SISOMB2009suppstatementArev3

Supporting Statement A
The Sister Study PHASE 2:
Environmental and Genetic Risk Factors for Breast Cancer
(NIH/NIEHS)

Submitted: 20 August 2009

Principal Investigator and Co-Project Officer:
Dale P Sandler PhD
Chief, Epidemiology Branch
National Institute of Environmental Health Sciences
PO Box 12233
Research Triangle Park NC 27709
Phone:
919-541-4668
Fax:
919-541-2511
Email:
sandler@niehs.nih.gov
Co-Principal Investigator:
Clarice Weinberg PhD
Chief, Biostatistics Branch
PO Box 12233
Research Triangle Park NC 27709
Phone:
919-541-4927
Fax:
919-541-4311
Email:
weinberg@niehs.nih.gov
Project Officer:
Paula S Juras PhD
Epidemiology Branch
National Institute of Environmental Health Sciences
PO Box 12233
Research Triangle Park NC 27709
Phone:
919-541-4668
Fax:
919-541-2511
Email:
juras@niehs.nih.gov

Sister Study—OMB 2009: Supporting Statement A

page 1 of 20

Summary
The Sister Study is a prospective cohort study of the environmental and genetic risk factors for breast
cancer and other diseases among 50,000 sisters of women who have had breast cancer. The long-term
design allows us to also study the impact of environmental and genetic factors on survival and outcomes
following a breast cancer diagnosis.
In the United States, approximately 180,000 new cases of breast cancer are diagnosed each year.
Breast cancer accounts for over 30% of all new cancer cases among women and 15% of cancer deaths.
The etiology of breast cancer is complex, with both genetic and environmental factors playing a role.
Currently established breast cancer risk factors account for less than half the variation in breast cancer
risk across the United States, and known breast cancer genes are found in fewer than 10% of breast cancer
patients. Although the concordance rate for monozygotic twins is less than 20% (underscoring the
importance of environmental contributors), sisters of women with breast cancer have, on average, a 2-fold
increase in risk for breast cancer themselves. By focusing on a genetically susceptible group, more precise
estimates of the contribution of environmental and other non-genetic factors to disease risk may be
possible and the power to study gene by environment interactions will be greatly enhanced.
Between August 2003 and August 2009, we enrolled just over 50,000 at-risk volunteers into a
prospective cohort study known as the Sister Study. Sister Study participants were recruited from across
the United States and Puerto Rico. They were between ages 35 and 74 at enrollment and had at least one
full or half sister who was diagnosed with breast cancer. Participants were recruited through a variety of
means including the media, breast cancer advocacy groups, medical practitioners, community partners,
and Sister Study participants themselves. Recruitment strategies were designed to enroll a cohort of
sisters that is ethnically, geographically, and socioeconomically diverse. At enrollment, participants
provided complete histories of personal and family health, reproductive health, diet, and environmental
and lifestyle exposures. They completed a home exam in which height, weight, waist circumference, and
blood pressure were measured by an examiner who also collected a blood sample. Whole blood, serum,

Sister Study—OMB 2009: Supporting Statement A

page 2 of 20

plasma, blood clots, blood spots, and cryopreserved lymphocytes were stored for future use. Participants
also provided a first morning urine sample, toenail clippings, and household dust wipe samples.
Now in Phase 2 of the Sister Study, the enrolled high-risk cohort will be followed actively for ten or
more years for the development of breast cancer and other diseases. We anticipated, on average, 300 new
cases of breast cancer to be diagnosed each year; thus after five years of follow-up, we will have
sufficient power, with about 1,500 new breast cancer cases, to address hypotheses regarding geneenvironment interactions that require large numbers of cases. Baseline data, banked blood, urine, and
toenail samples, as well as banked environmental samples will provide a rich resource for testing current
and future hypotheses regarding risks for breast cancer and a wide range of other medical conditions.
Sister Study participants will be contacted annually to track changes in their medical history. Every
two to three years, participants will be asked to provide more comprehensive health and exposure updates
and to provide additional information on other factors that may affect disease risk/survival such as stress
phenotype or diet. Over time, participants may be asked to provide additional biological or environmental
samples and will be invited to participate in more focused clinical studies of specific outcomes.
Participants who develop breast cancer will be asked to provide information about their diagnosis and
treatment and asked to sign medical release forms allowing us to request medical records, additional
details about diagnosis and treatment, and tumor tissue and/or diagnostic H&E slides from their health
care providers. Similarly, additional information to allow disease confirmation or validation of selfreported diagnoses will be sought from participants who report other diseases of interest such as asthma,
uterine fibroids, diabetes, thyroid disease, osteoporosis, rheumatoid arthritis and other autoimmune
diseases, neurodegenerative diseases, and other cancers. Nested case-control or case-cohort studies will
be carried out among sisters who develop cancer or other clinical outcomes of interest and a sample of
those who do not, to assess specific gene-environment interactions or other hypotheses. Patients who
develop breast cancer during the follow-up period for the Sister Study will continue to be followed to
study the role of environment and genes in survival following breast cancer diagnosis. Add-on studies

