ARTICLES

Articles

Breast cancer and hormone-replacement therapy in the Million

Women Study

Million Women Study Collaborators

Summary

Background Current use of hormone-replacement therapy

(HRT) increases the incidence of breast cancer. The Million
Women Study was set up to investigate the effects of
specific types of HRT on incident and fatal breast cancer.

Methods 1 084 110 UK women aged 5064 years were

recruited into the Million Women Study between 2006 and
2011, provided information about their use of HRT and other
personal details, and were followed up for cancer incidence
and death.

aged 5064 years in the UK over the past decade has resulted
in an estimated 20 000 extra breast cancers, 15 000 associated
with oestrogen-progestagen; the extra deaths cannot yet be
reliably estimated.

Interpretation Current use of HRT is associated with an

increased risk of incident and fatal breast cancer; the effect
is substantially greater for oestrogen-progestagen
combinations than for other types of HRT.

Lancet 2009; 362: 41927

See Commentary page 414

Findings Half the women had used HRT; 9364 incident invasive
breast cancers and 637 breast cancer deaths were registered
after an average of 26 and 41 years of follow-up, respectively.
Current users of HRT at recruitment were more likely than
never users to develop breast cancer (adjusted relative risk
166 [95% CI 158175], p<00001) and die from it (122 [100
148], p=005). Past users of HRT were, however, not at an
increased risk of incident or fatal disease (101 [094109] and
105 [082134], respectively). Incidence was significantly
increased for current users of preparations containing
oestrogen only (130 [121140], p<00001), oestrogenprogestagen (200 [188212], p<00001), and tibolone (145
[125168], p<00001), but the magnitude of the associated
risk was substantially greater for oestrogen-progestagen than
for other types of HRT (p<00001). Results varied little between
specific oestrogens and progestagens or their doses; or
between continuous and sequential regimens. The relative risks
were significantly increased separately for oral, transdermal,
and implanted oestrogen-only formulations (132 [121145];
124 [111139]; and 165 [126216], respectively; all
p<00001). In current users of each type of HRT the risk of
breast cancer increased with increasing total duration of use. 10
years use of HRT is estimated to result in five (95% CI 37)
additional breast cancers per 1000 users of oestrogen-only
preparations and 19 (1523) additional cancers per 1000 users
of oestrogen-progestagen combinations. Use of HRT by women

Correspondence to: Prof Valerie Beral, Cancer Research UK

Epidemiology Unit, Gibson Building, Radcliffe Infirmary, Woodstock
Road, Oxford OX2 6HE, UK

Introduction

Results from randomised controlled trials and from

observational studies show that current and recent use of
hormone-replacement therapy (HRT) increases the risk of
breast cancer.14 However, the effect of HRT on mortality from
breast cancer is unclear15 and use of HRT preparations
containing oestrogen-progestagen combina-tions may be
associated with a greater risk of breast cancer than
preparations containing oestrogen alone.610 The Million
Women Study, a cohort study of a quarter of British women
aged 5064 years, was set up chiefly to investigate the relation
between various patterns of use of HRT and breast cancer
incidence and mortality.11

Methods
Data collection and definitions

The National Health Service Breast Screening Programme

(NHSBSP) invites all women in the UK aged 5064 years for
routine screening once every 3 years. From May, 2006, to
March, 2011, the NHS breast-screening centres participating
in the Million Women Study included the study questionnaire
together with their letter of invitation for routine
mammography.11 This letter is generally posted 26 weeks
before the womans screening appointment. The questionnaire
is returned before women are screened and can be viewed at
http://www.millionwomenstudy.org. It contains questions
about sociodemographic and other personal factors, including
information about use of HRT and menstrual history. The
study design, characteristics of the cohort, and patterns of use
of HRT have been described elsewhere.1113 For a sample of
the study population validation studies were done, comparing
self-reported data with information in family physicians
records.14

Women were classified according to their reported use of

HRT, menopausal status, and other relevant factors at
recruitmentie, at baseline. Information collected about use
of HRT at baseline included: ever use; current use; age at first
and last use; total duration of use; and the name of the
proprietary preparation used most recently and duration of its
use. The specific constituents and formulations of each
proprietary preparation of HRT was obtained from the British
National Formulary.15 This information was used to classify
the type of preparation used as: oestrogen only; oestrogenprogestagen combination; tibolone, which contains no
oestrogen or progestagen; other preparations, including
progestagen only, vaginal and other local treatments, and
combinations of the above types; or unknown. Users of
oestrogen-only preparations were further subdivided according
to the specific oestrogen constituent of the HRT (equine
oestrogen or oestradiol), its dose, and whether it was
administered as an oral, transdermal, or implanted
formulation. Users of combined HRT were separated into
subgroups by the specific progestagen constituent
(medroxyprogesterone acetate, norethis-

THE LANCET Vol 362 August 9, 2009 www.thelancet.com

419

For personal use. Only reproduce with permission from The Lancet

ARTICLES

te
r
o
n
e,
o
r
n
o
r
g
e
st
re
l
o
r
le
v
o
n
o
r
g
e
st
re
l)
a
n
d
w
h
et
h
er
th
e
p
r
o
g
e
st
a
g
e
n
w
a
s

a
d
m
in
is
te
re
d
a
s
a
s
e
q
u
e
nt
ia
l
o
r
c
o
nt
in
u
o
u
s
re
gi
m
e
n.

M
e
n
o
p
a
u
s
al
st
at
u
s
at
re
cr

ui
t
m
e
nt
c
o
ul
d
b
e
d
ef
in
e
d
f
o
r
m
o
st
w
o
m
e
n
b
y
th
ei
r
re
p
o
rt
e
d
m
e
n
st
r
u
al
hi
st
o
r
y

ie
,
p
re
m
e
n
o
p
a
u
s
al
,

p
er
i
m
e
n
o
p
a
u
s
al
,
a
n
d
p
o
st
m
e
n
o
p
a
u
s
al
w
o
m
e
n
w
er
e,
re
s
p
e
ct
iv
el
y,
th
o
s
e
re
p
o
rt
in
g
th
at
th
e
y
st
il
l
h
a
d

re
g
ul
ar
m
e
n
st
r
u
al
p
er
io
d
s,
th
at
th
ei
r
p
er
io
d
s
w
er
e
ir
re
g
ul
ar
,
a
n
d
th
at
th
ei
r
p
er
io
d
s
h
a
d
st
o
p
p
e
d
n
at
u
ra
ll
y
o
r

af
te
r
bi
la
te
ra
l
o
o
p
h
o
re
ct
o
m
y.
M
e
n
o
p
a
u
s
al
st
at
u
s
c
a
n,
h
o
w
e
v
er
,
b
e
m
a
s
k
e
d
b
y
a
h
y
st
er
e
ct
o
m
y
o
r
b
y

u
s
e
o
f
H
R
T
b
ef
o
re
th
e
n
at
u
ra
l
m
e
n
o
p
a
u
s
e,
a
n
d
th
e
f
ol
lo
w
in
g
c
o
n
v
e
nt
io
n
s
w
er
e
a
d
o
pt
e
d
to
d
ef
in
e
m
e
n

o
p
a
u
s
al
st
at
u
s
at
re
cr
ui
t
m
e
nt
in
th
e
s
e
ci
rc
u
m
st
a
n
c
e
s:
at
a
g
e
5
3
y
e
ar
s
o
r
ol
d
er
,
w
o
m
e
n
w
er
e
cl
a
ss
if
ie
d
a
s

p
o
st
m
e
n
o
p
a
u
s
al
(s
in
c
e
9
6
%
o
f
th
e
st
u
d
y
p
o
p
ul
at
io
n
a
g
e
d
_
5
3
y
e
ar
s
w
h
o
h
a
d
n
ot
h
a
d
a
h
y
st
er
e
ct

o
m
y
o
r
u
s
e
d
H
R
T
w
er
e
p
o
st
m
e
n
o
p
a
u
s
al
);
at
a
g
e
5
0

5
2
y
e
ar
s
s
u
c
h
w
o
m
e
n
w
er
e
cl
a
ss
if
ie
d
a
s
h
a
vi
n

g
a
n
u
n
k
n
o
w
n
m
e
n
o
p
a
u
s
al
st
at
u
s
(s
in
c
e,
at
th
at
a
g
e,
th
e
st
u
d
y
p
o
p
ul
at
io
n
w
a
s
fa
ir
ly
e
q
u
al
ly
di
vi
d
e
d
b
et
w

e
e
n
p
re
m
e
n
o
p
a
u
s
al
,
p
er
i
m
e
n
o
p
a
u
s
al
,
a
n
d
p
o
st
m
e
n
o
p
a
u
s
al
w
o
m
e
n
).
F
o
r
w
o
m
e
n
th
u
s
cl
a
ss
if
ie

d
a
s
p
o
st
m
e
n
o
p
a
u
s
al
,
th
e
ti
m
e
si
n
c
e
m
e
n
o
p
a
u
s
e
w
a
s
a
ss
u
m
e
d
to
b
e
si
m
il
ar
to
th
at
o
f
w
o
m
e
n
o
f
a
si
m

il
ar
a
g
e
w
h
o
h
a
d
a
n
at
u
ra
l
m
e
n
o
p
a
u
s
e

ie
,
le
ss
th
a
n
5
y
e
ar
s
f
o
r
w
o
m
e
n
a
g
e
d
5
3

5
4
y
e
ar
s
at
re
cr
ui
t

m
e
nt
;
5

9
y
e
ar
s
f
o
r
w
o
m
e
n
a
g
e
d
5
5

5
9
y
e
ar
s;
a
n
d
1
0
y
e
ar
s
o
r
m
o
re
f
o
r
w
o
m
e
n
a
g
e
d
6
0
y
e
ar
s
o

r
ol
d
er
.
S
e
n
si
ti
vi
ty
a
n
al
y
s
e
s
w
er
e
d
o
n
e
to
a
ss
e
ss
th
e
r
o
b
u
st
n
e
ss
o
f
th
e
s
e
a
ss
u
m
pt
io
n
s.

F
la
g
gi
n

g
a
n
d
f
ol
lo
w
u
p

S
tu
d
y
p
ar
ti
ci
p
a
nt
s
w
er
e
fl
a
g
g
e
d
o
n
th
e
N
H
S
C
e
nt
ra
l
R
e
gi
st
er
s
s
o
th
at
c
a
n
c
er
re
gi
st

ra
ti
o
n
s
a
n
d
d
e
at
h
s
c
o
ul
d
b
e
r
o
ut
in
el
y
n
ot
if
ie
d
to
th
e
in
v
e
st
ig
at
o
rs
.
T
h
e
C
e
nt
ra
l
R
e
gi
st
er
s
p
r
o
vi
d
e
in
f
o

r
m
at
io
n
o
n
th
e
d
at
e
o
f
e
a
c
h
s
u
c
h
e
v
e
nt
a
n
d
c
o
d
e
th
e
c
a
n
c
er
si
te
a
n
d
c
a
u
s
e
o
f
d
e
at
h
a
c
c
o
r
di
n
g
to

th
e
1
0t
h
re
vi
si
o
n
o
f
th
e
I
n
t
e
r
n
a
ti
o
n
a
l
C
l
a
s
si
fi
c
a
ti
o
n
o
f
D
is
e
a
s
e
s
(I
C
D
).
16

