Accepted: 2 January 2018

DOI: 10.1111/all.13401

ORIGINAL ARTICLE
Skin and Eye Diseases

Epidemiology of atopic dermatitis in adults: Results from an

international survey

S. Barbarot1 | S. Auziere2 | A. Gadkari3 | G. Girolomoni4 | L. Puig5 |

E. L. Simpson6 | D. J. Margolis7 | M. de Bruin-Weller8 | L. Eckert9

1
Department of Dermatology, CHU Nantes,
Nantes, France Abstract
2
Kantar Health, Paris, France Background: There are gaps in our knowledge of the prevalence of adult atopic
3
Regeneron, Tarrytown, NY, USA dermatitis (AD).
4
Department of Medicine, Section of
Objective: To estimate the prevalence of AD in adults and by disease severity.
Dermatology, University of Verona, Verona,
Italy Methods: This international, cross-sectional, web-based survey was performed in
5
Hospital de la Santa Creu i Sant Pau, the United States, Canada, France, Germany, Italy, Spain, United Kingdom, and
Universitat Auto noma de Barcelona,
Barcelona, Spain
Japan. Adult members of online respondent panels were sent a questionnaire for
6
Department of Dermatology, Oregon AD identification and severity assessment; demographic quotas ensured population
Health & Science University, Portland, OR, representativeness for each country. A diagnosis of AD required subjects to be posi-
USA
7 tive on the modified UK Working Party/ISAAC criteria and self-report of ever hav-
Department of Dermatology, University of
Pennsylvania Perelman School of Medicine, ing an AD diagnosis by a physician. The proportion of subjects with AD who
Philadelphia, PA, USA
8
reported being treated for their condition was determined and also used to estimate
Department of Dermatology & Allergology,
University Medical Center Utrecht, Utrecht, prevalence. Severity scales were Patient-Oriented SCORAD, Patient-Orientated
The Netherlands Eczema Measure, and Patient Global Assessment.
9
Sanofi, Chilly-Mazarin, France
Results: Among participants by region, the point prevalence of adult AD in the over-
Correspondence all/treated populations was 4.9%/3.9% in the US, 3.5%/2.6% in Canada, 4.4%/3.5% in
Sebastien Barbarot, MD, PhD, Service de
the EU, and 2.1%/1.5% in Japan. The prevalence was generally lower for males vs
Dermatologie, Centre Hospitalier
Universitaire de Nantes, Nantes, France. females, and decreased with age. Regional variability was observed within countries.
Email: [email protected]
Severity varied by scale and region; however, regardless of the scale or region, propor-
Funding information tion of subjects reporting severe disease was lower than mild or moderate disease.
This study was funded by Regeneron
Conclusions: Prevalence of adult AD ranged from 2.1% to 4.9% across countries.
Pharmaceuticals, Inc. and Sanofi.
Severe AD represented a small proportion of the overall AD population regardless
of measure or region.

KEYWORDS
atopic dermatitis, epidemiology, prevalence, severity

1 | INTRODUCTION interactions between skin barrier defects and immune dysregulation,

with recent evidence suggesting that it is a systemic disorder.1,2 There
Atopic dermatitis (AD) is a chronic, complex, often relapsing inflamma- is a consistent association of AD with other atopic and allergic condi-
tory skin disease. The clinical presentation of AD includes pruritus, tions including asthma and atopic rhinitis, often in a progression
xerosis, and eczematous lesions, and its pathology is characterized by known as the atopic march.3 The burden of illness associated with

----------------------------------------------------------------------------------------------------------------------------------------------------------------------
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any
medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
© 2018 The Authors. Allergy published by John Wiley & Sons Ltd.

1284 | wileyonlinelibrary.com/journal/all Allergy. 2018;1284–1293.

