Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 25
1
Introduction
T
he prospect of developing breast cancer is a source of anxiety for
many women. Breast cancer remains the most common invasive can-
cer among women (aside from nonmelanoma skin cancers), account-
ing in 2011 for an estimated 230,480 new cases among women in the
United States and another 2,140 new cases among men (ACS, 2011). After
lung cancer, it is the second most common cause of mortality from cancer
for women, with about 39,520 deaths expected in the United States in
2011. Another 450 breast cancer deaths are expected among men in 2011
(ACS, 2011). Since the mid-1970s, when the National Cancer Institute
(NCI) began compiling continuous cancer statistics, the annual incidence of
invasive breast cancer rose from 105 cases per 100,000 women to 142 per
100,000 women in 1999 (NCI, 2011). Since then, however, the incidence
has declined. In 2008, the incidence of breast cancer was 129 cases per
100,000 women.
Further reduction of the incidence of breast cancer is a high priority,
but finding ways to achieve this is a challenge. As in most types of adult
cancer, breast cancer is thought to develop as a result of accumulated dam-
age induced by both internal and external triggers resulting in initial carci-
nogenic events. The affected cells and tissues then progress through multiple
stages, with accompanying alterations in the surrounding tissue likely play-
ing a role in whether the damage leads to a cancer. These events contrib-
uting to subsequent cancers may occur spontaneously as a by-product
of errors in normal processes, such as DNA replication, or potentially
through effects of environmental exposures. The early procarcinogenic
events from endogenous and exogenous processes may be sustained and
25
OCR for page 26
26 BREAST CANCER AND THE ENVIRONMENT
furthered by physiologic conditions such as obesity. It is likely that many
such procarcinogenic events may never be entirely preventable because,
although potentially modifiable, they are consequences of basic biologic
processes, such as oxidative damage to DNA from endogenous metabo-
lism, or stimulation of cell growth through normal hormonal processes.1
Although such biological “background” mutagenesis is unavoidable, highly
efficient protective pathways, such as DNA repair and immune surveillance,
are effective at reducing the impacts of procarcinongenic events (Loeb and
Nishimura, 2010; Bissell and Hines, 2011).
Although more needs to be learned about both the mechanisms by
which breast cancers arise and the array of factors that influence risk for
them, much has been established. Among the factors generally accepted as
increasing women’s risk are older age, having a first child at an older age
or never having a child, exposure to ionizing radiation, and use of certain
forms of postmenopausal hormone therapy (HT). Inherited mutations in
the BRCA1 and BRCA2 genes also markedly increase risk for breast can-
cer (and other cancers as well), but these mutations are rare in the general
population and account for only 5 to 10 percent of cases (ACS, 2011).
Even though aging, genetics, and patterns of childbearing account for
some of the risk for breast cancer, they are not promising targets for preven-
tive measures. More helpful would be identifying modifiable risk factors.
For example, the publication of findings from the Women’s Health Initiative
(Writing Group for the Women’s Health Initiative Investigators, 2002) con-
firming earlier indications that estrogen–progestin HT was contributing to
an increase in the risk of postmenopausal breast cancer was followed by a
rapid reduction in use of HT and in the incidence of invasive breast cancer.
As reflected in NCI data, the incidence in 2002 was 136 cases per 100,000
women, compared with 127 in 2003 (NCI, 2011). A portion of the decline
in breast cancer incidence since 1999 is attributed to this reduced use of HT
(e.g., Ravdin et al., 2007; Farhat et al., 2010). But there are long-standing
and still unresolved concerns that aspects of diet, ambient chemicals, or
other potentially modifiable environmental exposures may be contributing
to high rates of breast cancer.
At present, a large but incomplete body of evidence is available on the
relationship between breast cancer and the wide variety of external factors
that can be said to comprise the environment. Information on interactions
between genetic susceptibility and environmental factors is particularly
sparse. In contrast, knowledge of the complexity of breast cancer is grow-
ing, with the characterization of multiple tumor subtypes; the possibility
1 Loeb and Nishimura (2010, p. 4270) note that each normal cell in a person’s body may be
exposed to as many as 50,000 DNA-damaging events each day, and that oxygen free radicals
are a major source of DNA damage.
