Breast Cancer Research and Treatment

https://doi.org/10.1007/s10549-017-4655-0

EPIDEMIOLOGY

Assessment of potential risk factors for breast cancer in a population

in Southern Brazil
Juliana Zeni Breyer1 · Eliana Marcia Wendland1,2 · Natália Luiza Kops1,3 · Maira Caleffi1,4 · Luciano Serpa Hammes1

Received: 27 December 2017 / Accepted: 30 December 2017

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Abstract
Purpose  The aim of this study is to assess potential risk factors for breast cancer in a population in Southern Brazil and build
a multivariate logistic model using these factors for breast cancer risk prediction.
Methods  A total of 4242 women between 40 and 69 years of age without a history of breast cancer were selected at primary
healthcare facilities in Porto Alegre and submitted to mammographic screening. They were evaluated for potential risk factors.
Results  In all, 73 participants among the 4242 women had a breast cancer diagnosis during the follow-up of the project
(10 years). The multivariate analysis considering all the patients aged 40–69 years showed that older age (OR 1.08, 95% CI
1.04–1.12), higher height (OR 1.04, 95% CI 1.01–1.09), and history of previous breast biopsy (OR 2.66, 95% CI 1.38–5.13)
were associated with the development of breast cancer. Conversely, the number of pregnancies (OR 0.87, 95% CI 0.78–0.98)
and use of hormone replacement therapy (OR 0.39, 95% CI 0.20–0.75) were considered a protective factor. Additionally,
we performed an analysis separating the participants into groups of 40–49 and 50–69 years old, since a risk factor could
have a specific behavior in these age groups. No additional risk factors were identified within these age brackets, and some
factors lost statistical significance.
Conclusion  The risk prediction model indicates that the following variables should be assessed in this specific population:
age, height, having had previous breast biopsies, number of pregnancies, and use of hormone replacement therapy. These
findings may help to better understand the causal model of breast cancer in Southern Brazil.

Keywords  Breast cancer · Risk factors · Mortality · Etiology · Incidence

Introduction cause of cancer death in developed countries and the highest

cause of cancer death in developing countries [2].
Due to its high incidence and mortality, breast cancer is a Breast cancer incidence shows variations in different
public health problem worldwide. It is the most common populations [1]. Thus, the importance of knowing the risk
cancer in women both in the developed and less developed factors for breast cancer for a better understanding of such
world [1]. In 2017, it is estimated that there will be 252,710 variations is undisputable. This evidence suggests that there
new cases of breast cancer [2]. In addition, it is the second are different environmental factors that lead to different inci-
dence rates. The highest incidence of breast cancer in 2012
was in Northern America and Oceania, and the lowest inci-
* Natália Luiza Kops dence was in Asia and Africa [3]. Slightly more cases of
natalia.kops@hmv.org.br breast cancer were diagnosed in less developed countries
Juliana Zeni Breyer (53%) [1]. However, the highest proportion of breast cancer
juliana.zeni@hmv.org.br survivors were in developed countries [1].
1
Hospital Moinhos de Vento, Porto Alegre, Brazil In Brazil, excluding non-melanoma skin cancer, breast
2 cancer is the most frequent cancer among women. In 2016,
Department of Public Health, Federal University of Health
Sciences of Porto Alegre (UFCSPA), Porto Alegre, Brazil 57,960 new cases were estimated, with an incidence of
3 56.2 cases per 100,000 women [2]. The prevalence varies
Post‑Graduate Program in Endocrinology, Universidade
Federal do Rio Grande do Sul (UFRGS), Porto Alegre, Brazil between different regions of the country [2, 4, 5]. Women
4 living in Southern Brazil have a higher chance of having
Núcleo Mama Porto Alegre, Porto Alegre, Brazil

