British Journal of Dermatology 2001; 145: 258±263.

Sensitive skin: an epidemiological study

C.M.WILLIS, S.SHAW, O.DE LACHARRIEÁ RE,* M.BAVEREL,² L.REICHE, R.JOURDAIN,*
P.BASTIEN* AND J.D.WILKINSON
Department of Dermatology, Amersham Hospital, Amersham, Bucks HP7 0JD, U.K.
*L'OreÂal Research, Life Science, Prospective Clinical Research, Centre Charles Zviak, Clichy, France
²L'OreÂal Research, Applied Research, Centre Eugene Schuller, Clichy, France

Accepted for publication 28 March 2001

Summary Background There is a growing awareness that some individuals exhibit heightened skin sensitivity,
particularly on the face, and have a high incidence of adverse reactions to cosmetics and toiletries.
Objectives To carry out an epidemiological study to assess the prevalence of sensitive skin and
cosmetic-related adverse events in a U.K. population, and to examine possible factors that may be
associated with sensitive skin.
Methods Self-assessment questionnaires were sent out to 3300 women and 500 men, randomly
selected, who were over the age of 18 years and lived within a 10-mile radius of High Wycombe
(Bucks.). Fifty non-responder women were also questioned by telephone to ensure that the postal
responders were representative of the population as a whole.
Results The response rates were 62% for women and 52% for men, with the incidence of self-
reported skin sensitivity being 51´4% and 38´2%, respectively. Ten per cent of women and 5´8% of
men described themselves as having very sensitive skin. Fifty-seven per cent of women and 31´4% of
men had experienced an adverse reaction to a personal product at some stage in their lives, with
23% of women and 13´8% of men having had a problem in the last 12 months. Among the women,
symptoms of cosmetic-induced subjective sensory skin discomfort (burning, stinging, itching etc.)
occurred more commonly in the sensitive skin cohort (53%) than in those who regarded themselves
as non-sensitive (17%). An atopic diathesis in women did not appear to be a predictive factor for
sensitive skin, the incidence of self-perceived sensitive skin being equivalent for atopics (49%) and
non-atopics (51%). Furthermore, some 34% of atopic women described themselves as being non-
sensitive. Nevertheless, the incidence of atopy was higher among the women in the sensitive skin
group (49%) than among those in the non-sensitive group (27%). Dry skin and a predilection for
blushing/flushing were associated factors for sensitive skin.
Conclusions Our survey indicates that sensitive facial skin is a common problem for women and
men in the U.K. and points to the need for the development of personal products designed for this
skin phenotype.
Key words: atopy, cosmetics, epidemiology, sensitive skin

The concept of `sensitive skin' is one that is widely
appreciated by consumers and manufacturers of
personal products, and is becoming increasingly
recognized by dermatologists. Based on our current
understanding of the condition, we would define
sensitive skin as being that which exhibits a reduced
tolerance to frequent or prolonged use of cosmetics and
toiletries, with symptoms ranging from visible signs of
irritation such as erythema and scaling, through to
Correspondence: Dr Carolyn M.Willis.
E-mail: carolynwillis@sbnhst.ftech.co.uk
more subjective neurosensory forms of discomfort such
as stinging, burning, itching, tightness and smarting.1±3
Essentially a self-diagnosed condition, the level and
nature of symptoms needed to produce the self-
perception of sensitivity are difficult to quantify, but
will almost certainly be highly variable between
individuals. In 1991, Mills and Berger4 attempted to
categorize sensitive skin phenotypes, but acknowledged
that, at present, we have little understanding as to why
some people appear to have sensitive skin and/or react
adversely to cosmetics, while others do not. Certainly,