BARBAROT ET AL. | 1285

AD has been well characterized in the pediatric population, including countries (Kantar LightSpeed GMI, all countries except Japan; Research
recognition of the impact of AD on the family and caregivers as well Now, all countries except Japan; AIP, Japan; Netquest, Spain; Instantly,
as on the patients themselves.4,5 Similarly, among adult patients with US; and Asking Canadians, Canada). Recruitment was by broad-reach
AD, a multidimensional burden has been described that includes not portals, special interest sites, and direct emailing campaigns, and panel
only the skin symptoms associated with AD, but also sleep distur- members who completed the questionnaire received points redeem-
bances, impaired mental health, and reductions in quality-of-life and able for items in a prize catalogue. For inclusion, members of an online
work productivity.6-14 This burden is higher with greater AD sever- respondent panel were required to be 18–65 years old, inclusive, and
6,11,12,15-18
ity, which has been reported as moderate in 20%–37% and able to read and write the native country language.27
19,20
severe in 10%–34% of patients. To reduce selection bias, panelists were blinded to the research
The epidemiology of AD has focused on the pediatric popula- topic when invited. To ensure robustness of the data collected, the
tion.21 Limited information on the prevalence of AD among adults personal information provided was matched to a third-party data-
suggests variability that may be dependent on the population, dis- base to confirm validity; inactive e-mails and inactive members are
ease definitions, and methodology. The European Community Respi- regularly removed from the panels. The Internet Protocol address of
ratory Health Survey (ECRHS) study (N = 8206), which was based the respondent was verified against a known list of fraudulent ser-
on self-diagnosis in the adult population of 11 European countries vers to identify fraud at the time of registration. Individuals who
and the US, reported adult AD point prevalence rates that varied completed the survey in an unreasonably short time (<2 minutes for
from 0.3% (Switzerland) to 6.2% (Estonia).22 Among several US stud- those without self-reported AD; <4 minutes for those with self-
ies, reported point prevalence rates varied from 3.2% to 10.7% reported AD) were excluded. Individuals were excluded if their
depending on the population evaluated and the definition used.23-25 responses on severity scales lacked consistency (ie, very low scores
In the Japanese population, the point prevalence of adult AD was on 1 scale and very high scores on another). Interim quality checks
estimated to be 2.9%, with 1-year and lifetime prevalence rates of were conducted on the data after 1000, 2000, and 3000 respon-
3.0% and 3.3%, respectively.26 dents were recruited; in particular, consistency in the distribution of
The above epidemiologic data indicate distinct gaps in our AD severity was confirmed in panels within and across countries.
knowledge of the prevalence of AD in adults. Challenges that have The survey was conducted during the same period across all
contributed to these gaps include inconsistency in applying diagnos- countries (February 29 through April 13, 2016), and the maximum
tic criteria to epidemiologic evaluation, lack of a universally accepted total duration for questionnaire completion was 15 minutes. The
measure for assessment of severity, and less than adequate repre- questionnaire was administered in the native language of each coun-
sentativeness of the samples that would enable generalizability to try, and the outcome measures that were included were validated
the broader population (eg, the US sample in the ECRHS study was translations as made available by the developer of the measure.
represented by only a single city).22
The primary objective of this study was to fill these gaps by pro-
2.2 | Questionnaire and outcomes
viding global data on the prevalence of AD in representative samples
of adults from different countries using standardized diagnostic crite- The questionnaire consisted of 5 sections, of which the first cap-
ria and consistently applying established methods from previous tured demographic characteristics. The second section elicited infor-
studies. A secondary objective was to enable comparative and robust mation to determine the presence of AD using an automated
estimation of AD prevalence by disease severity in each country diagnosis algorithm that assessed whether the subjects were “non-
using validated patient-reported outcomes. AD” or “AD” (provided as online Supplementary material), and also
included questions on the presence of comorbid atopic conditions
(asthma, hay fever, chronic rhinosinusitis, food allergies, and allergic
2 | METHODS
keratoconjunctivitis) and family history of AD. Only subjects who
were classified as “AD” based on the algorithm were allowed to
2.1 | Study design
report disease severity. Sections 3–5 consisted of validated measures
This multinational, cross-sectional study was designed to represent for the assessment of disease severity.
the general populations of the US, Canada, European Union (EU) Table 1 shows the algorithm used for self-report of the presence
(France, Germany, Italy, Spain, United Kingdom [UK]), and Japan. of AD, which was based on questions from the International Study
Data collection was according to ethical codes of the European Soci- of Asthma and Allergies in Childhood (ISAAC)28 and the UK Working
ety for Opinion and Marketing Research (ESOMAR) and European Party criteria29,30 modified for self-completion. ISAAC has been vali-
Pharmaceutical Market Research Association (EphMRA), and was dated for population-based studies, and while the UK Working Party
compliant with the US Health Insurance Portability and Accountabil- criteria were validated for clinician use, the self-completed version
ity Act (HIPAA) of 1996; all subjects provided informed consent prior has not yet been validated. This combination was used as ISAAC
to participation. alone has variable sensitivity and specificity across populations,31
Data were collected through a web-based survey among subjects and it was considered that addition of the UK Working Party criteria
who were members of online respondent panels in their respective would provide more rigorous identification of AD.
1286 | BARBAROT ET AL.

T A B L E 1 UK Working Party criteria modified for self-diagnosis of atopic dermatitis

Requirement for
Criterion Question Origin AD identification
History of pruritic skin • Have you ever had an itchy rash which was coming and going for at least ISAAC21 Mandatory
condition 6 months?
• Have you had this itchy rash in the past 12 months?
History of dry skin In the last year, have you suffered from a dry skin in general? UK Working Party29,30 Must meet at
History of asthma Do you suffer or have you ever suffered from asthma (bouts of wheezing UK Working Party29,30 least 3 of 5
with coughing)?
Age of rash onset < 2 y At what age did this itchy rash first occur? ISAAC21
Flexural dermatitis Has this itchy rash at any time affected any of the following places: the ISAAC21
folds of the elbows, behind the knees, in front of the ankles, under the
buttocks or around the neck, ears, or eyes?
Visible dermatitis Questions based on pictures:
Have you observed signs of erythema/induration, papulation or edema/
lichenification/oozing and crusting on your skin in the past 12 months?