13
Vol.:(0123456789)

breast cancer, but no data are available to examine if this
difference is due to a different risk profile.
Knowing the risk factors of this population will permit
us to propose public health actions or increase awareness
in a determined group of high-risk women to decrease risk
and increase early diagnosis.
Although there are international standards for risk pre-
diction [6], the fact that there is a concomitantly similar
cancer incidence among countries with a higher level of
development and a discrepancy in the prevalence of classic
risk factors among these populations suggests that models
based on local data are needed.
It is clear that incidence has a great geographic vari-
ation, which suggests that the action of risk factors var-
ies substantially between the different populations. Thus,
studies on determining the factors for breast cancer in cer-
tain populations may contribute to improving public health
strategies and reduce morbimortality. Therefore, the objec-
tive of this study was to assess the potential risk factors for
breast cancer in a population sample of Southern Brazil.
Methods
Patient selection
This study is a subanalysis of the Project “Development
of Operational and Management Techniques in a Breast
Healthcare Service—Núcleo Mama Porto Alegre,” which
is a cohort study assessing the risk factors, prevalence,
incidence, and characteristics of breast cancer in a group
of 9176 women. This project has been described in detail
elsewhere [7]. Inclusion criteria were age greater than or
equal to 15 years old, no previous history of breast can-
cer, and living in the city of Porto Alegre in areas served
by primary healthcare units. This group was recruited
from 2004 to 2006. For this subproject, all women aged
40–69 years who were enrolled in the cohort were selected.
Therefore, 4242 women were selected.
Follow‑up of participants
In the recruitment period, participants underwent a clinical
questionnaire and physical examination at their primary
healthcare unit. The physical examination consisted of an
anthropometric assessment and clinical examination of the
breast by inspection and palpation.
After registering, the participants were referred for fol-
low-up in the reference center of the study (Núcleo Mama
Porto Alegre) in three situations:
13
Breast Cancer Research and Treatment
(a) Upon turning age 40. For example, participants who
were 35 years old at the time of registration would be
sent for their first follow-up visit 5 years later,
(b) Participants presenting any breast-related symptoms
reported at the time of registration or at any time before
age 40 or
(c) Participants who reported a family history of cancer
in the registration questionnaire were sent to Núcleo
Mama Porto Alegre for an appointment with a geneti-
cist for risk assessment, and participants considered to
be at a moderate or high risk started having follow-ups
every 6 months.
Participants who were sent to the reference center were
followed up on at annual appointments or, in the case of par-
ticipants who had been classified as moderate or high risk for
breast cancer, at semiannual appointments. In this follow-up,
participants had their clinical data updated and underwent a
new clinical breast examination and mammography.
Participants who had alterations in the clinical examina-
tion and a normal mammogram were referred for further
examination, in this case, a breast ultrasound.
Participants who had alterations in their clinical exami-
nation and mammogram were referred for further examina-
tion, which could be a new mammogram or ultrasound, as
prescribed by the physician.
Following complementary examination, a decision was
made as to whether a diagnostic procedure, such as breast
biopsy, aspiration puncture of the breast or even breast sur-
gery (when an ultrasound-guided biopsy was not feasible)
should be performed.
When surgery was indicated for the participant’s lesion,
whether benign or malignant, the surgery was performed
within the scope of the study. Participants diagnosed with
breast cancer had full access to surgical treatment at Núcleo
Mama Porto Alegre and were referred for further treatment,
radiotherapy and chemotherapy, when necessary. Later, they
continued undergoing follow-ups at Núcleo Mama Porto
Alegre.
Data collection instruments
All data collection instruments were pre-tested for their suit-
ability. All project interviewers were trained and underwent
a certification process consisting of simulated interviews and
interviews supervised by the researchers. A procedure man-
ual developed by the researchers was used for training and
as a reference regarding standards for the interviews and for
taking anthropometric measurements. The instruments used
in the project were periodically calibrated by INMETRO.
Participants who needed to be evaluated in the reference
center of the study (Núcleo Mama Porto Alegre) had their
clinical data updated and underwent a physical examination
Breast Cancer Research and Treatment