Sister Study—OMB 2009: Supporting Statement A

page 3 of 20

may collect serial biological samples from informative subgroups to evaluate preclinical biomarkers and
assess changes in biomarkers over time.
The Sister Study completed PHASE 1 enrollment in August 2009, concluding contact with the
public at large, and as of 31 August 2009 will discontinue the data collection using Sister Study
enrollment questionnaires and materials previously approved by OMB.
This application is to request renewed approval for PHASE 2 long-term clinical follow-up of the
enrolled cohort, including currently approved long-term clinical follow-up activities. Currently,
using the previously OMB-approved forms, 31,137 participants have completed or are in progress for
completing their first Biennial Update. Also 911 participants have reported incident cases of breast
cancer, and are in progress for completing incident breast cancer follow-up activities.

A. Justification
A.1. Circumstances Making the Collection of Information Necessary
The National Institute of Environmental Health Sciences (NIEHS) is responsible for conducting research
on chemical, physical, and biological factors in the environment that affect human health. The Sister
Study, with its focus on potential environmental causes of breast cancer, is supported by the mandate of
NIEHS as defined by US Code Title 42, Chapter 6A, Subchapter III, Part A, Section 281, as amended by
the Health Research Extension Act of 1985, which is “the conduct and support of research, training,
health information dissemination, and other programs with respect to factors in the environment that
affect human health, directly or indirectly.” [Attachment 1 deleted from final package].
Breast cancer is the most common cancer among women in developed countries. In the United States,
192,370 new cases of invasive breast cancer are anticipated in 2009. Many of the best-understood risk
factors are endogenous or are related to lifestyle choices that are not easily modified. Family history of
breast cancer is one of the most well established risk factors for the disease with an approximate 2-fold
increase in risk in first-degree relatives. No known genetic or environmental breast cancer risk factor or

Sister Study—OMB 2009: Supporting Statement A

page 4 of 20

combination of risk factors has high enough relative risks to account for the observed association with
family history.
The etiology of breast cancer is complex, with genetic and environmental factors likely playing a
combined role. Identifying gene-environment or gene-gene interactions will be important in
understanding breast cancer etiology and identifying prevention strategies. A number of genes already
have been identified as candidates for study. Large-scale studies are needed to confirm reported geneenvironment interactions and to test new biologically based hypotheses as both technology and our
understanding of mechanisms improve over time. The Sister Study offers a unique opportunity to gather
important epidemiological data that will make a difference in understanding the multi-factorial etiology of
breast cancer.
There is suggestive evidence, including studies of disease patterns in migrants, to support a role of the
environment in breast cancer risk, although specific environmental factors have not been clearly
elucidated. With the exception of recent studies of organochlorine pesticides and earlier studies of
irradiated populations, little work has been done in the area of non-lifestyle environmental factors. In part,
this may be due to the difficulty of retrospectively measuring most environmental exposures. Few studies
have examined occupational exposures among women, although there is evidence that some may play a
role in breast cancer etiology. The National Toxicology Program has identified at least 35 mammary
carcinogens, many of which have not been evaluated in human studies. Agents on this list include
pesticides and fumigants, chemicals used in manufacturing rubber, vinyl, polyurethane foams, benzenebased dyes, and some pharmaceuticals, as well as solvents.
The Sister Study is designed to investigate the independent and joint effects of genetic susceptibility
and environmental, biological and lifestyle factors on the risk of breast cancer and other diseases in a
cohort of sisters of women with breast cancer. In addition, we will have the opportunity to efficiently
assess risks for other diseases with similar risk factor profiles or which are otherwise important causes of
morbidity in women, such as cardiovascular disease, autoimmune disease, osteoporosis, and diabetes.

