BACKGROUND

Validation of the Severity of Pruritus Scale (SPS) for the Assessment of  
Pruritus in Atopic Dermatitis (AD)

Gil Yosipovitch,1 Eric Simpson,2 Andrew G. Bushmakin,3 Joseph C. Cappelleri,3 Thomas Luger,4 Sonja Ständer,4 Wynnis Tom,5 Katy Benjamin,6 William C. Ports,3 Anna M. Tallman,7 Huaming Tan,3 Robert A. Gerber3

1University of Miami, Miller School of Medicine, Miami, FL, USA; 2Oregon Health and Science University, Portland, OR, USA; 3Pfizer Inc., Groton, CT, USA; 4University Hospital Münster, Münster, Germany; 5Rady Children’s Hospital-San Diego, San Diego, CA, USA; 6ICON plc, Gaithersburg, MD, USA; 7Pfizer Inc., New York, NY, USA

Presented at the 2017 Fall Clinical Dermatology Conference; October 12-15, 2017; Las Vegas, NV

Table 1. Severity of Pruritus Scale
Instructions: Please think about your itching (or your child’s itching if you are 

completing for your child) over the past 24 hours and choose the category that 
best describes it.

Score Grade Definition

0 None No Itching

1 Mild Occasional, slight itching/scratching

2 Moderate
Constant or intermittent itching/scratching which is  

not disturbing sleep

3 Severe
Bothersome itching/scratching which is  

disturbing sleep

OBJECTIVES
• To assess the content validity of the SPS to ensure it is a clear and appropriate tool for the assessment of pruritus intensity
• To evaluate the psychometric properties (quantitative validation) of the SPS to ensure it is a valid and reliable measure of pruritus intensity

METHODS
Qualitative Content Validation of the SPS
• A combined concept elicitation and cognitive interview study was conducted to evaluate the content validity of the SPS in accordance with  

best practices6-10

• Eligible participants were aged ≥2 years, had a dermatologist confirmed diagnosis of AD within the preceding 12 months, and had experienced 
itching caused by AD

• Participants completed a brief background questionnaire and took part in a single one-on-one interview
 – For children aged 2-7 years, the caregiver completed the SPS and participated in the interview

• Because of the relative homogeneity of the US-only population and the single-item, single-concept nature of the SPS, a target sample size of  
15 subjects was considered adequate

Quantitative Psychometric Validation of the SPS
Data Set
• Data from the 2 identically designed, multicenter, randomized, double-blind, vehicle-controlled Phase 3 clinical trials investigating the safety 

and efficacy of crisaborole were pooled to assess the psychometric properties of the SPS (AD-301: NCT02118766; AD-302: NCT02118792)
 – 1522 patients aged ≥2 years with mild to moderate AD were included in the intent-to-treat population
 – Efficacy was assessed using the Investigator’s Static Global Assessment (ISGA; a 5-point rating scale measuring overall disease severity 
from clear [0] to severe [4]); the SPS; signs of AD (measured on a 4-point rating scale from none [0] to severe [3]); and health-related  
quality-of-life measures (Dermatology Life Quality Index [DLQI], Children’s Dermatology Life Quality Index [CDLQI], Dermatitis Family  
Impact Questionnaire [DFI])

Test-Retest Reliability
• Assessed via an intraclass correlation coefficient (ICC) using all available SPS observations from stable subjects between baseline/day 1 and 

day 8 (stable group was defined as having no change on ISGA between baseline/day 1 and day 8) 
 – ICC ≥0.70 was considered indicative of acceptable test-retest reliability; ICC ≥0.90 was considered indicative of excellent test-retest reliability11,12

Construct Validity
• Convergent validity was evaluated by calculation of Pearson correlations with the ISGA, quality-of-life instruments (DLQI, CDLQI, and DFI), 

and signs of AD
 – Evidence for convergent validity was based on a Pearson correlation ≥0.40 (correlations ≥0.50 were considered indicative of a strong 
association)

• Known-groups validity was assessed based on the difference in mean SPS scores between the “no disease group/clear” (ISGA = 0) and the 
“severe disease group” (ISGA = 4)
 – SPS scores as a function of ISGA were modeled using repeated-measures longitudinal analyses
 – The effect size was calculated as the difference in the mean divided by the baseline standard deviation (values of 0.20, 0.50, and 0.80 
standard deviation units were considered small, medium, and large, respectively)

Ability to Detect Change
• Evaluated using a repeated-measures longitudinal mixed model to estimate the relationship between SPS and ISGA scores

Clinically Important Difference
• Estimated using a repeated-measures longitudinal model linked to a 1-cateogry difference on the ISGA