and mammography. The collected data included the educa-
tional level (grouped according to years and type of educa-
tion) [8], skin color, presence of smoking, alcohol abuse,
history of breast biopsies, family history of cancer, age at
menarche (years), number of pregnancies, age at the birth
of their first child (years), age at the last pregnancy (years),
duration of breastfeeding (months), history of oophorectomy
and hysterectomy, age at oophorectomy and menopause
(years), use and duration of hormone replacement therapy
(month), use of hormonal contraceptive, and history of
depression.
Duplicate anthropometric measurements were made by
trained interviewers. Weight (kg) was measured using an
electronic scale (Urano, Brazil—180 kg), with patients wear-
ing light clothing and no shoes. Height (m) was measured
in the Frankfurt position using a wall-mounted stadiometer
(1-mm precision; Sanny, Brazil). Weight and height meas-
urements were used to calculate body mass index (BMI, kg/
m2) and were classified according to World Health Organi-
zation criteria: normal weight 18.5–24.9 kg/m2, overweight
25.0–29.9 kg/m2, and obesity ≥ 30 kg/m2 [9].
Quality tests were performed every 2 months by double
checking, and the data were evaluated later by the inter-
viewer–researcher agreement using the Kappa statistical test.
Ethical aspects
This study was approved by the Ethics Committee on
Human Research (Hospital Moinhos de Vento—protocol
no. 2004/2013), and all participants were informed about
the study and signed an Informed Consent Form.
Statistical analysis
Categorical variables were described using absolute frequen-
cies and percentage, while continuous variables were ana-
lyzed using the mean and standard deviations or the median
and interquartile range (25th and 75th percentiles). To com-
pare the proportions, the χ2 test and Fisher’s exact test were
used. Student’s t test or Mann–Whitney non-parametric tests
were used for continuous variables.
First, we performed a univariate analysis on each poten-
tial risk factor for breast cancer comparing the groups. Vari-
ables associated with the outcome in the bivariate analysis
with a p value < 0.20 were included in the logistic regression
model. Then, we conducted a multivariate analysis to assess
which risk factors were independently associated with the
occurrence of breast cancer.
Additionally, to evaluate whether the risk factors behaved
differently in different age groups, we stratified the model
between the age brackets of 40–49 years and 50–69 years
because some risk factors are influenced by age (e.g., hor-
monal factors).
We also stratified the model by the type of cancer, such
as invasive ductal carcinoma (IDC) and other types, since
IDC was the most common cancer type in the sample. For
this purpose, the following variables were used: age, weight,
height, breast cancer-positive family history, previous breast
biopsy, number of pregnancies, and use of hormone therapy.
Previous history of breast and ovarian cancer was not used,
as there would be no biological need because the breast
cancer-positive family history and previous breast biopsy
variables were included in the model. Use of hormone
therapy was included only in the model for women aged
50–69 years, because its use was negligible in the age group
from 40 to 49 years.
A significance level of 5% was set for the comparisons.
The analyses were carried out in ­SAS® v9.4 (Cary, United
States) [10].
Results
After a 10-year follow-up of the cohort, a total of 9176
women were registered over the age of 15, and 95 partici-
pants were diagnosed with breast cancer. For this study, only
women between 40 and 69 years old were selected, totaling
4242 women without a history of breast cancer, and of these,
73 participants were diagnosed with breast cancer. Table 1
describes the breast cancer cases and shows that most of
them (75.3%) consist of invasive ductal carcinoma.
The mean age of the participants without breast cancer
was 50.9 (7.6) years old, while the participants with breast
cancer were 55.8 (7.2) years (p < 0.001) (Table 2). Weight
was also different between the groups. The women with
breast cancer weighed approximately 4 kg more on average
than women without breast cancer (p = 0.035). In addition,
history of previous breast biopsies (p < 0.001), alterations in
previous breast biopsies (p = 0.001), and personal history of
breast and ovarian cancer were more prevalent in the group
with breast cancer (p = 0.007).
For a better selection of the risk factors that could be
included in the multivariate model, we made comparisons
by stratifying the groups into age brackets of 40–49 and
50–69 (data not shown). In the 40- to 49-year-old group, we
Table 1  Description of breast cancer cases among women aged
40–69 years in the NMPOA cohort
Type of cancer N %
Invasive ductal carcinoma 55 75.3
Ductal carcinoma in situ 11 15.1
Invasive lobular carcinoma 2 2.7
Mixed 2 2.7
Others 3 4.1
13
 Breast Cancer Research and Treatment

