DOI: 10.1111/jdv.

12894 JEADV

ORIGINAL ARTICLE

Methodology of the EDIFICE Melanoma survey

C. Touboul*
Kantar Health, Montrouge, France
*Correspondence: C. Touboul. E-mail: chantal.touboul@kantarhealth.com

Abstract
Background Melanoma accounts for a small proportion of skin cancers but is the most serious. The prognosis is, how-
ever, good if it is diagnosed early, before the metastatic stages. For these reasons, population awareness and knowl-
edge with regard to melanoma and the associated risk factors are critical to improving the effectiveness of melanoma
primary prevention campaigns.
Objectives The EDIFICE Melanoma survey was designed to cover a large number of topics related to sun exposure
and melanoma: knowledge and awareness of the risks associated with sun and articial ultraviolet exposure, knowledge
of melanoma and the associated risk factors, behaviour regarding protective measures in adults and children.
Methods The survey questionnaire was administered by experienced independent interviewers from the Kantar Health
Institute using the Computer-Assisted Telephone Interview technique. It was important to avoid the effects of contami-
nation in answers to questions about sun exposure habits. Questions about exposure habits were therefore asked
before those relating to knowledge of best practices for sun protection. A sample of 1502 individuals aged 18 years and
over was interviewed. The survey data were weighted to eliminate structural variations between the sample interviewed
and the reference population. Quantitative data were described by the means and standard deviations and categorical
data by the numbers in each category and corresponding percentages. Three multivariate logistic regression analyses
were conducted and expressed in terms of odds ratio and 95% condence intervals.
Results In addition to the classic questions relating to sociodemographic characteristics, the EDIFICE Melanoma study
questionnaire was organized around three themes: the opinion, attitude and knowledge of individuals regarding sun
exposure and melanoma. The survey questionnaire lasted approximately 20 min.
Conclusions This study provided a complete snapshot of opinions, attitudes and knowledge of French people in
2011 with regard to sun exposure.
Received: 21 July 2014; Accepted: 22 October 2014

Conict of interest
None declared.

Funding sources
EDIFICE surveys are funded by Roche SA.

Introduction French institute for public health surveillance (Institut National

Melanoma may only account for a small proportion of skin can-
cers but it is the most serious. If it is diagnosed early, however,
melanoma still has a good prognosis, especially when the
tumour has not yet spread to other organs. Indeed, 80% of
patients with non-metastatic melanoma are still alive after
5 years whereas this figure does not exceed 1020% in cases of
metastatic melanoma. Information and awareness campaigns
have been running for many years, both in France and abroad,
to ensure correct means of protection are used. Despite these
campaigns, of the seven most common cancers in the United
States, melanoma is the only one with an incidence rate that rose
between 2000 and 2009: +1.9% annually.1 In France, in 2010, the
de Veille Sanitaire) estimated the number of new cases diagnosed
to be approximately 8250, i.e., 3% of all cancers. However, it is
known that sun exposure habits have changed over the course of
the last few years due to greater awareness of the risks.2,3 Numer-
ous studies (certain dating back several years) have been con-
ducted on this topic: some concerning knowledge of the risks
related to personal protective habits;2,4 others, regarding the
protective habits of parents towards their children4 or habits of
adolescents for themselves;5,6 and yet more, concerning the use
of sun screens7,8 or tanning booths.911 However, few studies
have assessed all these aspects at the same time and in the same
population. The EDIFICE Melanoma survey, conducted at the

JEADV 2015, 29 (Suppl. 2), 25 2015 European Academy of Dermatology and Venereology
Methodology of the EDIFICE Melanoma survey 3

Table 1 Representativeness rates of the survey population (N = 1502) and of the adjusted population according to sociodemographic
characteristics
EDIFICE Melanoma sample, Final weighted sample based on French
unweighted (%) population aged 18 and over (%)
Gender
Men 46.8 47.6
Women 53.2 52.4
Age group of the individual (seven categories)
1819 years 1.7 3.0
2024 years 7.9 7.9
2534 years 15.7 15.6
3549 years 26.2 25.9
5064 years 26.0 25.2
65 years and over 22.5 22.4
PCS (professions and socioprofessional categories) of the individual
(eight categories)
Agricultural worker 0.7 1.1
Tradesperson, shopkeeper 3.3 3.4
Managerial, professional 9.8 9.4
Intermediate profession 15.4 14.1
Employee 18.2 17.2
Labourer 11.1 12.6
Retired 30.3 25.7
Other/not working 11.2 16.5
Urban unit aggregate sector (ve categories)
Rural 25.6 27.3
200020 000 inh. 16.5 17.4
20 000100 000 inh. 13.6 13.2
More than 100 000 inh. 28.8 27.5
Urban area of Paris 15.5 14.6
UDA (French Union of Advertisers) Region
Paris area 17.1 17.2
North 6.6 6.5
East 9.2 8.6
Eastern Paris Basin 7.0 8.0
Western Paris Basin 9.5 9.7
West 14.2 14.1
South West 11.6 11.4
South East 12.5 12.0
Mediterranean 12.3 12.6