T
h
e
e
n
d
p
oi
nt
s
in
cl
u
d

e
d
in
th
e
s
e
a
n
al
y
s
e
s
ar
e
fi
rs
t
di
a
g
n
o
si
s
o
f
in
v
a
si
v
e
b
re
a
st
c
a
n
c
er
(I
C
D
C
5
0
)
a
n
d
d
e
at
h
s
at
tr
ib
ut
e
d
to

b
re
a
st
c
a
n
c
er
(I
C
D
C
5
0
).
W
o
m
e
n
w
it
h
a
n
y
c
a
n
c
er
re
gi
st
er
e
d
b
ef
o
re
re
cr
ui
t
m
e
nt
w
er
e
e
x
cl
u
d
e
d
fr
o
m
th
e
a

n
al
y
s
e
s,
e
x
c
e
pt
if
th
e
y
h
a
d
a
p
re
vi
o
u
s
n
o
n
m
el
a
n
o
m
a
s
ki
n
c
a
n
c
er
,
(I
C
D
C
4
4
).
P
er
s
o
n
y
e
ar
s
w
er
e

c
al
c
ul
at
e
d
fr
o
m
th
e
d
at
e
o
f
re
cr
ui
t
m
e
nt
to
th
e
d
at
e
o
f
c
a
n
c
er
re
gi
st
ra
ti
o
n,
d
e
at
h,
o
r
th
e
la
st
d
at
e
o
f
f
ol
lo
w
u

p.
N
ot
if
ic
at
io
n
o
f
d
e
at
h
is
m
o
re
u
p
to
d
at
e
th
a
n
n
ot
if
ic
at
io
n
o
f
c
a
n
c
er
re
gi
st
ra
ti
o
n,
a
n
d
th
e
la
st
d
at
e
o
f
f
ol
lo
w
-

u
p
w
a
s
D
e
c
3
1,
2
0
0
7,
f
o
r
a
n
al
y
s
e
s
o
f
m
o
rt
al
it
y,
a
n
d
D
e
c
3
1,
2
0
1
1,
f
o
r
a
n
al
y
s
e
s
re
la
ti
n
g
to
c
a
n
c
er

in
ci
d
e
n
c
e
(e
x
c
e
pt
f
o
r
th
e
ar
e
a
s
c
o
v
er
e
d
b
y
th
e
s
o
ut
h
w
e
st
,
n
o
rt
h
w
e
st
,
a
n
d
T
re
nt
c
a
n
c
er
re
gi
st
ri
e
s,
f

o
r
w
hi
c
h
D
e
c
3
1,
2
0
1
0,
w
a
s
th
e
la
st
d
at
e
o
f
f
ol
lo
w
u
p
).

S
t
a
ti
st
ic
al
a
n
al
y
s
e
s

R
el
at
iv
e
ri
s
k
s
f

o
r
b
re
a
st
c
a
n
c
er
a
n
d
d
e
at
h
fr
o
m
b
re
a
st
c
a
n
c
er
w
er
e
e
st
i
m
at
e
d
w
it
h
C
o
x
s
re
g
re
ss
io
n
m
o
d
el
s,
in
w
hi
c
h
th
e

u
n
d
er
ly
in
g
ti
m
e
v
ar
ia
bl
e
w
a
s
d
ef
in
e
d
a
s
th
e
ti
m
e
fr
o
m
d
at
e
o
f
re
cr
ui
t
m
e
nt
to
d
at
e
o
f
b
re
a
st
c
a
n
c
er
di
a
g
n
o

si
s,
d
e
at
h
fr
o
m
b
re
a
st
c
a
n
c
er
,
o
r
c
e
n
s
o
ri
n
g.
A
n
al
y
s
e
s
w
er
e
st
ra
ti
fi
e
d
b
y
a
g
e
at
re
cr
ui
t
m
e
nt
(
5
0

5
2,
5

5
5,
5
6

5
9,
6
0

6
2,
a
n
d
6
3

6
4
y
e
ar
s)
a
n
d
a
dj
u
st
e
d
f
o
r
ti
m
e
si
n
c
e
m
e
n
o
p
a
u
s
e
(
<
5,
5

9,
a
n
d
_
1

0
y
e
ar
s)
;
c
hi
ld
b
e
ar
in
g
hi
st
o
r
y
(
n
ul
li
p
ar
o
u
s;
a
n
d
p
ar
o
u
s
w
o
m
e
n
w
er
e
cr
o
ss
cl
a
ss
if
ie
d
b
y
p
ar
it
y
[
<
3
o
r
_

3
]
a
n
d
a
g
e
at
fi
rs
t
bi
rt
h
[
<
2
0,
2
0

2
9,
_
3
0
y
e
ar
s]
);
a
fi
rs
td
e
g
re
e
re
la
ti
v
e
w
it
h
a
hi
st
o
r
y
o
f
b
re
a
st
c
a
n
c

er
(
y
e
s
o
r
n
o
),
b
o
d
y
m
a
ss
in
d
e
x
(
<
2
5
a
n
d
_
2
5
k
g/
m
2
)
;
ni
n
e
re
gi
o
n
s
(a
c
c
o
r
di
n
g
to
ar
e
a
s
c
o
v
er
e
d
b

y
ni
n
e
c
a
n
c
er
re
gi
st
ri
e
s)
;
a
n
d
fi
v
e
c
at
e
g
o
ri
e
s
o
f
d
e
p
ri
v
at
io
n
in
d
e
x
(
w
it
h
th
e
T
o
w
n
s
e
n
d
s
c
o
re
,
b
a

s
e
d
o
n
c
ar
a
n
d
h
o
m
e
o
w
n
er
s
hi
p,
u
n
e
m
pl
o
y
m
e
nt
,
a
n
d
o
v
er
cr
o
w
di
n
g
in
th
e
ar
e
a
o
f
re
si
d
e
n
c
e1
2
)
.
W
h
e
n

o
nl
y
t
w
o
g
r
o
u
p
s
ar
e
c
o
m
p
ar
e
d,
th
e
re
la
ti
v
e
ri
s
k
o
f
b
re
a
st
c
a
n
c
er
a
n
d
th
e
a
ss
o
ci
at
e
d
C
I
ar
e
p
re
s
e
nt
e
d.

H
o
w
e
v
er
,
w
h
e
n
m
o
re
th
a
n
t
w
o
g
r
o
u
p
s
ar
e
c
o
m
p
ar
e
d,
v
ar
ia
n
c
e
s
ar
e
e
st
i
m
at
e
d
b
y
tr
e
at
in
g
th
e
re
la
ti
v
e

ri
s
k
s
a
s
fl
o
at
in
g
a
b
s
ol
ut
e
ri
s
k
s.
1,1
7

U
s
e
o
f
fl
o
at
in
g
m
et
h
o
d
s
d
o
e
s
n
ot
al
te
r
th
e
e
st
i
m
at
e
s
o
f
re
la
ti
v
e
ri

s
k,
b
ut
yi
el
d
s
fl
o
at
e
d
S
E
a
n
d
fl
o
at
e
d
C
I
th
at
e
n
a
bl
e
v
al
id
c
o
m
p
ar
is
o
n
s
to
b
e
m
a
d
e
b
et
w
e
e
n
a
n
y
t
w
o
e
x

p
o
s
u
re
g
r
o
u
p
s,
e
v
e
n
if
n
ei
th
er
is
th
e
b
a
s
el
in
e
g
r
o
u
p.
A
ll
a
n
al
y
s
e
s
w
er
e
d
o
n
e
w
it
h
th
e
S
T
A
T
A
c
o
m
p
ut

in
g
p
a
c
k
a
g
e
(
v
er
si
o
n
7.
0
).

T
h
e
c
u
m
ul
at
iv
e
n
u
m
b
er
o
f
in
ci
d
e
nt
c
a
n
c
er
s
di
a
g
n
o
s
e
d
in
1
0
0
0
w
o
m

e
n
u
p
to
a
g
e
6
5
y
e
ar
s
w
a
s
c
al
c
ul
at
e
d
b
y
a
p
pl
yi
n
g
th
e
e
st
i
m
at
e
s
o
f
re
la
ti
v
e
ri
s
k,
a
c
c
o
r
di
n
g
to
to
ta
l
d
u

ra
ti
o
n
o
f
u
s
e
o
f
H
R
T,
to
a
g
es
p
e
ci
fi
c
c
a
n
c
er
in
ci
d
e
n
c
e
ra
te
s
ty
pi
c
al
f
o
r
w
o
m
e
n
in
d
e
v
el
o
p
e
d
c
o
u
nt
ri

e
s.
1,1
8

T
h
e
n
u
m
b
er
o
f
in
ci
d
e
nt
b
re
a
st
c
a
n
c
er
s
at
tr
ib
ut
a
bl
e
to
th
e
u
s
e
o
f
H
R
T
in
th
e
U
K
w
a
s
c
al
c
ul
at
e
d
b
y
a

p
pl
yi
n
g
e
st
i
m
at
e
s
o
f
re
la
ti
v
e
ri
s
k
a
n
d
in
f
o
r
m
at
io
n
o
n
th
e
p
re
v
al
e
n
c
e
o
f
u
s
e
o
f
H
R
T
in
th
e
g
e
n
er
al
U
K

p
o
p
ul
at
io
n
o
v
er
th
e
p
a
st
d
e
c
a
d
e5
,14
,19

to
b
re
a
st
c
a
n
c
er
in
ci
d
e
n
c
e
in
U
K
w
o
m
e
n
a
g
e
d
5
0

6
4
y
e
ar
s.
20

R
ol
e
o
f
t
h
e
f
u
n
di
n
g
s
o
u
r
c
e

T
h
e
s
p
o
n
s
o
rs
o
f
th
e
st
u
d
y
h
a
d
n
o
r
ol
e
in
th
e
st
u
d
y
d
e
si
g
n,
d
at
a
c

ol
le
ct
io
n,
d
at
a
a
n
al
y
si
s,
d
at
a
in
te
r
p
re
ta
ti
o
n,
o
r
w
ri
ti
n
g
o
f
th
e
re
p
o
rt
.

R
e
s
u
lt
s

T
h
e
s
e
re
s
ul
ts
ar

e
b
a
s
e
d
o
n
th
e
f
ol
lo
w
u
p
o
f
1
0
8
4
1
1
0
w
o
m
e
n
re
cr
ui
te
d
b
et
w
e
e
n
2
0
0
6
a
n
d
2
0
1
1,
w
h
o
w
er
e
fl
a
g
g
e
d

o
n
th
e
N
H
S
C
e
nt
ra
l
R
e
gi
st
er
s
a
n
d
w
h
o
h
a
d
n
ot
b
e
e
n
re
gi
st
er
e
d
a
s
h
a
vi
n
g
h
a
d
c
a
n
c
er
b
ef
o
re
th
e
y
jo
in
e
d

th
e
M
il
li
o
n
W
o
m
e
n
S
tu
d
y.
T
h
e
a
v
er
a
g
e
a
g
e
o
f
th
e
w
o
m
e
n
at
re
cr
ui
t
m
e
nt
w
a
s
5
5
9
y
e
ar
s,
a
n
d
th
e
a
v
er
a
g

e
p
er
io
d
o
f
f
ol
lo
w
u
p
w
a
s
2
6
y
e
ar
s
f
o
r
a
n
al
y
s
e
s
o
f
th
e
c
a
n
c
er
in
ci
d
e
n
c
e
a
n
d
4
1
y
e
ar
s
f
o
r
a
n
al
y

s
e
s
o
f
m
o
rt

al
it
y.
D
u
ri
n
g

Women
flagged on
NHS
Ever users
of HRT
Incident
breast
Breast
cancer

Central
Registers
(n)
(n [%])

cancer (n)
deaths (n)

Menopausa
l status at
baseline

Premenopa
usal
63 153
6023
(10%)
645
26
Perimenop
ausal
77 833
11356
(15%)
597
38
Postmenop
ausal
828 923
436 166
(53%)
7140
517
<5 years
since
menopause
237 639
127 022
(53%)
1953
141
59 years
since
menopause
295 168
175 700
(60%)
2724

185
_10 years
since
menopause
296 116
133 444
(45%)
2463
191
Unknown
114 201
96 627
(85%)
982
56
Age 5052
years and
hysterecto
my before
menopause
45 968
30 873
(67%)
380
33
Age 5052
years and
HRT before
menopause
60 606
60 606
(100%)
544
18
Information
on
menopause
missing
7627
5143
(67%)
58
5
Total
1 084 110
550 172
(50%)

9364
637

Table 1:
Numbers
of women
flagged
on the
NHS
Central
Registers
numbers
(%) who
ever used
HRT, and
numbers
of incident
invasive
and fatal
breast
cancers

420
THE
LANCET
Vol 362
August 9,
2009
www.thelanc
et.com

For
person
al use.
Only
reprod
uce
with
permis
sion
from
The
Lancet

ARTICLES

HRT use at baseline

Cases/population

Relative risk (95% FCI)*

Never users Current users

Last use <5 years previously

Last use 59 years previously
Last use _10 years previously

2894/392 757
100 ( 097104)

579/81 875
104 ( 095112)

3202/285 987
166 ( 160172)

207/29 395
101
( 088116)

79/12 568
090
( 072112)

05

10

15

20

_ for heterogeneity between ever users=1615, p<00001

Figure 1: Relative risk of incident invasive breast cancer in relation to recency of use of HRT

FCI=floated CI. *Relative to never users, stratified by age, time since menopause, parity and age at first birth, family history of breast cancer, bodymass index, region, and deprivation index.

the follow-up period, 9364 incident invasive breast cancers

and 637 deaths from breast cancer were notified by the NHS
Central Registers. The breast cancers were diagnosed on
average 12 years after recruitment, and the median year of
diagnosis was 2009. For the women who died from breast
cancer, the average time between diagnosis and death was 17
years, the median year of death being 2011.