Table 2  Description of the sample of women aged 40–69 years in the NMPOA cohort

Variables With breast cancer Without breast cancer p value
Value N Value N

Age at baseline (years) 55.8 (7.2) 50.9 (7.6) < 0.0012

Educational level
 Incomplete primary education 58.9 43 63.4 2571 0.832a
 Complete primary education 17.8 13 16.8 682
 Incomplete secondary education 4.1 3 4.2 172
 Complete secondary education or higher education 19.2 14 15.6 631
Race
 White 78.1 57 81.2 3082 0.602c
 Others 21.9 16 18.8 714
Weight (kg) 73.3 (15.9) 69.6 (14.6) 0.035b
Height (cm) 1.57 (0.1) 1.56 (0.1) 0.068b
Body mass index (kg/m2) 29.6 (6.5) 28.7 (5.8) 0.185b
 18–24 20.6 13 23.6 843 0.348a
 25–30 36.5 23 42.3 1513
 > 30 42.9 27 34.1 1219
Presence of smoking 26.0 19 29.8 1243 0.564c
Alcohol abuse – – 0.3 13 0.999c
Positive family history 21.9 16 14.9 622 0.136c
Previous breast biopsy 18.1 13 5.5 229 < 0.001c
Number of previous breast biopsies 1.2 (0.4) 1.2 (0.5) 0.842b
Alteration in previous breast biopsies 6.9 5 0.9 39 0.001c
Personal history positive for breast or ovarian cancer 5.6 4 1.0 41 0.007c
Age at menarche (years) 12.6 (1.7) 13.9 (9.2) 0.232b
History of pregnancy
 Yes 94.5 69 95.2 3962 0.778c
 No 5.5 4 4.8 198
Age at birth of first child (years) 21.8 (5.1) 21.5 (5.1) 0.672b
Number of pregnancies 3.7 (2.3) 4.3 (2.7) 0.111b
Age at last pregnancy (years) 31.7 (6.4) 32.1 (6.2) 0.610b
Breastfeeding time (months) 18 (3–40) 9 (3–30) 0.214d
Having had a hysterectomy 15.4 6 25.2 323 0.225c
Age at hysterectomy (years) 44.5 (2.9) 42.8 (7.1) 0.566b
Having had an oophorectomy 15.4 6 17.0 194 0.960c
Age at oophorectomy (years) 38.0 (8.2) 39.2 (9.4) 0.771b
Presence of menopause 68.5 50 75.4 2184 0.226c
Age at menopause (years) 46.9 (5.6) 46.3 (5.7) 0.522b
Use of hormone replacement therapy 19.2 14 30.3 530 0.056c
Length of use of hormone replacement therapy (months) 12 (4–60) 24 (6–48) 0.667d
Use of hormonal contraception
Yes 82.2 60 75.2 3137 0.219c
No 17.8 13 24.8 1032
Length of use of hormonal contraception 24 (15–180) 72 (24–168) 0.587d
History of depression present 44.7 21 44.9 1050 0.999c