end of the summer of 2011, between 28 September and 20
October, covered all these themes.
Material and methods
This nationwide survey was conducted by telephone among a
representative sample of subjects living in France, aged 18 and
older. The survey questionnaire was administered by trained
independent interviewers from the Kantar Health Institute using
the Computer-Assisted Telephone Interview (CATI) technique.
Each interview lasted 20 min on average. A sample population
of 1502 individuals aged 18 years and over was interviewed. The
95% confidence level for this sample size, and for a 50%
response, was 2.5%. Representativeness of the survey sample
for gender, age, profession and double stratification by geo-
graphical area and community size as compared to the French
general population, was ensured by use of the method of quo-
tas,12 based on the statistics of the French Employment Survey
conducted in 2009 and updated in 2011 by the French national
institute for statistics and economic studies (Insee). The survey
data was weighted to eliminate structural variations between the
sample interviewed and the reference population. The controlled
variables were as follows: age, gender, socioprofessional category
and geographical area of residence of the person interviewed.
The weighting effectiveness was 98.8% with a minimum weight

JEADV 2015, 29 (Suppl. 2), 25 2015 European Academy of Dermatology and Venereology
4 Touboul
of 0.85 and a maximum weight of 1.33. Table 1 compares the
structures of the two populations.
Statistical analysis
Monovariate analysis Quantitative data were described by
the means and standard deviations (SD) and categorical data
by the numbers in each category and corresponding percent-
ages.
All comparisons between two populations were carried out
using Students t test for quantitative data, and the Z test and the
Chi-square test for the comparison of percentages and numbers,
respectively, in the case of categorical data. Differences were con-
sidered statistically significant when the probability value was
less than 0.05 (bilateral test).
Three multivariate logistic regression analyses were con-
ducted. They were expressed in terms of odds ratio (OR) and
95% confidence intervals (CI) and performed using the R soft-
ware, version 12.5.1 (packages stats and MASS).13,14
Questionnaire
In addition to the classic questions relating to sociodemographic
characteristics (gender, age and socioprofessional category of the
interviewee, region and population of city of residence), the
EDIFICE Melanoma study questionnaire was organized around
three themes: the opinion, attitude and knowledge of individuals
regarding sun exposure and melanoma.
It is a known fact that asking an interviewee a question de
facto provides him/her with information and can thus influ-
ence his/her opinion in the context of a questionnaire.15 It is
therefore important to be mindful of the order in which the
questions are asked as the results of a survey can be altered by
this.15 For our questionnaire, it was important to avoid the
effects of contamination (or even the effects of social desirabil-
ity) in responses to questions about sun exposure habits. For
instance, it can be assumed that a person would have more diffi-
culty admitting to exposing themselves to the sun between
12 pm and 4 pm if a few minutes beforehand they had been told
that there was a risk associated with doing so. It is for this reason
that it was decided that questions about exposure habits would
be asked before those about knowledge of best practices for sun
protection.
The survey questionnaire lasted around 20 min. After several
questions relating to assessment of the sociodemographic profile,
the interviewer asked two general questions that introduced the
subject of the survey. The first question consisted of seven state-
ments about the effects of sun exposure: three related to the ben-
eficial effects of the sun (protects the skin, gives skin a fresh
appearance and protects from certain cancers) and four ques-
tions related to the harmful effects (has an ageing effect on skin,
causes skin cancer, causes allergic reactions and reduces the
skins regenerative capacity). The second question was about
JEADV 2015, 29 (Suppl. 2), 25
tanning and contained seven positive statements (gives skin a
fresh appearance, makes you prettier, more handsome, more
attractive, gives you a healthy glow, makes you look younger,
shows that you benefited from your vacation, makes you look
energetic and full of life). For each of these questions, a Likert
four-point scale was suggested (completely agree, somewhat
agree, somewhat disagree, completely disagree).
The questionnaire then continued with a series of questions
on exposure habits whether for leisure, vacation or work, protec-
tion habits for themselves but also for their children (regardless
of whether they still lived at home or not), and on protection
habits (clothing, sunglasses, hat and sunscreen) for themselves
and their children. Subsequently, the questionnaire focused on
knowledge of ways to protect the skin from the sun, as well as
knowledge about melanoma: risk factors and warning signs.