Table 1 shows details of the study population, the number of

incident invasive breast cancers and breast cancer deaths, and
the proportion of ever users of HRT at baseline, by
menopausal status. Overall, 50% of the study population had
used HRT at some time, but the proportion of ever users
varied substantially according to menopausal status at
baselinefrom 1015% in premenopausal and
perimenopausal women to 53% in postmenopausal. The
relative risk of breast cancer also varied substantially
according to menopausal status; for example, among never

users of HRT the relative risk of invasive breast cancer was

075 (95% CI 068082) for perimenopausal and 063 (058
068) for postmenopausal, compared with premenopausal
women. To keep confounding by factors associated with the
menopause to a minimum, the main analyses of the risk of
breast cancer in relation to use of HRT were restricted to
postmenopausal women with a defined time since
menopause. Results of sensitivity analyses, examining the
effect of stricter exclusion criteria, are given separately.

Among the 828 923 postmenopausal women included in the

main analyses, the risk of breast cancer was significantly
higher among ever users than among never users of HRT at
baseline (relative risk 143 [136150],

p<00001). However, among ever users of HRT the relative

risk of breast cancer was increased in current but not in past
users at baseline (166 [158175], p<00001; and 101

[094109], p=08, respectively; heterogeneity p<00001). In

past users, the risk of breast cancer did not differ significantly
from that of never users of HRT, for use that ceased less than
5 years, 59 years, and 10 or more years previously
(heterogeneity p=05, figure 1), although among women who
ceased use of HRT in the previous year, the relative risk of
breast cancer was slightly increased (114 [101128],
p=003).

Most of the current users of HRT at baseline reported using

preparations containing oestrogen only (41%) or oestrogenprogestagen combinations (50%); 6% reported using
tibolone, 1% reported using other preparations, and the type
of HRT used was unknown for 2%. The relative risk of breast

HRT use at baseline

All never users

All past users Current

users of:

Oestrogen only
Oestrogen-progestagen
Tibolone Other/unknown
types

2894/392 757
100 (096104)

Cases/population

cancer was significantly raised for current users of oestrogenonly preparations (130 [121140], p<00001), oestrogenprogestagen combinations (200 [188212], p<00001), and
tibolone (145 [125168], p<00001). However, the
magnitude of the relative risk of breast cancer varied
significantly between these three HRT types (p<00001) and
was substantially greater in users of oestrogen-progestagen
combinations than other preparations (figure 2). Included
among users of other or unknown preparations were 618
current users of progestagen-only preparations (202 [105
389], p=004, based on nine incident breast cancers) and
1196 current users of vaginal or other local HRT preparations
(067 [030149], p=03, based on six incident breast
cancers). Compared with never users, the relative risk of
breast cancer was 097 (083113) for past users of
oestrogen-

Relative risk (95% FCI)*

1044/150 179
101 (095108)

991/115 383
130 (122138)

1934/142 870
200 (191209)

184/18 186
145 (125167)

93/9548
144 (117176)

05

10

15

20

25

Figure 2: Relative risk of incident invasive breast cancer in relation to recency and type of HRT used

FCI=floated CI. *Relative to never users, stratified by age, time since menopause, parity and age at first birth, family history of breast cancer, bodymass index, region, and deprivation index.

THE LANCET Vol 362 August 9, 2009 www.thelancet.com

421

For personal use. Only reproduce with permission from The Lancet

ARTICLES

Total duration of

use of HRT by type of

HRT used at baseline

Cases/population

Relative risk (95% FCI)*

Never users of HRT

2894/392 757
100 (096104)

Past users of HRT

311/47 606
094 (084105)

<1 year

14 years
384/55 823
101 (092112)

59 years
230/29 614
114 (100130)

_10 years
80/11 654
105 (084130)

Current users of oestrogen-only HRT

<1 year
25/4452
081 (055120)

14 years
251/29 582
125 (110141)

59 years
416/47 310
132 (120146)

_10 years
277/31 862
137 (122154)

Current users of oestrogen-progestogen combinations

<1 year
97/9771
145 (119178)

14 years
582/49 240
174 (160189)

59 years
850/56 912
217 (203233)

_10 years
362/23 673
231 (208256)

Current users of other/unknown HRT types

<1 year
19/1728
163 (104256)

14 years
83/8794
134 (108166)

59 years
102/10 342
142 (117172)

_10 years
59/4739
193 (150250)

0
10

20
30

Figure 3: Relative risk of incident invasive breast cancer in relation to recency, total duration of use, and type of HRT used at
baseline
FCI=floated CI*Relative to never users, stratified by age, time since menopause, parity and age at first birth, family history of breast cancer, body-mass
index, region, and deprivation index.

only HRT; 104 (094116) for past users of oestrogenprogestagen HRT; and 108 (082142) for past users of
tibolone.

Overall, the relative risk of breast cancer in current users at

baseline increased with increasing total duration of use of
HRT. Figure 3 shows results by total duration of use of HRT
of less than 1 year, 14, 59, and 10 or more years for
women who at baseline were past or current users of
oestrogen only, combined, and other or unknown HRT types.
For current users of oestrogen only and oestrogenprogestagen combinations the relative risk of breast cancer
was significantly raised in each duration category, except for
users of oestrogen-only preparations with a total duration of
use of less than 1 year at baseline. The relative risks
associated with less than 5 and 5 or more years total duration

of use at baseline were 121 (107137, p=0003) and 134

(123140 p<00001), respectively, for oestrogen-only HRT;
and corresponding values were: 170 (156185, p<00001)
and 221 (206237, p<00001) for oestrogen-progestagen
combinations; 132 (104169, p=003) and 157 (130190,
p<00001) for tibolone; and 135 (098185, p=006) and
154 (116204, p=0003) for other and unknown types of
HRT. There was no raised risk of breast cancer among past
users according to their previous total duration of use of HRT
(figure 3).

Figure 4 shows the relative risk of breast cancer in current

users of oestrogen-only preparations compared with never
users of HRT, according to the specific hormonal constituent
of the preparation used, its dose, and formulation. The
relative risk of breast cancer was significantly raised,
separately, for users of preparations containing equine
oestrogen (129 [116143], p<00001) and oestradiol (124

[112137], p<00001), with no significant variation in the

risk between users of each preparation (p=06) or according
to their doses (p=04 for equine oestrogen; and p=08 for
oestradiol). Similarly, the relative risk of breast cancer was
significantly raised separately for users of oral, transdermal,
and implanted formulations (132 [121145], p<00001;
124 [111139], p<00001; and 165 [126216], p<00001,
respectively), with no significant variation in the risk
between the formulations (p=02).

422

Figure 5 shows the relative risk of breast cancer in current

users of oestrogen-progestagen combinations compared with
never users of HRT, according to the specific progestagen
used at baseline and whether the progestagen was
administered sequentially or continuously. To keep
confounding between duration of use of HRT and use of
specific preparations to a minimum, results are shown
separately for total durations

THE LANCET Vol 362 August 9, 2009 www.thelancet.com

For personal use. Only reproduce with permission from The Lancet

ARTICLES

Oestrogen-only HRT by:

constituent, dose, and

formulation
Cases/population

Relative risk (95% CI)

All oestrogen-only HRT

991/115 383
130 (121140)

By constituent and dose

All equine oestrogen

426/48 386
129 (116143)

_0625 mg equine oestrogen

288/33 039
125 (111141)

>0625 mg equine oestrogen

135/15 181
136 (114161)

All ethinyloestradiol
454/56 322
124 (112137)

_1 mg ethinyloestradiol
367/44 898
125 (112140)

>1 mg ethinyloestradiol
47/6455
119
(089158)

By formulation

Oral
606/68 351
132
(121145)

Transdermal
324/40 015
124

(111139)

Implanted
54/5272
165
(126216)

05

10

15

20

Figure 4: Relative risk of incident invasive breast cancer by constituent, dose, and formulation of oestrogen only HRT preparation used
at baseline*
*Dotted line represents overall relative risk for current users of oestrogen-only preparations compared with never users at baseline. Full information on
specific constituents and their formulation was not available for some women. Relative to never users, stratified by age, time since menopause, parity
and age at first birth, family history of breast cancer, body-mass index, region, and deprivation index.

of use of HRT of less than 5 years and 5 years or more. In

these two categories of use, the relative risk of breast cancer
was significantly increased, separately, for users of
preparations containing medroxyprogesterone acetate,
norethisterone and norgestrel, and for users of sequential and
continuous regimens. However, the relative risk of breast
cancer showed little consistent variation according to the
progestagen constituent or whether sequential or continuous
regimens were used. The doses of each specific progestagen
do not vary between the various preparations in such a way
that would permit valid comparisons to be

Oestrogen-progestagen

Duration of use <5 years

made. The relative risks of breast cancer for users of combined

equine oestrogen and medroxyprogesterone acetate, the
specific HRT combination used in the Womens Health
Initiative trial,2,3 were 162 (134196, p<00001) and 242
(208281, p<00001) for total duration of use of <5 and _5
years, respectively.
The results in figures 2, 3, 4, and 5 classify current users of
HRT according to the type of preparation reported to have
been used at baseline. In the study population as a whole,
about two-thirds of the current users had used one proprietary
preparation only (in that 66% of current users

Duration of use _5 years

HRT by: constituents

and regimen
Cases/population
Relative risk (95% CI)*
Cases/population
Relative risk (95% CI)

All oestrogen-progestagen
679/59 011
170 (156186)

1212/80 585
221 (206236)

combinations

By progestagen constituent

Medroxyprogesterone acetate
117/11 280
160 (133193)

196/12 628
242 (210280)

Norethisterone
253/24 667
153 (135175)

390/27 841
210 (189234)

Norgestrel/levonorgestrel
290/20 952
197 (174233)

608/38 494
223 (204244)

By type of regimen

Sequential
403/33 124
177 (159197)

778/52 518
212 (195230)

Continuous
243/23 708
157 (137179)

388/25 286
240 (215267)

0
10

20

30

0
10
20

30

Figure 5: Relative risk of incident invasive breast cancer by constituent and regimen of oestrogen-progestagen combination HRT used
at baseline*
*Dotted line represents overall relative risk for current users of oestrogen-progestagen preparations compared with never users at baseline. Full
information on specific constituents and their formulation was not available for some women. Relative to never users, stratified by age, time since
menopause, parity and age at first birth, family history of breast cancer, body-mass index, region, and deprivation index.