258 q 2001 British Association of Dermatologists


some pre-existing skin conditions do appear to pre-
dispose to sensitive skin, atopic eczema being arguably
the best known.3,4
Despite the significance of sensitive skin, particularly
to the manufacturers of personal products, little
epidemiological evidence exists with respect to its
prevalence. The limited data available suggest that
adverse reactions to personal products are relatively
common, epidemiological studies from the U.K. and
the Netherlands, respectively, reporting 1- and 5-year
prevalence figures of about 12%.5,6 Therefore, in order
to assess the scale of the problem more accurately and
to improve our understanding of the factors that may
be associated with sensitive skin, we undertook an
epidemiological study within the Amersham/High
Wycombe (Bucks., U.K.) local population. The
approach adopted was that of a postal self-assessment
questionnaire sent out to individuals randomly selected
from the electoral register. Both men and women were
targeted, with a variety of questions being asked about
skin and scalp sensitivity, as well as history of atopy and
other characteristics, which may relate to sensitive skin.
Subjects and methods
Survey
After receiving approval from the Wycombe Local
Research Ethics Committee, a questionnaire was posted
to 3300 women and 500 men over the age of 18 years
who lived within a 10-mile radius of High Wycombe.
All subjects were randomly selected from the local
Figure 1. Bar chart showing the frequencies
with which the responding populations of
men and women exhibited characteristics
associated with reactive skin and scalp.
q 2001 British Association of Dermatologists, British Journal of Dermatology, 145, 258±263
SENSITIVE SKIN 259
electoral register, with assistance in the design,
posting and subsequent computer scanning of returned
questionnaires being given by a local market research
company. Fifty randomly selected women who failed to
return their questionnaires (`non-responders') were
surveyed by telephone interview some 6 weeks later,
using the same questionnaire.
Statistical methods
In the analysis of the data obtained, basic frequencies
for men and women were calculated. The results for
a range of characteristics in those women who
classified themselves as having sensitive skin were
then compared with those who felt they did not
have sensitive skin. Subsequently, the relationship
between sensitive skin and atopy in the female
population was examined in more detail. The influence
of age on skin sensitivity could not be assessed, as our
local ethics committee would not permit the inclusion of a
specific age-related question in the postal questionnaire.
The Value test (V-test) was the method of choice to
determine statistical significance. This test is used as
a statistical criterion to measure the significance
between percentages (qualitative variables) or means
(quantitative variables) of two populations.7 The
criterion is derived from the hypergeometric distribution
that measures the discrepancy between percentages or
means as a number of SDs of a Gaussian distribution.
Variables with V-tests of more than 2´0 are judged
statistically significant at the 5% level (non-adjusted)
260 C.M.WILLIS et al.

Figure 2. Bar chart showing the frequencies

of positive responses by male and female
responders to questions concerning the
prevalence of adverse reactions to personal
products and incidence of subjective sensory
discomfort.

Table 1. Comparison between female sensitive skin subpopulation (those who responded `yes' to the question: `Do you regard yourself as having
sensitive skin?') and female non-sensitive skin subpopulation (those who responded `no') with respect to different variables, the latter being
arranged in descending order of statistical significance, as determined by the Value test (V-test). Variables with V-tests of . 2´0 are judged
statistically significant at the 5% level
% in population % in
with sensitive population with
Characteristic % global skin non-sensitive skin
Variable modality (female) (n ˆ1046; 50´8%) (n ˆ 1012; 49´2%) V-test n
Is your facial skin easily irritated? Yes 33Š´77 61Š´09 5Š´53 28Š´24 695
Have you ever experienced an adverse Yes 56Š´75 80Š´11 32Š´61 22Š´18 1168
reaction to a cosmetic product?
Do some cosmetic products make Yes 32Š´94 53Š´25 11Š´95 20Š´49 678
your skin burn or irritated?
Do some cosmetic products make Yes 32Š´82 53Š´63 11Š´31 19Š´80 696
your skin itch within 30 min?
Do some cosmetic products make Yes 35Š´86 53Š´73 17Š´39 17Š´46 738
your skin sting within 30 min?
Have you suffered an adverse Yes 22Š´93 35Š´85 9Š´58 14Š´50 472
reaction to a cosmetic in the last 12 months?
Have you ever suffered from eczema Yes 33Š´62 46Š´94 19Š´85 13Š´11 692
or dermatitis?
Do you have any other problem with Yes 30Š´56 41Š´87 18Š´87 11Š´39 629
your skin, not covered by other questions?
Do you regard yourself as having Yes 23Š´13 32Š´70 13Š´24 10Š´56 476
a sensitive scalp?
Do you suffer from eczema or Yes 14Š´33 21Š´80 6Š´61 10Š´00 295
dermatitis at this moment?
Do some hair products make Yes 31Š´24 40Š´92 21Š´23 9Š´65 643
your scalp itch, sting or smart?
Have you ever suffered from Yes 33Š´28 42Š´73 23Š´51 9Š´26 685
asthma or hay fever?
Did you suffer from eczema Yes 10Š´98 16Š´44 5Š´34 8Š´17 226
or dermatitis as a child?
How would you describe your skin? Dry 26Š´43 34Š´13 18Š´47 8Š´05 544
Do you blush and flush particularly easily? Yes 41Š´35 48Š´47 33Š´99 6Š´64 851
Do you tend to suffer from dandruff? Yes 24Š´54 28Š´01 20Š´95 3Š´68 505

q 2001 British Association of Dermatologists, British Journal of Dermatology, 145, 258±263

 SENSITIVE SKIN 261

Figure 3. Bar chart showing the incidence

of features associated with atopy in the
sample populations of men and women.

and are potentially discriminatory between items of the Results

variable to be characterized.
Statistical analysis was carried out using SPAD
(Version 4´00; CISIA, Montreuil, France) and
SPSS (Version 8´00; SPSS Inc., Chicago, IL, U.S.A.)
software.
Of the 3300 questionnaires sent out to women, 2058
were returned, a response rate of just over 62%. The
response rate among the men was just under 52% (258
questionnaires returned). Occasionally, individuals