Data expressed as the mean (standard deviation), median (P25–P75) or prevalence

a 2
 χ test
b
 Student’s t test
c
 Fisher’s exact test
d
 Mann–Whitney. α = 0.05

13
Breast Cancer Research and Treatment

found the following different risk factors: height (p = 0.016), Table 4  Association between the type of breast cancer and the risk
having had a previous breast biopsy (p = 0.033), and the factors among participants aged 40–69 years in the NMPOA cohort
presence of menopause (p = 0.022). In the 50- to 69-year- Variables IDC OR (95% CI) Other OR (95% CI)
old group, the risk factors were age (p = 0.052), weight
Age (years) 1.08 (1.04–1.12) 1.07 (1.01–1.15)
(p = 0.058), a previous breast biopsy (p = 0.003), changes
Weight (kg) 1.01 (0.99–1.03) 1.02 (0.99–1.05)
in previous breast biopsies (p = 0.004), positive personal
Height (cm) 1.05 (1.01–1.10) 1.04 (0.96–1.12)
history for breast or ovarian cancer (p = 0.039), number
Positive family history 0.92 (0.42–2.01) 2.06 (0.74–5.76)
of pregnancies (p  =  0.024), having had a hysterectomy
Previous biopsy 1.73 (0.74–4.03) 5.95 (2.10–16.86)
(p = 0.028), the presence of menopause (p = 0.012), hor-
Number of pregnancies 0.79 (0.68–0.92) 1.04 (0.88–1.24)
mone replacement therapy (p = 0.033), and the use of a
Use of hormone replace- 0.24 (0.10–0.58) 0.98 (0.35–2.72)
hormonal contraceptive (p = 0.036) (data not shown).
ment therapy
Multivariate analysis showed that age (OR 1.08, 95% CI
1.04–1.12), height (OR 1.04, 95% CI 1.01–1.09), and having IDC invasive ductal carcinoma, OR odds ratio, Other ductal carci-
had previous breast biopsies (OR 2.66, 95% CI 1.38–5.13) noma in  situ, invasive lobular carcinoma, mixed or other rare types.
Logistic regression model
were found to increase the risk of breast cancer, while the
number of pregnancies (OR 0.87, 95% CI 0.78–0.98) and
the use of hormone replacement therapy (OR 0.39, 95% Discussion
CI 0.20–0.75) acted as a protective factor (Table 3). When
analyzed by age group, height and previous breast biopsies Breast cancer is one of the most researched neoplasms in
alone increased the risk of breast cancer among women aged the world, but there is still much disagreement over the risk
40–49 years. Among women aged 50–69 years, having had factors for the development of the disease. This study has
previous breast biopsies increased the risk of breast cancer, assessed the potential risk factors for breast cancer in a pop-
while the number of pregnancies and the use of hormone ulation in Southern Brazil. The risk prediction model indi-
replacement therapy was shown to be a protective factor. cates that age, height, having had previous breast biopsies,
The multivariate analysis by the type of breast cancer number of pregnancies, and the use of hormone replace-
showed that age (OR 1.08, 95% CI 1.04–1.12) was a risk ment therapy should be assessed in this specific population
factor for IDC cancer, while the number of pregnancies (OR of women aged 40–69 years.
0.79, 95% CI 0.68–0.92) and the use of hormone replace- Age remains as one of the key factors in the development
ment therapy (OR 0.24, 95% CI 0.10–0.58) proved to be of breast cancer [4, 6, 11]. Although it can occur in young
protective factors (Table 4). For other types of cancer, only women, approximately 75% of cases occur after age 50 [12].
age (OR 1.07, 95% CI 1.01–1.15) and having had a previous The median age of breast cancer diagnosis for women in the
breast biopsy (OR 5.95, 95% CI 2.10–16.86) proved to be U.S. is 62; however, this could vary by race and ethnicity
risk factors. [13]. This result was demonstrated in our study, in which
the mean age of women with breast cancer was 55.8 (7.2)
years old.
Numerous studies have assessed the relationship between
weight, height, BMI, and breast cancer. The consensus of
these studies is that obesity may increase the risk for breast

Table 3  Association between Variables 40–69 years old OR 40–49 years old OR (95% CI) 50–69 years old
the risk factors and breast (95% CI) OR (95% CI)
cancer, stratified by age group
Age (years) 1.08 (1.04–1.12) 1.16 (0.95–1.42) 1.05 (1.01–1.11)
Weight (kg) 1.01 (1.00–1.03) 1.00 (0.97–1.04) 1.02 (1.01–1.03)
Height (cm) 1.04 (1.00–1.09) 1.10 (1.02–1.18) 1.03 (0.98–1.08)
Positive family history 1.20 (0.65–2.24) 1.45 (0.39–5.41) 1.18 (0.58–2.37)
Previous biopsy 2.66 (1.38–5.13) 4.11 (1.06–15.94) 2.61 (1.24–5.50)
Number of pregnancies 0.87 (0.78–0.98) 0.94 (0.72–1.22) 0.86 (0.76–0.98)
Use of hormone replace- 0.39 (0.20–0.75) – 0.44 (0.22–0.85)
ment therapy

