JEADV ISSN 1468-3083

ORIGINAL ARTICLE
Blackwell Publishing Ltd

Development and validation of a questionnaire to evaluate how

a cosmetic product for oily skin is able to improve well-being in
women
E Segot-Chicq,* D Compan-Zaouati, P Wolkenstein, S Consoli, C Rodary, V Delvigne,** V Guillou, F Poli
LOral Recherche, Chevilly Larue, France
CHU H.Mondor, Crteil, France
Consultant Dermatologist, Paris, France
Institut G. Roussy, Villejuif, France
** Lancme, Paris, France

Keywords
cosmetic skincare, oily skin, psychological
questionnaire
*Corresponding author, LOral Recherche,
Centre de Chevilly, 188-200 rue Paul Hochart,
FR 94550 Chevilly Larue, France, tel.
+33 1 4979 5248; fax +33 1 4979 5142; E-mail:
esegot@rd.loreal.com
Received: 20 April 2006,
accepted 20 December 2006
DOI: 10.1111/j.1468-3083.2007.02193.x

Abstract
Background Skin diseases are known to negatively affect self-image and to
have detrimental psychosocial effects. Oily skin is a cosmetic skin problem that
women often describe as invalidating.
Objective To develop and validate a questionnaire to assess the psychological
and psychosocial effects of oily skin condition in women and the outcome of a
targeted cosmetic skincare treatment.
Methods We developed and validated a concise 18-item questionnaire [oily
skin self-image questionnaire (OSSIQ)] to assess perception, behavioural, and
emotional consequences associated with oily skin condition. The questionnaire
was then used to assess the effects of a skincare treatment for oily skin and
compare them with sebum level measurements.
Results The 18-item questionnaire clearly distinguished the oily skin group
from the control group. Responsiveness, reliability, and construct validity showed
satisfactory performance. The questionnaire provided a relevant assessment of
the psychological benefits associated with the skincare programme.
Conclusion The OSSIQ is a valid tool that can be used to monitor the benefits
of cosmetic skincare treatments

Introduction
Skin disorders can have a significant effect on quality of
life, and numerous studies have shown the extent to
which self-perception of physical appearance can influence
psychological state and social interactions.13 Oily skin is a
condition characterized by open pores and a shiny appearance that results from sebum overproduction by the
sebaceous glands and its release from the follicular ostia at
the skin surface.4 The features of oily skin are disturbing to
women with this condition and are perceived to be a serious
cosmetic problem. Oily skin condition can lead to negative
self-perception and is detrimental to social interactions.
The increased interest in the accurate measurement
of the psychosocial effect of dermatological events in
affected patients has led to the development of a number

of standardized quality of life (QoL) questionnaires.58 The

development of a proper questionnaire requires an appreciation of subtle differences in culture, language, and
psychological factors that may influence QoL assessment.
It must be brief, simple, unambiguous, easy to complete,
and have shown psychometric properties.9,10 Many studies have used life quality scales to evaluate dermatological
treatments in a clinical or healthcare context.11 On the
other hand, very few papers refer to QoL approach for the
appraisal of cosmetic products. Decorative cosmetics with
adequate coverage properties were shown to improve
QoL in patients with disfiguring skin diseases.12 The influence of hairdressing on psychological mood of women
also was investigated.13 However, the effect of cosmetic
skincare designed to or intended to correct unpleasant or
disliked appearance of otherwise healthy skin condition

2007 LOral Recherche

JEADV 2007, 21, 11811186 Journal compilation 2007 European Academy of Dermatology and Venereology

1181

Well-being benet of cosmetic treatment

Segot-Chicq et al.

has not yet been addressed using such tools. As oily skin
seems to have a very negative effect in women, and as
cosmetic routines by improving skin condition, as well as
by stimulating narcissistic feelings are likely to modify the
psychological behaviour of patients, we developed a specific questionnaire to assess the psychosocial effect of
skincare cosmetics on this disliked healthy skin condition.
The purpose of this study was to develop and validate a
French designed questionnaire to assess the psychological
and behavioural reactions to oily skin condition and to
test the suitability of the questionnaire to measure the
benefits of a cosmetic programme.