Figure 4. Pie-charts showing the inter-

relationships between sensitive skin in
women (defined as a `yes' answer to `Do you
regard yourself as having a sensitive skin?')
and atopy (defined as a `yes' answer to `Did
you suffer from eczema or dermatitis as a
child?' and/or a `yes' answer to `Have you
ever suffered from asthma or hay fever?').
Although atopics make up almost half of the
female sensitive skin cohort, not all female
atopics have sensitive facial skin.

q 2001 British Association of Dermatologists, British Journal of Dermatology, 145, 258±263

262 C.M.WILLIS et al.
failed to provide answers to all questions. To correct for
this, all frequencies represent the percentage of the
population who answered the particular question.
Statistical analysis of the telephone survey of 50 non-
responders indicated that the group who returned their
questionnaires was representative of the sample popu-
lation as a whole, there being no evidence of bias in
favour of those with sensitive skin or a history of
cosmetic-induced skin problems (data not shown).
Figure 1 gives the basic frequencies with which a
positive (`yes') response was given to questions
relating to selected skin and scalp characteristics. Just
over half of the women and about 38% of the
men regarded themselves as having a sensitive skin.
Ten per cent of the total female population reported
having a `very sensitive' skin phenotype, the corre-
sponding figure for men being just under 6%.
Similar proportions of men and women described
themselves as having a sensitive scalp (about one-
quarter). Dandruff was more commonly reported
among men than women.
Figure 2 gives the frequencies with which women
and men reported adverse reactions to cosmetic and
hair products. Over half of the female responders had
experienced an adverse reaction to cosmetics at some
stage in their lives, with over 30% of women and over
20% of men reporting symptoms of subjective sensory
discomfort on the skin and scalp.
Statistical comparison between women with sensitive
skin and those who viewed themselves as non-sensitive
revealed that the former group had a significantly
greater incidence of adverse reactions to cosmetic
products, both in terms of lifetime and 12-month
prevalence (Table 1). Symptoms associated with
subjective sensory skin discomfort, such as burning,
stinging and itching, were significantly more common
among those with sensitive skin (Table 1).
Figure 3 displays the frequencies with which clinical
signs of an atopic diathesis were exhibited by men
and women. Over one-third of the individuals would
be classified as atopic, on the basis of a positive
history of childhood eczema and/or a positive history
of asthma or hay fever. In the analysis of sensitive
vs. non-sensitive women, all of the individual
manifestations of atopy were significantly more
prevalent among the sensitive skin group (Table 1).
However, among the women who regarded themselves
as having sensitive skin, 51% did not have an atopic
diathesis, while one-third of those who were atopic did
not consider themselves as having sensitive skin
(Fig. 4).
q 2001 British Association of Dermatologists, British Journal of Dermatology, 145, 258±263
Of the remaining data collected on facial skin
characteristics, dry skin was found to be present to a
significantly greater extent in those who perceived
themselves as having sensitive skin (Table 1), as was
the tendency to blush or flush readily (Table 1).
Discussion
Our survey was conducted to assess the level of skin
sensitivity and adverse reactions to cosmetic pro-
ducts in men and women, and to learn more about
factors that predispose or are associated with
sensitive skin in women. Although the sample
population was taken from one regional area
alone, extrapolation of the data to the U.K.
population as a whole should be permissible, as
the demographic profile of the region is known to be
representative.8 Furthermore, our incidence figure of
about 33% for the presence of atopy in the sample
population closely mirrors that of epidemiological
studies conducted in western Europe,9±11 again
suggesting that we have a representative population.
Self-diagnosis of sensitive skin should also be
regarded as reliable; studies conducted consequent
to this survey showed that self-perceived sensitive
skin correlates well with many of the predictive
tests for susceptibility to irritation (manuscript in
preparation).
The data obtained from the questionnaire provide
strong evidence that sensitive skin is indeed a common
phenomenon in the U.K., with about 50% of women
and 40% of men regarding themselves as having a
sensitive skin. Well over a half of women and over a
third of men who responded to the questionnaire had
experienced an adverse reaction to a cosmetic product
at some stage in their life, with subjective sensory skin
discomfort being a commonly reported phenomenon.
Our prevalence figure for the preceding 12-month
period of about 18´5% (men and women combined)
suggests that the level of adverse reactions to
cosmetic products in the general population may be
on the increase, a consumer study conducted in the
U.K. during the 1970s having described a 12%
incidence of unwanted effects over a similar time
period.6
In the detailed analysis of sensitive vs. non-sensitive
women, it was shown statistically that certain features
and characteristics were over-represented in the former
group. Burning, itching and stinging all occurred more
commonly, confirming the link between heightened
neurosensory discomfort and the self-perception of