Clinically Important Response
• Estimated using a repeated-measures longitudinal model with the change in SPS score from baseline as the outcome and a newly created 

static global impression of change (SGIC) anchor as the predictor (SGIC was based on categorizing the change from baseline in ISGA scores 
as worse [–1], same [0], or better [1])

RESULTS

Qualitative Content Validation of the SPS
• 14 individuals participated in the content validation study (Table 2)
• 9 interviews were conducted in US English, and 5 were conducted 

in US Spanish

Table 2.  Participant Characteristics 
Children
2-7 Years  

of Age
n = 5

Children
8-11 Years  

of Age
n = 4

Adolescents 
and Adults

n = 5

Patients characteristics

Sex
Female 4 1 3

Language
English 3 3 3

Age, mean (SD, range), 
years

5.0  
(1.2, 3-6)

9.3  
(1.5, 8-11)

27.8  
(17.9, 17-59)

Race
White
Black

4
1

2
2

5
0

Ethnicity
Hispanic/Latino 3 1 2

AD severity at interview 
(patient/caregiver reported)

Almost clear
Mild
Moderate
Severe

3
0
1
1

1
0
3
0

2
1
3
0

Currently receiving AD 
treatment (any)

5 4 5

Characteristics of caregivers of children 2-7 years of age (n = 5)

Sex
Female 4 — —

Language
English 3 — —

Age, mean (SD, range), 
years

34.6  
(4.5, 28-38)

— —

Race
White
Black

4
1

—
—

—
—

Ethnicity
Hispanic/Latino 3 — —

Working status
Homemaker
Full-time work
Part-time work

2
2
1

—
—
—

—
—
—

Education level
High school
Some college
Master’s degree

1
3
1

—
—
—

—
—
—

AD, atopic dermatitis; SD, standard deviation. 

Concept Elicitation
• The most prevalent symptom was itch, with 79% (n = 11) of 

participants reporting it spontaneously and 21% (n = 3) reporting  
it after probing

• Other reported signs included change in skin color (50%, n = 7),  
dry skin (36%, n = 5), and change in skin texture (29%, n = 4)

• Concept saturation analysis showed that only 2 new concepts were 
reported by the fifth interview, and all concepts were reported by the 
ninth interview

Cognitive Interview
• All participants correctly interpreted the SPS instructions and found 

them easy to understand (Table 3)
• Most participants found the scale easy to complete and correctly 

interpreted the meaning of the questions and response options and 
the phrase “sleep disturbance” (Table 3)

Table 3. Summary of the Cognitive Interviewing Results

Question Results Overview

Interpretation of  
SPS instructions

• 14 (100%) participants correctly interpreted the 
SPS instructions

Easy or difficult to  
understand the  
instructions

• 14 (100%) participants found the SPS instructions 
easy to understand

Easy or difficult to  
complete the scale

• 12 (86%) participants reported that the scale was 
easy to complete

Interpretation of item

• 13 (93%) participants correctly interpreted the 
meaning of the question 

 – 1 child (age 8 years) did not interpret the 
meaning of the question correctly

Interpretation of  
response options

• 11 (79%) participants interpreted the response 
options consistently and in agreement with the 
provided definitions

 – 1 caregiver (US English) interpreted the 
meaning of each response option as frequency 
instead of severity

 – 1 child (age 8 years, US Spanish) defined 
“moderate” as “night”

 – 1 child (age 8 years, US English) did not 
differentiate between “mild” and “moderate”

Meaning of  
“sleep disturbance”

• 13 (93%) participants had no issues interpreting 
the terms “disturbing sleep”

 – 1 adult (US English) had difficulty interpreting 
the terms “disturbing sleep”

Recall period

• 11 (79%) participants correctly interpreted the  
24-hour recall period

 – 1 adult reported on itch “in general”

 – 1 adult thought about itch severity “a few  
weeks ago”

 – 1 child (age 11) interpreted the question as 
frequency during the past week rather than  
the past 24 hours

How participant  
arrived at response

• No issues were raised regarding how patients 
arrived at their answers

SPS, Severity of Pruritus Scale.