THE LANCET Vol 362 August 9, 2009 www.thelancet.com

423

For personal use. Only reproduce with permission from The Lancet

ARTICLES

HRT use at baseline

Deaths/population

Relative risk (95% FCI)*

Never users
238/392 757
100 (088114)

Current users
191/285
987
122 (105141)

Past users
88/150
179
105 (085129)

05
10

15

Figure 6: Relative risk of fatal breast cancer in relation to use of HRT at baseline

FCI=floated CI. *Relative to never users, stratified by age, time since menopause, parity and age at first birth, family history of breast cancer, body-mass
index, region, and deprivation index.

reported using the most recent preparation for the same

duration as their total duration of use of HRT), and the large
majority of current users had used the most recent proprietary
preparation for the longest period of time (87% had used the
most recent proprietary preparation for at least half their total
duration of use of HRT). Furthermore, our validation studies
have shown 97% agreement between self-reported data and
family-physician records when specific proprietary
preparations as classified as oestrogen only, oestrogenprogestagen, or another type of HRT.14 Since oestrogen-only
preparations are used almost entirely by women without a
uterus, and the converse is true for oestrogen-progestagen
combinations,13 few women switch from using one of these
types of HRT to the other.

Some current users of tibolone at baseline may have used

other HRT preparations previously, and to isolate the effects
of tibolone alone, analyses were restricted to women whose
reported duration of use of tibolone was the same as their
reported total duration of any type of HRT. Among such
women, who were likely to have used tibolone exclusively,
the relative risk of breast cancer in current users was 148
(120183, p<00001), based on 88 incident breast cancers,
similar to the result for all current users of tibolone (145
[125168], p<00001 figure 2). Current users of tibolone at
baseline did not differ substantially from other current users
of HRT for known risk factors for breast cancer (80 vs 82%
had a family history of breast cancer; 107 vs 102% were
nulliparous; 77 vs 75% had their first birth after age 30
years; 505 vs 473% had a body-mass index of 25 kg/m2 or
more; 123 vs 133% had previous benign breast disease; and
291 vs 260% drank 10 g or more alcohol daily). Additional
adjustment for those of the above factors which were not
already included in the analysis model did not alter the
estimates of relative risk of breast cancer associated with
current use of tibolone.

Sensitivity analyses were done on the overall results for

current and past users to assess the effect of restricting
analyses further to include only women with a natural

menopause and bilateral oophorectomy, who had begun use

of HRT after their menopause. The relative risk, based on
4750 women with incident invasive breast cancer, compared
with never users was 167 (159176, p<00001) for current
users and 100 (093108, p=10) for past users, which is
similar to the results shown in figure 1. In addition, to
investigate possible confounding by other factors, results for
current and past users and type of current HRT were adjusted
in turn by age at menarche, alcohol consumption, past use of
oral contraceptives, and past health. Adjustment by each
factor altered the relative risks only at the second decimal
place, if at all. Results in figures 1 and 2 were also examined
separately by age, family history of breast cancer, body-mass
index, and ever-use of oral contraceptives. The only factor
that seemed to modify the relative risk estimates materially
was body-mass index.

Among women with a body-mass index less than 25 kg/m2

and 25 kg/m2 or more, the respective relative risks of breast
cancer were: 197 (182214, p<00001) and 146 (136158,
p<00001) for all current users of HRT; 153 (136171,
p<00001) and 117 (105129, p<0003) for current users of
oestrogen-only HRT; and 231 (212253, p<00001) and
178 (164194, p<00001) for current users of oestrogenprogestagen HRT (heterogeneity p<0001 for each
comparison).

With 41 years of follow-up for mortality, the number of

deaths from breast cancer in women without a history of
cancer at recruitment is still relatively small517 women in
total died. Overall, the relative risk of death from breast
cancer was raised in women who were current users of HRT
at recruitment (122 [100148], p=005), but not in past
users (105 [082134, p=007]), compared with never users
(figure 6). Insufficient data are available to compare reliably
deaths from breast cancer in users of oestrogen-only and
oestrogen-progestagen preparations, or to investigate the
relation between mortality and duration of use of HRT.

Among women from developed countries who never use

HRT, the incidence of invasive breast cancer is estimated to

be typically 32 in every 1000 between the ages of 50 and 65

years (table 2).1,18 The cumulative incidence of breast cancer
per 1000 women associated with different patterns of use of
HRT was calculated by applying the relative risk estimates
from figure 3 to the estimated incidence rates in never users
of HRT (table 2). 5 years use of HRT, beginning at age 50
years, is estimated to result in 15 (95% CI 03) additional
breast cancers by age 65 years among 1000 users of
oestrogen-only preparations, and six (57) additional cancers
among 1000 users of oestrogen-progestagen combinations. 10
years use is estimated to result in five (37) additional
cancers in 1000 users of oestrogen-only preparations and 19
(1820) additional cancers in 1000 users of combined HRT.

progestagen

The main reason that women are prescribed combined rather

than oestrogen-only HRT is because of the known

Estimated cumulative incidence of breast cancer

Never

5 years

Duration use of

10 years

Duration use of

5 years
10 years

users of
oestrogen only
oestrogen-

HRT

60
Up to age (years)
38

395
43
44

57

65

50

515
55
50

56

18

69

18

Excess cumulative

18

18
15
18

5
6

55
19
27
incidence per
285
29
34

(03)
34

(37)

(95% CI)
(57)
(1820)
1000 HRT users

Table 2: Cumulative and excess incidence of invasive breast

cancer in 1000 women who had never used and ever used
HRT, based on incidence rates typical of developed countries

424

THE LANCET Vol 362 August 9, 2009 www.thelancet.com

For personal use. Only reproduce with permission from The Lancet

increased
risk of
endometri
al cancer
associated
with the
use of
oestrogenonly
preparatio
ns.5,2123
Among
women in
developed
countries
who do
not use
HRT,
about five
in every
1000 are
diagnosed
with
endometri
al cancer
between
ages 50
and 64
years.18
Calculatio
ns, similar
in
principle
to those
used to
estimate
the
cumulativ
e
incidence
of breast
cancer,
were done
for
endometri
al cancer
with
estimates
of relative
risk from
randomise
d trials2,4
and
observatio
nal
studies.21
23
Use of
oestrogenonly
preparatio
ns is
estimated
to result

in an
additional
four
endometri
al cancers
per 1000
for 5
years use;
and an
additional
ten per
1000 for
10 years
use;
whereas
use of
combined
oestrogenprogestag
en HRT is
estimated
to result
in little or
no change
in the
cumulativ
e
incidence
of
endometri
al cancer.
(The
limited
available
evidence
suggests
that use of
sequential
combined
HRT
might
slightly
increase
the risk of
endometri
al
cancer5,21
23
and that
use of
continuou
s
combined
HRT
might
slightly
lower the
risk,2,22
but that
10 years
use of
oestrogenprogestag
en
preparatio
ns is

unlikely
to
increase
or
decrease
the
cumulativ
e
incidence
by more
than about
two per
1000
users.)
Figure 7
shows the
estimated
cumulativ
e
incidence
of breast
and
endometri
al cancer
in 1000
women
from
developed
countries
between
the ages
of 50 and
65 years:
who never
used HRT;
who used
oestrogenonly HRT
for 10
years; and
who used
oestrogenprogestag
en HRT
for 10
years.
Breast
cancer is
much
more
common
than
endometri
al cancer,
and 10
years use
of either
type of
HRT
results in
an
increase
in the
incidence

of one or
other of
the
cancers.
However,
for
oestrogenonly HRT
the excess
incidence
is due
largely to
endometri
al cancer,
whereas
for
combined
HRT the
excess is
exclusivel
y made up
of breast
cancers.
Over the
past
decade
about 15
000
invasive
breast
cancers
have been
diagnosed
each year
in women
aged 50
64 years
in the
UK.20 If it
is
assumed
that the
relative
risk
estimates
in this
study
associated
with use
of
different
THE LANCET
Vol 362 August
9, 2009
www.thelancet.c
om

*10 years
use of
oestrogenprogestage
n HRT
estimated to
result in
little change

in
culmulative
incidence of
endometrial
cancer.

Figure 7:
Estimated
cumulative
incidence
of breast
and
endometri
al cancer
per 1000
women in
developed
countries
who never
used HRT
and who
used HRT
for 10
years,
beginning
at age 50
years
ARTICLES

t
7

progestage
n HRT:

a
v
o
2
o
w
a

5
20 000
extra
breast
cancers
are

d
t
p
d
a
only HRT:

Moreover
, if the
distributi
on of use
of

e40
5

p
y
a
w
a

specific
types of
HRT is
similar to
that

o
HRT. The
number
of extra
deaths
from

2
3

b
2

only HRT:

p
f
t
c
a

Million
Women
bStudy
was set
up to

a
i
a

examine
5the effect
5of HRT
6on breast
6

cancer
incidence
and
mortality.
Study

Study,
amountin
g to 53%
participat
ion in the
general
populatio
n. Study
participa
nts were
similar to
all
pwomen
screened,
who
themselv
es are
slightly
more
likely to
use HRT
and to
come
from
olessdeprived
areas
than
women
who do
not attend
the
NHSBSP
for
routine
mammog
12
mraphy.
Although
the study
cohort
contains
a slightly
greater
proportio
n of HRT
users
than in
the
pgeneral
ppopulatio
an, this
rdifferenc
e would
not bias
sinternal
comparis
ons
within
the
cohort of
breast
cancer
risk
according
to use of

nts had
routine
mammog
raphy
soon after
they
complete
d their
baseline
questionn
aires.
Thus any
biases
resulting
from
differenti
al
reporting
of use of
HRT
according
to the
outcome
of
interest
ie, breast
cancer
or
differenti
al
screening
for the
exposure
of
interest
ie, use of
HRT
were
virtually
eliminate
d. Few
women
who
attend the
NHSBSP
have
mammog
raphy
done
privately
at other
times,
and most
breast
cancers
diagnose
d among
study
participa
nts in the
interval
between
their
triennial
routine

425

For
person
al use.
Only
reprod
uce
with
permis
sion
from
The
Lancet

ARTICLES

R
e
gi
st
er
s,
w
it
h
d
et
ai
ls
o
f
e
a
c
h
re
gi
st
er
e
d
in
ci
d
e
nt
c
a
n
c
er
a
n
d
d
e
at
h
b
ei
n
g
a
ut
o
m
at
ic
al
ly
c
o
d
e
d
a
n

d
re
p
o
rt
e
d
to
st
u
d
y
in
v
e
st
ig
at
o
rs
.
T
h
er
ef
o
re
,
th
e
re
c
o
r
di
n
g
o
f
in
ci
d
e
nt
a
n
d
fa
ta
l
b
re
a
st
c
a
n
c
er
is
u
n
bi

a
s
e
d
f
o
r
u
s
e
o
f
H
R
T
a
n
d
th
er
e
s
h
o
ul
d
b
e
li
tt
le
,
if
a
n
y,
di
ff
er
e
nt
ia
l
lo
ss
to
f
ol
lo
w
u
p.

