Topical Treatment Utilization for Patients With Atopic Dermatitis in the  

United States and Budget Impact Analysis of Crisaborole Ointment, 2% 

Ryan Clark,1 Duygu Bozkaya,1 Mark Levenberg,2 Steven Faulkner,3 Timothy W. Smith,3 Robert A. Gerber4 
1
Xcenda, Palm Harbor, FL, USA; 

2Pfizer Inc., Collegeville, PA, USA; 3Pfizer Inc., New York, NY, USA; 4Pfizer Inc., Groton, CT, USA 

BACKGROUND 
• Atopic dermatitis (AD) is a chronic inflammatory skin disease affecting between 

6-13% of the population in the United States (US).1,2
• AD is often treated with prescription topical corticosteroids (TCS) and topical 

calcineurin inhibitors (TCI).
 – Concerns regarding local cutaneous atrophy, striae formation, and systemic 

side effects limit broader user of TCS.3,4
 – TCI are associated with application-site reactions and carry boxed warning for 

risk of malignancy (eg, skin and lymphoma).3-6
• Crisaborole ointment is a nonsteroidal PDE4 inhibitor for the treatment of mild to 

moderate AD.
 – Clinical studies have demonstrated that crisaborole is effective7 and has a 

relatively low incidence of treatment-related/treatment-emergent adverse 
events in patients with mild to moderate AD.7,8 Application site pain (mainly 
reported as burning or stinging) was the only treatment-related adverse event 
that occurred in >1% of patients (crisaborole: 4.4%; vehicle: 1.2%).7

OBJECTIVES
• The objectives of this study were:

1. To evaluate recent real-world treatment utilization and costs for TCS and TCI 
in the US.

2. To estimate the pharmacy budget impact of crisaborole ointment, 2% over 
a 2-year period in patients ≥2 years of age with AD, from a third-party US 
payer perspective.

METHODS
Real-world treatment utilization and costs (RWTUC) 

• The analysis assessed annual treatment patterns and associated costs of 
prescription topical therapies in patients who had a diagnosis of AD in 2015, 
using the Commercial and Medicare Supplemental claims from the Truven Health 
Analytics MarketScan® Research Databases from 1/1/2015 through 12/31/2015.

• All prescription utilization of TCS and TCI was assessed, and patients were 
classified by strategy based on how many different treatment categories they filled 
during 2015. 

 – To capture the use of combination therapy by AD patients comprised of 
TCI and distinct categories of TCS potency, 9 mutually exclusive treatment 
strategies were defined: 
1. TCS I-II only (representing high potency TCS)
2. TCS III-IV only (representing medium potency TCS)
3. TCS V-VII only (representing low potency TCS)
4. TCS I-VII (any multiple excluding above)
5. TCI only
6. TCI + TCS I-II only
7. TCI + TCS III-IV only
8. TCI + TCS V-VII only
9. TCI + TCS I-VII (any multiple excluding above)
Each strategy with a TCS included at least one of the TCS strengths referred 
to in the group, but did not represent the order of the treatments received. 
The patient may have used more than one type generic drug within the same 
strength category. 

• Patients with continuous enrollment for the entire calendar year 2015 were 
included if they had 1 or more diagnosis for AD (ICD-9 = 691.8 or ICD-10 = 
L20.*) at any time during 2015. Patients <2 years of age at time of diagnosis, 
having invalid medication quantity (METQTY <1 or >10 X tube size), or having 
invalid days’ supply (DAYSSUP <1) were excluded. 

• MarketScan data for TCS/TCI pricing represents the amount eligible for payment 
under the plan terms after applying rules such as discounts, but before applying 
copayments and deductible. In the absence of pharmacy claims for crisaborole, 
wholesale acquisition cost (WAC) is used in the BIM.

Crisaborole Budget impact model (BIM)

• Model structure
 – The incremental pharmacy budget impact due to changes in market share 

is calculated as the difference in costs between an environment reflecting 
the current shares of topical AD treatments without crisaborole, and an 
environment with projected shares including crisaborole.
• The number of patients on each treatment in each scenario (ie, “current 

environment” and “new environment”) and year is determined based 
on the size of the target population and market shares. The model then 
calculates the total costs based on annual per treatment per patient costs 
and number of patients on each treatment (Figure 1).