The question on visible dermatitis asked about the presence of Quota apportionment was used prior to data collection to ensure
erythema, induration, papulation, edema, lichenification, oozing, and that sampled subjects were representative of the general adult popu-
crusting with the assistance of pictures from the Eczema Area and lation of the countries.37 Hard quotas were set for gender,
32
Severity Index (EASI) scale. Subjects were identified with AD on age-group, and region, with soft quotas for current occupation of
the basis of giving a mandatory answer of “yes” to the questions the head of household (France, the UK, Italy, Spain, Japan) and
“Have you ever had an itchy rash which was coming and going for income level (Germany, US, Canada). A minor weighting adjustment
at least 6 months” and “Have you had this itchy rash in the past 12 was applied at the country level where some deviations have been
months?,” plus a positive response to at least 3 of 5 other criteria observed between the soft quota objectives and the final sample
(Table 1) and were required to have self-reported a physician diag- structure.
nosis of AD based on the question: “Have you ever been diagnosed Populations were characterized with regard to demographics and
with atopic dermatitis or atopic eczema by a doctor?” AD severity using descriptive statistics. The point prevalence of AD
Disease severity was assessed using 3 validated measures includ- was estimated based on meeting both of the individual criteria (ie,
ing the Patient-Oriented Scoring of Atopic Dermatitis (PO- subjects both positive to modified UK Working Party criteria in the
SCORAD),33 the Patient-Orientated Eczema Measure (POEM),34 and past 12 months and reporting having ever received a physician’s
the Patient Global Assessment (PGA). On PO-SCORAD, severity was diagnosis of AD). Additionally, prevalence was estimated using a
based on 0–24, 25–49, and ≥50, which by consensus represent the prevalent population defined as subjects who met the UK Working
severity of mild, moderate, and severe, respectively.35 Severity thresh- Party criteria and received a physician diagnosis who also reported
old scores on POEM were 0-7 (mild), 8-16 (moderate), and >16 (sev- being treated for their AD.
36
ere); subjects self-reported PGA as mild, moderate, or severe. Differences between countries were evaluated using bivariate
A series of stand-alone questions were also included on the spe- analyses with Z-tests for categorical variables, and paired-sample t
cialties of the physicians managing AD; consultations with healthcare tests for continuous variables with 95% confidence intervals (CIs). All
professionals for AD treatment; whether the subject has sought analyses were conducted using DAISIE version 2.4.25 (ADN, Paris,
treatment and is currently taking medication; AD medications used; France).
and consumption of tobacco and alcohol (not reported here).

3 | RESULTS
2.3 | Statistical analysis
3.1 | Populations
Based on the assumption of a 3% prevalence, a sample size of
20 000 US subjects and 10 000 in each of the other countries was The attrition flow resulting in the final sampled populations is shown
determined to provide a prevalence estimate with a precision of in Figure S1. The demographic characteristics of these populations
0.24% (US) and 0.33% (other countries). This prevalence assump- (Table 2) were representative of the individual countries. Regional
tion would also enable identification of 600 and 300 AD subjects in distributions within each country reflected the study goals as being
the US and other countries, respectively, for assessment of severity representative of each country, as did the ranges of income levels
with a precision of 4.0% and 5.7%. (data not shown).
BARBAROT ET AL. | 1287

T A B L E 2 Demographic characteristics of the sampled populations by country

a) US b) Canada c) France d) Germany e) Italy f) Spain g) UK h) Japan
Variable (n = 19 986) (n = 10 004) (n = 9964) (n = 9971) (n = 9897) (n = 9924) (n = 10 001) (n = 10 911)
Sex (%)
Male 38.5 39.1 45.5 47.8 46.5 49.8 44.5 48.8
Female 61.5 60.9 54.5 52.2 53.5 50.2 55.5 51.2
Age range, %
18–24 y 15.6 9.6 13.1 12.3 11.9 9.0 11.8 14.6
25–34 y 20.1 18.2 20.1 20.7 20.4 21.3 22.2 17.2
35–44 y 21.1 19.5 20.7 19.1 26.5 28.5 22.0 23.9
45–54 y 23.6 27.9 24.1 26.3 25.6 24.3 24.1 22.2
55–64 y 19.5 24.9 22.0 21.6 15.6 16.9 19.9 22.1

F I G U R E 1 Prevalence of adult atopic

dermatitis in the sampled populations by
country. A significantly higher point
prevalence vs other countries is indicated
by the superscript letters (P < .05)

prevalence (5.1% vs 4.6%) (Figure 2A). Spain had the greatest differ-
3.2 | Prevalence
ence between sexes, 9.3% among females vs 5.1% among males
In the overall population, the 12-month adult prevalence of AD (P < .05). The prevalence of AD was generally lower among older
was 4.9% (95% confidence interval [CI]: 4.6%, 5.2%) in the US, age-groups (P < .05) with a peak prevalence most frequently
3.5% (95% CI: 3.1%, 3.9%) in Canada, 4.4% (95% CI: 4.2%, 4.6%) observed in the 25- to 34-year or 35- to 44-year age-groups and
in the EU with individual country ranges of 2.2% (95% CI: 1.9%, decreasing prevalence in the 45- to 54-year and 55- to 64-year age-
2.5%) for Germany to 8.1% (95% CI: 7.5%, 8.6%) for Italy, and groups (Figure 2B).
2.1% (95% CI: 1.8%, 2.3%) in Japan (Figure 1). There were regio- As shown in Table 3 regarding the presence of the individual UK
nal differences across countries within Europe (Figure 1) as well Working Party criteria in the overall populations, dry skin was the
as across regions within a country (Table S1). A higher point most frequently reported criterion (38.8%–65.5%) except in France
prevalence was observed in southern European countries (Italy (27.0%), where itchy skin had a higher prevalence (28.9%). Itchy skin,
and Spain) compared with the other countries (P < .05) (Figure 1). generally the second most frequent criterion, was highest in Italy
In the US, the Midwest region was associated with the lowest (46.4%) and was significant vs the other countries (all P < .05).
prevalence, and the difference was significant relative to the other Asthma was highest in Spain (26.0%; P < .05 vs all other countries)
regions (Table S1). and lowest in Japan (10.5%). While the onset of rash <2 years of
Positivity on the UK Working Party criteria ranged from 4.3% age was low overall, the highest proportion was in France (9.6%;
(Japan) to 16.7% (Italy) and was higher than the proportion of P < .05 vs all other countries).
subjects who reported being diagnosed by a physician for all
countries except Spain and Japan (Figure 1). The proportion of
3.3 | Severity
subjects who reported having been diagnosed by a physician was
<10% except for the US (10.6%), Italy (12.4%), and Spain (17.6%) Differences in severity distribution were observed across the scales
(Figure 1). and across countries (Figure 3). Although the PGA consistently
Females had a higher AD prevalence except in the UK, where resulted in the lowest proportions of severe (2%–8%) relative to PO-
the prevalence was the same in males and females (2.5%), and in the SCORAD (10%–21%) and POEM (8%–17%), severe AD was generally
US, where males had a numerically but not significantly higher stable within a particular scale across the countries. The US had the
1288 | BARBAROT ET AL.