OCR for page 27
27
INTRODUCTION
that critical events in the origins of breast cancer can occur very early in
life; the variety of pathways through which breast cancer risks may be
shaped; and the potential significance of both the timing of exposures and
the way combinations of factors determine the effect on risks for different
types of breast cancer. This growing knowledge has stimulated a transition
in breast cancer research. The new perspectives on breast cancer highlight
the limitations of the current understanding of the disease, and innovative
ideas are beginning to influence the design and analysis of epidemiologic
studies, experimental studies in animals, and mechanistic studies of breast
cancer biology, all directed toward elucidating how external factors may
influence the etiology of breast cancer.
This report presents the results of a study commissioned to review the
current evidence on environmental risk factors for breast cancer, consider
gene–environment interactions in breast cancer, explore evidence-based
actions that might reduce the risk of breast cancer, and recommend research
in these areas.
STUDY CHARGE AND COMMITTEE ACTIVITIES
This study resulted from a request to the Institute of Medicine (IOM)
by Susan G. Komen for the Cure and its Scientific Advisory Board. Komen
for the Cure funds research on prevention, diagnosis, and treatment of
breast cancer, and also provides educational information and support ser-
vices for the public and health care providers. The Statement of Task for
the IOM study appears in Box 1-1.
The members of the study committee were selected to contribute
expertise in epidemiology, toxicology, risk assessment, biostatistics, molec-
ular carcinogenesis, gene–environment interactions, communication of
health messages, environmental health science, exposure assessment, and
health care. The committee includes a member from the patient advocacy
community.
The committee met in person five times from April 2010 through
February 2011 and conducted additional deliberations by conference call.
During these meetings and calls, the committee reviewed and discussed the
existing research literature on the topics central to its charge and developed
and revised this report. At three of its meetings, the committee held public
sessions during which it heard presentations by researchers, representatives
of advocacy organizations, and members of the public.
The committee also commissioned work on two topics. One project
was a review of data available to assess temporal changes in the potential
for exposure to a selected set of chemicals and other environmental agents.
The agents included in this paper have been discussed in the research lit-
erature and the popular press as possible contributors to increased risk for
OCR for page 28
28 BREAST CANCER AND THE ENVIRONMENT
B OX 1-1
Study Charge
In response to a request from Susan G. Komen for the Cure®, the
Institute of Medicine will assemble a committee to:
1. Review the evidentiary standards for identifying and measuring
cancer risk factors;
2. Review and assess the strength of the science base regarding
the relationship between breast cancer and the environment;
3. Consider the potential interaction between genetic and environ-
mental risk factors;
4. Consider potential evidence-based actions that women could
take to reduce their risk of breast cancer;
5. Review the methodological challenges involved in conducting
research on breast cancer and the environment; and
6. Develop recommendations for future research in this area.
In addition to reviewing the published literature, the committee will
seek input from stakeholders, in part by organizing and conducting a
public workshop to examine issues related to the current status of eviden-
tiary standards and the science base, research methods, and promising
areas of research. The workshop will focus on the challenges involved
in the design, conduct, and interpretation of research on breast cancer
and the environment. The committee will generate a technical report with
conclusions and recommendations, as well as a summary report for the
lay public.
breast cancer. This work served as an information resource for the com-
mittee and helped to identify some data presented in Chapter 4. The other
project resulted in a paper examining temporal changes in the United States
in exposure to ionizing radiation, with a particular focus on exposure from
medical imaging (see Appendix F, available electronically at http://www.nap.
edu/catalog.php?record_id=13263).
APPROACH TO THE STUDY
The committee began its work with recognition of the potentially vast
scope of the study task and the need to develop a perspective and approach
that could lead to a useful and timely report. The committee sought to focus
its attention in areas that it considered to be the most significant and the
most pertinent to the charge placed before it.
For purposes of this report, the committee interpreted “environment”
broadly, to encompass all factors that are not directly inherited through
OCR for page 29
29
INTRODUCTION
DNA. As a result, this definition includes elements that range from the
cellular to the societal: the physiologic and developmental course of an
individual, diet and other ingested substances, physical activity, microbial
agents, physical and chemical agents encountered at home or at work,
medical treatments and interventions, social factors, and cultural practices.