opaque plastic strip after a 30-s contact with the skin.16,17

It provides a measurement of the amount of lipids on the
skin (i.e. of the oily status of the skin). As it is known that
sebum output fluctuates over the day time, the sebum
measurements for each volunteer were made at the same
hour of the day and under controlled temperature and
hygrometric conditions to get rid of these environmentinduced variations. The oily skin group consisted of 98
women with high sebum level (over 150 g cm2). Mean
age was 26 years (range, 1835). The normal skin group
involved 102 women whose sebum level was within the
range 70 to 150 g cm2. Mean age of the group was also
26 years (range, 1839).

Materials and methods

Study design
Development of the oily skin self-image
questionnaire
The oily skin self-image questionnaire (OSSIQ) was
developed through a number of testing steps and refined
for optimal performance. As a first approach, a 29-item
questionnaire was designed by a panel of experts (dermatologists and psychologists) based on the French version
of a previous QoL evaluation tool used in dermatology,
Skindex,14,15 and the description of self-perception and
discomfort as expressed by 35 women with oily skin.
This first draft questionnaire was then submitted to an
independent group of 16 women with oily skin condition.
The comments of this test group were then used to
generate an improved version of the 29-item questionnaire, which was approved by the panel of experts. Each
item had a scoring range with five possible grades (from
0 = never to 4 = always). Three multiple-item subscales
reflecting symptoms (4 items), emotional status (15 items),
and behaviour (9 items) were defined a priori from the
questionnaire items as well as the associated scores. These
subscale scores correspond to the sum of related item
responses standardized on a percentage scale from 0%
(well-being) to 100% (discomfort). An additional item
was considered to be neutral. The improved 29-item
questionnaire was tested on 200 women (98 with oily
skin and 102 with normal skin) and subsequently reduced
to a validated 18-item questionnaire based on statistical
results (poor sensitivity or low discriminatory value of
certain items) and following experts agreement.

Selection criteria for test groups

Two groups of female volunteers were selected according
to individual sebum level, which was evaluated using
standard measurement (Sebumeter SM810, Courage
and Khazaka). This device affords direct photometric
reading of the amount of lipids collected on a probe of

We conducted an open study that involved (i) validation

of the questionnaire using oily skin and normal skin
groups and (ii) testing the suitability of the questionnaire
to evaluate a specific daily skincare programme in the oily
skin group.

Validation of the OSSIQ

The 29-item questionnaire was completed twice within
the first week of the study (day 0 and day 7) by both
the oily skin group and the control group. We assessed
its responsiveness, reliability, and construct validity in
order to eliminate items with poor sensitivity or low
discriminatory value.14,15

Evaluation of a daily skincare programme in

women with oily skin
Women in the oily skin group applied a specific daily
skincare programme, previously tested for efficacy
(unpublished data), after the first week of the study. They
sequentially used a rinse-off cleansing gel, a toner, and
a skincare fluid (containing mattifying agents, a salicylic
acid derivative, and a zinc derivative in hydroalcoholic
bases Pure Focus, Lancme), which they applied on the
whole face twice daily for 4 weeks (day 7 to day 35). The
OSSIQ was completed again at the end of the daily care
programme (day 35), and sebum levels were measured
with a Sebumeter SM810 (Courage and Khazaka).
Volunteers were required to come for each sebum
measurements and questionnaire completion at the same
time of the day to get rid of circadian variations.

Recorded data
The following data sets were collected: From both
groups: day 0, sebum level measurements and completed

2007 LOral Recherche

1182

JEADV 2007, 21, 11811186 Journal compilation 2007 European Academy of Dermatology and Venereology

Segot-Chicq et al.

questionnaire; day 7, completed questionnaire. From the

oily skin group: day 35, sebum level measurements and
completed questionnaire.