sensitive skin. Why, in physiological terms, some
individuals have a greater predilection for this type of
response is not fully understood, but our data are
consistent with the results of provocation studies
which show that those with self-perceived sensitive
skin are significantly more susceptible to the neuro-
genic effects of certain chemical irritants such as lactic
acid.12±14
An atopic diathesis was shown to be significantly
more prevalent among women who regarded
themselves as having sensitive skin, about 50% of
the cohort reporting a personal history of atopic
eczema and/or asthma/hay fever, as compared
with 27% among the non-sensitive group. Although
the association between atopy and skin susceptibility
to irritation is established,4 a predictive link
between atopic diathesis and sensitive skin is not
so evident. Our results demonstrated that the
presence of atopy does not lead in all cases to
sensitive skin; half of the sensitive skin group did
not have a history of atopy and one-third of atopic
women regarded themselves as having non-sensitive
skin.
In addition to atopic dermatitis, other skin
disorders, such as rosacea and acne, have been
linked to heightened skin reactivity,3,4 and certainly
over a third of the total sample population reported
having skin problems other than those covered by
the questionnaire. Our study also confirmed the
association between sensitive skin and both dry
skin and the propensity for blushing/flushing, charac-
teristics, which respectively indicate a tendency
to barrier impairment and heightened vascular
reactivity.15,16
In summary, our survey has shown that sensitive
skin, manifested primarily as subjective sensory skin
discomfort, is common among women and men in the
U.K. There is clearly scope for improvements in the
formulation of cosmetic products specifically designed
for this type of skin.
q 2001 British Association of Dermatologists, British Journal of Dermatology, 145, 258±263
SENSITIVE SKIN 263
Acknowledgments
The authors acknowledge the assistance and advice
given by Christiane Montastier, Colette Hourseau,
GenevieÁve Loussouarn and Monique Romeu.
References
1 Maibach HI, Engasser PG. Cosmetic intolerance syndrome.
Semin Dermatol 1986; 5: 273±6.
2 Simion FA, Rau AH. Sensitive skin: what it is and how to
formulate for it. Cosmet Toiletries 1994; 109: 43±9.
3 Amin S, Maibach HI. Cosmetic intolerance syndrome: patho-
physiology and management. Cosmet Dermatol 1996; 9: 34±42.
4 Mills OH, Berger RS. Defining the susceptibility of acne-prone and
sensitive skin populations to extrinsic factors. Dermatol Clin 1991;
9: 93±7.
5 de Groot AC, Nater JP, van der Lende R, Rijcken B. Adverse effects
of cosmetics and toiletries: a retrospective study in the general
population. Int J Cosmet Sci 1988; 9: 255±9.
6 Consumers Association. Reactions of the Skin to Cosmetic and
Toiletry Products. London: Consumers Association, 1979.
7 Morineau A. Note sur la caracteÂrisation d'une classe et les
valeurs-tests. Bull Techn Cent Statist Inform Appl 1984; 2: 20±7.
8 Young E, Stoneham MD, Petruckevitch A , Barton J , Rona R.
A population study of food intolerance. Lancet 1994; 343: 1127±30.
9 de Groot AC, Beverdam E, Tjong Ayong C et al. The role of contact
allergy in the spectrum of adverse effects caused by cosmetics.
Contact Dermatitis 1988; 19: 195±201.
10 JoÄgi R, Janson C, BjoÈrnsson E et al. Atopy and allergic disorders
among adults in Tartu, Estonia compared with Uppsala, Sweden.
Clin Exp Allergy 1998; 28: 1072±80.
11 WuÈthrich B, Schindler C, Leuenberger P, Ackermann-Liebrich U.
Prevalence of atopy and pollinosis in the adult population of
Switzerland (SAPALDIA Study). Int Arch Allergy Immunol 1995;
106: 149±56.
12 Frosch PJ, Kligman AM. A method of appraising the stinging
capacity of topically applied substances. J Soc Cosmet Chem 1977;
28: 197±209.
13 Issachar N, Gall Y, Borell MT, Poelman M-C. pH measurements
during lactic acid stinging test in normal and sensitive skin.
Contact Dermatitis 1997; 36: 152±5.
14 Muizzuddin N, Marenus KD, Maes DH. Factors defining sensitive
skin and its treatment. Am J Contact Dermatitis 1998; 9: 170±5.
15 Draelos ZD. Sensitive skin: perceptions, evaluation, and treat-
ment. Am J Contact Dermatitis 1997; 8: 67±78.
16 Seidenari S, Francomano M, Mantovani L. Baseline biophysical
parameters in subjects with sensitive skin. Contact Dermatitis
1998; 38: 311±15.