Sister Study—OMB 2009: Supporting Statement A

page 5 of 20

By focusing on sisters, the study has several unique advantages. Not only are the sisters of women
with breast cancer at greater risk for breast cancer themselves (nearly a two-fold risk based on other
studies), they will have a higher prevalence of any genes that prove to be related to breast cancer risk and
they should have a higher frequency of some risk factors due to behaviors and experiences they share
with their sisters who have already developed breast cancer. The study is especially well suited to address
hypotheses related to shared childhood and prenatal exposures, in addition to factors later in life. Our
study will have greater statistical power than a similarly sized general population cohort (see Sections B).
Furthermore, based on current experience, sisters of women with breast cancer are highly motivated to
participate in breast cancer research. Participation and retention rates are higher than in other studies and
the quality of data is high. These features will enhance both validity and statistical power to detect
associations. The study is large enough to investigate many gene-environment interactions after five years
of follow-up. Less common genes and rare exposures can be studied with continued follow-up of the
cohort.
The prospective design offers several advantages over the retrospective case-control studies that have
been most widely used to assess gene-environment interactions. The most often cited advantage of the
case-control design is efficiency; in order to accrue the same number of cases in a population-based
cohort, a substantially larger number of women must be studied. However, in retrospective studies,
exposures must be ascertained after the occurrence of disease. Although certain exposures can be studied
retrospectively using questionnaires or, occasionally, using occupational or medical records, numerous
potential biases can limit the ability to make causal inferences. In the Sister Study, we ascertain exposures
and collected biological and environmental samples prior to the onset of disease. By banking samples for
later analysis, we minimize the costs of measuring exposure, since only new cases and a sub-sample of
non-cases will be analyzed.
We enrolled a cohort of 50,000 high-risk women (ages 35 to 74) who are the cancer-free sisters of
women who have had breast cancer. This cohort will be followed annually for at least ten years.
Comprehensive baseline questionnaire data, fasting blood, first morning urine, household dust and toenail

Sister Study—OMB 2009: Supporting Statement A

page 6 of 20

samples were collected. Study questionnaires were designed to collect information on known, suspected,
and speculative risk factors in order to maximize the chance of detecting environmental risk factors of
concern to women. In order to address new hypotheses, it was necessary to also collect comprehensive
data on what is currently known, to be able to account for these factors in other analyses. In contrast to
previous cohort studies where environmental exposures are either ignored or addressed superficially, we
aimed to collect sufficient data so that environmental risks can be conclusively identified, or in the event
that risks are not confirmed, we can say that we have thoroughly investigated the possibility. Brief annual
updates record changes in contact information, environment and health. Bi- or triennial questionnaires
address new hypotheses that arise. Our study will assess hypotheses that derive from the scientific
literature, but we will also collect data on poorly studied exposures of concern to women. The incident
cases that develop over time will be followed for the duration of the study, allowing us to also study the
role of lifestyle, environmental exposures, and genes in prognosis and survival among women with
cancer.
The Sister Study is an NIEHS intramural study designed to allow for trans-NIH and extramural
collaboration. The study was developed in response to the heightened concern expressed by numerous
women’s and environmental groups about the possibility of increased breast cancer risk due to
environmental causes, and the perceived lack of serious attention that had been paid to such concerns. In
addition to consulting with scientific colleagues both within the government and at universities, we
solicited the input of groups such as the National Breast Cancer Coalition and other advocacy and interest
groups and consumers as we planned the Sister Study and assessed its feasibility. Our Steering
Committee, Scientific Advisory Board, and Recruitment and Retention Advisory Board are diverse
groups of professionals in the various areas of this complex effort. They include experts in epidemiology,
breast cancer, biological specimen management, laboratory science, as well as representatives from
various organizations focused on minorities, breast cancer support, and women.