Statistical analysis
Validation of the OSSIQ (based on Day 0 responses)
Face validity: refusal, rate of missing items; Responsiveness: ability to discriminate with normal skin using data at
day 0 for normal skin women (MannWhitney Wilcoxon
test for subscales and Fishers exact test for items);
Reliability: Cronbachs coefficient, nonparametric Spearmans correlation between day 0 and day 7 responses
(testretest); Structure validity: multi-trait, multi-method
analysis and principal component analysis (PCA) with
Varimax rotation; External validity: ability to discriminate
between different sebum levels (KruskalWallis test).

Evaluation of a daily skincare programme in women

with oily skin
Comparison between day 0 and day 35 measurements of
sebum level: paired t-test or validated subscales of the
questionnaire (paired Wilcoxon test) in the oily skin
group. Comparison between day 35 data in oily skin
group and day 0 data in normal skin group: Students ttest on the same parameters (sebum level and subscales
of the questionnaire). All comparisons were done with a
two-sided 5% level. Analyses were conducted with SAS
8.2, SPSS 11.5 and MAP-R softwares.

Results and discussion

A group of 200 female volunteers were involved in
psychological assessment (using the OSSIQ) and sebum
measurements. Approximately half of the volunteers had
an oily skin condition (sebum level over 150 g cm2),
whereas the other half had normal skin sebum level (70
150 g cm2). The average age for both groups was 26 years
(range, 1839).

Well-being benet of cosmetic treatment

naire, which presented at least two of the following

weaknesses: poor responsiveness (9 times), no convergent validity (4 times), no significant difference with
normal skin (10 times). This first phase led to define
a more concise 18-item questionnaire. The questions
retained in the 18-item final OSSIQ are listed in Table 1
(together with a non-validated translation). The questionnaire items can be divided into three groups or subscales
representing symptoms (S1S3), behaviour (B1B5),
or emotional status (E1E9), with one additional item
considered as neutral.

Validation of the 18-item questionnaire

We assessed the effectiveness of the shortened questionnaire to measure psychological attributes in terms of
responsiveness, reliability, construct validity, and correlation
with sebum measurements.

Responsiveness
In general, responsiveness was high, with the exception
of four items (B2, B3, E8, and E9) for which the maximal
response (discomfort) category was not used. A floor
effect was observed for four items (B2, B3, B4, and E7): at
least 50% of the volunteers gave a minimal response
(0 = never). We found that the ability to discriminate
between oily skin and normal skin groups could be clearly
shown for 10 items in the questionnaire (B1, B2, B3, B4,
E1, E3, E5, E6, E7, and E8; see Table 1 for related items).
Results show that scores for both emotional and behavioural
factors were significantly different (P < 0.01) between oily
skin and normal skin (control) groups (see Table 2). The
oily skin group had an average emotional score of 33.7
22.3 compared with 24.7 17.9 (P < 0.05) in the control
group and an average behaviour score of 16.9 17.3
compared with 8.63 12.1 (P < 0.05) in the control
group. These results suggest that the shortened 18-item
questionnaire is able to distinguish oily skin group from
control group; women with oily skin had consistently
higher OSSIQ scores, implying that they are emotionally
and behaviourally affected by their skin condition.

Questionnaire shortening
Women in both the oily skin group (n = 98) and control
group (n = 102) were asked to complete a 29-item
questionnaire about their self-perception of skin appearance and social consequences (see Methods). All items in
the questionnaire were completed, showing high face
validity. QoL questionnaires must be short, simple, and
informative in order to be effective. On the basis of the
results from the first phase of testing and with the experts
approval, we removed 11 items from the original question-

Reliability and construct validity

The results of OSSIQ questionnaire testing show that
subscale scores were reliable after 1 week (testretest:
correlation of day 0 and day 7 subscale scores; Spearmans
r 0.70; no significant difference between the two scores,
P > 0.06; see Table 3) and were internally consistent
(Cronbachs = 0.720.92). Structure validity of the
questionnaire was shown by multi-trait multi-method
analysis (see Table 4). Moreover, from PCA with Varimax

2007 LOral Recherche

JEADV 2007, 21, 11811186 Journal compilation 2007 European Academy of Dermatology and Venereology

1183

Well-being benet of cosmetic treatment

Segot-Chicq et al.

