bodily pain scale
CO
(SF-36 BPS), and measure of intermittent and constant osteoarthritis pain
(ICOAP), Arthritis Care Res (Hoboken) 63 (Suppl. 11) (2011) S240–52.
[11] E.J. Gallagher, M. Liebman, P.E. Bijur, Prospective validation of clinically im-
portant changes in painseverity measured on a visual analog scale, Ann Emerg
Med 38 (6) (2001) 633–638.
[12] C.A. Marco, A.P. Marco, Assessment of pain, in: S.H. Thomas (Ed.), Emer-
gency department analgesia: an evidence based guide, Cambridge University
Press, Cambridge, UK, 2008, pp. 10–18.
[13] S.A. McLean, R.M. Domeier, H.K. DeVore, E.M. Hill, R.F. Maio, S.M. Fred-
UN
eriksen, The feasibility of pain assessment in the prehospital setting, Prehosp
Emerg Care 8 (2) (2004) 155–161.
[14] G.B. Langley, H. Sheppeard, The visual analogue scale: its use in pain measure-
ment, Rheumatol Int 5 (1985) 145–148.
[15] C.L. von Baeyer, Children's self-reports of pain intensity: scale selection, limita-
tions and interpretation, Pain Res Manag 11 (3) (2006) 157–162.
[16] W.H. Oldenmenger, C.C. van der Rijt, Feasibility of assessing patients' accept-
able pain in a randomized controlled trial on a patient pain education program,
Palliat Med 31 (6) (2017) 553–558.
[17] J.L. Goulet, C. Brandt, S. Crystal, D.A. Fiellin, C. Gibert, A.J. Gordon, et al.,
Agreement between electronic medical record-based and self-administered pain
numeric rating scale, Med Care 51 (3) (2013) 245–250.
[18] R.H. Gracely, P. McGrath, R. Dubner, Ratio scales of sensory and affective ver-
bal pain descriptors, Pain 5 (1978) 5–18.
[19] M.D. Ferraz, M.R. Quaresma, L.R.L. Aquino, et al., Raliability of pain scales in
the assessment of literate and illiterate patients with rheumatoid arthritis, J
F
Rheumatol 17 (1990) 1022–1024.
[20] G.H. Guyatt, A.D. Oxman, G.E. Vist, R. Kunz, Y. Falck-Ytter, P.
Alonso-Coello, et al., GRADE: an emerging consensus on rating quality of evi-
OO
dence and strength of recommendations, BMJ 336 (7650) (2008) 924–926.
[21] Effective Practice and Organisation of Care (EPOC), Suggested risk of bias cri-
teria for EPOC reviews, In: EPOC resources for review authors, Norwegian
Knowledge Centre for the Health Services, Oslo, 2016.
[22] Cochrane Handbook for Systematic Reviews of Interventions: The Cochrane
Collaboration, Available from: http://www.handbook.cochrane.org/, 2011.
[23] M. Leigheb, M. Sabbatini, M. Baldrighi, E.A. Hasenboehler, L. Briacca, F.
Grassi, et al., Prospective analysis of pain and pain management in an emer-
gency department, Acta Biomed 88 (4-S) (2017) 19–30.
PR
[24] L.V. Pereira, A. Pereira Gde, L.A. Moura, R.R. Fernandes, Pain intensity among
institutionalized elderly: a comparison between numerical scales and verbal de-
scriptors, Rev Esc Enferm USP 49 (5) (2015) 804–810.
[25] K.A. Herr, K. Spratt, P.R. Mobily, G. Richardson, Painintensity assessment in
older adults: use of experimental pain to compare psychometric properties and
usability of selected pain scales with younger adults, Clin J Pain 20 (4) (2004)
207–219.
[26] L.J. Ware, C.D. Epps, K. Herr, A. Packard, Evaluation of the revised Faces pain
scale, verbal descriptor scale, numeric rating scale, and Iowa pain thermometer
in older minority adults, Pain Manag Nurs 7 (3) (2006) 117–125.
