2 Toselli S, Gualdi-Russo E, Marzouk D et al. Psychosocial health among
immigrants in central and southern Europe. Eur J Pub Health 2014; 24
(Suppl 1): 26–30.
3 Gilliver SC, Sundquist J, Li X et al. Recent research on the mental health of
immigrants to Sweden: a literature review. Eur J Pub Health 2014; 24(Suppl
4 van Berlaer G, Bohle Carbonell F, Manantsoa S et al. A refugee camp in
the centre of Europe: clinical characteristics of asylum seekers arriving in
Brussels. BMJ Open 2016; 6: e013963.
5 Hannigan A, O’Donnell P, O’Keeffe M et al. WHO Health Evidence Net-
work Synthesis Reports. How do variations in deﬁnitions of “migrant” and
their application inﬂuence the access of migrants to health care services?
WHO Regional Ofﬁce for Europe, Copenhagen, 2016.
6 Trovato A, Reid A, Takarinda KC et al. Dangerous crossing: demographic
and clinical features of rescued sea migrants seen in 2014 at an outpatient
clinic at Augusta Harbor, Italy. Conﬂ Health 2016; 10: 14.
Melanoma and chronic exposure
to contraceptives containing
microdoses of ethinylestradiol in
young women: a retrospective
study from the Research on
Adverse Drug Events and
Reports (RADAR) project
comprising a large Midwestern
U.S. patient population
An association between the use of contraceptives containing
exogenous oestrogen compounds and subsequent diagnosis for
malignant melanoma (MM) has been suspected for decades.
This is, in part, due to the ﬁnding that oestrogen stimulates
and the observation that the incidence of MM is
greater in women vs. men before the age of 50, but lower than in
men after the age of 50,
corresponding with the average age of
menopause when oestrogen levels dramatically decrease.
Prior studies assessing the relationship between the incidence
of MM and exposure to exogenous oestrogen provide conﬂicting
Moreover, the vast majority of previous reports do
not specify the oestrogen-based dosage form, nor the actual dose
of the oestrogenic compound, and typically rely on patient com-
pleted questionnaires that are subject to recall bias. Further, over
recent decades, the dosage of oestrogenic compounds in hor-
monal contraceptives has greatly decreased
compared to studies
of the earliest oestrogenic compounds where an association with
MM was reported.
Given widespread and chronic use, the aim
of this study was to determine whether exposure to contracep-
tives containing ethinylestradiol (EE) at microdoses (40mcg/day
or less) showed an association with MM within a large, urban,
Midwestern U.S. patient population served by an NCI-desig-
nated Comprehensive Cancer Center.
We searched the NMEDW (Northwestern Medicine Enter-
prise Data Warehouse), a medical record data repository for a
large, Midwestern urban medical centre (>4 million patients).
The study population consisted of women 18–40 years old, who
had a ﬁrst clinic encounter (index date) anytime between Jan-
uary 2001 and December 2011 and ≥5 years of clinic follow-up
(prior to January 2017). Inclusion criteria included exposure to
≥12 months of an EE microdose (<40mcg/day as oral tablet,
vaginal ring or transdermal patch). Exclusion criteria included
diagnosis of MM, ≤12 months of EE exposure and exposure to
unknown doses or doses >40mcg/day. The control population
consisted of women in the same age group with no MM diagno-
sis at index date and no EE exposure. ICD-9 and ICD-10 codes
were used to detect data for patients with MM (172 and C43,
respectively). Data for race and age were also collected. Fisher’s
exact test was used to assess for an association.
The characteristics of the study populations (EE-exposed and
non-EE-exposed, n = 77 293) are shown in
Table 1. No signiﬁ-
cant association between EE exposure and MM diagnosis was
detected (Fisher’s exact test P = 0.3).
There was no evidence for an association between exposure to
microdose EE contraceptives and subsequent diagnosis for MM.
These ﬁndings are in contrast to the early studies that reported
an association between those who ever used oestrogen-based
contraceptives and MM,
or an association between chronic use
Table 1 Patient characteristics of EE-exposed and non-EE-exposed populations
Characteristic EE-exposed (n = 2425) Non-EE-exposed (n = 74 868)
Age at index date (years), mean Æ SD 27.1 Æ 4.8 28.7 Æ 5.7
Range 18–40 18–40
Follow-up time (months), mean Æ SD 111.6 Æ 33 105.5 Æ 32.8
Range 60–191 60–190
Time to event (months), mean Æ SD 58.7 Æ 29.3 N/A
Range 37–92 N/A
Persons who developed MM (n) 3 194
Prevalence of MM (%) 0.1 0.3
© 2017 European Academy of Dermatology and Venereology
2018, 32, e86–e121
Letters to the Editor