• Model inputs
 – Diagnosis and treatment rates, current market share, number of prescriptions 

per year, and payment (cost) per prescription were based on the real-world 
utilization analysis (see Results section).

 – The percentages of patients who are 2 years and older (97.53%) were based 
on US census data.9

 – Crisaborole uptake rates of 4.7% from TCS and 20.2% from TCI, with an annual 
increase of 1% in year 2, were assumed, based on a published BIM for a TCI.10

 – In the absence of current crisaborole utilization data, average number of 
prescriptions per patient from TCS I-II only, TCS III-IV only, TCS V-VII only, and 
TCI only groups was taken from the RWTUC analysis (1.46 prescriptions per 
year per patient). Cost per crisaborole prescription is $580.00 based on WAC 
(one 60 gram tube of crisaborole per prescription).11

Figure 1. Budget Impact Diagram
Current environment (without crisaborole) 

Plan size 

Patients diagnosed with AD 

Patients 2 years 

Patients receiving topical AD treatment 

Pharmacy costs 

Plan size 

Patients diagnosed with AD 

Patients 2 years 

Patients receiving topical AD treatment 

Pharmacy costs 

New environment (with crisaborole) 

BUDGET IMPACT (difference) 

Change 

in market 

share 

Change in 

pharmacy 

costs 

 

Key: AD – atopic dermatitis. 

• Model outputs
 – The results are presented over 2 years in terms of total pharmacy budget and 

pharmacy cost per-member per-month (PMPM) without discounting.
 – Two AD populations were evaluated separately: patients receiving TCI or TCS 

alone or in combination (“TCS/TCI population”) and patients receiving TCI alone or 
in combination with TCS (“TCI population”; excludes TCS alone).

RESULTS
Real-world treatment utilization and costs

• More than 70% of the population was 18-64 years old; patients in the 2-11 age 
group have the highest AD diagnosis rate (2.4%) (Table 1). 

• Among all diagnosed patients, 50%-65% received TCS and/or TCI treatment, 
while only 2%-7% received TCI treatment with or without TCS (Table 1).

 – Based on these results, for a health plan with 1 million members, the target 
population size for the BIM was estimated as 4,706 when the TCS/TCI 
population was considered and 414 when the TCI population was considered. 

• In the TCI/TCS population, the majority of the patients received medium potency 
TCS only (40.54%); while TCI population mostly used TCI alone (25%) or TCI 
combined with medium potency TCS (22.09%) (Table 2).

• The annual number of prescriptions per patient ranged from 1.36 to 6.41, with an 
annual cost per patient range of $53.11-$1,465.03 (Table 2, Figure 2).

 – Patients receiving TCS had similar annual number of prescriptions (1.36-1.49 
for patients receiving TCS without TCI; 2.99-3.74 for patients receiving TCS 
with TCI), with a decreasing trend towards low potency TCS.

 – TCIs were associated with higher per-prescription costs (~$470), and 
medium-potency TCS were less costly (~$50), while low and high-potency 
TCS had similar costs (~$180).

 – Patients using multiple types of TCS potencies in a year had the highest 
number of prescriptions and overall costs.

Crisaborole BIM

• For the TCI/TCS population, 285 and 332 patients were estimated to receive 
crisaborole in years 1 and 2, respectively (Figure 3). 

 – The total budget impact of crisaborole over 2 years in the TCS/TCI population 
was $350,946 (PMPM = $0.015), with a total budget increase of $162,106 in 
year 1 (PMPM = $0.014) and $188,841 in year 2 (PMPM = $0.016). 

• For the TCI population, 84 and 88 patients were estimated to receive crisaborole 
in years 1 and 2, respectively (Figure 3). 

 – The total budget impact of crisaborole in the TCI population was a savings of 
$22,871, with a total budget decrease of $11,160 in year 1 and $11,712 in year 
2 (each PMPM = −$0.001) (Figure 3). 

• For both populations, budget impact was most sensitive to changes in crisaborole 
cost and annual usage in one-way sensitivity analyses.

