Mammography utilization in women aged 40–49 years:
the French EDIFICE survey

Xavier Pivot

a,b

, Franc¸ois Eisinger

c,d

, Jean-Yves Blay

e,f

, Yvan Coscas

g

,

Anne Calazel-Benque

h

, Je´roˆme Viguier

i

, Claire Roussel

j

and Jean-Franc¸ois More`re

k,l

In France, mammography screening is offered to women

aged between 50 and 74 years. EDIFICE, the iterative
nationwide survey, collected data on a national level

about consumers utilization of available cancer screening
procedures. This analysis compared data from a subset

of 241 women aged between 40 and 50 years with that
of 488 women aged between 50 and 74 years. Multivariate

analysis showed the following significant factors to be
linked with screening attendance for women younger

than 50 years: age, with a threshold value at 44 years;
awareness of the recommended screening period

(2 years); consultation within the last 12 months with
a general practitioner or a gynaecologist; and long

standing local programme for colon cancer screening.
Notwithstanding the debate on risk/benefit of breast

cancer screening for women in their 40s, we observed
more differences with regard to attendance between

women aged 40–45 years versus that of 46–74 years than
between women aged 40–49 years versus that of 50–74

years. The issue that is unanswered as yet is, do women

make a kind of heuristic for starting breast cancer
screening, leading them to a threshold choice of 45 years

or alternatively to a collective answer of standard age
minus 5 years!

European Journal of Cancer Prevention

20:S16–S19

c

2011 Wolters Kluwer Health | Lippincott

Williams & Wilkins.

European Journal of Cancer Prevention

2011, 20:S16–S19

Keywords: breast cancer, cancer screening, France, mammography
utilization

a

University Hospital Jean Minjoz,

b

INSERM U645, Besanc¸on,

c

Paoli-Calmettes

Institute,

d

INSERM UMR 912, Marseille,

e

Le´on Be´rard Cancer Center,

f

INSERM

U590, Centre L Be´rard & Conticanet, Lyon,

g

Porte de St Cloud Clinic, Boulogne-

Billancourt,

h

Clinique du Parc, Toulouse,

i

Cancer Screening Department,

University Hospital Centre of Tours,

j

Roche S.A., Neuilly-sur-Seine,

k

Avicenne

University Hospital, Bobigny and

l

University Paris XII, Paris, France

Correspondence to Xavier Pivot, Department of Medical Oncology, University
Hospital Jean Minjoz, 25030 Besanc¸on Cedex, France
Tel: + 33 3 81 66 86 93; fax: + 33 3 81 66 88 08;
e-mail: [e-mail]

Background

In France, mammography screening has been offered to
women over the age of 50 years since 1989 in pilot areas,
and throughout the country since 2003, based on one
mammography every 2 years for all women aged between
50 and 74 years. EDIFICE, the iterative nationwide
survey, collected data on a national level about con-
sumers’ utilization of available cancer screening proce-
dures (through organized programmes or individual
initiatives) and general practitioners’ (GPs’) behaviour
towards cancer screening (Eisinger et al., 2008a). The
overall results for breast cancer among women aged
between 50 and 74 years and targeted by the national
screening programme provided data which have clarified
numerous beliefs and assumptions with regard to both
commitment and resistance to breast cancer screening
(Pivot et al., 2008). Elsewhere in this supplement, More

`re

et al. (More

`re et al., 2011) compared the evolution of

behaviour between 2005 and 2008 in this population of
women targeted by the national screening programme.
The aim of this analysis is to compare data on declared
screening attendance from women aged between 40
and 50 years with that of women aged between 50 and
74 years.

The utility curves for breast cancer screening according
to age are not sharp. In an article in this supplement,
Eisinger et al. examine the behaviour of women over the
age of 75 years as reported in the EDIFICE survey
(Eisinger et al., 2011). This analysis focuses on a subset
of women aged between 40 and 50 years. The usefulness
of mammography screening in women in their 40s, with
a standard risk level, is highly controversial. The recent
update from the US Preventive Services Task Force
(2009) is a reminder of the issue highlighted by Fletcher
in 1997 (‘Alice in the wonderland of breast cancer
screening’). In addition to assumptions, beliefs and
scientific data, the reality of the behaviour of women,
which is naturally affected by the background of both the
individual and the society in which they live, is likely to
enlighten and fuel the debate.