Sister Study—OMB 2009: Supporting Statement A

page 7 of 20

A.2. Purpose and Use of the Information
Information collected in this study will be used to further scientific understanding of the effects of
environmental exposures in women who are genetically at risk for developing breast cancer and to
address questions of concern to women at risk for breast cancer. Epidemiologists and biostatisticians at
NIEHS and their collaborators at other institutions will be responsible for testing the hypotheses of
interest and disseminating results through the scientific literature. Results will be published in medical
and epidemiologic journals as well as basic science journals when appropriate. Results will be presented
at scientific meetings and at meetings of breast cancer advocates and other interested groups. Data will be
used to assess current hypotheses regarding risk factors for breast cancer, to generate new hypotheses for
subsequent analyses in the Sister Study, and specifically to identify preventable risk factors or
combinations of risk factors. In addition to scientists and clinicians who will use this information in
developing prevention strategies and to advise their patients, results will be reported to the women who
participate in the study and to other women through the media, our website, and other Sister Study
publications and newsletters. Results of the study may figure in future risk assessments and evaluations
of the carcinogenicity of specific environmental agents and could be used in the development of exposure
guidelines or standards, should important environmental risks be uncovered. Thus the results will be of
use to Public Health officials, other scientists, physicians, elected and appointed officials, and women and
their families.
The type and amount of information we collected at baseline before women develop breast cancer,
and at subsequent yearly or bi/triennial intervals, fulfill many scientific and clinical needs. For breast
cancer, many of the exposures of interest, including endogenous hormone levels, micronutrients, and even
some environmental exposures, are measured most accurately in biological samples collected before the
onset of disease or treatment and their associated symptoms and biological and lifestyle changes. The
cohort design allows us to collect data on exposures, including biological exposure measures, diet and
lifestyle, before the onset of disease. Forms for the currently approved long-term clinical follow-up data
collection activities are in Attachment 2.1 and 2.2.

Sister Study—OMB 2009: Supporting Statement A

page 8 of 20

Brief self-administered forms are used annually to update changes in contact information and health
status. Bi/triennial updates record changes in health, lifestyle, occupational and environmental exposures
and address new hypotheses. The first Biennial Update focused on stress. As technology for self-collected
environmental samples improves, women may be asked to provide such samples (for example, using inhome water test kits) as part of the bi/triennial follow-up.
Women who develop breast cancer or other conditions of interest during the course of follow-up are
asked to allow us to obtain medical records and tissue samples from their health care providers.
Since the last review, we have completed full enrollment activities (data and sample collection) for
the cohort, having enrolled 50,000 women whose sisters had breast cancer. Participants have completed
Annual and Biennial Updates according to schedule. Average response rates for Update activities remain
>90%. In addition, several reports have been published or submitted, describing factors under
investigation:
Kim S, Parks CG, DeRoo LA, Chen H, Taylor JA, Sandler DP. Obesity and weight gain in adulthood and telomere
length. Cancer Epid Biom Prev 2009;18:816-20.
Parks CG, Miller DB, Andrew ME, Cawthon RM, DeRoo LA, Sandler DP. Telomere length, current perceived
stress, and urinary stress biomarkers in women. Cancer Epidem Biom Prev 2009;18:551-60.
Spector D, Mishel M, Skinner CS, DeRoo LA, VanRipper M, Sandler DP. Breast cancer risk perception and
lifestyle behaviors among white and black women with a family history. Cancer Nursing 2009, in press.
Weinberg CR, Shore DL, Umbach DM, Sandler DP. Using risk-based sampling to enrich cohorts for endpoints,
genes and exposures. Am J Epidemiol, 2007;166:447-455.
Xu Q, Parks CG, DeRoo LA, RM Cawthon, Sandler DP, Chen H. Multivitamin use and telomere length among
women. Am J Clin Nutrition 2009;89:1-7.
Crowder K, Sandler DP, Parks C. Performance of a case classification algorithm for self-reported rheumatoid
arthritis in relation to demographic factors, healthcare utilization, and smoking. Submitted
D’Aloisio AA, Baird DD, DeRoo LA, Sandler DP. Association of intrauterine and early life exposures with uterine
leiomyomata in young women in the Sister Study. Submitted

Sister Study—OMB 2009: Supporting Statement A

page 9 of 20

Kim S, DeRoo LA, Sandler DP. Sociodemographic and lifestyle factors associated with sleep duration. Submitted
Kim S, DeRoo LA, Sandler DP. Eating patterns and nutritional characteristics associated with sleep duration.
Submitted
Parks CG, DeRoo LA, Miller DA, McCanlies EM, Cawthon RM, Sandler DP. Work schedule and telomere length
in The Sister Study. Submitted

A.3. Use of Information Technology and Burden Reduction
Web-based completion of annual updates is available. Bi/triennial follow-up forms may also be made
available on the Web.
Computer Assisted Telephone Interview, or CATI, is a special data collection approach designed to
reduce the burden to respondents and improve quality control. The Sister Study uses this technology to
complete bi/triennial updates. This technology allows several advantages over the traditional pencil and
paper method. First, it requires less paper. Second, there is no “mail wait” to get the information from
participants. Also, the telephone interview requires little reading for the participant (only of help guides
included in the enrollment Kit), an important factor when a segment of the population has low educational
level or poor eyesight. Last, data extraction is more efficient with the CATI system as compared to the
keyed entry method because skip patterns are automated and response inconsistencies can be queried at
the time of the interview.
At enrollment during Phase 1 (now completed) name, address, SSN, date of birth, and medical
information were collected. Personal identifiers are stored encrypted and separately from all other data.
Now in Phase 2, PII are used to address Update materials, and to request medical records from physicians
(with participant authorization). A Privacy Impact Assessment was completed for the Sister Study
information management system.
A.4. Efforts to Identify Duplication and Use of Similar Information
The information we collect is not available from other sources. There is little consensus in the
scientific community on how the environment impacts breast cancer. While some studies have addressed