This perspective was a foundation for the committee’s work; application of
it in its broadest sense is something that the committee hopes will expand
the scope of future research. For some readers, this interpretation will differ
from their association of the phrase “environmental risk factors” primar-
ily with pollutants and other products of industrial processes (Baralt and
McCormick, 2010). Furthermore, throughout the report the term “breast
cancer” is used to refer to disease in humans and “mammary cancer” or
“mammary tumor” to refer to disease in animals.
The committee explored the available evidence concerning breast can-
cer risks associated with a varied but limited collection of specific sub-
stances and factors (Chapter 3), and it also reviewed the many challenges
that researchers have had to contend with in studying breast cancer, includ-
ing those pertaining to gene–environment interactions (Chapter 4). But in
its examination of the relation between breast cancer and the environment,
the committee chose to highlight an approach that emphasizes the biologic
mechanisms through which environmental factors may be operating and the
importance of the changing picture over the life course (Chapter 5). This
perspective played a major role in shaping the committee’s conclusions and
recommendations.
A Life Course Perspective
Breast cancer is primarily (but far from exclusively) a disease of adult
women who are approaching or have reached menopause. In 2009, approx-
imately 90 percent of new cases in U.S. women were diagnosed at age 45 or
older (ACS, 2009). But the breast undergoes substantial changes from the
time it begins developing in the fetus through old age, especially in response
to hormonal changes during puberty, pregnancy, lactation, and menopause.
With the timing of these developmental events related to risk for some types
of breast cancer, there has been growing interest in exploring whether the
timing of a variety of environmental exposures also is important in under-
standing what influences breast cancer risks. In Chapter 5, the committee
has sought to link its examination of the mechanisms of carcinogenesis with
a life course perspective on when and how those pathologic pathways may
be particularly relevant in relation to when and how environmental expo-
sures occur. Attention was paid to growing evidence for critical windows
of susceptibility (e.g., periods with rapid cell proliferation or maturation)
OCR for page 30
30 BREAST CANCER AND THE ENVIRONMENT
when specific mechanisms that increase the likelihood of a breast cancer
developing may be more likely to be activated.
Identifying Environmental Risks for Breast Cancer
Trying to determine which environmental exposures may be influenc-
ing rates of breast cancer poses substantial challenges, many of which are
discussed in Chapter 4. Cancer is a complex disease, and its “causes” are
generally harder to trace than the bacteria and viruses that cause infectious
diseases. People who are never exposed to the measles virus will never get
measles. But the impact of removing a particular environmental exposure
associated with breast cancer is less clear because many other factors can
still contribute to the development of breast cancer. The role of underlying
susceptibility from inherited genes appears to involve both rare variants
and common ones, but it is still not well characterized. Moreover, people
are exposed to a complex and changing mix of environmental agents over
the course of a lifetime, so discerning the effects of an individual agent, or
knowing which components of the mixture may influence the development
of disease or how the mixture’s components may interact with each other
or with genes, is not straightforward.
Observational epidemiologic studies are a critical tool for learning
about elevated risks, but they can be difficult to do well. They typically are
the basis for demonstrating correlations between risk factors and outcomes,
but establishing a causal inference is much more difficult. The challenges
in establishing causality in such studies include difficulties with exposure
measurement and accounting for undetected or poorly measured differences
that may exist between the groups designated as exposed and unexposed.
Furthermore, the timing and duration of observational studies may affect
whether sufficient time has elapsed to detect differences in the incidence of
a cancer that may not appear until many years after an exposure. Random-
ized controlled trials, which assign participants to a specific exposure or
a comparison condition, are easier to interpret. However, for ethical and
methodological reasons, such studies are rarely possible, especially when
the goal is to determine whether the exposure is associated with an adverse
event.
Experimental studies in animal models and in vitro systems offer an
important opportunity to study the effects of well-defined exposures and to
explore mechanisms of carcinogenicity in ways that are not possible in epi-
demiologic studies. They can signal potential hazards to human health that
cannot be identified in other ways, but their results have to be interpreted
with an understanding of differences across species and the comparability
of an experimental exposure to the conditions encountered in the human
population.