Table 1 The 18-item shortened questionnaire. The validated French questions are shown together with corresponding, non-validated English translations.
Volunteers were asked to choose one of ve possible answers: never (0), rarely (1), sometimes (2), often (3), and always (4).
The questions are grouped together within symptoms, behaviour, and emotional status subscales. Subscale scores are the sum of scores for each relevant
item then standardized on a percentage scale from 0% (well-being) to 100% (discomfort)
French version

English translation

Symptoms
Ma peau me procure une sensation dinconfort
Jai la peau sensible
Ma peau est irrite

S1
S2
S3

My skin gives me a feeling of discomfort

I have sensitive skin
My skin is irritated

Behaviour
Ltat de ma peau gne ma vie sociale
Jai tendance rester chez moi cause de laspect de ma peau
Ltat de ma peau gne mes contacts avec mes proches
Laspect de ma peau perturbe mes relations avec les autres
Jessaie de dissimuler ma peau au regard des autres

B1
B2
B3
B4
B5

My skin condition affects my social life

I tend to stay at home because of the appearance of my skin
My skin condition affects my contact with relatives
The appearance of my skin affects my relations with others
I try to keep my skin out of peoples sight

Emotional status
Ma peau nest pas belle
Ma peau me dprime
Je ne prends pas plaisir regarder ma peau
Je ne prends pas plaisir toucher ma peau
Je ne suis pas satisfaite de ma peau
Je suis complexe par ltat de ma peau
Ma peau mobsde (emplit mes penses)
Mon problme de peau me stresse
Mon problme de peau ne me rend pas sduisante

E1
E2
E3
E4
E5
E6
E7
E8
E9

My skin is not pretty

My skin makes me feel depressed
I dont take pleasure in looking at my skin
I dont take pleasure in touching my skin
I am not happy with my skin
I am hung up about my skin condition
I am obsessed by my skin
My skin problem is stressful
My skin problem makes me unattractive

Neutral
Lentretien de ma peau est contraignant

Taking care of my skin is constraining

Table 2 Compared emotional and behavioural scores of groups of women with oily skin vs. with normal skin. The oily skin group was evaluated before and
after treatment with a specic skincare programme. The OSSIQ distinguishes between oily skin and control groups and measures the effects of a treatment
that signicantly decreases sebum level

Oily skin before treatment (n = 98)

Oily skin after treatment (n = 86)
Normal skin (n = 102)

Emotional score

Behavioural score

g cm2)
Sebum level (

33.7 22.3*
23.9 19.5
24.7 17.9

16.9 17.3*
9.7 12.5
8.63 12.1

212 45*
161 63
100 27

mean SD.
*, signicant comparison (P < 0.05) with oily skin after treatment.
, signicant comparison (P < 0.05) with normal skin.

Table 3 Reliability in the oily skin group (n = 98)

Retest

Subscale

No. of
items

Scale score at day 0*

(n = 98)

Scale score at day 7*

(n = 90)

Spearmans correlation
coefcient

Comparison between
day 0 and day 7

Symptoms
Emotional
Behavioural

3
9
5

40.2 20.6
33.7 22.3
16.9 17.3

37.9 21.1
31.8 22.6
14.7 16.8

0.79
0.89
0.70

P = 0.06
P = 0.16
P = 0.48

*mean SD.
, Wilcoxon paired test.

2007 LOral Recherche

1184

JEADV 2007, 21, 11811186 Journal compilation 2007 European Academy of Dermatology and Venereology

Segot-Chicq et al.