Sister Study—OMB 2009: Supporting Statement A

page 10 of 20

environmental factors like diet, pesticides, and electromagnetic fields, no conclusive evidence exists
because of limits in sample size and/or study design. Although there are cohort studies, such as Harvard’s
Nurses Health Study, that do address risk factors for breast cancer, none of these cohorts has collected
substantial information on environmental and occupational exposures, and none includes biological and
environmental samples for all of the participants. All of the existing cohorts focus on diet and other
lifestyle factors. Large-scale prospective studies such as the Sister Study are needed to validate some of
the already reported gene-environment interactions and to test new biologically-based hypotheses as both
the technology and our understanding of synergistic mechanisms improve over time.
As noted above, there are important advantages to the Sister Study design. First, these sisters are at
increased risk of breast cancer, likely due to shared genetic and environmental factors. Their family
history increases the expected number of new cancers, the frequency of relevant gene polymorphisms,
and the frequency of relevant exposures, making the study efficient compared to unenriched cohort
designs. Second, because the study is prospective, blood samples and risk factor information were
collected prior to diagnosis. A third advantage is that sisters are highly motivated, which should improve
data quality and completeness and reduce loss to follow-up.
We are unaware of any duplication of this project with any other project now underway at other
organizations. Several prospective cohort studies--for example, The Nurses’ Health Study, Canadian
National Breast Screening Study, New York University Women’s Health Study, Iowa Women’s Health
Study—have investigated breast cancer in women, but none have focused primarily on the geneenvironment link, especially in terms of the broader external environment.
A.5. Impact on Small Businesses or Other Small Entities
We are asking physicians of women who develop breast cancer or certain other conditions to provide
confirmation of the diagnosis (with written permission from the women). We are minimizing the time
and effort this will require by asking only for answers to key questions about procedures and findings
and/or simply sending copies of the medical record. We are also requesting samples of tissue from stored
diagnostic pathology specimens. In order to standardize the process and reduce the burden, we ask

Sister Study—OMB 2009: Supporting Statement A

page 11 of 20

physicians to mail the tissue samples to us, and we offer to store the samples, to be returned to the
physician on request.
Although the number of women in the study is large, only a fraction (approximately 300 per year)
will develop breast cancer, and the diagnoses will be spread over many years. Since the study is
nationwide, it is not likely that any one medical practice will be contacted more than once or twice. The
chances of being the physician of a woman in our study is greater for doctors whose practice is located in
large cities and/or is affiliated with major medical centers. We estimate that the information we are
requesting will take only 15 minutes to provide and in many cases can easily be provided by assistants in
the physician’s office.
A.6. Consequences of Collecting the Information Less Frequently
Annual updates take ~10 minutes, and bi/triennial self-administered questionnaires are 75 minutes or
less. Annual contact cannot be done less frequently because the analysis relies on exposure and healthstatus changes over time, and ascertaining cases close to the time of diagnosis is important. A
participant’s recall diminishes greatly with time, and death may occur. Annual contact is necessary to
preserve reliability and completeness and will facilitate maintenance of the cohort and tracing of those
few who are lost to follow-up.
A.7. Special Circumstances Relating to the Guidelines of 5 CFR 1320.5
There are no special circumstances relating to the guidelines of 5 CFR 1320.5 and the project fully
complies.
A.8. Comments in Response to the Federal Register Notice and Efforts to Consult Outside Agency
A 60-Day Federal Register Notice was published 10 July 2009 on page 33259. There were no public
comments.
Early discussions were held with special interest groups to gain information on to what extent women
who are at higher risk of developing breast cancer were willing to participate in a new longitudinal study
requiring a long-term commitment and intense initial data collection. These meetings involved advocacy
groups including cancer-free sisters, breast cancer survivors, Latina women, African-American women,