OCR for page 31
31
INTRODUCTION
Reviewing Evidence on Specific Risk Factors
The literature on risk factors for cancer in general and breast cancer
in particular is large and varied. In the United States, the Environmental
Protection Agency (EPA) and the National Toxicology Program (NTP) in
the National Institute of Environmental Health Sciences have programs
to review the evidence on the carcinogenicity of various substances.2 The
International Agency for Research on Cancer (IARC), which is part of the
World Health Organization, is a focal point for major international col-
laboration in such reviews.3 In addition, a collaborative project between the
World Cancer Research Fund International and the American Institute for
Cancer Research has an ongoing program to review evidence on diet, physi-
cal activity, and cancer (WCRF/AICR, 2007).4 All of these review programs
consider evidence concerning breast cancer (or mammary cancers in animal
studies) when it is available, but it is not their focus. Reviews specifically
concerning breast cancer have also been conducted. These reviews include
one conducted by the California Breast Cancer Research Program (2007)
and a review sponsored by Komen for the Cure and conducted by the Silent
Spring Institute (e.g., Brody et al., 2007; Rudel et al., 2007).
Assembling a comprehensive review of evidence on the relation between
a complete set of environmental factors and breast cancer was not feasible
for this study. Instead, the committee chose to focus on a limited selection
of various types of environmental factors and potential routes of exposure.
These factors are discussed in Chapter 3. The committee’s aim was to
characterize the available evidence and identify where substantial areas of
uncertainty exist.
Observations About Risk
One component of the committee’s task was to comment on actions
that can be taken to reduce the risk of breast cancer. Opportunities for
action are discussed in Chapter 6, but it is important to emphasize from the
outset the challenge of interpreting evidence regarding risk and risk reduc-
tion. The widely quoted estimate that women in the United States have a
1-in-8 chance of being diagnosed with breast cancer during their lifetimes
2Information on the EPA and NTP review programs is available at http://www.epa.gov/
ebtpages/pollcarcinogens.html and http://ntp.niehs.nih.gov/?objectid=7 2016262-BDB7-CEBA-
FA60E922B18C2540.
3 Information on IARC reviews is available at http://www.iarc.fr/ and http://monographs.
iarc.fr/index.php.
4 Information on the review by the World Cancer Research Fund International and the
American Institute for Cancer Research is available at http://www.wcrf.org/cancer_research/
expert_report/index.php.
OCR for page 32
32 BREAST CANCER AND THE ENVIRONMENT
can be restated as approximately a 12 percent lifetime risk of developing
invasive breast cancer (NCI, 2010). The risk can also be presented for
shorter, more comprehensible intervals. For example, among white women
who are 50 years old, 2.4 percent are likely to be diagnosed with invasive
breast cancer over the next 10 years (NCI, 2010). This 10-year risk is 2.2
percent for 50-year-old black women, 2.0 percent for Asian women, and
1.7 percent for Hispanic women. For 70-year-olds, the 10-year risks are 3.9
percent for white women, 3.2 percent for black women, and 2.4 percent for
both Asian and Hispanic women. Estimates for longer follow-up periods
(e.g., 20 or 30 years) will only increase those risks. Within average values
such as these, there are always groups of women whose particular charac-
teristics give them a higher or lower 10-year risk.
These estimates of risk are a critical reference point for understanding
the implications of findings from epidemiologic studies on factors associ-
ated with increased or decreased risk of breast cancer. These findings are
typically reported in terms of relative risk, which reflects a comparison
between the risk in a population exposed to a particular factor and that
in a similar population that is not exposed. Thus, a relative risk of 2.0 (a
doubling of risk) might mean that for women with that risk factor, the
10-year risk of breast cancer is 5 percent rather than 2.5 percent. Similarly,
a relative risk of 0.5 for a protective factor means that women with that
characteristic may have a 10-year risk of 1.3 percent rather than 2.5 per-
cent. These examples are offered to illustrate the scale of the change in risk
implied by typical epidemiologic findings; they are not a formal analysis.
From a public health perspective, another important piece of informa-
tion is the prevalence of the risk factor in the population. Finding that an
environmental factor is associated with a large relative risk may still mean
that it accounts for few cases of disease if the disease or the exposure is
rare in that population. Alternatively, an environmental exposure that is
associated with only a small increase in risk may be contributing to a large
number of cases if the exposure is very common in the population. How-
ever, if the exposure is so common that there is little variability across the
population (virtually everyone is exposed), it can be extremely difficult to
identify the contribution from that exposure.