Logistic regression model, OR odds ratio

13

cancer in women aged 50 years or older [14–17]. Links
between greater adiposity and increased risk of many can-
cers are biologically plausible considering that obesity is
related to a vast array of metabolic and physiological dys-
functions. Our study found a significant difference in body
weight between the groups with and without breast cancer;
however, this difference was not found after adjustment,
possibly because approximately 80% of the women were
overweight or obese [18]. Therefore, nutritional strategies
should be considered for this population [19, 20].
Family history of breast cancer is also a well-established
risk factor in the literature [21]. Breast cancer history in first-
degree relatives is considered an independent risk factor for
the development of the disease. Mutations in the BRCA-1
and BRCA-2 genes are responsible for most cases of breast
cancer in relatives [22]. In our study, there was no statisti-
cally significant difference between the groups studied.
In this study, hormone replacement therapy was shown
to be protective against breast cancer in women aged
50–69 years. Several studies do not corroborate this find-
ing, although a recent meta-analysis of seven prospective
cohort studies has shown the effect modification by hormone
therapy use in postmenopausal breast cancer, where patients
who never received hormone therapy showed an ~ 20%
increased risk per standard deviation of BMI, waist circum-
ference, and hip circumference compared to those patients
that did receive hormone therapy [17]. We conducted several
subanalyses of this group to try to identify whether any con-
founding factor was present, which was not demonstrated.
We should consider this result with caution, especially
because studies suggest that hormone replacement therapy
increases the risk of breast cancer when used for five years
or longer [23]. In addition, our study did not investigate
the type of hormone therapy and the total time of use, but
instead we looked at whether it was used or not. This factor
needs further exploration in other cohorts and populations.
It is important to note that risk prediction analyses should
include the steps performed in this study: first in a univariate
analysis to explore the simple association of the risk factors,
and then in a multivariate analysis to effectively assess the
actual association of the risk factors with the outcome. In
addition, when there is a possibility that risk factors have
different behaviors in subgroups, it is essential to conduct an
exploratory analysis of the risk factors in these subgroups.
Although there are studies in the literature about the vali-
dation of new models for breast cancer risk prediction, we
did not find studies on the validation of breast cancer risk
models specifically for the population of Rio Grande do Sul,
and the prevalence varies between different regions of the
country [2].
Among the well-established risk factors in the literature,
only age, height, and having had previous breast biopsies
were associated with an increased risk of breast cancer
13
Breast Cancer Research and Treatment
development when a multivariate analysis was carried out.
Additionally, the number of pregnancies was considered a
protective factor. When stratifying by type of cancer, only
age was considered a risk factor for the development of the
histological type, IDC. Therefore, these factors should be
prioritized in the assessment of the risk of breast cancer
development.
It is not possible to carry out preventive interventions on
some major risk factors for the development of breast cancer,
such as age, family history, and genetic mutation. However,
there are risk factors that should be submitted to interven-
tions for the breast cancer prevention, such as the overweight
and obesity risk factors.
Breast cancer remains a public health problem in the
world despite all the investments that have been made, which
shows that preventive measures have been mostly ineffec-
tive. There are no population-specific measures yet to con-
trol and prevent the disease. The greatest advances in this
area have been aimed solely at early diagnosis and treatment
of the disease. Thus, studies like ours are very important to
understand the distribution of risk factors that lead to the
development of breast cancer in different locations and may
help develop more effective practices to control the inci-
dence of breast cancer in a specific population.
Conclusion
Risk factors for breast cancer are not well understood yet.
Life habits have been changing over the years and many
of these habits may be contributing to the increased inci-
dence of the disease. More robust studies are necessary to
elucidate the risk factors involved in the increasing number
of breast cancer cases in specific populations, thus allow-
ing greater efforts to prevent such cases and reducing the
high morbimortality rates in the female population. In our
study, the risk prediction model indicates that the follow-
ing variables should be assessed in this specific population
aged 40–69 years: age, height, having had previous breast
biopsies, number of pregnancies, and the use of hormone
replacement therapy.
Funding Information  This study was funded by Brazilian Ministry of
Health along with the Moinhos de Vento Hospital through the Program
for Supporting the Institutional Development of the Public Health Sys-
tem (PROADI-SUS).
Compliance with ethical standards 
Conflict of interest  The authors declare that they have no conflicts of
interest.
Informed consent  Informed consent was obtained from all individual
participants included in the study.
Breast Cancer Research and Treatment