Quantitative Psychometric Validation of the SPS
Test-Retest Reliability
• The ICC value for a single SPS measurement was estimated to be 0.54 
• Reliability improved with the use of multiple SPS measurements

 – The average of 2 SPS measurements (representing average 
pruritus over 1 day) provided an ICC value of 0.70 (indicative of 
acceptable test-retest reliability)

 – The average of 14 SPS measurements (representing average 
pruritus over a 1-week period) provided an ICC value of 0.94 
(indicative of excellent test-retest reliability)

• AD is a chronic inflammatory skin disease characterized by the 
development of eczematous lesions1

• Pruritus is a significant feature of AD and is believed to be 
responsible for much of the burden associated with the disease2

• Crisaborole ointment is a nonsteroidal phosphodiesterase 4 inhibitor 
for the treatment of mild to moderate AD3

• The safety and efficacy of crisaborole was established in 2 Phase 3  
clinical trials conducted in the United States (AD-301: NCT02118766;  
AD-302: NCT02118792)4

• Pruritus was assessed within these trials using the SPS, a 4-point 
rating scale, ranging from 0 (none: no itching) to 3 (severe, 
bothersome itching/scratching which disturbs sleep) that was 
adapted from the Atopic Dermatitis Severity Index5 to quantify 
pruritus intensity within a 24-hour period (Table 1)

Construct Validity
• Convergent validity was supported by strong correlation (Pearson 

correlation ≥0.50) between SPS scores and ISGA, DLQI, CDLQI, 
and DFI instruments, and correlations of ≥0.40 with 4 of 5 of the 
signs of AD at day 29 (Table 4)

Table 4. Correlation Between the SPS and All Instruments 
(Pearson correlation coefficient [r value]a)

Baseline Day 8 Day 15 Day 22 Day 29

ISGA 0.22 0.36 0.42 0.48 0.50

Erythema  
(redness)

0.16 0.30 0.38 0.39 0.42

Induration/ 
papulation

0.16 0.30 0.36 0.40 0.44

Exudation (oozing 
or crusting)

0.18 0.29 0.29 0.36 0.35

Excoriation  
(evidence of 
scratching)

0.25 0.36 0.42 0.45 0.41

Lichenification  
(epidermal  
thickening)

0.16 0.31 0.33 0.35 0.40

DLQI 0.46 — — -— 0.59

CDLQI 0.47 — — — 0.58

DFI 0.38 — — — 0.53

CDLQI, Children’s Dermatology Life Quality Index; DFI, Dermatitis Family Impact questionnaire; DLQI, Dermatology Life Quality 
Index; ISGA, Investigator’s Static Global Assessment; SPS, Severity of Pruritus Scale. 
aP < 0.0001 for all correlations.

• A 4-category difference in the ISGA (between the “no disease 
group/clear” and “severe disease group”) was associated with 
a difference in SPS score of 0.80 (continuous anchor) and 0.87 
(categorical anchor), indicating that the SPS can distinguish 
between groups known to be different (Table 5)

Table 5. Known-Groups Validity of the SPS in Relation to 
Scores on the ISGA

Data and Model

Differences in Mean SPS Scores on 
ISGA Between the “Severe Disease 
Group” and the “No Disease Group” 

Difference  
(95% CI) ES

Pooled studies (AD-301 and AD-302)a
0.80

(0.73-0.88)
1.03

Pooled studies (AD-301 and AD-302)b
0.87

(0.73-1.0)
1.12

Study AD-301a
0.75

(0.64-0.85)
0.96

Study AD-301b
0.89

(0.71-1.07)
1.14

Study AD-302a
0.86

(0.75-0.97)
1.10

Study AD-302b
0.84

(0.65-1.04)
1.08

ES, effect size; ISGA, Investigator’s Static Global Assessment; SPS, Severity of Pruritus Scale. 
ES calculated as difference/standard deviation of 0.78 (based on pooled data from post hoc analyses). 
aISGA as a continuous anchor.
bISGA as a categorical anchor.

Ability to Detect Change
• The relationship between SPS scores and ISGA scores provides 

evidence of sensitivity to change over time (Table 6, Figure 1)

Table 6. Ability to Detect Change in the SPS in Relation to  
ISGA Scores

ISGA Score
Mean SPS Score (95% CI) 
(ISGA continuous predictor)

Mean SPS Score (95% CI) 
(ISGA categorical predictor)

Pooled studies AD-301 and AD-302

0 0.79 (0.75-0.84) 0.84 (0.79-0.90)

1 0.99 (0.96-1.03) 0.99 (0.95-1.03)

2 1.19 (1.16-1.23) 1.16 (1.13-1.20)

3 1.40 (1.36-1.44) 1.42 (1.38-1.47)

4 1.60 (1.54-1.65) 1.71 (1.59-1.83)

Study AD-301

0 0.82 (0.75-0.89) 0.88 (0.80-0.96)

1 1.01 (0.95-1.06) 0.99 (0.94-1.05)

2 1.19 (1.14-1.24) 1.17 (1.11-1.22)

3 1.38 (1.32-1.44) 1.40 (1.34-1.47)

4 1.57 (1.49-1.64) 1.76 (1.59-1.93)

Study AD-302

0 0.77 (0.70-0.83) 0.81 (0.72-0.89)

1 0.98 (0.93-1.03) 0.98 (0.92-1.04)

2 1.20 (1.15-1.24) 1.16 (1.11-1.21)

3 1.41 (1.36-1.47) 1.44 (1.39-1.50)

4 1.63 (1.55-1.70) 1.65 (1.47-1.83)
ISGA, Investigator’s Static Global Assessment; SPS, Severity of Pruritus Scale.