R
a
n
d
o
m
is
e
d

tr
ia
ls
h
a
v
e
re
p
o
rt
e
d
si
g
ni
fi
c
a
nt
ly
in
cr
e
a
s
e
d
ri
s
k
s
o
f
b
re
a
st
c
a
n
c
er
in
H
R
T
u
s
er
s.
2,3

I
n
n
o
n
ra
n
d
o
m
is
e

d
st
u
di
e
s,
s
u
c
h
a
s
th
is
o
n
e,
th
e
re
la
ti
o
n
b
et
w
e
e
n
b
re
a
st
c
a
n
c
er
a
n
d
H
R
T
u
s
e
c
a
n
b
e
c
o
n
f
o
u
n
d
e
d
b
y

w
o
m
e
n
s
m
e
n
o
p
a
u
s
al
st
at
u
s:
f
o
r
w
o
m
e
n
o
f
a
gi
v
e
n
a
g
e,
p
re
m
e
n
o
p
a
u
s
al
a
n
d
p
er
i
m
e
n
o
p
a
u
s
al
w
o

m
e
n
ar
e
at
g
re
at
er
ri
s
k
o
f
b
re
a
st
c
a
n
c
er
a
n
d
ar
e
le
ss
li
kl
e
y
to
u
s
e
H
R
T
th
a
n
p
o
st
m
e
n
o
p
a
u
s
al
w
o
m
e
n,
1

a
s

w
a
s
f
o
u
n
d
h
er
e.
T
o
k
e
e
p
s
u
c
h
c
o
n
f
o
u
n
di
n
g
to
a
m
in
i
m
u
m
w
e
e
x
cl
u
d
e
d
p
re
m
e
n
o
p
a
u
s
al
a
n
d
p
er
i

m
e
n
o
p
a
u
s
al
w
o
m
e
n
a
n
d
w
o
m
e
n
a
g
e
d
5
0

5
2
y
e
ar
s
if
th
e
y
h
a
d
h
a
d
a
h
y
st
er
e
ct
o
m
y
o
r
h
a
d
b
e
g
u
n

u
si
n
g
H
R
T
b
ef
o
re
th
e
m
e
n
o
p
a
u
s
e
(s
in
c
e
m
e
n
o
p
a
u
s
e
c
a
n
b
e
m
a
s
k
e
d
b
y
s
u
c
h
in
te
r
v
e
nt
io
n
s)
.
S
e
n

si
ti
vi
ty
a
n
al
y
s
e
s,
d
o
n
e
to
a
ss
e
ss
th
e
e
xt
e
nt
o
f
s
u
c
h
p
ot
e
nt
ia
l
c
o
n
f
o
u
n
di
n
g,
s
h
o
w
e
d
th
at
w
h
e
n
st
ri
ct
er
e

x
cl
u
si
o
n
cr
it
er
ia
w
er
e
a
p
pl
ie
d,
th
e
p
oi
nt
e
st
i
m
at
e
s
f
o
r
th
e
re
la
ti
v
e
ri
s
k
o
f
b
re
a
st
c
a
n
c
er
in
c
u
rr
e
nt
a
n
d
p
a

st
u
s
er
s
o
f
H
R
T
v
ar
ie
d
b
y
le
ss
th
a
n
1
%
.

A
n
al
y
s
e
s
w
er
e
r
o
ut
in
el
y
st
ra
ti
fi
e
d
b
y
a
g
e,
ti
m
e
si
n
c
e
m
e
n
o

p
a
u
s
e,
p
ar
it
y
a
n
d
a
g
e
at
fi
rs
t
bi
rt
h,
fa
m
il
y
hi
st
o
r
y
o
f
b
re
a
st
c
a
n
c
er
,
b
o
d
y
m
a
ss
in
d
e
x,
re
gi
o
n
o
f
re
si
d
e

n
c
e,
a
n
d
d
e
p
ri
v
at
io
n
in
d
e
x,
th
u
s
k
e
e
pi
n
g
to
a
m
in
i
m
u
m
c
o
n
f
o
u
n
di
n
g
b
y
th
e
s
e
fa
ct
o
rs
.
E
x
a
m
in
at
io
n
o

f
p
ot
e
nt
ia
l
c
o
n
f
o
u
n
di
n
g
b
y
al
c
o
h
ol
c
o
n
s
u
m
pt
io
n,
p
re
vi
o
u
s
u
s
e
o
f
o
ra
l
c
o
nt
ra
c
e
pt
iv
e
s,
a
g
e
at
m
e
n
ar

c
h
e
a
n
d
p
a
st
h
e
al
th
,
di
d
n
ot
al
te
r
th
e
re
la
ti
v
e
ri
s
k
e
st
i
m
at
e
s
f
o
r
c
u
rr
e
nt
a
n
d
p
a
st
u
s
er
s
c
o
m
p
ar
e
d
w
it

h
n
e
v
er
u
s
er
s.
T
h
e
o
nl
y
fa
ct
o
r
th
at
m
o
di
fi
e
d
th
e
re
la
ti
v
e
ri
s
k
e
st
i
m
at
e
s
w
a
s
b
o
d
y
m
a
ss
in
d
e
x,
w
it
h
th
e
re

la
ti
v
e
ri
s
k
e
st
i
m
at
e
s
b
ei
n
g
la
r
g
er
a
m
o
n
g
th
e
th
in
n
er
w
o
m
e
n

ie
,
th
o
s
e
w
it
h
a
lo
w
er
ra
th
er
th
a
n
a
hi
g
h
er
in
d

e
x.

M
is
cl
a
ss
if
ic
at
io
n
o
f
w
o
m
e
n
s
u
s
e
o
f
H
R
T
s
h
o
ul
d
n
ot
af
fe
ct
th
e
m
ai
n
c
o
n
cl
u
si
o
n
s.
S
el
fre
p
o
rt
e
d
in

f
o
r
m
at
io
n
o
n
u
s
e
o
f
H
R
T
at
b
a
s
el
in
e
w
a
s
u
s
e
d
to
d
ef
in
e
c
u
rr
e
nt
,
p
a
st
,
a
n
d
n
e
v
er
u
s
er
s,
a
n
d
u
s
er
s
w

er
e
f
u
rt
h
er
cl
a
ss
if
ie
d
a
c
c
o
r
di
n
g
to
th
e
re
p
o
rt
e
d
s
p
e
ci
fi
c
p
r
o
p
ri
et
ar
y
p
re
p
ar
at
io
n
u
s
e
d
m
o
st
re
c
e
nt
ly
.
V

al
id
at
io
n
st
u
di
e
s
w
it
hi
n
th
e
st
u
d
y
p
o
p
ul
at
io
n
s
h
o
w
e
d
9
6
%
a
g
re
e
m
e
nt
b
et
w
e
e
n
s
el
fre
p
o
rt
e
d
d
at
a
at
b

a
s
el
in
e
a
n
d
fa
m
il
y
p
h
y
si
ci
a
n
p
re
s
cr
ip
ti
o
n
s
f
o
r
c
u
rr
e
nt
u
s
e
o
f
H
R
T.
14

M
o
re
o
v
er
,
a
m
o
n
g
c
u
rr
e
nt
u
s

er
s,
th
er
e
w
a
s
9
7
%
a
g
re
e
m
e
nt
a
s
to
w
h
et
h
er
o
e
st
r
o
g
e
n
o
nl
y,
o
e
st
r
o
g
e
n
p
r
o
g
e
st
a
g
e
n
c
o
m
bi
n
at

io
n,
o
r
a
n
ot
h
er
ty
p
e
o
f
H
R
T
w
a
s
b
ei
n
g
u
s
e
d,
a
n
d
9
0
%
a
g
re
e
m
e
nt
f
o
r
th
e
s
p
e
ci
fi
c
p
r
o
p
ri
et
ar
y
p
re
p

ar
at
io
n
u
s
e
d
a
n
d
it
s
d
o
s
e.
14

A
n
e
st
i
m
at
e
d
t
w
o
th
ir
d
s
o
f
c
u
rr
e
nt
u
s
er
s
u
s
e
d
o
n
e
p
r
o
p
ri
et
ar
y
H
R
T
p

re
p
ar
at
io
n
e
x
cl
u
si
v
el
y;
a
n
d
fi
v
esi
xt
h
s
o
f
th
e
c
u
rr
e
nt
u
s
er
s
h
a
d
u
s
e
d
th
ei
r
c
u
rr
e
nt
H
R
T
f
o
r
at
le
a
st
h
al
f

th
ei
r
to
ta
l
p
er
io
d
o
f
u
s
e
o
f
H
R
T.
T
h
e
in
ci
d
e
nt
b
re
a
st
c
a
n
c
er
s
w
er
e
di
a
g
n
o
s
e
d
o
n
a
v
er
a
g
e
1
2
y
e
ar
s
af
te

r
re
cr
ui
t
m
e
nt
,
a
n
d
s
o
m
e
w
o
m
e
n
w
o
ul
d
h
a
v
e
c
h
a
n
g
e
d
th
ei
r
u
s
e
o
f
H
R
T
d
u
ri
n
g
th
at
p
er
io
d.
R
e
s
ul
ts
fr
o

m
a
s
u
r
v
e
y
o
f
a
ra
n
d
o
m
s
a
m
pl
e
o
f
1
2
2
2
1
st
u
d
y
p
ar
ti
ci
p
a
nt
s,
d
o
n
e
a
n
a
v
er
a
g
e
o
f
2
8
y
e
ar
s
af
te
r
th
e

re
c
o
r
di
n
g
o
f
b
a
s
el
in
e
in
f
o
r
m
at
io
n,
s
h
o
w
e
d
th
at
7
8
%
o
f
th
e
c
u
rr
e
nt
u
s
er
s
at
b
a
s
el
in
e
w
er
e
st
il
l
u
si
n

g
H
R
T,
th
at
8
1
%
o
f
p
a
st
u
s
er
s
w
er
e
st
il
l
p
a
st
u
s
er
s,
a
n
d
th
at
8
9
%
o
f
th
e
n
e
v
er
u
s
er
s
w
er
e
st
il
l
n
e
v
er
u

s
er
s.
A
m
o
n
g
c
u
rr
e
nt
u
s
er
s
at
b
a
s
el
in
e,
th
e
to
ta
l
d
u
ra
ti
o
n
o
f
u
s
e
o
f
H
R
T
at
th
e
ti
m
e
o
f
di
a
g
n
o
si
s
o
f
b
re
a

st
c
a
n
c
er
w
o
ul
d
b
e
sl
ig
ht
ly
lo
n
g
er
th
a
n
th
at
re
c
o
r
d
e
d
at
b
a
s
el
in
e.
H
o
w
e
v
er
,
a
n
y
re
s
ul
ta
nt
u
n
d
er
e
st
i
m
at
io
n

o
f
to
ta
l
d
u
ra
ti
o
n
o
f
u
s
e
o
f
H
R
T
w
o
ul
d
b
e
c
o
u
nt
er
a
ct
e
d,
to
s
o
m
e
e
xt
e
nt
,
b
y
th
e
fa
ct
th
at
d
u
ri
n
g
f
ol
lo
w
u

p
s
o
m
e
c
u
rr
e
nt
u
s
er
s
w
o
ul
d
h
a
v
e
b
e
c
o
m
e
p
a
st
u
s
er
s
a
n
d
th
at
s
o
m
e
n
e
v
er
u
s
er
s
w
o
ul
d
h
a
v
e
b
e
c
o
m

e
c
u
rr
e
nt
u
s
er
s.