F I G U R E 2 Point prevalence by
demographic characteristics. (A) Sex.
(B) Age

T A B L E 3 Presence of the individual items of the UK Working Party criteria modified for self-completion in the overall adult populations. A
significantly higher prevalence of the criterion vs other countries is indicated by the superscript letters (P < .05)
a) US b) Canada c) France d) Germany e) Italy f) Spain g) UK h) Japan
Criterion (n = 19 986) (n = 10 004) (n = 9964) (n = 9971) (n = 9897) (n = 9924) (n = 10 001) (n = 10 911)
Onset of rash < 2 y old 3.0dh 3.8adfgh 9.6abdefgh 0.9 3.8adfgh 2.6dh 2.9dh 1.4d
Itchy skin condition in past 12 mo 22.6dg 24.4adg 28.9abdfgh 18.6 46.4abcdfgh 23.8adg 21.2d 25.3adfg
bdgh gh abdgh gh abcdfgh bdgh h
Flexural dermatitis 15.7 14.4 17.1 13.9 33.6 16.6 12.8 10.4
Visible dermatitis 18.4bdgh 16.7dgh 19.7abdgh 13.3 24.3abcdfgh 21.0abcdgh 14.2h 13.0
cdeh acdegh dh h dh abcdegh cdeh
Presence of asthma 19.4 21.4 16.9 12.8 16.9 26.0 19.1 10.5
Dry skin 62.2cefgh 61.1cefgh 27.0 62.5bcefgh 57.5cfh 47.2ch 57.2cfh 38.8c

highest proportion of subjects with severe AD regardless of scale.

3.4 | Diagnosis and management
France and the southern European countries Italy and Spain had
higher proportions of mild AD relative to the UK and Germany; Ger- Family practitioners/general practitioners were the primary diagnos-
many consistently had the lowest proportion of mild AD on each ing specialty in the UK (66%) and Canada (52%), in contrast to der-
scale. Severity ratings in Japan showed considerable variability matologists in all other countries (51%–72%) (Figure 4). Other
depending on the scale. specialties were scarcely represented (≤10%), except for pediatricians
BARBAROT ET AL. | 1289

F I G U R E 3 Atopic dermatitis severity by country based on different assessment scales among the prevalent population (ie, those who met
UK Working Party criteria and reported a physician’s diagnosis). PO-SCORAD, Patient-Oriented Scoring of Atopic Dermatitis; POEM, Patient-
Orientated Eczema Measure; PGA, Patient Global Assessment

F I G U R E 4 Specialty of physicians who

had made the diagnosis of atopic
dermatitis (AD) in subjects who reported
having been diagnosed with AD by a
physician

in Japan (16%). Among subjects positive for UK Working Party crite- the accuracy of self-report by incorporating both the validated
ria, less than half (48.2%) reported having received the diagnosis of ISAAC scale and a self-reported modification of the validated UK
AD from a physician. Among the prevalent AD subjects, that is, Working Party criteria. The dual use of these scales would be
those with meeting the UK Working Party criteria and received a expected to reduce misclassification and detection bias, and mini-
physician diagnosis of AD, the majority across countries reported mize issues associated with self-report that have previously been
treatment for their AD, which ranged from 69.3% in France to recognized,38 thereby also increasing the epidemiologic and clinical
80.2% in the UK (Figure 5). relevance of the study.
When subjects who met the AD criteria (ie, UK Working Party Results of the study show that individual country point preva-
criteria and reported a physician diagnosis) and who also reported lence of AD in adults, which range from 2.1% (Japan) to 8.1%
being treated for their disease were considered as the prevalent (Italy), is substantial and is consistent with the 2%–10% reported
population, the prevalence of adult AD was 3.9% in the US, 2.6% in by the World Allergy Organization.20 However, the observed rates
Canada, 3.5% in the EU, and 1.5% in Japan. are higher than the 12-month prevalence rates reported among
specific countries in the multinational ECRHS,22 especially in
southern Europe and the US. These disparities may reflect
4 | DISCUSSION methodologic differences, including the limited study sites evalu-
ated in the ECRHS that in some cases were restricted to only a
This study provides comparative estimates of the adult AD preva- few regions of the countries surveyed (eg, the US and Germany
lence from several industrialized countries in North America, Europe, were only represented by 1 and 2 cities, respectively), and reliance
and Asia. Notably, the sample populations were large and represen- in ECRHS only on the ISAAC questionnaire to identify the popula-
tative of the individual countries by virtue of the use of demographic tion, which was smaller and more selected that also enabled clini-
quotas. As such, this represents the first international study to cal evaluation. Clinical evaluation is impractical in a larger
gather AD prevalence data in adults stratified by severity from gen- multinational study of representative populations such as those
eral populations. The study was also specifically designed to improve evaluated in the current study.
1290 | BARBAROT ET AL.