Well-being benet of cosmetic treatment

Table 4 Construct validity in the oily skin group (n = 98)

Subscale

No. of
items

Internal-consistency
reliability (Cronbachs
coefcient)

Item internal
consistency
(%success)*

Item discriminant
validity (%success)

Item discriminant
validity (%success)

Symptoms
Emotional
Behavioural

3
9
5

0.72
0.92
0.80

100
100
100

50
50
55

83
80
100

*, item scale correlation with the hypothesized subscale, corrected for overlap = 0.40.
, item scale correlation signicantly higher (>2 SE) with the hypothesized subscales than with others.
, item scale correlation higher with the hypothesized subscales than with others.

rotation, three main factors (corresponding to the three

a priori subscales) accounting for 64% of variance were
extracted. The first factor (associated with Emotional
items) contributed to 29% of the variation.

External validation: correlation with

instrumental sebum measurements
We found that both emotional (P = 0.002) and behavioural (P = 0.001) scores were in good agreement with
the severity of oily skin condition as shown by further
distinguishing two subgroups within oily skin volunteers:
one with sebum level between 150 g cm2 and 200 g
cm2 (n = 44) and one with sebum level over 200 g cm2
(n = 54). The higher the sebum level, the higher the
OSSIQ score (see fig. 1).

Evaluation of the outcome of a daily skincare

programme using the 18-item questionnaire
In order to assess the performance of the OSSIQ as an
evaluation tool, we used the final 18-item OSSIQ to
determine emotional and psychological parameters in
women with oily skin who applied a dedicated skincare
programme. Eighty-six of 98 volunteers completed the
4-week skincare treatment for oily skin. We measured
sebum levels at the beginning and at the end of the
treatment course and used the 18-item questionnaire to
assess patient well-being (see Table 2).
At day 0, women in the oily skin group were measurably different from control group in terms of sebum level
(average 212 g cm2 vs. 100 g cm2 P < 0.05) as well as
emotional score (33.7 vs. 24.7, P < 0.05) and behavioural
score (16.9 vs. 8.63, P < 0.05). Following the 4-week skincare treatment, the oily skin group showed a marked
reduction in sebum level (decreased to average 161 g
cm2) as well as a markedly improved OSSIQ scores. The
emotional and behavioural test scores of women after the
oily skincare treatment were almost as low as normal skin
control group (23.9 and 9.7, respectively).

g. 1 OSSIQ emotional and behavioural subscale scores measured in

women with different sebum levels (mean SD).

Interestingly, we found a poor correlation of sebum

variation with emotional or behavioural score variation
before and after treatment. It thus appears that although
the OSSIQ scores are in good accordance with objective
measurements of sebum level, their modification by
the cosmetic treatment are two independent results,
the biological and the psychological effects being two
independent benefits. Whereas the sebum measurements
assess the efficacy of the cosmetic routine in correcting the
skin oiliness, the OSSIQ scores mirror the actual changes

2007 LOral Recherche

JEADV 2007, 21, 11811186 Journal compilation 2007 European Academy of Dermatology and Venereology

1185

Well-being benet of cosmetic treatment

Segot-Chicq et al.

induced in self-perception and confidence by positive

alteration of self-image from use of the daily cosmetic
routine. This positive alteration seems to be parallel to the
improvement of skin condition, but their lack of correlation suggests a difference of responsiveness of the subjects
to the biological and to the psychological effects, which
could be worthy of further study. Women enrolled in the
study were regular users of cosmetic products; thus, the
mere effect of unusual daily skincare cannot be the only
explanation for this positive effect on self perception, but
efficacy in improving skin condition, together with individual sensitivity to this skin condition, also contributes to
the observed effect. This psychological effect being an
additional benefit of the cosmetic skincare, it is useful to
have a tool for its measurement.