Interviewees were also asked to describe their self-monitoring
habits for skin and moles, for identifying melanoma, breast,
colorectal and prostate cancer. Lastly, questions leading to calcu-
lation of the individuals risk factors were asked: phototype, pro-
longed periods of exposure to strong sun, family history of
melanoma and use of UV tanning booths.
The biases linked to this survey are those traditionally associ-
ated with a declarative survey: social desirability response biases
and memorization biases. Given that the survey was carried out
a few weeks after the end of summer, this bias was limited. It can
therefore be assumed that apart from statements concerning
protection habits for children aged 15 years and over, the state-
ments made were very close to reality.
Conclusion
This study produced a complete snapshot of opinions, atti-
tudes and the knowledge of French people in 2011 with regard
to sun exposure. The data has provided important informa-
tion on the profile of individuals who do or do not expose
themselves to the sun, those who, during exposure to the sun,
do or do not use protective measures, and analyses the differ-
ences in behaviour regarding protection for themselves com-
pared with protective measures for their children. The data
has also provided a clear insight into the correlation between
perceived risk and known risk of melanoma, and also into the
factors influencing the use of tanning lamps. Finally, the origi-
nality of this study was the comparison of the responses pro-
vided by parents of young children (<15 years) with those
provided by parents of children over the age of 15, and the
subsequent analysis of the differences in behaviour between
the sun-protection practices adopted for children nowadays
(children under the age of 15) and those used for children in
the past (children over the age of 15).
Acknowledgements
Medical writing assistance was provided by Potentiel dAction
and funded by Roche.
2015 European Academy of Dermatology and Venereology
Methodology of the EDIFICE Melanoma survey
References
1 Howlader NNA, Krapcho M, Neyman N et al. SEER Cancer Statistics
Review, 19752009 (Vintage 2009 Populations). National Cancer Insti-
tute, Bethesda, MD; [WWW document] April 2012. based on November
2011 SEER data submission. URL http://seer.cancer.gov/csr/
1975_2009_pops09/(last accessed 17 November 2014).
2 Robinson JK, Rigel DS, Amonette RA. Trends in sun exposure knowl-
edge, attitudes, and behaviors: 1986 to 1996. J Am Acad Dermatol 1997;
37(2 Pt 1): 179186.
3 Cancer trends progress report2009/2010 update. Bethesda MD:
National Cancer Institute, NIH, DHHS, [WWW document] 2010. URL
http://progressreport.cancer.gov/2009/ (last accessed 17 November 2014).
4 Bien prot
eger les enfants du soleil: une nouvelle campagne pour renforcer
la vigilance des parents (Adequate sun protection for children: a new
campaign to increase parental vigilance) [press release]. Institut national
du cancer, [WWW document] April 2008. URL http://www.e-cancer.fr/
soleilmodedemploi/pdf/actu_juin08/CP_melanome_INCa_avril_2008.
pdf (last accessed 17 November 2014).
5 Cokkinides V, Weinstock M, Glanz K, Albano J, Ward E, Thun M.
Trends in sunburns, sun protection practices, and attitudes toward sun
exposure protection and tanning among US adolescents, 1998-2004.
Pediatrics 2006; 118: 853864.
6 Banks BA, Silverman RA, Schwartz RH, Tunnessen WW Jr. Attitudes of
teenagers toward sun exposure and sunscreen use. Pediatrics 1992; 89: 4042.
7 Green AC, Williams GM, Logan V, Strutton GM. Reduced melanoma
after regular sunscreen use: randomized trial follow-up. J Clin Oncol
2011; 29: 257263.
JEADV 2015, 29 (Suppl. 2), 25
5
8 Suppa M, Argenziano G, Moscarella E et al. Selective sunscreen applica-
tion on nevi: frequency and determinants of a wrong sun-protective
behaviour. J Eur Acad Dermatol Venereol 2014; 28: 348354.
9 Colantonio S, Bracken MB, Beecker J. The association of indoor tanning
and melanoma in adults: systematic review and meta-analysis. JAMA
Dermatol 2014; 70: 847857.e18.
10 Boniol M, Autier P, Boyle P, Gandini S. Cutaneous melanoma attribut-
able to sunbed use: systematic review and meta-analysis. BMJ 2012; 345:
e4757.
11 Wehner MR, Chren MM, Nameth D et al. International prevalence of
indoor tanning: a systematic review and meta-analysis. JAMA Dermatol
2014; 150: 390400.
12 Deville J. A theory of the quota surveys. Survey Methodol 1991; 17:
163181.
13 Venables WN, Ripley BD. Modern Applied Statistics with S, 4th edn.
Springer, New York; 2002. ISBN 0-387-95457-0.
14 R Development Core Team. R: A Language and Environment for Statisti-
cal Computing. R Foundation for Statistical Computing, Vienna, Austria,
2012. ISBN 3-900051-07-0. URL http://www.R-project.org/ (last accessed
17 November 2014).
15 Gr
emy J-P. Les exp
eriences francaises sur la formulation des ques-
tions denqu^ete: R
esultats dun premier inventaire (French experience
on the wording of survey questions: Primary results). Revue francaise
de sociologie. Published by Sciences Po University Press. [WWW
document] URL http://www.jstor.org/stable/3321549. 1987;28(4):
567-99.
2015 European Academy of Dermatology and Venereology