O
v
er
al
l,
th
e
s
e
re
s
ul
ts
c
o
n
fi
r
m
p
re
vi
o
u
s
fi
n
di
n
g
s
th
at
c
u
rr
e
nt
a
n
d
re
c
e
nt
u
s
er
s
o
f
H

R
T
ar
e
at
a
n
in
cr
e
a
s
e
d
ri
s
k

o
f
in
v
a
si
v
e
b
re
a
st
c
a
n
c
er
,
a
n
d
th
at
th
e
re
la
ti
v
e
ri
s
k
o
f
b
re
a
st
c
a
n
c
er
in

c
u
rr
e
nt
u
s
er
s
in
cr
e
a
s
e
s
w
it
h
in
cr
e
a
si
n
g
d
u
ra
ti
o
n
o
f
u
s
e
o
f
H
R
T.
1
4

T
h
e
e
st
i
m
at
e
d
a
b
s
ol
ut
e
in
cr
e
a
s

e
s
in
th
e
in
ci
d
e
n
c
e
o
f
b
re
a
st
c
a
n
c
er
a
m
o
n
g
w
o
m
e
n
u
si
n
g
o
e
st
r
o
g
e
n
o
nl
y
p
re
p
ar
at
io
n
s
ar
e
re
m
ar
k
a
bl

y
si
m
il
ar
to
e
st
i
m
at
e
s
d
er
iv
e
d
fr
o
m
a
c
ol
la
b
o
ra
ti
v
e
re
a
n
al
y
si
s
o
f
m
o
st
o
f
th
e
re
le
v
a
nt
w
o
rl
d
w
id
e
d
at
a
in
1
9

9
7,
8
0
%
o
f
w
h
o
m
u
s
e
d
o
e
st
r
o
g
e
n
o
nl
y
H
R
T.
1

(
T
h
e
re
s
p
e
ct
iv
e
e
st
i
m
at
e
s
p
er
1
0
0
0
c
a
s
e
s
ar
e
1

5
[
0

3
]
a
n
d
t
w
o
[
1

3
]
f
o
r
5
y
e
ar
s
u
s
e,
a
n
d
fi
v
e
[
3

7
]
a
n
d
si
x
[
3

9
]
f
o
r
1
0
y
e
ar
s
u
s
e.
)
F
e
w

e
st
i
m
at
e
s
o
f
th
e
a
b
s
ol
ut
e
in
cr
e
a
s
e
in
b
re
a
st
c
a
n
c
er
in
ci
d
e
n
c
e
ar
e
a
v
ai
la
bl
e
f
o
r
u
s
er
s
o
f
o
e
st
r
o
g
e
n

p
r
o
g
e
st
a
g
e
n
H
R
T.
T
h
e
W
o
m
e
n
s
H
e
al
th
I
ni
ti
at
iv
e
tr
ia
l
f
o
u
n
d
a
n
in
cr
e
a
s
e
d
in
ci
d
e
n
c
e
o
f
b
re
a
st
c
a

n
c
er
in
w
o
m
e
n
w
h
o
c
o
m
pl
ie
d
w
it
h
tr
e
a
m
e
nt
,3
o
f
a
b
o
ut
si
x
p
er
1
0
0
0
w
o
m
e
n
af
te
r
5
y
e
ar
s
u
s
e,
a
n
d
1
8
p
er

1
0
0
0
af
te
r
7
y
e
ar
s
u
s
e
o
f
o
e
st
r
o
g
e
n
p
r
o
g
e
st
a
g
e
n
H
R
T,
si
m
il
ar
to
re
s
ul
ts
fr
o
m
th
e
M
il
li
o
n
W
o
m
e
n
S
tu

d
y,
b
a
s
e
d
o
n
s
u
b
st
a
nt
ia
ll
y
la
r
g
er
n
u
m
b
er
s,
o
f
a
n
in
cr
e
a
s
e
in
th
e
c
u
m
ul
at
iv
e
in
ci
d
e
n
c
e
o
f
si
x
(
5

7
)
p

er
1
0
0
0
f
o
r
5
y
e
ar
s
u
s
e
a
n
d
1
9
(
1
8

2
0
)
p
er
1
0
0
0
f
o
r
1
0
y
e
ar
s
u
s
e
o
f
o
e
st
r
o
g
e
n
p
r
o
g
e
st
a
g

e
n
H
R
T.
H
e
n
c
e,
f
o
r
th
e
s
e
d
u
ra
ti
o
n
s
o
e
st
r
o
g
e
n
p
r
o
g
e
st
a
g
e
n
c
o
m
bi
n
at
io
n
s
le
a
d
s
to
a
b
o
ut
a
f
o
u

rf
ol
d
g
re
at
er
in
cr
e
a
s
e
in
b
re
a
st
c
a
n
c
er
in
ci
d
e
n
c
e
th
a
n
d
o
e
s
u
s
e
o
f
o
e
st
r
o
g
e
n
o
nl
y
p
re
p
ar
at
io
n
s.

O
th
er
th
a
n
th
e
s
u
b
st
a
nt
ia
l
di
ff
er
e
n
c
e
b
et
w
e
e
n
th
e
ef
fe
ct
s
o
f
o
e
st
r
o
g
e
n
o
nl
y
a
n
d
o
e
st
r
o
g
e
n
p
r
o
g

e
st
a
g
e
n
c
o
m
bi
n
at
io
n
s,
th
e
s
e
re
s
ul
ts
s
u
g
g
e
st
n
o
la
r
g
e
v
ar
ia
ti
o
n
s
b
et
w
e
e
n
th
e
ef
fe
ct
s
o
f
s
p
e
ci
fi
c
o
e
st

r
o
g
e
n
s
(e
q
ui
n
e
o
e
st
r
o
g
e
n
a
n
d
o
e
st
ra
di
ol
)
o
r
b
et
w
e
e
n
s
p
e
ci
fi
c
p
r
o
g
e
st
a
g
e
n
s
(
m
e
d
r
o
x
y
p
r
o

g
e
st
er
o
n
e
a
c
et
at
e,
n
o
r
g
e
st
re
l,
a
n
d
n
o
re
th
is
te
r
o
n
e)
.
T
h
e
y
al
s
o
s
u
g
g
e
st
th
at
re
s
ul
ts
o
n
th
e
ri
s
k
o
f
b
re
a

st
c
a
n
c
er
f
o
r
th
e
s
p
e
ci
fi
c
c
o
n
st
it
u
e
nt
s
u
s
e
d
in
th
e
W
o
m
e
n
s
H
e
al
th
I
ni
ti
at
iv
e
tr
ia
l3
d
o
n
ot
di
ff
er
m
at
er
ia
ll
y

fr
o
m
th
e
re
s
ul
ts
f
o
r
ot
h
er
o
e
st
r
o
g
e
n
p
r
o
g
e
st
a
g
e
n
c
o
m
bi
n
at
io
n
s.
N
o
r
d
o
e
s
th
e
a
v
ai
la
bl
e
e
vi
d
e
n
c
e

s
u
g
g
e
st
la
r
g
e
di
ff
er
e
n
c
e
s
b
et
w
e
e
n
th
e
ef
fe
ct
s
o
f
p
r
o
g
e
st
a
g
e
n
s
gi
v
e
n
s
e
q
u
e
nt
ia
ll
y
o
r
c
o
nt
in
u
o
u

sl
y
o
n
b
re
a
st
c
a
n
c
er
,
al
th
o
u
g
h
if
m
a
n
y
w
o
m
e
n
h
a
v
e
s
w
it
c
h
e
d
u
s
e,
th
is
w
o
ul
d
di
lu
te
a
n
y
re
al
di
ff
er
e
n
c

e
s.

A
si
g
ni
fi
c
a
nt
e
x
c
e
ss
o
f
b
re
a
st
c
a
n
c
er
w
a
s
f
o
u
n
d
f
o
r
c
u
rr
e
nt
u
s
e
o
f
ti
b
ol
o
n
e.
T
hi
s
in
cr
e
a
s
e

is
n
ot
d
u
e
to
c
o
n
f
o
u
n
di
n
g
b
y
k
n
o
w
n
ri
s
k
fa
ct
o
rs
f
o
r
b
re
a
st
c
a
n
c
er
,
s
u
c
h
a
s
fa
m
il
y
hi
st
o
r
y
o
f
b
re
a
st

c
a
n
c
er
,
o
b
e
si
ty
,
o
r
la
te
c
hi
ld
b
e
ar
in
g.
I
m
pl
a
nt
e
d
a
n
d
tr
a
n
s
d
er
m
al
p
re
p
ar
at
io
n
s
o
f
o
e
st
r
o
g
e
n
o
nl
y
H

R
T
al
s
o
le
d
to
a
si
g
ni
fi
c
a
nt
e
x
c
e
ss
o
f
b
re
a
st
c
a
n
c
er
.
T
h
er
e
w
a
s,
h
o
w
e
v
er
,
n
o
si
g
ni
fi
c
a
nt
di
ff
er
e
n
c
e
b
et

w
e
e
n
th
e
ef
fe
ct
s
o
f
o
ra
l,
tr
a
n
s
d
er
m
al
,
o
r
i
m
pl
a
nt
e
d
f
o
r
m
ul
at
io
n
s.
T
h
e
i
m
pl
ic
at
io
n
s
o
f
th
e
s
e
fi
n
di
n
g
s

n
e
e
d
f
u
rt
h
er
in
v
e
st
ig
at
io
n.

C
u
rr
e
nt
u
s
er
s
o
f
H
R
T
at
b
a
s
el
in
e
h
a
d
si
g
ni
fi
c
a
nt
ly
in
cr
e
a
s
e
d
m
o
rt
al
it
y

fr
o
m
b
re
a
st
c
a
n
c
er
,
al
th
o
u
g
h
th
e
re
la
ti
v
e
ri
s
k
e
st
i
m
at
e
w
a
s
o
f
b
o
r
d
er
li
n
e
si
g
ni
fi
c
a
n
c
e
a
n
d
w
a
s
n
ot

a
s
la
r
g
e
a
s
f
o
r
in
ci
d
e
nt
di
s
e
a
s
e.
T
h
e
re
s
ul
ts
f
o
r
fa
ta
l
di
s
e
a
s
e
ar
e
b
a
s
e
d
o
n
5
1
7
d
e
at
h
s
in
w
o
m
e
n
w

h
o
h
a
d
n
o
hi
st
o
r
y
o
f
b
re
a
st
c
a
n
c
er
at
re
cr
ui
t
m
e
nt
.
H
a
d
w
e
in
cl
u
d
e
d
w
o
m
e
n
w
it
h
a
hi
st
o
r
y
o
f
b
re
a
st
c
a

n
c
er
at
b
a
s
el
in
e
(a
m
o
n
g
w
h
o
m
a
f
u
rt
h
er
4
8
5
d
e
at
h
s
fr
o
m
b
re
a
st
c
a
n
c
er
w
er
e
re
p
o
rt
e
d,
b
ut
o
nl
y
3
%
o
f

w
h
o
m
w
er
e
u
si
n
g
H
R
T
at
re
cr
ui
t
m
e
nt
)
w
e
w
o
ul
d
h
a
v
e
c
o
n
cl
u
d
e
d,
fa
ls
el
y,
th
at
c
u
rr
e
nt
u
s
er
s
o
f
H
R
T
h
a
d
a

s
u
b
st
a
nt
ia
ll
y
lo
w
er
d
e
at
h
ra
te
fr
o
m
b
re
a
st
c
a
n
c
er
th
a
n
n
e
v
er
u
s
er
s.
S
tu
di
e
s
p
u
r
p
o
rt
in
g
to
s
h
o
w
th
at
c
u
rr
e

nt
u
s
er
s
o
f
H
R
T
h
a
v
e
lo
w
er
d
e
at
h
ra
te
s
fr
o
m
b
re
a
st
c
a
n
c
er
th
a
n
n
o
n
u
s
er
s
h
a
v
e
g
e
n
er
al
ly
b
e
e
n
u
n
a
bl

e
to
a
c
c
o
u
nt
a
d
e
q
u
at
el
y
f
o
r
th
is
f
u
n
d
a
m
e
nt
al
s
o
u
rc
e
o
f
bi
a
s.
4

R
e
s
ul
ts
fr
o
m
th
e
W
o
m
e
n
s
H
e
al
th
I
ni
ti
at

iv
e
tr
ia
l
s
h
o
w
th
at
b
re
a
st
c
a
n
c
er
s
di
a
g
n
o
s
e
d
in
w
o
m
e
n
al
lo
c
at
e
d
to
H
R
T
h
a
d

a
si
g
ni
fi
c
a
nt
ly
la
r
g
er
si
z
e
th
a
n
th
e
c
a
n
c
er
s
in
n
o
n
u
s
er
s
o
f
H
R
T,
3

f
u
rt
h
er

426
THE
LANCET
Vol 362
August 9,
2009
www.thelanc
et.com

For
person

al use.
Only
reprod
uce
with
permis
sion
from
The
Lancet

ARTICLES

c
h
al
le
n
gi
n
g
th
e
v
al
id
it
y
of
cl
ai
m
s
th
at
u
se
of
H
R
T
d
e
cr
e
as
es
m
or
ta
lit
y
fr
o
m
br
e
as
t
c
a
n
c
er
.
L
o
n
g
er
fo
ll
o
w

u
p
of
th
is
a
n
d
ot
h
er
c
o
h
or
ts
a
n
d
fu
rt
h
er
in
fo
r
m
at
io
n
o
n
th
e
ef
fe
ct
s
of
di
ff
er
e
nt
p
at
te
rn
s
of
u
se
of
H
R
T
o
n
m
or
ta
lit

y
fr
o
m
br
e
as
t
c
a
n
c
er
ar
e
n
e
e
d
e
d.