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

RESULTS (continued)
Table 1. Patient Demographics, and Diagnosis and Treatment Rates from RWTUC Analysis*
Age category (years) 2-11 12-17 18-64 65+ Total

Age distribution among patients ≥2 years 11.39% 8.47% 71.27% 8.87% 100%
Number of patients by age** 111,087 82,608 695,096 86,509 975,300
AD diagnosis rate by age 2.40% 1.14% 0.54% 0.69%
Number of patients diagnosed with AD by age** 2,666 942 3,754 597 7,958

TCI/TCS
% receiving TCI/TCS among diagnosed 50.30% 60.25% 65.14% 58.99%
Number of AD patients receiving TCI/TCS treatment** 1,341 567 2,445 352 4,706

TCI Only
% receiving TCI among diagnosed 4.60% 7.13% 5.61% 2.28%
Number of AD patients receiving TCI treatment** 123 67 211 14 414

Key: AD – atopic dermatitis; BIM – budget impact model; RWTUC – real-world utilization and costs; TCI – topical calcineurin inhibitor; TCS – topical corticosteroid.
* Results are used as inputs for the BIM.
** Estimated for a health plan with 1 million members, where 97.53% of the population is 2 years of age and older.

Table 2. Current market share, number of prescriptions per year, and payment per prescription from RWTUC Analysis*
Current market 

share for TCI/ 

TCS population

Current market 

share for TCI  

population

Rx per patient (annual) Pay per Rx

TCI TCS I-II TCS III-IV TCS V-VII TCI TCS I-II TCS III-IV TCS V-VII

1. TCS I-II only 22.84% n/a n/a 1.69 n/a n/a n/a $153.83 n/a n/a
2. TCS III-IV only 40.54% n/a n/a n/a 1.42 n/a n/a n/a $37.45 n/a
3. TCS V-VII only 9.63% n/a n/a n/a n/a 1.36 n/a n/a n/a $177.91
4. TCS I-VII** 18.19% n/a n/a 1.52 1.52 0.80 n/a $162.64 $50.66 $175.50
5. TCI only 2.20% 25.00% 1.36 n/a n/a n/a $463.44 n/a n/a n/a
6. TCI + TCS I-II 1.42% 16.11% 1.52 2.22 n/a n/a $478.04 $194.23 n/a n/a
7. TCI + TCS III-IV 1.94% 22.09% 1.48 n/a 1.82 n/a $466.32 n/a $62.22 n/a
8. TCI + TCS V-VII 0.65% 7.37% 1.44 n/a n/a 1.55 $462.31 n/a n/a $191.90
9. TCI + TCS I-VII** 2.59% 29.43% 1.68 1.82 1.84 1.08 $485.45 $183.16 $64.64 $184.88
Total 100% 100%

Key: AD – atopic dermatitis; BIM – budget impact model; n/a - not applicable; RWTUC – real-world utilization and costs; TCI – topical calcineurin inhibitor; TCS – topical corticosteroid.
* Results are used as inputs for the BIM.
** Any multiple, excluding the strategies above.

Figure 2. Annual Per-Patient Prescriptions and Costs

1.69 1.42 1.36 3.84 1.36 3.74 3.30 2.99 6.41 

$259.61 
$53.11 $242.75 

$463.78 
$632.54 

$1,157.44 

$803.22 
$961.98 

$1,465.03 

$0 

$200 

$400 

$600 

$800 

$1,000 

$1,200 

$1,400 

$1,600 

0 

1 

2 

3 

4 

5 

6 

7 

1. TCS 

I-II only

2. TCS

III-IV only

3. TCS

V-VII only

4. TCS 

I-VII 

5. TCI 

only 

6. TCI +

TCS I-II

7. TCI +

TCS III-IV

8. TCI +

TCS V-VII

TCI +

TCS I-VII 

A
n

n
u

a
l 
R

x
 c

o
s
t 

p
e
r
 p

a
ti

e
n

t 

A
n

n
u

a
l 
n

u
m

b
e
r
 o

f
 R

x
s
 p

e
r
 p

a
ti

e
n

t 

Annual Rx/patient Annual cost/patient 

Key: Rx – prescription; TCI – topical calcineurin inhibitor; TCS – topical corticosteroid.