In France, there are currently no economic constraints
on breast cancer screening using the mammogram test,
despite the absence of official recommendations for
women in their 40s; when a woman undergoes a mammo-
graphy outside the national screening programme she
is reimbursed by the national health insurance. There
are no financial incentives for GPs to recommend or to

S16

Supplement article

0959-8278

c 2011 Wolters Kluwer Health | Lippincott Williams & Wilkins

DOI: 10.1097/01.cej.0000391565.41383.ba

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

discourage breast cancer screening. Thus, a woman’s deci-
sion to undergo screening or not, actually reflects her own
preference and is not affected by economic bias.

Materials and methods

The methodology used in the EDIFICE survey is
described elsewhere in this supplement in the article
by Roussel and Touboul (Roussel and Touboul, 2011).
This study focuses on an additional sample of 241
women, under the age of 50 years, with no personal
history of cancer, and who were not included within the
systematic screening programme. This subset was com-
pared with 488 women aged between 50 and 74 years,
with no personal history of cancer and who were targeted
by the national screening programme.

Statistical analysis

The statistical method is described by Roussel and Touboul
elsewhere in this supplement (Roussel and Touboul, 2011).
The precision of the results was ± 6.3% for a 95%
confidence interval (CI) in the 488 women eligible in the
national French screening plan and the 241 younger
women. The population of women below the age of 50
years was compared with the population of women eligible
for the national French screening plan. Comparison of
distribution according to the behaviour of women was
carried out using the Student’s t-test for quantitative data,
and the Z-test and the w

2

test for the comparison of

percentages and numbers, respectively, in the case of
categorical data. Logistic regression was used to identify
factors explaining the behaviour to undergo a breast cancer
screening and the risk to withdraw over time. Multivariate
logistic regression analyses were expressed in terms of odds
ratio (OR) and 95% CI, and carried out using the SAS
software, version 8.2, SAS Institute Inc., Cary, North
Carolina, USA (proc FREQ and proc LOGISTIC proce-
dures).

Results

The characteristics of both populations questioned in the
survey (subset under 49 years and those aged between 50
and 74 years) are summarized in Table 1. In women aged
below 50 years and who reported having undergone a mam-
mogram, the mean number of tests was 3.2. The mean age
declared at first mammography was 37.1 years (standard
deviation: 7.0). The reasons for the first mammogram
among the women aged below 50 years were symptom
management in 23%, familial history in 14% (interestingly,
3% of women declared that the affected person herself was
at the root of screening) and medical advice in 58% (GP in
12% and gynaecologist in 45%).

In the monovariate analysis of the younger subset of
women (40–49 years) in comparison with the women
eligible according to the national French screening plan
(50–74 years) and who reported undergoing a screening

mammography, the following characteristics were signifi-
cantly different: high socioeconomic level (> h3000/month,
OR = 2.59, 95% CI = 1.72–3.89); high educational level
(OR = 1.65, 95% CI = 1.14–2.4); single (OR = 1.46, 95%
CI = 0.97–2.2); greater frequency of gynaecological exam-
ination (OR = 8.85, 95% CI = 5–14.68); a higher rate of a
history of breast cancer in relatives (OR = 1.49, 95%
CI = 0.9–2.35) and a higher level of motivation or concern
towards

screening

(OR = 1.68,

95%

CI = 0.76–3.72)

(Fig. 1).

In the multivariate analysis, the following significant
factors were linked with screening attendance for women
aged below 50 years (never users versus at least one
mammogram): age, with a threshold value at 44 years
(OR = 10, 95% CI = 3.8–25) for women aged between 45
and 49 years versus those aged between 40 and 44 years
(Fig. 2); awareness of the recommended screening period
(2 years) (adjusted OR = 3.69, 95% CI = 1.5–9.05);
consultation within the last 12 months by a GP or a
gynaecologist (adjusted OR = 3.28, 95% CI = 1.32–8.14)
and long-lasting local programme for colon cancer screen-
ing (adjusted OR = 4.17, 95% CI = 1.21–14.40) (Fig. 3).