Conclusion
The development of this 18-item OSSIQ provides a
reliable and valid tool to measure psychological reactions
brought about by a proper skincare treatment able to
modify skin condition. The close agreement with instrumental measurements supports the relevance and
usefulness monitoring the additional beneficial effects of a
product or multistep skincare programme. These results
suggest that carefully designed questionnaires, validated
in different languages and adapted to different cultures,15
can be effectively applied to evaluate both the psychosocial and psychological effect of various non-diseased
skin problems and to assess the influence or suitability of
cosmetic care products aimed at restoring well-being of
users.

References
1 Parsad D, Dogra S, Kanwar AJ. Quality of life in patients
with vitiligo. Health Qua Life Outcomes 2003; 1: 58.
2 Lasek RJ, Chren MM. Acne vulgaris and the quality of life
of adult dermatology patients. Arch Dermatol 1998; 134:
454458.
3 Leary MR, Rapp SR, Herbst KC, Exum ML, Feldman SR.
Interpersonal concerns and psychological difficulties of
psoriasis patients: effects of disease severity and fear of
negative evaluation. Health Psychol 1998; 17: 530536.

4 Thiboutot D. Regulation of human sebaceous glands. J Invest

Dermatol 2004; 123: 112.
5 Morgan M, McCreedy R, Simpson J et al. Dermatology
quality of life scales a measure of the impact of skin
diseases. Br J Dermatol 1997; 136: 202206.
6 Anderson RT, Rajagopalan R. Development and validation
of a quality of life instrument for cutaneous diseases. J Am
Acad Dermatol 1997; 37: 4150.
7 Chren MM, Lasek RJ, Quinn LM, Mostow MN, Zyzanski SJ.
Skindex, a quality-of-life measure for patients with skin
disease: reliability, validity, and responsiveness. J Invest
Dermatol 1996; 107: 707713.
8 Finlay AY, Khan GK. Dermatology Life Quality Index
(DLQI) a simple practical measure for routine clinical use.
Clin Exp Dermatol 1994; 19: 210216.
9 Nunnally JC, Bernstein ICH (eds). Psychometric theory, 3rd
edn. Mc Graw Hill, New York NY, 1994.
10 Fallissard B (ed.). Mesurer la Subjectivit En Sant. Perspective
Mthodologique et Statistique, 1st edn. Masson, Paris, France,
2001.
11 Anderson R, Rajagopalan R. Responsiveness of the
dermatology-specific Quality of life (DSQL) instrument to
treatment for acne vulgaris in a placebo-controlled clinical
trial. Qual Life Res 1998; 7: 723734.
12 Boehncke W-H, Ochsendorf F, Paeslack I, Kaufmann R,
Zollner TM. Decorative cosmetics improve the quality of life
in patients with disfiguring skin diseases. Eur J Dermatol
2002; 13: 577580.
13 Picot-Lemasson A, Decocq G, Aghassian F, Leveque JL.
Influence of hairdressing on the psychological mood of
women. Inter J Cosmet Sci 2001; 23: 161164.
14 Chren MM, Lasek RJ, Sahay AP, Sands LP. Measurement
properties of Skindex-16: a brief quality-of-life measure for
patients with skin diseases. J Cutan Med Surg 2001; 5: 105
110.
15 Leplege A, Ecosse E, Zeller J, Revuz J, Wolkenstein P.
Version franaise du Skindex (Skindex-France): adaptation
et valuation des proprits psychomtriques. Ann Dermatol
Venereol 2003; 130: 177183.
16 Pierard GE, Pierard-Franchimont C, Marks R, Paye M,
Rogiers V. EEMCO guidance for the in vivo assessment of
skin greasiness. The EEMCO Group. Skin Pharmacol Appl
Skin Physiol 2000; 13: 372389.
17 Dickstein S, Zlotogorski A, Avriel E, Katz M, Harms M.
Comparison of the Sebumeter and the Lipometer. Bioeng
Skin 1987; 3: 197.

2007 LOral Recherche

1186

JEADV 2007, 21, 11811186 Journal compilation 2007 European Academy of Dermatology and Venereology