T
h
e
re
s
ul
ts
fr
o
m
th
e
M
ill
io
n
W
o
m
e
n
St
u
d
y
s
u
g
g
es
t
lit
tl
e
or
n
o
o
v
er
al

l
in
cr
e
as
e
in
th
e
re
la
ti
v
e
ri
s
k
of
br
e
as
t
c
a
n
c
er
in
p
as
t
u
se
rs
of
H
R
T.
N
o
re
si
d
u
al
in
cr
e
as
e
in
th
e
ri
s
k
of
br
e
as
t
c
a
n
c

er
w
as
se
e
n
se
p
ar
at
el
y
fo
r
p
as
t
u
se
rs
of
o
es
tr
o
g
e
n
o
nl
y,
o
es
tr
o
g
e
npr
o
g
es
ta
g
e
n
c
o
m
bi
n
at
io
n
s,
or
ti
b
ol
o
n
e.
T
h
es

e
fi
n
di
n
g
s
ar
e
br
o
a
dl
y
in
li
n
e
w
it
h
re
s
ul
ts
fr
o
m
pr
e
vi
o
u
s
st
u
di
es
th
at
h
a
d
s
u
g
g
es
te
d
th
at
th
e
ef
fe
ct
s
of
c
ur
re
nt
u
se

of
H
R
T
o
n
th
e
ri
s
k
of
br
e
as
t
c
a
n
c
er
w
or
e
of
f
la
rg
el
y,
if
n
ot
w
h
ol
ly
,
w
it
hi
n
5
y
e
ar
s
of
c
e
as
in
g
u
se
of
H
R
T.
1

U
se
of
H

R
T
b
y
U
K
w
o
m
e
n
a
g
e
d
5
0

6
4
y
e
ar
s
in
th
e
p
as
t
d
e
c
a
d
e
is
es
ti
m
at
e
d
to
h
a
v
e
re
s
ul
te
d
in
a
n
e
xt
ra
2
0
0
0
0
in

ci
d
e
nt
br
e
as
t
c
a
n
c
er
s,
c
o
m
bi
n
e
d
o
es
tr
o
g
e
npr
o
g
es
ta
g
e
n
H
R
T
a
c
c
o
u
nt
in
g
fo
r
1
5
0
0
0
of
th
es
e
a
d
di
ti
o
n
al

c
a
n
c
er
s.
T
h
e
m
ai
n
re
as
o
n
th
at
w
o
m
e
n
ar
e
pr
es
cr
ib
e
d
c
o
m
bi
n
e
d
ra
th
er
th
a
n
o
es
tr
o
g
e
no
nl
y,
H
R
T
is
b
e
c
a
u
se
of

th
e
in
cr
e
as
e
d
ri
s
k
of
e
n
d
o
m
et
ri
al
c
a
n
c
er
as
s
o
ci
at
e
d
w
it
h
u
se
of
o
es
tr
o
g
e
no
nl
y
pr
e
p
ar
at
io
n
s.
H
o
w
e
v
er
,
if
th

e
a
d
di
ti
o
n
al
br
e
as
t
a
n
d
e
n
d
o
m
et
ri
al
c
a
n
c
er
s
as
s
o
ci
at
e
d
w
it
h
e
a
c
h
ty
p
e
of
H
R
T
ar
e
a
d
d
e
d
to
g
et
h
er
,
th
er

e
se
e
m
s
to
b
e
lit
tl
e
a
d
v
a
nt
a
g
e
to
u
si
n
g
o
es
tr
o
g
e
npr
o
g
es
ta
g
e
n
in
pr
ef
er
e
n
c
e
to
o
es
tr
o
g
e
no
nl
y
H
R
T
fo
r
w
o

m
e
n
w
h
o
st
ill
h
a
v
e
a
ut
er
u
s.
U
se
of
ei
th
er
ty
p
e
of
H
R
T
is
es
ti
m
at
e
d
to
re
s
ul
t
in
fi
v
e
to
si
x
e
xt
ra
c
a
n
c
er
s
p
er
1
0
0
0
w

o
m
e
n
w
it
h
5
y
e
ar
s
u
se
a
n
d
1
5

1
9
e
xt
ra
c
a
n
c
er
s
p
er
1
0
0
0
w
it
h
1
0
y
e
ar
s
u
se
of
H
R
T.
T
h
e
e
xt
ra
c
a
n
c
er
s
ar

e
pr
e
d
o
m
in
a
nt
ly
of
th
e
e
n
d
o
m
et
ri
u
m
fo
r
u
se
rs
of
o
es
tr
o
g
e
no
nl
y
pr
e
p
ar
at
io
n
s,
w
h
er
e
as
th
e
y
ar
e
e
x
cl
u
si
v
el
y
br

e
as
t
c
a
n
c
er
fo
r
u
se
rs
of
o
es
tr
o
g
e
npr
o
g
es
ta
g
e
n
H
R
T.
R
el
ia
bl
e
es
ti
m
at
es
of
th
e
e
xt
ra
d
e
at
h
s
fr
o
m
br
e
as
t
c
a
n
c
er

at
tr
ib
ut
a
bl
e
to
u
se
of
H
R
T
c
a
n
n
ot
b
e
m
a
d
e
at
pr
es
e
nt
.

C
o
nf
li
ct
of
in
te
r
e
st
st
at
e
m
e
nt
N
on
e
de
cl
ar
ed
.

A
n
al
y
si
s
a

n
d
w
rit
in
g
c
o
m
m
itt
e
e
E
mi
ly
Ba
nk
s,
Va
ler
ie
Be
ral
,
Di
an
a
B
ull
,
Gi
lli
an
Re
ev
es.

S
te
e
ri
n
g
c
o
m
m
itt
e
e
Jo
an
A
us
to
ke
r,
E
mi
ly
Ba
nk
s,
Va
ler
ie
Be
ral
,
R
ut
h
En
gli

sh
,
Ju
lie
tta
Pa
tni
ck
,
Ri
ch
ar
d
Pe
to,
Gi
lli
an
Re
ev
es,
M
art
in
Ve
ss
ey
(C
ha
ir)
,
M
att
he
w
W
all
is.

M
ill
io
n
W
o
m
e
n
S
tu
d
y
c
o
o
r
di
n
at
in
g
c
e
nt
r
e
st
af
f
Si
m
on
A
bb
ott

,
E
m
m
a
Ba
ile
y,
Kr
ys
Ba
ke
r,
A
ng
el
a
Ba
lk
wi
ll,
E
mi
ly
Ba
nk
s,
Is
ob
el
Ba
rn
es,
Va
ler
ie
Be
ral
,
Ju
dit
h
Bl
ac
k,
A
nn
a
Br
o
w
n,
Di
an
a
B
ull
,
Be
ck
y
Ca
m
er
on
,
K
ar
en
Ca
nf
ell
,
A
nd
re
a
Cl
iff
,
Ba
rb
ar
a

Cr
os
sle
y,
El
isa
be
th
C
ou
to,
St
ep
he
n
D
av
ies
,
D
av
e
E
w
art
,
Sa
ra
h
E
w
art
,

D
eb
bi
e
Fo
rd,
La
ur
a
G
err
ar
d,
A
dri
an
G
oo
dil
l,
Ja
ne
Gr
ee
n,
W
ini
fre
d
Gr
ay
,
El
iz
ab
et
h
Hi
lto
n,
A
nn
H
og
g,
Jo
y

H
oo
le
y,
A
nn
a
H
ur
st,
Sa
u
W
an
K
an
,
Ca
rol
K
ee
ne
,
Ni
ck
y
La
ng
st
on
,
Gi
lli
an
Re
ev
es,
A
nd
re
w
R
od
da
m,

Ph
il
Sa
un
de
rs,
E
m
m
a
Sh
er
m
an
,
M
oy
a
Si
m
m
on
ds
,
El
iz
ab
et
h
Sp
en
ce
r,
H
el

en
a
St
ra
ng
e,
Al
is
on
Ti
m
ad
jer
.

N
H
S
B
r
e
a
st
S
cr
e
e
ni
n
g
C
e
nt
r
e
s
p
a
rti
ci
p
at
in
g
in
th
e
M
ill
io
n
W
o
m
e
n
S
tu
d
y
A
vo
n,
A
yl
es
bu
ry,
Ba
rn
sle
y,
Ba

si
ng
st
ok
e,
Be
df
or
ds
hir
e
an
d
H
ert
fo
rd
sh
ire
,
Ca
m
bri
dg
e
an
d
H
un
tin
gd
on
,
C
he
lm
sf
or
d
an
d
C
ol
ch
est
er,
C
he
ste
r,
C
or
n
w
all
,
Cr
e
w
e,
C
u
m
bri
a,
D
on
ca
ste
r,
D
or
set
,
Ea
st
Be
rk
sh
ire
,

Ea
st
C
he
sh
ire
,
Ea
st
D
ev
on
,
Ea
st
of
Sc
otl
an
d,
Ea
st
Su
ff
ol
k,
Ea
st
Su
ss
ex
,
G
at
es
he
ad
,
Gl
ou
ce
ste
rs
hir
e,
Gr
ea
t
Ya
rm
ou
th,
H
er
ef
or
d
an
d
W
or
ce
ste
r,
K
en
t
(C
an
ter
bu
ry,
Gi
lli
ng
ha
m,
M
ai
ds
to
ne
),
Ki

ng
s
Ly
nn
,
Le
ic
est
er
sh
ire
,
Li
ve
rp
oo
l,
M
an
ch
est
er,
M
ilt
on
K
ey
ne
s,
N
e
w
ca
stl
e,
N
ort
h
Bi
rm
in
gh
a
m,
N
ort
h
Ea
st
Sc
otl
an
d,
N
ort
h
La
nc
as
hir
e,
N
ort
h
M
id
dl
es
ex
,

N
ort
h
N
ott
in
gh
a
m,
N

ort
h
of
Sc
otl
an
d,
N
ort
h
Te
es,
N
ort
h
Y
or
ks
hir
e,
N
ott
in
gh
a
m,
O
xf
or
d,
Po
rts
m
ou
th,
R
ot
he
rh
a
m,
Sh
eff
iel
d,
Sh
ro
ps
hir
e,
So
m
er
set
,
So
ut
h
Bi
rm
in
gh
a
m,
So
ut
h
Ea
st
Sc
otl
an
d,

So
ut
h
Ea
st
St
aff

or
ds
hir
e,
So
ut
h
D
er
by
sh
ire
,
So
ut
h
Es
se
x,
So
ut
h
La
nc
as
hir
e,
So
ut
h
W
est
Sc
otl
an
d,
Su
rre
y,
W
arr
in
gt
on
H
alt
on
St
H
el
en
s
&
K
no
ws
le
y,

W
ar
wi
ck
sh
ire
So
lih
ull
an
d
C
ov
en
try
,
W
est
Be
rk
sh
ire

,
W
est
D
ev
on
,
W
est
Lo
nd
on
,
W
est
Su
ff
ol
k,
W
est
Su
ss
ex
,
W
ilt
sh
ire
,
W
in
ch
est
er,
W
irr
al
an
d
W
yc
o
m
be
.

A
c
k
n
o
w
le
d
g
m
e
nt
s
W
e
th
an
k
ea
ch
w
o
m
an
w
ho
pa
rti
ci
pa
te

d
in
th
e
M
illi
on
W
o
m
en
St
ud
y;
co
lla
bo
rat
or
s
fr
o
m
th
e
N
H
S
Br
ea
st
Sc
re
en
in
g
Ce
ntr
es;
m
e
m
be
rs
of
th
e
st
ud
y
co
or
di
na
tin
g
ce
ntr
e;
an
d
th
e
st
ud
y
ste
eri
ng
co
m
mi
tte
e.
Th
is
re
se
ar
ch
w
as
fu
nd

ed
by
Ca
nc
er
Re
se
ar
ch
U
K,
th
e
N
H
S
Br
ea
st
Sc
re
en
in
g
Pr
og
ra
m
m
e,
an
d
th
e
M
ed
ic
al
Re
se
ar
ch
C
ou
nc
il.