F I G U R E 5 Proportion of subjects
meeting UK Working Party criteria and
received a physician diagnosis who
reported being treated for their atopic
dermatitis

The estimated point prevalence in Japan (2.1%) and Germany well as extrinsic factors that vary across regions within a country such
39,40
(2.2%) was comparable to previous country-specific studies. In as behavioral or cultural dynamics and climatic elements. However, in
the US, the point prevalence of 4.9% was slightly higher than the countries where the prevalence of AD was higher, the proportion of
3.2% population prevalence reported in a study using the criteria of severe forms was generally much lower, with a tendency to equalize
eczema in conjunction with having a history of asthma and/or a 1- the burden of severe AD among countries. The number of confound-
year history of hay fever as a proxy for AD.24 The difference ing factors increases the complexity of explaining regional differences.
between these 2 numbers may be accounted for by the fact that the The results reported here also confirm previous correlations of
previous study relied on the presence of asthma or hay fever to con- AD prevalence with sex and age;39,40,43 prevalence was higher
firm AD diagnostics. However, when the current study considered among females relative to males except in the US and was generally
treated subjects (79.4%; n = 773) as the prevalent population, the lower among older age-groups across all countries.
prevalence is 3.9% (ie, 773/19 986), a percentage point lower than Mild or moderate severity were the most common individual
the overall point prevalence (4.9%) and closer to the 3.2% reported severity presentations, with generally low proportions of severe AD.
previously. Similarly, the point prevalence in Italy, 8.1%, was higher While other studies have reported moderate AD to be the most
than the prevalence of 3.1% based on using a proxy of eczema with common presentation,19,20,44 2 studies have reported a high preva-
concurrent asthma and/or hay fever, and which also reported a high lence of severe AD, 51% and 55% in a UK and an Italian study,
rate of adult onset.41 In regard to adult onset, it should be noted respectively, although the former used alternate cutoff scores and
that the current data did not enable assessment of the age of dis- the latter used a different assessment scale.44,45
ease onset. Additionally, the generally low report of rash onset The observed distributions according to severity grading were
<2 years old (0.9%–9.6%) may reflect, at least in part, subjects not dependent on the scales used, likely reflecting the fact that they do
remembering their childhood disease. not necessarily assess the same AD-related constructs. The consis-
Higher point prevalence in Italy (8.1%) and Spain (7.2%) relative tently higher proportions of severe AD on PO-SCORAD and POEM
to other countries (2.1%–4.9%) may reflect differences in presenta- relative to PGA likely reflect what is being measured. Both PO-
tion and diagnosis as indicated by the observed differences in pro- SCORAD and POEM rely on a more clinical and granular approach
portions of subjects endorsing the individual UK Working Party to assessing specific symptoms, including itch that substantially con-
criteria; subjects in Italy reported a higher rate of itchy skin and tributes to the severity rating, and effects that also include nonskin
those in Spain had a higher rate of asthma. Furthermore, both coun- symptoms such as sleep disturbance. The PGA, in contrast, uses a
tries had the highest proportions of subjects who reported having more holistic approach, which may underestimate the impact of itch,
been diagnosed by a physician, which may reflect a greater access to and is also likely to introduce variability and limitations similar to
health care that could have also potentially contributed to the higher what has been described for Investigator’s Global Assessment mea-
prevalence in these countries. sures.46 However, the epidemiologic and clinical relevance of this
Similar to what has previously been suggested by results from the holistic approach represented by the PGA should also be considered,
22
ECRHS, some regional variability in adult AD prevalence was as it may be informative of individuals who are more likely to seek
observed within each country, with Italy and Spain having the highest medical attention because of poorer global health status.
inter-region variability. The regional variability for Italy was also in The greatest variability in distribution across scales was observed
concordance with a previous study that showed higher prevalence in Japan, and this may indicate potential issues in the comprehensi-
rates in cities in Mediterranean regions relative to those in a more bility of the questions. There is currently no standardized patient-
continental (northern) climate.41 While regional variability is not well reported measure of AD severity, and thus, the observed variability
understood, several variables may be proposed as factors that may further emphasizes the need for such standardization, as has been
contribute to the observed variability, including genetic factors,42 as recommended by the Harmonizing Outcome Measures for Eczema
BARBAROT ET AL. | 1291