Sister Study—OMB 2009: Supporting Statement A

page 12 of 20

lesbian women, and other minorites. Support for this study was overwhelmingly enthusiastic and the
majority of comments were favorable. These discussions gave us specific and valuable feedback that was
incorporated into our study protocol.
Efforts to consult both within and outside NIEHS are summarized in Attachment 3. Also listed in
Attachment 3 are the current members of the Sister Study Scientific Advisory Board. Numerous
researchers both within the NIH and in the extramural community were consulted during the planning of
this study. Among the scores of scientists who were consulted during the early planning for the Study
were, from the Division of Cancer Etiology and Genetics at the NCI: Drs. Patricia Hartge, Sholom
Wacholder, Shelia Zahm, Aaron Blair, Jeffrey Streuwing, Louise Brinton, Rashmi Sinha, and Patricia
Stewart; from the CDC: Drs. Thomas Sinks, Michelle Marcus (currently at Emory University), Heidi
Blank, and Elizabeth Whelan (NIOSH); and from the National Institute of Aging, Dr. Tamara Harris.
Other outside consultants include Drs. Pamela Horn Ross (Northern California Cancer Center -dietary
phytoestrogens), Gladys Block (NCI - food frequency questionnaire), Julia Brody (Silent Spring Institute
– exposure assessment) and Patricia Moorman (Duke University - medications and hormone
replacement).
Our study was formally reviewed within the NIEHS intramural community, NIEHS advisory boards,
and externally by federal and academic experts. Many continue to serve on the Sister Study Scientific
Advisory and/or Recruitment and Retention Advisory Boards. As the study progresses, the archived data
and samples will facilitate collaborations both within the NIH and in the extramural community.
Advisory to the study investigators in regular monthly meetings are the Sister Study Steering
Committee, composed of NIEHS research staff Honglei Chen PhD (919-541-3782), Lisa DeRoo PhD
(919-541-0799), Jane Hoppin ScD (919-541-7622), Stephanie London MD DrPH (919-541-5772),
Christine Parks PhD (919-541-2577), and Jack Taylor MD PhD (919-541-4631), together with SSS senior
staff, Polly Armsby, Deborah Bittner, Cynthia Kleeberger.

Sister Study—OMB 2009: Supporting Statement A

page 13 of 20

A.9. Explanation of Any Payment or Gift to Respondents
During PHASE 1, participants received a prepaid 120-minute phone card valued at less than $5 as an
incentive to complete the interviews and specimen collection at enrollment. While the <$5 value was not
equivalent to the value of the effort required of participants, it helped to convey our appreciation of that
effort.
Other tokens such as newsletters, magnets and bookmarks with the study logo are provided
periodically to enhance participation and retention. Increasing response rates and retention will improve
the quality of the scientific data we are collecting by minimizing response bias. Monetary and other
incentives have been shown to significantly increase response rates.
A.10. Assurance of Confidentiality Provided to Respondents
Procedures to protect the confidentiality of the study population and the data collected include the
following:
•

The data constitute a system of records under the Privacy Act System (#09-25-0134). A copy of
the Federal Register Notice of System of Records of December 29, 1993 is included in
Attachment 4.

•

Each participant was assigned a study ID number. The ID alone is used to identify biological
samples and all data forms. Only the ID number is entered into the database and used in the
analysis of data. Subjects’ names and addresses are stored separately.

•

Any information linking subject’s ID number to subject’s name are kept in locked physical files
or password-protected, restricted access electronic files at the North Carolina office of Social and
Scientific Systems, Inc. (SSS), the NIEHS Epidemiology Branch Support Services Contractor.

•

Employees of SSS undergo background checks, ethics training and sign a Pledge of
Confidentiality (see Attachment 5).

•

Only Sister Study research personnel have access to study data.

•

Study results will be published in summary form only — no individual results will be published
or shared.

•

Shared samples and data will be provided without identifiers and study ID numbers will be
scrambled to prevent accidental identification of participants.

•

A Certificate of Confidentiality has been secured for this study (see Attachment 6).

Sister Study—OMB 2009: Supporting Statement A

page 14 of 20

•

The proposal was initially reviewed by the Institutional Review Board of NIEHS on March 14,
2002 (see Attachment 7). Also attached is an informal communication of approval for the most
recent IRB Continuing Review (13 August 2009; Attachment 8; formal documentation not yet
received) and Informed Consent form (completed in PHASE 1; Attachment 9).

•

Informed consent forms spelled out the steps taken to protect privacy. Similar information was
provided verbally at the time of enrollment, again prior to the CATI interview, and on the
website.