References 14. Hunter DJ, Willett WC (1993) Diet, body size, and breast cancer.
Epidemiol Rev 15:110–132
15. Suzuki Y, Tsunoda H, Kimura T, Yamauchi H (2017) BMI change
1. Breast cancer statistics | World Cancer Research Fund Interna-
and abdominal circumference are risk factors for breast cancer,
tional. http​://www.wcrf​.org/int/canc​er-fact​s-figu​res/data​-spec​ific​
even in Asian women. Breast Cancer Res Treat. http​s://doi.
-canc​ers/brea​st-canc​er-stat​isti​cs. Accessed 5 Sep 2017
org/10.1007​/s105​49-017-4481​-4
2. INCA—Instituto Nacional de Câncer—Estimativa 2016. http​://
16. Elkum N, Al-Tweigeri T, Ajarim D et al (2014) Obesity is a sig-
www.inca​.gov.br/esti​mati​va/2016​/. Accessed 5 Sep 2017
nificant risk factor for breast cancer in Arab women. BMC Cancer
3. Fact Sheets by Population. http:​ //globo​ can.​ iarc.​ fr/Pages​ /fact_​ shee​
14:788. http​s://doi.org/10.1186​/1471​-2407​-14-788
ts_popu​lati​on.aspx​. Accessed 13 Sep 2017
17. Freisling H, Arnold M, Soerjomataram I et al (2017) Comparison
4. Renck DV, Barros F, Domingues MR et al (2014) Equity in access
of general obesity and measures of body fat distribution in older
to breast cancer screening in a mobile mammography program
adults in relation to cancer risk: meta-analysis of individual par-
in southern Rio Grande do Sul State, Brazil. Cad Saúde Públi
ticipant data of seven prospective cohorts in Europe. Br J Cancer
30:88–96. http​s://doi.org/10.1590​/0102​-311X​0001​7113​
116:1486–1497. http​s://doi.org/10.1038​/bjc.2017​.106
5. da Santos SS, Melo LR, Koifman RJ, Koifman S (2013) Breast
18. Park J, Morley TS, Kim M et al (2014) Obesity and cancer–mech-
cancer incidence and mortality in women under 50 years of age
anisms underlying tumour progression and recurrence. Nat Rev
in Brazil. Cad Saúde Públ 29:2230–2240. http​s://doi.org/10.1590​
Endocrinol 10:455–465. http​s://doi.org/10.1038​/nren​do.2014​.94
/0102​-311x​0003​0713​
19. Gandini S, Merzenich H, Robertson C, Boyle P (1990) Meta-
6. Breast Cancer Risk and Prevention. http​s://www.canc​er.org/canc​
analysis of studies on breast cancer risk and diet: the role of fruit
er/brea​st-canc​er/risk​-and-prev​enti​on.html​. Accessed 5 Sep 2017
and vegetable consumption and the intake of associated micronu-
7. Caleffi M, Ribeiro RA, Filho DLD et al (2009) A model to opti-
trients. Eur J Cancer Oxf Engl 36:636–646
mize public health care and downstage breast cancer in limited-
20. Inumaru LE, da Silveira ÉA, Naves MMV (2011) Risk and pro-
resource populations in southern Brazil. (Porto Alegre Breast
tective factors for breast cancer: a systematic review. Cad Saúde
Health Intervention Cohort). BMC Publ Health 9:83. http​s://doi.
Públ 27:1259–1270. http​s://doi.org/10.1590​/S010​2-311X​2011​
org/10.1186​/1471​-2458​-9-83
0007​0000​2
8. Associação Brasileira de Empresas de Pesquisa (ABEP) (2012)
21. Sepandi M, Akrami M, Tabatabaee H et al (2014) Breast cancer
Critério de Classificação Econômica Brasil
risk factors in women participating in a breast screening pro-
9. WHO | Obesity and overweight. In: WHO. http​://www.who.int/
gram: a study on 11,850 Iranian females. Asian Pac J Cancer
medi​acen​tre/fact​shee​ts/fs31​1/en/. Accessed 7 Sep 2017
Prev APJCP 15:8499–8502
10. SAS (2016) SAS Institute Inc. 2016. SAS/STAT​® 14.2 User’s
22. Fackenthal JD, Olopade OI (2007) Breast cancer risk associated
Guide. Cary, NC: SAS Institute Inc
with BRCA1 and BRCA2 in diverse populations. Nat Rev Cancer
11. U.S. Breast Cancer Statistics. In: Breastcancer.org. http​://www.
7:937–948. http​s://doi.org/10.1038​/nrc2​054
brea​stca​ncer​.org/symp​toms​/unde​rsta​nd_bc/stat​isti​cs. Accessed 6
23. Collaborative Group on Hormonal Factors in Breast Cancer
Sep 2017
(1997) Breast cancer and hormone replacement therapy: col-
12. National Breast and Ovarian Cancer Centre (2009) Breast cancer
laborative reanalysis of data from 51 epidemiological studies of
risk factors: a review of the evidence. National Breast and Ovarian
52,705 women with breast cancer and 108,411 women without
Cancer Centre, Surry Hills, NSW
breast cancer. Lancet Lond Engl 350:1047–1059
13. Cancer Statistics Review, 1975–2014—SEER Statistics. http​s://
seer​.canc​er.gov/csr/1975​_2014​/. Accessed 22 Oct 2017

13