(HOME) initiative for defining and assessing AD.47 Of note, HOME CONFLICTS OF INTEREST
has also endorsed POEM as the core outcome instrument for mea-
S. Barbarot has received research grants from Pierre Fabre Labora-
suring patient-reported symptoms.48
tory and Fondation pour la dermatite atopique; personal fees from
Geographic differences were observed with regard to severity
Bioderma, Laboratoire La Roche Posay, Sanofi-Genzyme, Novalac,
prevalence. Milder disease was observed among countries that have
Ferring; and nonfinancial support from Abbvie, Novartis, Janssen. A.
generally a similar latitude with more uniform hours of daylight
Gadkari is an employee of and stockholder in Regeneron Pharmaceu-
exposure and an overall continental climate (France, Italy, Spain, and
ticals, Inc. S. Auziere is an employee of Kantar Health, who received
Japan) relative to a greater prevalence of moderate and severe sub-
funding from Sanofi to conduct the study. E. Simpson’s institution
jects in countries with a more northerly (Germany, UK, Canada) or
has received grants/research funding from Amgen, Inc, Anacor Phar-
diverse (US) range of latitudes and climate. This is consistent with a
maceuticals Inc., Celgene Corporation, Chugai Pharma US, LLC, Eli
review of published data that indicated the low humidity and low
Lilly and Company, Galderma Research and Development, Genen-
temperatures of more northerly climates not only lead to a decrease
tech Inc., MedImmune LLC, Novartis Pharmaceuticals Corporation,
in skin barrier function but also an increased prevalence and risk of
Pfizer Inc., Regeneron Pharmaceuticals, Inc., Sanofi, Tioga Pharma-
AD flares.49
ceuticals Inc, and is a consultant for Regeneron Pharmaceuticals, Inc.,
Sanofi, Anacor, Celgene Corporation, Galderma Research and Devel-
4.1 | Strengths and limitations opment, Genentech, MedImmune LLC, Pfizer Inc., AbbVie, Dermira,
and Valeant. M. de Bruin-Weller is a consultant for Regeneron Phar-
Strengths of this study are its large population and selection of sub-
maceuticals, Inc., Sanofi Genzyme; an advisory board member for
jects for broad representation of the populations and regions of each
AbbVie, Anacor, Regeneron Pharmaceuticals, Inc., Sanofi Genzyme;
country, providing external validity and greater generalizability, and
and principal investigator for AbbVie, Regeneron Pharmaceuticals,
its use of validated measures for identifying AD and its severity.
Inc., Roche, Sanofi Genzyme. G. Girolomoni has been principal inves-
Additionally, this study provides a consistent method of measuring
tigator in clinical trials sponsored by and/or and has received per-
AD prevalence across continents. Selection bias represents a limita-
sonal fees from AbbVie, Abiogen, Almirall, Amgen, Bayer, Biogen,
tion that was manifested by the use of an online survey, which pre-
Celgene, Eli-Lilly, Galderma, Hospira, Janssen, Leo Pharma, Merck,
supposes computer literacy and Internet access. Selection bias may
MSD, Mundipharma, Novartis, Pfizer, Pierre Fabre, Regeneron Phar-
also result from potential differences between subjects who agreed
maceuticals, Inc., Samsung, Sandoz, Sanofi, and Sun Pharma. L. Puig
to participate and those who did not. Another limitation is that older
has received consultancy honoraria from and participated in clinical
subjects (ie, >65 years) were excluded. Online panel members in this
trials sponsored by Regeneron Pharmaceuticals, Inc., and Sanofi. L.
older age-group may not necessarily have been representative of the
Eckert is an employee of and stockholder in Sanofi.
general population given access to the Internet is generally lower for
this age-group. As the study is based on self-report, another limita-
tion is recall bias. AUTHOR CONTRIBUTIONS

The authors were responsible for all content and editorial decisions
and received no honoraria related to the development of this publi-
5 | CONCLUSIONS
cation. All authors had full access to all the data in this study and
take complete responsibility for the integrity of the data and accu-
A population-based comparison of prevalence using validated meth-
racy of the data analysis. All named authors meet the International
ods shows an adult AD prevalence generally within a fairly narrow
Committee of Medical Journal Editors (ICMJE) criteria for authorship
range across North America, Europe, and Japan (2.1%-4.9%),
for this manuscript, take responsibility for the integrity of the work
although variations were observed across countries as well as across
as a whole, and have given final approval for the version to be
regions within a country. Except for a higher prevalence among
published.
males in the US, the association of AD with female sex was con-
firmed, as well as the lower prevalence in older age-groups. The
severity distribution was observed to vary based on the outcome ORCID
measure used, suggesting a need for standardization of severity
assessment. Patient Global Assessment consistently provided the S. Barbarot http://orcid.org/0000-0002-6629-9100
lowest rates of the severe AD (2%–8%), which formed a small pro-
portion of the overall AD population regardless of scale or region.
REFERENCES

1. Gittler JK, Shemer A, Suarez-Farinas M, et al. Progressive activation

ACKNOWLEDGMENTS
of T(H)2/T(H)22 cytokines and selective epidermal proteins charac-
Medical writing support was provided by E. Jay Bienen, PhD, and terizes acute and chronic atopic dermatitis. J Allergy Clin Immunol.
2012;130:1344-1354.
funded by Sanofi and Regeneron Pharmaceuticals.
1292 | BARBAROT ET AL.