The biological and environmental samples collected will be stored indefinitely in a secure building
for future testing and may be disposed of at any time at the Investigator’s discretion. Specimens are
labeled with ID number only. These and related issues were explained in the Informed Consent
documents. Specimens shared with outside researchers will be assigned a new identification number; the
code linking the new and the old identification number will be known only to the NIEHS contractor
responsible for the Sister Study field work. This new identification number will not be linked to any
identifying information. Identifying information, such as name, social security number, or address will
not be shared with other researchers. Samples will only be shared for scientifically valid studies that meet
approved scientific and ethical standards. Samples and data that are shared can be used only for the
specific research described in an approved research proposal and may not be used for other purposes
without approval from the Sister Study investigators.
Participants may elect to leave the study at any time. As explained in the Informed Consent
documents, no new data will be collected from patients who elect to drop out, but the data already
provided will continue to be used in some analyses unless a written request to destroy data and specimens
is received. Screening data on women who are found to be ineligible were not retained by the Sister
Study.
A.11. Justification for Sensitive Questions
Breast cancer is a complex disease likely caused by many factors. The currently approved first
Biennial Update includes a form that focuses on psychological stress and related measures. Many of the

Sister Study—OMB 2009: Supporting Statement A

page 15 of 20

questions address sensitive and personal issues such as personal experience with violence, physical and
psychological abuse, and discrimination, and stress related to the sister’s breast cancer diagnosis. In order
to carry out a comprehensive analysis of stress, other measures focus on social support, personality traits
and depression. The justification for each of these scales and their source and derivation is fully
described in Attachment 10. In contrast to other updates, these forms are self-administered; no telephone
option is available except in the rare case when a participant has difficulty reading or completing forms.
As described in Attachment 10, stress may play an important direct or indirect role in risk for breast
cancer and other health outcomes. It is likely however, that the relationship is complex and a
comprehensive approach, including assessing acute and chronic stress at different time periods and
accounting for factors that modify response to stress, is required. In addition to compelling scientific
evidence, we learned through our early focus groups that many women believe that stress plays a role in
risk for breast cancer and other diseases, and we have been urged, by the various constituencies that have
endorsed the Sister Study, to study this topic of high interest to women.
Information is collected directly from participants. Participation is voluntary, and respondents can
withdraw from the study at any time. Participants may refuse to answer specific individual questions,
including those they find to be too sensitive or personal. All information is kept confidential to the extent
provided by law. At no time will any individualized genetic results be given out. We have a Certificate of
Confidentiality in place for this study. Participant informed consent forms are attached in Attachment 9.
A.12. Estimates of Hour Burden Including Annualized Hourly Costs
For the remainder of the study, women will be contacted once each year to update contact information
and health status (5-10 minutes per response); and asked to complete short (60-75 minutes, total) updates
every two-to-three years. Women diagnosed with breast cancer or other health outcomes of interest are
asked to provide additional information about their diagnosis (20 minutes per response) and their doctors
will be contacted to provide medical records related to diagnosis and treatments (15 minutes per
response). The annual reporting burden is as follows: Estimated Number of Respondents: 50,000 study

Sister Study—OMB 2009: Supporting Statement A

page 16 of 20

participants and 2100 medical office staff. Estimated Number of Responses per Respondent: See table
below:
Activity (3-yrs) Estimated

Estimated

Average Burden Estimated Total

Number of
Respondents

Responses per
Respondent

Annual Updates

50,000

2

0.085

8,500

Bi/Trienniel Follow-Up

50,000

1

1.25

62,500

1800

1

0.33

594

300

1

0.33

99

2100

1

0.25

525

Incident BC Case Follow-Up
Incident Other Case Follow-Up
Incident Case Medical Office Contact

Hours per
Response

Burden Hours
Requested

TOTAL

72,218

Average Burden Hours Per Response: 0.7 hour; and Estimated Total Burden Hours Requested: 72,218
(over 3 years). The average annual burden hours requested is 24,073. The annualized cost to
respondents is estimated at $14 (assuming $20 hourly wage X 0.7 hour). There are no Capital Costs to
report. There are no Operating or Maintenance Costs to report.
A.13. Estimate of Other Total Annual Cost Burden to Respondents or Recordkeepers
There is no other total annual cost burden to respondents or recordkeepers.
A.14. Annualized Cost to the Federal Government
The estimated cost of contracting out information collection is $95,159,703. Federal Salary Support is
$300,000 per year. Cost of archiving samples is about $350,000 per year, including purchase of freezers.
Therefore, total cost is $104,259,703 over 14 years, with an annualized cost to the Federal Government of
$7,447,122.