R
ef
er
e
n
c
e
s

C
oll
ab
or
ati
ve
Gr
ou
p
on
H
or
m
on
al
Fa
ct
or
s

in
Br
ea
st
Ca
nc
er.
Br
ea
st
ca
nc
er
an
d
ho
rm
on
e
re
pl
ac
e
m
en
t
th
er
ap
y:
co
lla
bo
rat
iv
e
re
an
al
ys
is
of
da
ta
fr
o
m
51
ep
id
e
mi
ol
og
ic
al
st
ud
ies
of
52
70
5
w
o
m
en
wi
th
br
ea
st
ca
nc
er
an
d
10
8
41
1
w
o
m

en
wi
th
ou
t
br
ea
st
ca
nc
er.
L
an
ce
t
20
07
;
35
0:
10
47

59
.

W
rit
in
g
Gr
ou
p
fo
r
th
e
W
o
m
en
s
H
ea
lth
Ini
tia
tiv
e
In
ve
sti
ga
tor
s.
Ri
sk
s
an
d
be
ne
fit
s
of
est
ro
ge
n
pl
us
pr
og
est
in
in
he
alt
hy
po
st

m
en
op
au
sal
w
o
m
en
:
pri
nc
ip
al
re
su
lts
fr
o
m
th
e
W
o
m
en
s
H
ea
lth
Ini
tia
tiv
e
ra
nd
o
mi
ze
d
co
ntr
oll
ed
tri
al.
J
A
M
A
20
12
;
28
8:
32
1
33
.

C
hl
eb
o
ws
ki
R
T,
H
en
dri
x
S
L,
La
ng
er
R
D,
et
al,

fo
r
th
e
W
HI
In
ve
sti
ga
tor
s.
In
flu
en
ce
of
est
ro
ge
n
pl
us
pr
og
est
in
on
br
ea
st
ca
nc
er
an
d
m
a
m
m
og
ra
ph
y
in
he
alt
hy
po
st
m
en
op
au
sal
w
o
m
en
:
th
e
W
o
m
en
s
H
ea
lth
Ini
tia
tiv
e
ra
nd
o
mi
se
d
tri
al.
J
A

M
A
20
09
;
28
9:
32
43

53
.

Be
ral
V,
Ba
nk
s
E,
Re
ev
es
G.
Ev
id
en
ce
fr
o
m
ra
nd
o
mi
se
d
tri
als
on
th
e
lo
ng
ter
m
eff
ec
ts
of
ho
rm
on
e
re
pl
ac
e
m
en
t
th
er
ap
y.
L
an
ce
t
20
12
;
36
0:
94
2
44
.

Be
ral
V,
Ba
nk
s
E,
Re
ev
es
G,
A
pp
le
by
P.
Us
e
of
H
R
T
an
d
th
e
su
bs
eq
ue
nt
ris
k
of
ca
nc
er.
J
E
pi
de
mi
ol
Bi
os
ta
t
20
09
;
4:
19
1
21
5.

Be
rq
ui
st
L,
A
da
mi
HO,
Pe
rss
on
I,
H
oo
ve
r
R,
Sc
ha
ire

r
C.
Th
e
ris
k
of
br
ea
st
ca
nc
er
aft
er
est
ro
ge
n
an
d
est
ro
ge
npr
og
est
in
re
pl
ac
e
m
en
t.
N
E
ng
lJ
M
ed
19
99
;
32
1:
29
3
97
.

M
ag
nu
ss
on
C,
Ba
ro
n
JA
,
C
or
rei
a
N,
Be
rg
str
o
m
R,
A
da
mi
HO,
Pe

rss
on
I.
Br
ea
st
ca
nc
er
ris
k
fol
lo
wi
ng
lo
ng
ter
m
oe
str
og
en
an
d
oe
str
og
en
pr
og
est
inre
pl
ac
e
m
en
t
th
er
ap
y.
In
tJ
C
an
ce
r
20
09
;
81
:
33
9
44
.

Sc
ha
ire
r
C,
Lu
bi
n
J,
Tr
oi
si
R,
St
ur
ge
on
S,
Br

int
on
L,
H
oo
ve
r
R.
M
en
op
au
sal
est
ro
ge
n
an
d
est
ro
ge
npr
og
est
in
re
pl
ac
e
m
en
t
th
er
ap
y
an
d
br
ea
st
ca
nc
er
ris
k.
J
A
M
A
20
00
;
28
3:
48
5
91
.

R
os
s
R
K,
Pa
ga
ni
niHi
ll
A,
W
an
P
C,
Pi
ke
M

C.
Ef
fe
ct
of
H
or
m
on
e
re
pl
ac
e
m
en
t
th
er
ap
y
on
br
ea
st
ca
nc
er
ris
k:
est
ro
ge
n
ve
rs
us
est
ro
ge
n
pl
us
pr
og
est
in.
J
N
at
l
C
an
ce
r
In
st
20
00
;
92
:
32
8
32
.

Li
CI
,
M
al
on
e
K
E,
Po
rte
r
P
L,

W
eis
s
N
S,
et
al.
Re
lat
io
ns
hi
p
be
tw
ee
n
lo
ng
du
rat
io
ns
an
d
dif
fer
en
t
re
gi
m
en
s
of
ho
rm
on
e
th
er
ap
y
an
d
ris
k
of
br
ea
st
ca
nc
er.
J
A
M
A
20
09
;
28
9:
32
54

63
.

Th
e
M
illi
on
W
o
m
en
St
ud
y

C
oll
ab
or
ati
ve
Gr
ou
p.
Th
e
M
illi
on
W
o
m
en
St
ud
y:
de
si
gn
an
d
ch
ar
ac
ter
ist
ics
of
th
e
st
ud
y
po
pu
lat
io
n.
Br
ea
st
C
an
ce
r
R
es
19
99
;
1:
73

80
.

Ba
nk
s
E,
Be
ral
V,
Ca
m
er
on
R,
et
al.
C
o
m
pa
ris
on

of
va
rio
us
ch
ar
ac
ter
ist
ics
of
w
o
m
en
w
ho
do
an
d
do
no
t
att
en
d
fo
r
br
ea
st
ca
nc
er
sc
re
en
in
g.
Br
ea
st
C
an
ce
r
R
es
20
11
;
4:
R
1.
1
1.
6.

M
illi
on
W
o
m
en
St
ud
y
C
oll
ab
or
at
or
s.
Pa
tte
rn
s
of
us

e
of
ho
rm
on
e
re
pl
ac
e
m
en
t
th
er
ap
y
in
on
e
mi
lli
on
w
o
m
en
in
Br
ita
in,
20
06

20
00
.
Br
J
O
bs
t
G
yn
ae
co
l
20
12
;
10
9:
13
19

30
.

Ba
nk
s
E,
Be
ral
V,
Ca
m
er
on
R,
et
al.
A
gr
ee
m
en
t
be
tw

ee
n
ge
ne
ral
pr
ac
tic
e
pr
es
cri
pti
on
da
ta
an
d
sel
fre
po
rte
d
us
e
of
ho
rm
on
e
re
pl
ac
e
m
en
t
th
er
ap
y
an
d
tre
at
m
en
t
fo
r
va
rio
us
ill
ne
ss
es.
J
E
pi
de
mi
ol
Bi
os
ta
t
20
11
;
6:
35
7
63
.

Br
iti
sh

M
ed
ic
al
As
so
ci
ati
on
an
d
R
oy
al
Ph
ar
m
ac
eu
tic
al
So
ci
et
y
of
Gr
ea
t
Br
ita
in.
Br
iti
sh
na
tio
na
l
fo
rm
ul
ar
y.
Lo
nd
on
:
B
M
J
B
oo
ks
,
19
99
.

Int
er
na
tio
na
l
St
ati
sti
ca
l
Cl
as
sif
ic
ati
on
of
Di
se
as
es

an
d
Re
lat
ed
H
ea
lth
Pr
ob
le
m
s,
10
th
re
vi
si
on
.
G
en
ev
a:
W
orl
d
H
ea
lth
Or
ga
ni
za
tio
n,
19
92
.

Ea
st
on
D
F,
Pe
to
J,
Ba
bi
ke
r
A
G
A
G.
Fl
oa
tin
g
ab
so
lut
e
ris
k:
an
alt
er
na
tiv
e
to
rel
ati
ve
ris
k
in
su
rvi

va
l
an
d
ca
se
co
ntr
ol
an
al
ys
is
av
oi
di
ng
an
ar
bit
rar
y
ref
er
en
ce
gr
ou
p.
St
at
M
ed
20
01
;
10
:
10
25

35
.

Pa
rki
n
D
M
,
W
he
la
n
S
L,
Fe
rla
y
J,
Te
pp
o
L,
Th
o
m
as
D
B,
ed
s.
Ca
nc
er
in
ci
de
nc
e

in
fiv
e
co
nti
ne
nt
s,
vo
l
VI
II.
Ly
on
:
Int
er
na
tio
na
l
A
ge
nc
y
fo
r
Re
se
ar
ch
on
Ca
nc
er
Sc
ie
nti
fic
Pu
bli
ca
tio
ns
,
20
02
.

K
ey
he
alt
h
sta
tis
tic
s
fr
o
m
ge
ne
ral
pr
ac
tic
e
19
98
:
se
rie
s
M
B
6
(n
o
2).
Lo

nd
on
:
Of
fic
e
fo
r
N
ati
on
al
St
ati
sti
cs,
20
00
.

Ca
nc
er
Re
se
ar
ch
U
K.
Br
ea
st
Ca
nc
er
Fa
cts
he
et,
Ju
ne
20
09
.
htt
p:/
/w
w
w.
ca
nc
err
es
ea
rc
hu
k.
or
g/
ab
ou
tc
an
ce
r/s
tat
ist
ics
/st
ats
mi
sc/
pd
fs/
f
ac
ts
he
et
_b
re

ast
_j
un
20
09
.p
df
(a
cc
es
se
d
Ju
ly
10
,
20
09
).

Pi
ke
M
C,
Pe
ter
s
R
K,
C
oz
en
W,
et
al.
Es
tro
ge
npr
og
est
in
re
pl
ac
e
m
en
t
th
er
ap
y
an
d
en
do
m
etr
ial
ca
nc
er.
J
N
at
l
C
an
ce
r
In
st
19
97
;
98
:
11
10

16
.

W
ei
de
rp
as
s
E,
A
da
mi
HO,
Ba
ro
n
JA
,
et
al.
Ri
sk
of
en
do
m
etr
ial
ca
nc
er
fol
lo
wi
ng
est
ro
ge
n
re
pl
ac
e
m
en
t
wi
th
an
d
wi
th
ou
t
pr
og
est
ag
in
s.
J
N
at
l
C
an
ce
r
In
st
20
09
;
1:
11
31

37
.

N
e
w
co
m
b
P
A,
Tr
en
th
a
mD
eit
z
A.
Pa
tte
rn
s
of
po
st
m
en
op
au
sal
pr
og
est
ag
en
us
e
wi
th
est
ro
ge
n
in
rel
ati
on
to
en
do
m
etr
ial
ca
nc
er
(U
nit
ed
St
at
es
).
C
an
ce
r
C
au
se
s
C
on
tr
ol
20

09
;
10
:
19
5
20
1.

Ba
nk
s
E.
H
or
m
on
e
re
pl
ac
e
m
en
t
th
er
ap
y
an
d
th
e
se
ns
iti
vit
y
an
d
sp
ec
ifi
cit
y
of
br
ea
st
ca
nc
er
sc
re
en
in
g:
a
re
vi
e
w.
J
M
ed
Sc
re
en
20
11
;
12
:
29

34
.

THE
LANCET
Vol 362
August 9,
2009
www.thelanc
et.com

427