2. Brunner PM, Silverberg JI, Guttman-Yassky E, et al. Increasing 21. Suarez-Varela MM, Alvarez LG, Kogan MD, et al. Diet and preva-
comorbidities suggest that atopic dermatitis is a systemic disorder. J lence of atopic eczema in 6 to 7-year-old schoolchildren in Spain:
Invest Dermatol. 2016;137:18-25. ISAAC phase III. J Investig Allergol Clin Immunol. 2010;20:469-475.
3. Dharmage SC, Lowe AJ, Matheson MC, Burgess JA, Allen KJ, Abram- 22. Harrop J, Chinn S, Verlato G, et al. Eczema, atopy and allergen expo-
son MJ. Atopic dermatitis and the atopic march revisited. Allergy. sure in adults: a population-based study. Clin Exp Allergy.
2014;69:17-27. 2007;37:526-535.
4. Balkrishnan R, Housman TS, Grummer S, et al. The family impact of 23. Hanifin JM, Reed ML. A population-based survey of eczema preva-
atopic dermatitis in children: the role of the parent caregiver. Pediatr lence in the United States. Dermatitis. 2007;18:82-91.
Dermatol. 2003;20:5-10. 24. Silverberg JI, Hanifin JM. Adult eczema prevalence and associations
5. Chamlin SL, Chren MM. Quality-of-life outcomes and measurement with asthma and other health and demographic factors: a US popula-
in childhood atopic dermatitis. Immunol Allergy Clin North Am. tion-based study. J Allergy Clin Immunol. 2013;132:1132-1138.
2010;30:281-288. 25. Silverberg JI, Garg NK, Paller AS, Fishbein AB, Zee PC. Sleep distur-
6. Fivenson D, Arnold RJ, Kaniecki DJ, Cohen JL, Frech F, Finlay AY. bances in adults with eczema are associated with impaired overall
The effect of atopic dermatitis on total burden of illness and quality health: a US population-based study. J Invest Dermatol. 2015;135:
of life on adults and children in a large managed care organization. J 56-66.
Manag Care Pharm. 2002;8:333-342. 26. Muto T, Hsieh SD, Sakurai Y, et al. Prevalence of atopic dermatitis
7. Barbeau M, Lalonde H. Burden of Atopic dermatitis in Canada. Int J in Japanese adults. Br J Dermatol. 2003;148:117-121.
Dermatol. 2006;45:31-36. 27. DiBonaventura MD, Wagner JS, Yuan Y, L’Italien G, Langley P, Ray
8. Dalgard FJ, Gieler U, Tomas-Aragones L, et al. The psychological bur- Kim W. Humanistic and economic impacts of hepatitis C infection in
den of skin diseases: a cross-sectional multicenter study among der- the United States. J Med Econ. 2010;13:709-718.
matological out-patients in 13 European Countries. J Invest Dermatol. 28. Asher MI, Keil U, Anderson HR, et al. International Study of Asthma
2015;135:984-991. and Allergies in Childhood (ISAAC): rationale and methods. Eur Respir
9. Bender BG, Leung SB, Leung DY. Actigraphy assessment of sleep J. 1995;8:483-491.
disturbance in patients with atopic dermatitis: an objective life qual- 29. Williams HC, Burney PG, Hay RJ, et al. The U.K. Working Party’s
ity measure. J Allergy Clin Immunol. 2003;111:598-602. Diagnostic Criteria for Atopic Dermatitis. I. Derivation of a minimum
10. Wittkowski A, Richards HL, Griffiths CE, Main CJ. Illness perception set of discriminators for atopic dermatitis. Br J Dermatol.
in individuals with atopic dermatitis. Psychol Health Med. 2007;12: 1994;131:383-396.
433-444. 30. Williams HC, Burney PG, Pembroke AC, Hay RJ. Validation of the
11. Yano C, Saeki H, Ishiji T, et al. Impact of disease severity on sleep U.K. diagnostic criteria for atopic dermatitis in a population setting.
quality in Japanese patients with atopic dermatitis. J Dermatol Sci. U.K. Diagnostic Criteria for Atopic Dermatitis Working Party. Br J
2013;72:195-197. Dermatol. 1996;135:12-17.
12. Yano C, Saeki H, Ishiji T, et al. Impact of disease severity on work 31. Strina A, Barreto ML, Cunha S, et al. Validation of epidemiological
productivity and activity impairment in Japanese patients with atopic tools for eczema diagnosis in Brazilian children: the ISAAC’s and UK
dermatitis. J Dermatol. 2013;40:736-739. Working Party’s criteria. BMC Dermatol. 2010;10:11.
13. Simpson EL, Bieber T, Eckert L, et al. Patient burden of moderate- 32. Hanifin JM, Thurston M, Omoto M, Cherill R, Tofte SJ, Graeber M.
to-severe atopic dermatitis: insights from a phase 2b clinical trial of The eczema area and severity index (EASI): assessment of reliability
dupilumab in adults. J Am Acad Dermatol. 2016;74:491-498. in atopic dermatitis. EASI Evaluator Group. Exp Dermatol.
14. Whiteley J, Emir B, Seitzman R, Makinson G. The Burden of Ato- 2001;10:11-18.
pic Dermatitis in U.S. Adults: results from the 2013 National 33. Stalder JF, Barbarot S, Wollenberg A, et al. Patient-Oriented
Health and Wellness Survey. Curr Med Res Opin. 2016;32:1645- SCORAD (PO-SCORAD): a new self-assessment scale in atopic der-
1651. matitis validated in Europe. Allergy. 2011;66:1114-1121.
15. Haeck IM, ten Berge O, van Velsen SG, de Bruin-Weller MS, Brui- 34. Charman CR, Venn AJ, Williams HC. The patient-oriented eczema
jnzeel-Koomen CA, Knol MJ. Moderate correlation between quality measure: development and initial validation of a new tool for mea-
of life and disease activity in adult patients with atopic dermatitis. J suring atopic eczema severity from the patients’ perspective. Arch
Eur Acad Dermatol Venereol. 2012;26:236-241. Dermatol. 2004;140:1513-1519.
16. Kim DH, Li K, Seo SJ, et al. Quality of life and disease severity are 35. Wollenberg A, Oranje A, Deleuran M, et al. ETFAD/EADV Eczema
correlated in patients with atopic dermatitis. J Korean Med Sci. task force 2015 position paper on diagnosis and treatment of atopic
2012;27:1327-1332. dermatitis in adult and paediatric patients. J Eur Acad Dermatol
17. Simpson E, Guttman-Yassky E, Margolis DJ, et al. Chronicity, comor- Venereol. 2016;30:729-747.
bidity and life course impairment in atopic dermatitis: insights from a 36. Charman CR, Venn AJ, Ravenscroft JC, Williams HC. Translating
cross-sectional study in US Adults. Poster presented at the 25th Patient-Oriented Eczema Measure (POEM) scores into clinical prac-
European Academy of Dermatology and Venereology, September tice by suggesting severity strata derived using anchor-based meth-
28–October 2, 2016, Vienna, Austria. 2016. ods. Br J Dermatol. 2013;169:1326-1332.
18. Guttman-Yassky E, Simpson E, Margolis D, et al. Patient-reported 37. Deville JC. A Theory of Quota Surveys. Surv Methodol. 1991;17:163-
disease burden in adults with atopic dermatitis: a US cross-sectional 181.
study. Poster presented at the 25th Congress of the European Acad- 38. Mortz CG, Andersen KE, Dellgren C, Barington T, Bindslev-Jensen C.
emy of Dermatology and Venereology, Vienna, Austria, 28 Septem- Atopic dermatitis from adolescence to adulthood in the TOACS cohort:
ber-2 October 2016. 2016. prevalence, persistence and comorbidities. Allergy. 2015;70:836-845.
19. Kim JP, Chao LX, Simpson EL, Silverberg JI. Persistence of atopic 39. Saeki H, Tsunemi Y, Fujita H, et al. Prevalence of atopic dermatitis
dermatitis (AD): a systematic review and meta-analysis. J Am Acad determined by clinical examination in Japanese adults. J Dermatol.
Dermatol. 2016;75:681-687. 2006;33:817-819.
20. The World Allergy Organization (WAO). The World Allergy 40. Wolkewitz M, Rothenbacher D, Low M, et al. Lifetime prevalence of
Organization (WAO) White Book on Allergy 2013 Update. 2013. self-reported atopic diseases in a population-based sample of elderly
http://www.worldallergy.org/UserFiles/file/WhiteBook2-2013-v8.pdf subjects: results of the ESTHER study. Br J Dermatol. 2007;156:693-
. Accessed on December 18, 2017. 697.
BARBAROT ET AL. | 1293