Sister Study—OMB 2009: Supporting Statement A

page 17 of 20

Budget Category
Labor
Federal Salary Support

Total Cost

1,839,982

4,200,000

300,000

743,831

53,131

23,411,052

1,672,218

1,322,739

94,481

2,773,604

198,115

4,950,037

353,574

5,325,269

380,376

37,846

2,703

728,750

52,054

1,762,628

125,902

8,460,263

604,305

31,500

2,250

4,900,000

350,000

19,852,440

1,418,031

$104,259,703

7,447,122

Equipment
Subcontracts
Telephone
Printing
Supplies
Postage
Local Travel
Clinics/Hospital
Respondent Incentives
Other (recruiting, travel, consultants, rent)
Honoraria
Sample Storage
Indirect Costs and Fees

Total Cost to Federal Government

Average Annual Cost

25,759,744

A.15. Explanation for Program Changes or Adjustments
Changes in burden and annualized cost merely reflect the anticipated progress and the next scheduled
phases of the longitudinal study. With enrollment now complete, total burden request decreases from
194,131 hours over 3 years (in 2006 revision) to 72,218 hours over 3 years (for 2009 revision). This does
not however represent a change in Protocol — merely completion of the Enrollment Phase of the
Protocol, and normal as-planned progression of follow-up.
A.16. Plans for Tabulation and Publication and Project Time Schedule
The primary goal of the study is to identify environmental and familial risk factors for breast cancer
and other diseases by studying a cohort of sisters of women who have had breast cancer. The Sister Study
is not designed around one particular a priori hypothesis, but is designed to allow us to address a number
of hypotheses regarding gene-environment interactions and risk for breast cancer. Current hypotheses
regarding environment-gene interactions will be addressed in the early years of the study. The reports

Sister Study—OMB 2009: Supporting Statement A

page 18 of 20

generated from this study will include the following risk factors that may be influenced by the action of
genes with known polymorphisms:
A.16-1
Factor

Genetic Marker of Interest

Cigarette smoke

CYP1A2, NAT2, GSTM1, CYP1A1, DNA repair polymorphisms,
CYP2A6, CYP2C9

Exogenous hormones

CYP17, CYP1A2, CYP1A1, estrogen receptor polymorphisms

Hormonal risk factors

CYP17, aromatase, hormone receptor polymorphisms

Oxidative stress

Genes involved in oxidative stress

Melatonin

Polymorphism screening

Nutritional and dietary factors

Vitamin D metabolism and receptor polymorphisms, CYP1A2, NAT2

Grilled meat consumption

GSTM1

Alcohol

ADH, ALDH, CYP2E1

Sunlight exposure

Vitamin D metabolism and polymorphisms

Calcium, calcium channel blockers

Vitamin D metabolism receptor, estrogen metabolism and receptor
polymorphisms

Clearly, by the time enough cases have accrued, scientific understanding of biologic mechanisms and
genes will have advanced considerably. A prospective study such as the Sister Study is designed to
respond to new hypotheses as they emerge.
A.16 - 2
Project Time Schedule
Activity

Time Schedule

Began vanguard phase of enrollment

August 2003

Began nationwide enrollment

October 2004

Completed PHASE 1 Enrollment questionnaires and
collect specimens on 50,000 women

August 2009

Began first Annual Update (self-administered
questionnaire)

June 2005

Began first Bi/Triennial Update

March 2008

Analyses

Began mid-2007

Publication

Began 2007

Sister Study—OMB 2009: Supporting Statement A

page 19 of 20

A.17. Reason(s) Display of OMB Expiration Date is Inappropriate
We request continued approval to display the OMB control number without the expiration date on
printed forms. This request is based on the precedent (see OMB No. 0935-0104: Medical Expenditure
Panel Survey: Survey About Your Diabetes Care) that this is a longitudinal study scheduled to last for 14
or more years, and for which individual enrollment and follow-up activities span across a number of
years, thus across OMB expiration dates. Annual update forms as well as two entire parts of the 3-part
bi/triennial update will be used throughout the length of the study. Therefore, the cost of multiple printing
cycles, merely to change OMB date, rather than taking advantage of cost savings realized with larger
batch printing of approved materials (hundreds of thousands of pages) that undergo little or no contextual
change is inordinately costly to the government. Nonetheless, these items would continue to be included
in each 3-year revision package sent to OMB for continuation of approval.
A.18. Exceptions to Certification for Paperwork Reduction Act Submissions
There are no exceptions to certification for this submission.

Sister Study—OMB 2009: Supporting Statement A

page 20 of 20