Figure 3. BIM Results

285 332 

4,292 4,031 3,989 

414 389 385 

0 

1,000 

2,000 

3,000 

4,000 

5,000 

Current Year 1 Year 2 

Number of patients per treatment per year 

Crisaborole TCS TCI ± TCS 

A

TCI Population 

84 88 

414 
330 326 

0 

100 

200 

300 

400 

500 

Current Year 1 Year 2 

Number of patients per treatment per year 

Crisaborole TCI ± TCS 

D

$0.109 $0.109 $0.123 $0.125 

$0.014 $0.016 

$0.000 

$0.050 

$0.100 

$0.150 

Year 1 Year 2 

PMPM budget impact over 2 years: $0.015 

Without crisaborole With crisaborole 

B

$1,311,174 $1,311,174 $1,473,279 $1,500,014 

$162,106 $188,841 

$0 

$500,000 

$1,000,000 

$1,500,000 

$2,000,000 

Year 1 Year 2 

Total budget impact over 2 years: $350,946 

Without crisaborole With crisaborole 

C

$405,793 $405,793 $394,633 $394,081 

-$11,160 -$11,712 

$0 

$100,000 

$200,000 

$300,000 

$400,000 

$500,000 

Year 1 Year 2 

Total budget impact over 2 years: -$22,871 

Without crisaborole With crisaborole 

F 

TCS/TCI Population

$0.034 $0.034 $0.033 $0.033 

-$0.001 -$0.001 

$0.000 

$0.010 

$0.020 

$0.030 

$0.040 

Year 1 Year 2 

PMPM budget impact over 2 years: -$0.001 

Without crisaborole With crisaborole 

E 

Key: AD – atopic dermatitis; BIM – budget impact model; TCI – topical calcineurin inhibitor; TCS – topical corticosteroid.

REFERENCES
1. Hanifin JM et al. Dermatitis. 2007 Jun;18(2):82-91. 
2. Silverberg JI. JAMA Dermatol. 2014 Dec;150(12):1380. 
3. Eichenfield LF et al. J Am Acad Dermatol. 2014;71:116-132.
4. Schneider L et al. J Allergy Clin Immunol. 2013;131:295-299.

5. Siegfried EC et al. Am J Clin Dermatol. 2013;14:163-178.
6. Walling HW et al. Clin Cosmet Investig Dermatol. 2010;3:99-117.
7. Paller AS et al. J Am Acad Dermatol. 2016 Sep;75(3):494-503.e6. 
8. Eichenfield LF et al. J Am Acad Dermatol. 2017 Aug 17. 

9. U.S. Census Bureau, Population Division. http://factfinder.census.gov. 
Accessed April 20, 2017.

10. Chang J et al. J Manag Care Pharm. 2005 Jan-Feb;11(1):66-73.
11. Red Book Online®, Micromedex 2.0, Truven Health Analytics, Inc. 

Accessed April 20, 2017. 

KEY LIMITATIONS
• Costs associated with medical resources besides topical AD medications were 

excluded since data supporting differential non-medication resource use with 
crisaborole does not currently exist. Additionally, preliminary analysis of claims 
data indicated that the medical visit frequencies (except those for outpatient visits) 
are similar in patients with and without AD.

• As crisaborole is assumed to replace existing topical AD prescription treatments and 
is not expected to grow the number of patients treated for AD, patients untreated 
with prescription therapy by default do not receive crisaborole

• The model does not exclude any patients based on disease severity given that the 
data source to populate the model (MarketScan claims data) does not differentiate 
patients by disease severity.

• If there is crisaborole uptake, it replaces all topical AD prescriptions in a year.
Note: This analysis does not imply comparable efficacy, safety, or product 
interchangeability. No head-to-head clinical studies have been conducted 
to assess the efficacy or safety of crisaborole vs any TCS, TCI or any other 
prescription medications.

CONCLUSIONS
• Current utilization of topical prescription medications for AD have a material 

impact on pharmacy budget regardless of treatment choice.
 – Annual per-patient costs vary widely depending on the mechanism of 

action (TCI vs TCS) and TCS strength.
• Adoption of crisaborole results in a modest pharmacy budget impact or 

savings from the perspective of a US payer.


	FC17PosterPfizerClarkTopicalTreatmentCrisaboroleOintment.pdf