[27] Y. Yazici Sayin, N. Akyolcu, Comparison of pain scale preferences and pain in-
tensity according to pain scales among Turkish patients: a descriptive study, Pain
Manag Nurs 15 (1) (2014) 156–164.
[28] A. Williamson, B. Hoggart, Pain: a review of three commonly used pain rating
scales, J Clin Nurs 14 (7) (2005) 798–804.
[29] M.J. Hjermstad, P.M. Fayers, D.F. Haugen, A. Caraceni, G.W. Hanks, J.H.
Loge, et al., European palliative care research collaborative (EPCRC). Studies
comparing numerical rating scales, verbal rating scales, and visual analogue
scales for assessment of pain intensity in adults: a systematic literature review, J
Pain Symptom Manage 41 (6) (2011) 1073–1093.
[30] M. Bahreini, M. Jalili, M.A. Moradi-Lakeh, Comparison of three self-report pain
scales in adults with acute pain, J Emerg Med 48 (1) (2015) 10–18.
[31] L. Aziato, F. Dedey, K. Marfo, J.A. Asamani, J.N. Clegg-Lamptey, Validation of
three pain scales among adult postoperative patients in Ghana, BMC Nurs
14 (2015) 42.
[32] K.E. Göransson, U. Heilborn, J. Selberg, S. von Scheele, T. Djärv, Pain rating in
the ED—a comparison between 2 scales in a Swedish hospital, Am J Emerg
Med 33 (3) (2015) 419–422.
[33] M.O. Edelen, D. Saliba, Correspondence of verbal descriptor and numeric rating
scales for pain intensity: an item response theory calibration, J Gerontol A Biol
Sci Med Sci 65 (7) (2010) 778–785.
[34] E. Pratici, S. Nebout, N. Merbai, J. Filippova, D. Hajage, H. Keita, An observa-
tional study of agreement between percentage pain reduction calculated from vi-
sual analog or numerical rating scales versus that reported by parturients during
labor epidural analgesia, Int J Obstet Anesth 30 (2017) 39–43.
[35] L. Gagliese, N. Weizblit, W. Ellis, V.W. Chan, The measurement of postopera-
tive pain: a comparison of intensity scales in younger and older surgical patients,
Pain 117 (3) (2005) 412–420.
[36] P.E. Bijur, C.T. Latimer, E.J. Gallagher, Validation of a verbally administered
numerical ratingscale of acute pain for use in the emergency department, Acad
Emerg Med 10 (4) (2003) 390–392.
[37] K. Herr, K.F. Spratt, L. Garand, L. Li, Evaluation of the Iowa pain thermometer
and other selected pain intensity scales in younger and older adult cohorts using
controlled clinical pain: a preliminary study, Pain Med 8 (7) (2007) 585–600.
[38] L.J. Taylor, J. Harris, C.D. Epps, K. Herr, Psychometric evaluation of selected
pain intensity scales for use with cognitively impaired and cognitively intact
older adults, Rehabil Nurs 30 (2) (2005) 55–61.
[39] L. Li, X. Liu, K. Herr, Postoperative pain intensity assessment: a comparison of
four scales in Chinese adults, Pain Med 8 (3) (2007) 223–234.
[40] L. Li, K. Herr, P. Chen, Postoperative pain assessment with three intensity scales
in Chinese elders, J Nurs Scholarsh 41 (3) (2009) 241–249.
[41] A.K. Ismail, M.A. Abdul Ghafar, N.S. Shamsuddin, N.A. Roslan, H. Kaharud-
din, N.A. Nik Muhamad, The assessment of acute pain in pre-hospital care using
verbal numerical rating and visual analogue scales, J Emerg Med 49 (3) (2015)
287–293.
[42] A.O. Akinpelu, O.O. Olowe, Correlative study of 3 pain rating scales among ob-
stetric patients, Afr J Med Med Sci 31 (2) (2002) 123–126.