Virtually all of the epidemiologic evidence regarding breast cancer
risk is drawn from population-level analyses. As a result, the conclusions
reached on the basis of that evidence apply to an exposed population. With
current knowledge, it is not possible to apply those conclusions to predict
which individuals within that population are most likely to develop breast
cancer. Nevertheless, an understanding of population-based estimates of
risk can help people make personal choices that may lead to better health
outcomes.
OCR for page 33
33
INTRODUCTION
TOPICS BEYOND THE SCOPE OF THE STUDY
Several topics were defined as falling beyond the scope of the study.
With the focus on environmental risk factors for breast cancer, the commit-
tee chose to devote little attention to the established associations between
increased risk for breast cancer and reproductive events such as younger
age at menarche, older age at first birth, lack of lactation, and older age at
menopause. The committee also chose not to evaluate the established asso-
ciations between breast cancer risk and higher birth weight and attained
stature. Although some of them might fall under the committee’s very broad
definition of environmental factors, they were not the focus of its review.
Background is provided on many of these other factors in Chapter 2, and
the possibility that some environmental exposures may have an indirect
influence on risk for breast cancer because they may affect the timing of
these reproductive events is discussed in Chapter 5.
The committee also agreed that the nature and effectiveness of breast
cancer screening, diagnosis, and treatment were generally beyond the scope
of the study. It noted but did not analyze the impact of increased mammog-
raphy and changes in screening practices since the 1970s on the observed
incidence of breast cancer. The paper commissioned by the committee on
medical sources of exposure to ionizing radiation took into account the
contribution of mammography. The committee did not examine the appro-
priateness of screening recommendations or practices.
The committee decided as well that its charge called for a focus on
risk for the initial occurrence of breast cancer and not on recurrence or
factors that might be associated with the risk of recurrence. Although envi-
ronmental exposures may well influence the risk of recurrence, that risk is
also influenced by characteristics of tumors at the time of diagnosis and
subsequent treatment and follow-up practices. Consideration of clinical
practice in the treatment of women (and men) with diagnosed breast can-
cers is substantially different from the study’s primary focus on prevention
of breast cancer through improved understanding of and response to envi-
ronmental risks. Similarly, the committee concluded that its charge called
for a focus on the incidence of breast cancer and not mortality. Influences
on breast cancer mortality patterns include factors that affect diagnosis and
treatment that are separate from the effects of environmental exposures on
the incidence of the disease.
The committee did not explicitly assess environmental risk factors for
male breast cancer, beyond the general assumption that some of the risk
factors identified through studies in women may also be relevant to the
development of breast cancer in men.
OCR for page 34
34 BREAST CANCER AND THE ENVIRONMENT
THE COMMITTEE’S REPORT
This report reviews the current evidence on the biology of breast cancer,
examines the challenges of studying environmental risk factors, and pres-
ents the committee’s findings and research recommendations from its review
of evidence on environmental risk factors. Specifically, Chapter 2 provides
important background for evaluating factors influencing breast cancer risk
with a brief review of the biology of breast cancer and trends in incidence
in the United States, along with discussion of the kinds of studies used to
investigate breast cancer and environmental exposures. Chapter 3 presents
the committee’s review of evidence on selected environmental risk factors.
Chapter 4 discusses the variety of challenges that complicate the study of
environmental risk factors for breast cancer, as well as gene–environment
interactions. Chapter 5 examines mechanisms of carcinogenesis and links
them to a life course perspective on breast development and the potential
for environmental factors to influence risk for breast cancer. In Chap-
ter 6, the committee examines opportunities for evidence-based action to
reduce risks for breast cancer and also considers the challenges of avoid-
ing the unintentional introduction of new risks. Chapter 7 concludes the
report with the committee’s recommendations for future research efforts.
Included as appendixes are agendas for the committee’s public sessions
(Appendix A), biographical sketches of committee members (Appendix B),
a summary of weight-of-evidence categories used by major organizations
that evaluate cancer risks (Appendix C), a table summarizing reports of
population attributable risks for breast cancer (Appendix D), a glossary
(Appendix E), and the paper commissioned on exposure to ionizing radia-
tion (Appendix F).