Clinically Important Difference
• The estimated clinically important difference for SPS was 0.20 

(95% CI, 0.18-0.22)
 – The close functional relationship when using the ISGA as a 
categorical variable and continuous predictor supports the 
linearity assumption of the relationship between ISGA and SPS 
scores (Figure 1)

Figure 1. SPS score as a function of the ISGA as a continuous 
anchor and a categorical anchor.

0.50

0.75

1.00

1.25

1.50

2.00

0 1 2 3 4

ISGA

M
e

a
n

 S
P

S
 S

c
o

r
e

1.75

SPS score
(ISGA as a continuous anchor)

SPS score 
(ISGA as a categorical anchor)

ISGA, Investigator’s Static Global Assessment; SPS, Severity of Pruritus Scale.

Clinically Important Response
• A 1-category difference in SGIC corresponded to 0.19 points  

(95% CI, 0.16-0.22) in SPS score
 – Using the anchor as a categorical variable provided a close 
functional relationship to the results using the anchor as a 
continuous predictor, supporting the linearity assumption for the 
relationship between change from baseline of the SPS score and 
SGIC score (Figure 2)

• The responder definition was estimated as a decrease of 0.19 points 
from baseline in SPS score, linked to a 1-category difference 
between the SGIC categories

Figure 2. SPS change score as a function of the SGIC as  
a continuous anchor and a categorical anchor predictor.

–1.0

–0.9

–0.8

–0.5

–1 0 1

SGIC

M
e

a
n

 C
h

a
n

g
e

 F
r
o

m
 B

a
s

e
li

n
e

 i
n

 S
P

S
 S

c
o

r
e

–0.7

–0.6

Change from baseline in SPS score
(SGIC as a continuous anchor)

Change from baseline in SPS score 
(SGIC as a categorical anchor)

SGIC, static global impression of change; SPS, Severity of Pruritus Scale.

CONCLUSIONS
• The results of the qualitative content validation analysis 

confirm that itch is a significant symptom of AD and that the 
SPS is easy to understand and to complete in both US English 
and US Spanish

• The results of the quantitative analysis confirm the validity 
of the SPS, demonstrating that it has acceptable test-retest 
reliability provided ≥2 SPS measurements are used, has good 
convergent and known-group validity, and has an ability to 
adequately detect change

• A clinically important difference and a clinically important 
response were also identified, which could be used in future 
investigations that use the SPS to assess pruritus severity

REFERENCES
1. Weidinger S, Novak N. Lancet. 2016;387(10023):1109-1122.
2. Eichenfield LF et al. J Am Acad Dermatol. 2014;70(2):338-351.
3.  Eucrisa (crisaborole) ointment, 2%, for topical use [prescribing information]. Palo Alto, CA: Anacor Pharmaceuticals, Inc.; 2016.
4. Paller AS et al. J Am Acad Dermatol. 2016;75(3):494-503.e6.
5. Van Leent EJ et al. Arch Dermatol. 1998;134(7):805-809.
6.  US Department of Health and Human Services. Guidance for Industry Patient-Reported Outcome Measures: Use in Medical 

Product Development to Support Labeling Claims. December 2009. https://www.fda.gov/downloads/drugs/guidances/
ucm193282.pdf. Accessed September 20, 2017.

7. Patrick DL et al. Value Health. 2011;14(8):978-988.
8. Patrick DL et al. Value Health. 2011;14(8):967-977.
9. Brod M et al. Qual Life Res. 2009;18(9):1263-1278.

10. Lasch KE et al. Qual Life Res. 2010;19(8):1087-1096.
11. Nunnally JC, Bernstein IH. Psychometric Theory. 3rd ed. New York: McGraw-Hill; 1994.
12. Reeve BB et al. Qual Life Res. 2013;22(8):1889-1905.

ACKNOWLEDGMENTS
Editorial/medical writing support under the guidance of the authors was provided by Jemimah 
Walker, PhD, and Corey Mandel, PhD, at ApotheCom, San Francisco, CA, USA, and was funded 
by Pfizer Inc., New York, NY, USA, in accordance with Good Publication Practice (GPP3) guidelines 
(Ann Intern Med. 2015;163:461-464).


	FC17PosterPfizerYosipovitchValidation.pdf