Discussion

This EDIFICE survey highlights the high rate (78%) of
women aged below 50 years and who reported having
undergone a screening mammograph. This is a higher
proportion than the rate reported in the literature. In the
US Behavioral Risk Factor Surveillance System survey

Table 1

Characteristics of 729 women

Women eligible for the national screening

plan (n = 729)

Aged between 50

and 74 years

n = 488

Aged between 40

and 50 years

n = 241

Women who reported having had a

mammography

n = 460

n = 188

Mean age (standard deviation)

60.4 ( ± 7.0) years

45.2 ( ± 2.8) years

Marital status

Married/cohabitation

73%

80%

Incomes

2955

2317h

Study levels

At least Baccalaureat

23%

33%

Smoker

17%

27%

Alcohol consumer

11%

5%

Consultation with a physician during

the last 12 months

90%

88%

Consultation with a GP during the

last 12 months

86%

82%

Consultation with a gynaecologist

during the last 12 months

43%

58%

Cancer of any type known among

relatives

83%

88%

Breast cancer among relatives

41%

53%

Feel concerned by cancer

81%

76%

Feel concerned by breast cancer

74%

72%

Awareness of the recommended

screening period (2 years)

69%

55%

Long-standing local colon cancer

screening programme

27%

38%

GP, general practitioner.

Mammography utilization in women aged 40–49 years

Pivot et al.

S17

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

(2000), 69.6% of women aged between 40 and 49 years
had reported having undergone a mammogram in the past
2 years (Cronin et al., 2009). In 2000, the rates in US were
between 65.5 and 68.3%, according to the National Health
Interview Survey (Swan et al., 2003). In this study, the
multivariate analysis would appear to suggest that commu-
nication with regard to the breast cancer screening national
plan significantly influenced early screening. One can con-
sider that the significant value of the awareness of the
recommended screening period (adjusted OR = 3.69)
reflects a commitment for screening and therefore for an
early test.

The existence of a large organized programme to detect
cancer, such as colon cancer screening programme

(Eisinger et al., 2008b), increased the sensitivity to
screening benefit and also had a positive impact on
screening attendance (adjusted OR = 4.17). The role of
the GP or a gynaecologist (adjusted OR = 3.28) remained
crucial and it is a major factor to successively improve
screening programmes (Fig. 3) (Eisinger et al., 2008a).
Nevertheless, women’s behaviour as reported cannot be
seen as the ‘right’ answer to a tough and sensitive
controversy. It could not even be used as a proxy for a
balance between risk and perceived benefit, but only as a
perceived risk/perceived benefit global output (Andersen
et al., 2003).

The threshold issue to identify the threshold age to start
screening for breast cancer is highly controversial. Risk of
being affected and benefit of screening are indeed age-
dependant, but with a continuous and curvilinear pattern.
There is no clear answer at the individual level and even
at the collective public health level. The choice of an
intervention is dichotomic: to perform or not, while risk is
essentially a quantitative concept. Guidelines for risk
management therefore face the problem of how to cope
with, on one hand, a dichotomised choice and, on the
other, a continuous risk curve. Definitively, when looking
at curves showing the proportion of women reporting
having undergone a screening mammography (Fig. 2), the
actual age threshold in lay practice appears to be 45 years
rather than 50 years. Notwithstanding the debate on the
risk/benefit of breast cancer screening for women in their
40s, we observed a greater difference for attendance
between women aged 40–45 years versus those aged
46–74 years than between women aged 40–49 years versus
those aged 50–74 years. The real and as yet unanswered

Fig. 1

High socioeconomic level
(

> 3000 /month)

High educational level

Single

2.59

1.65

1.46

8.57

Gynaecological exam within the last 12
months

1.49

Breast cancer among relatives

1.68

Motivated or concerned by cancer screening

3

2

1

0

1

2

3

4

5

6

7

8

9

10

11

Characteristics shown to be significantly different in the monovariate analysis between the younger subset of women in comparison with the women
eligible for the national French screening plan.

Fig. 2

100

80

60

40

20

0

40–44 45–49 50–54 55–59

Age

60–64 65–69 70–74

> 75

66

92

96

93

100

91

78

88

Proportion of women who reported having undergone a mammography
for screening.