REFERENCES
ACS (American Cancer Society). 2009. Breast cancer facts and figures 2009–2010. Atlanta,
GA: ACS. http://www.cancer.org/Research/CancerFactsFigures/BreastCancerFactsFigures/
index (accessed November 17, 2010).
ACS. 2011. Breast Cancer facts and figures 2011–2012. Atlanta, GA: ACS. http://www.cancer.
org/acs/groups/content/@epidemiologysurveilance/documents/document/acspc-030975.
pdf (accessed November 15, 2011).
Baralt, L. B., and S. McCormick. 2010. A review of advocate–scientist collaboration in fed-
erally funded environmental breast cancer research centers. Environ Health Perspect
118(12):1668–1675.
Bissell, M. J., and W. C. Hines. 2011. Why don’t we get more cancer? A proposed role of the
microenvironment in restraining cancer progression. Nat Med 17(3):320–329.
Brody, J. G., K. B. Moysich, O. Humblet, K. R. Attfield, G. P. Beehler, and R. A. Rudel.
2007. Environmental pollutants and breast cancer: Epidemiologic studies. Cancer 109(12
Suppl):2667–2711.
OCR for page 35
35
INTRODUCTION
California Breast Cancer Research Program. 2007. Identifying gaps in breast cancer research:
Addressing disparities and the roles of the physical and social environment. http://cbcrp.
org/sri/reports/ identifyingGaps/index.php (accessed October 25, 2011).
Farhat, G. N., R. Walker, D. S. Buist, T. Onega, and K. Kerlikowske. 2010. Changes in inva-
sive breast cancer and ductal carcinoma in situ rates in relation to the decline in hormone
therapy use. J Clin Oncol 28(35):5140–5146.
Loeb, L. A., and S. Nishimura. 2010. Princess Takamatsu Symposium on DNA repair and
human cancers. Cancer Res 70(11):4269–4273.
NCI (National Cancer Institute). 2010. SEER cancer statistics review, 1975–2007. Edited
by S. F. Altekruse, C. L. Kosary, M. Krapcho, N. Neyman, R. Aminou, W. Waldron, J.
Ruhl, N. Howlader, Z. Tatalovich, H. Cho, A. Mariotto, M. P. Eisner, D. R. Lewis, K.
Cronin, H. S. Chen, E. J. Feuer, D. G. Stinchcomb, and B. K. Edwards. Bethesda, MD:
NCI. http://seer.cancer.gov/csr/1975_2007/ (accessed January 6, 2011).
NCI. 2011. SEER cancer statistics review, 1975–2008. Edited by N. Howlader, A. M. Noone,
M. Krapcho, N. Neyman, R. Aminou, W. Waldron, S. F. Altekruse, C. L. Kosary, J.
Ruhl, Z. Tatalovich, H. Cho, A. Mariotto, M. P. Eisner, D. R. Lewis, H. S. Chen, E. J.
Feuer, K. A. Cronin, and B. K. Edwards. Bethesda, MD: NCI. (Based on November
2010 SEER data submission, posted to the SEER website, 2011.) http://seer.cancer.gov/
csr/1975_2008/ (accessed June 1, 2011).
Ravdin, P. M., K. A. Cronin, N. Howlader, C. D. Berg, R. T. Chlebowski, E. J. Feuer, B. K.
Edwards, and D. A. Berry. 2007. The decrease in breast-cancer incidence in 2003 in the
United States. N Engl J Med 356(16):1670–1674.
Rudel, R. A., K. R. Attfield, J. N. Schifano, and J. G. Brody. 2007. Chemicals causing mam-
mary gland tumors in animals signal new directions for epidemiology, chemicals testing,
and risk assessment for breast cancer prevention. Cancer 109(12 Suppl):2635–2666.
WCRF/AICR (World Cancer Research Fund/American Institute for Cancer Research). 2007.
Food, nutrition, physical activity, and the prevention of cancer: A global perspective.
Washington, DC: AICR.
Writing Group for the Women’s Health Initiative Investigators. 2002. Risks and benefits of
estrogen plus progestin in healthy postmenopausal women: Principal results from the
Women’s Health Initiative randomized controlled trial. JAMA 288(3):321–333.
OCR for page 36