41. Pesce G, Marcon A, Carosso A, et al. Adult eczema in Italy: preva- measures for atopic eczema/dermatitis clinical trials (HOME initia-
lence and associations with environmental factors. J Eur Acad Derma- tive). Br J Dermatol. 2016;175:69-79.
tol Venereol. 2015;29:1180-1187. 49. Engebretsen KA, Johansen JD, Kezic S, Linneberg A, Thyssen JP.
42. Liang Y, Chang C, Lu Q. The Genetics and Epigenetics of Atopic The effect of environmental humidity and temperature on skin bar-
Dermatitis-Filaggrin and Other Polymorphisms. Clin Rev Allergy rier function and dermatitis. J Eur Acad Dermatol Venereol.
Immunol. 2016;51:315-328. 2016;30:223-249.
43. Ronmark EP, Ekerljung L, Lotvall J, et al. Eczema among adults: preva-
lence, risk factors and relation to airway diseases. Results from a large-
scale population survey in Sweden. Br J Dermatol. 2012;166:1301-1308.
44. Zeppa L, Bellini V, Lisi P. Atopic dermatitis in adults. Dermatitis. SUPPORTING INFORMATION
2011;22:40-46.
Additional Supporting Information may be found online in the
45. Baron SE, Morris PK, Dye L, Fielding D, Goulden V. The effect of
dermatology consultations in secondary care on treatment outcome supporting information tab for this article.
and quality of life in new adult patients with atopic dermatitis. Br J
Dermatol. 2006;154:942-949.
46. Futamura M, Leshem YA, Thomas KS, Nankervis H, Williams HC,
Simpson EL. A systematic review of Investigator Global Assessment How to cite this article: Barbarot S, Auziere S, Gadkari A,
(IGA) in atopic dermatitis (AD) trials: many options, no standards. J et al. Epidemiology of atopic dermatitis in adults: Results
Am Acad Dermatol. 2016;74:288-294. from an international survey. Allergy. 2018;73:1284–1293.
47. Schmitt J, Spuls P, Boers M, et al. Towards global consensus on out-
https://doi.org/10.1111/all.13401
come measures for atopic eczema research: results of the HOME II
meeting. Allergy. 2012;67:1111-1117.
48. Chalmers JR, Simpson E, Apfelbacher CJ, et al. Report from the
fourth international consensus meeting to harmonize core outcome