S18

European Journal of Cancer Prevention

2011, Vol 20 Supplement 1

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

issue is whether women make a kind of heuristic for
starting breast cancer screening, leading them to choose
an age threshold of 45 years, or alternatively whether they
are living in a world in which bargaining is always positive
and their collective answer is to choose the standard age
minus 5 years!

In all cases, the behaviour of women aged between 40 and
49 years in terms of attendance for breast cancer screen-
ing suggests that it is highly feasible to start an early
screening programme. This study supports the idea of
developing an experimental programme to assess the
benefit and the risk for breast cancer screening in women
aged between 40 and 49 years. The French National
Cancer Institute will propose such an investigational pro-
gramme in three French regions within the coming year.

Acknowledgements

This study was directly supported by Roche S.A. Data used
came from different groups (CFES, INPES, EDIFICE),
the two former funded directly by the Ministry of Health,
and the latter by Roche S.A.. The authors thank Ray Hill of
inScience Communications, a Wolters Kluwer business, who
helped in copyediting and journal styling of the manuscript
before submission. The authors wish to thank Laurent Cals
for his valuable contribution to the EDIFICE survey.
Franc

¸ois Eisinger, Je

´ro

ˆme Viguier, Jean-Yves Blay, Yvan

Coscas, Jean-Franc

¸ois More

`re and Xavier Pivot have

received honoraria from Roche S.A. There is however no

conflict of interest with regard to the content of this
contribution.

References

Andersen MR, Smith R, Meischke H, Bowen D, Urban N (2003). Breast cancer

worry and mammography use by women with and without a family history in a
population-based sample. Cancer Epidemiol Biomarkers Prev 12:314–320.

Cronin KA, Miglioretti DL, Krapcho M, Yu B, Geller BM, Carney PA, et al. (2009).

Bias associated with self-report of prior screening mammography. Cancer
Epidemiol Biomarkers Prev 18:1699–1705.

Eisinger F, Blay JY, Morere JF, Rixe O, Calazel-Benque A, Cals L, et al. (2008a).

Cancer screening in France: subjects’ and physicians’ attitudes. Cancer
Causes Control 19:431–434.

Eisinger F, Cals L, Calazel-Benque A, Blay JY, Coscas Y, Dolbeault S, et al.

(2008b). Impact of organised programs on colorectal cancer screening.
BMC Cancer 8:104–111.

Eisinger F, Viguier J, Blay J-Y, More`re J-F, Coscas Y, Roussel C, Pivot X (2011).

Uptake of breast cancer screening in women aged over 75 years: a
controversy to come? Eur J Cancer Prev 20 (Suppl 1):S13–S15.

Fletcher SW (1997). Whither scientific deliberation in health policy recommenda-

tions? Alice in the Wonderland of breast-cancer screening. N Engl J Med
336

:1180–1183.

More`re J-F, Pivot X, Viguier J, Blay J-Y, Calazel-Benque A, Coscas Y, et al. (2011).

Breast cancer screening in women aged 50–74 years: is there room for
improvement? Eur J Cancer Prev 20 (Suppl 1):S8–S12.

Roussel C, Touboul C (2011). Large population survey: strengths and limits.

Methodology of the EDIFICE survey. Eur J Cancer Prev 20 (Suppl 1):S5–S7.

Pivot X, Rixe O, More`re J, Coscas Y, Cals L, Namer M, et al. (2008). Breast

cancer screening in France: results of the EDIFICE survey. Int J Med Sci
5

:106–112.

Swan J, Breen N, Coates RJ, Rimer BK, Lee NC (2003). Progress in cancer

screening practices in the United States: results from the 2000 National
Health Interview Survey. Cancer 97:1528–1540.

US Preventive Services Task Force (2009). Screening for breast cancer: US

Preventive Services Task Force recommendation statement. Ann Intern Med
151

:716–726.

Fig. 3

10

4.17

3.69

Age (45–49 vs. 40–44 years)

Long-standing local programme
for colon cancer screening

Gynaecological exam within the
last 12 months

Awareness of the recommended
screening period for breast cancer

3.28

3

2

1

0

1

2

3

4

5

6

7

8

9

10

11

Significant factors in the multivariate analysis linked with screening attendance for women aged below 50 years.

Mammography utilization in women aged 40–49 years

Pivot et al.

S19

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.