Diet

HHS Public Access
Author manuscript
Cancer Epidemiol Biomarkers Prev. Author manuscript; available in PMC 2023 November
Author Manuscript
01.
Published in final edited form as:
Cancer Epidemiol Biomarkers Prev. 2023 May 01; 32(5): 678–686. doi:10.1158/1055-9965.EPI-22-1198.
Diet Quality and All-Cause Mortality in Women with Breast

Cancer from the Breast Cancer Family Registry
Danielle E. Haslam1, Esther M. John, PhD2,3, Julia A. Knight4,5, Zhongyu Li1, Saundra S.
Buys6, Irene L. Andrulis4,7, Mary B. Daly8, Jeanine M. Genkinger9,10, Mary Beth Terry9,10,
Fang Fang Zhang1
1Friedman School of Nutrition Science and Policy, Tufts University, Boston, Massachusetts
Author Manuscript
2Departments of Epidemiology & Population Health and Medicine (Oncology), Stanford University
of School of Medicine, Stanford, California
3Stanford Cancer Institute, Stanford University School of Medicine, Stanford, California
4Lunenfeld-Tanenbaum Research Institute, Sinai Health System, Toronto, Ontario, Canada
5Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada
6Huntsman Cancer Institute at the University of Utah Health Sciences Center
7Deparment of Molecular Genetics, University of Toronto, Toronto, Ontario, Canada
8Clinical Genetics, Fox Chase Cancer Center, Philadelphia, Pennsylvania, Salt Lake City, Utah
9Mailman School of Public Health, Columbia University
Author Manuscript
10Herbert Irving Comprehensive Cancer Center, New York City, New York
Abstract
Background: The impact of diet on breast cancer survival remains inconclusive. We assessed
associations of all-cause mortality with adherence to the four diet quality indices: Healthy Eating
Index-2015 (HEI-2015), Alternative Healthy Eating Index (AHEI), Alternative Mediterranean Diet
(aMED), and Dietary Approaches to Stop Hypertension (DASH).
Methods: Dietary intake data were evaluated for 6,157 North American women enrolled in the
Breast Cancer Family Registry who had been diagnosed with invasive breast cancer from 1993
to 2011 and were followed through 2018. Pre-diagnosis (n=4,557) or post-diagnosis (n=1,600)
dietary intake was estimated through a food frequency questionnaire. During a median follow-up
Author Manuscript
time of 11.3 years, 1,265 deaths occurred. Cox proportional hazards models were used to estimate
multivariable-adjusted hazard ratios (HR) and 95% confidence intervals (CI).
Corresponding Author: Fang Fang Zhang, MD, PhD, Friedman School of Nutrition Science and Policy, Tufts University, 150
Harrison Avenue, Boston MA 02111, Fax: (617) 636-3727; fang_fang.zhang@tufts.edu.
Author Contributions: FFZ, DEH, and EMJ conceptualized and designed the current study; EMJ, JAK, ILA, SSB, MBD, and
MBT designed the parent study and collected the data; JMG provided the dataset for the current analysis; DEH performed data
analysis and drafted the initial manuscript; ZL performed additional analysis during manuscript revision. All authors reviewed and
revised the manuscript, and approved the final manuscript as submitted.
Conflict of Interest Disclosures: None to disclose.
Haslam et al. Page 2
Results: Women in the highest vs. lowest quartile of adherence to the HEI-2015, AHEI, aMED,
Author Manuscript
and DASH indices had a lower risk of all-cause mortality. HR (95% CI) were 0.88 (0.74–1.04,
Ptrend=0.12) for HEI-2015; 0.82 (0.69–0.97, Ptrend=0.02) for AHEI; 0.73 (0.59–0.92, Ptrend=0.02)
for aMED; and 0.78 (0.65–0.94, Ptrend=0.006) for DASH. In subgroup analyses, the associations
with higher adherence to the four indices were similar for pre- or post-diagnosis dietary intake and
were confined to women with a body mass index <25 kg/m2 and women with hormone receptor
positive tumors.
Conclusions: Higher adherence to the HEI-2015, AHEI, aMED, and DASH indices was
associated with lower mortality among women with breast cancer.
Impact: Adherence to a healthy diet may improve survival of women with breast cancer.
Keywords
Author Manuscript
breast cancer; breast cancer survivors; diet quality; epidemiology; mortality
Introduction
An estimated 3.8 million women are breast cancer survivors in the U.S. as of 2019 (1).
Despite improved survival, breast cancer survivors are at risk for excess morbidity due
to obesity and cardiometabolic conditions, cancer recurrence, and premature death (2,3).
Maintaining a healthy diet and lifestyle may present opportunities to decrease these risks and
improve survival.
Evidence is still limited on whether adherence to a particular recommended diet or dietary

pattern can improve survival in women with breast cancer (4–9). Previous studies that
focused on individual foods and nutrients did not take into consideration the inherent
Author Manuscript
interactions between foods and nutrients consumed (4,10). A person’s overall diet quality or
dietary intake pattern, on the other hand, better captures the overall effects of its constituent
foods and nutrients (11,12). A healthy intake pattern is the core recommendation of the
current Dietary Guidelines for Americans (2020–2025) (12). Therefore, assessing adherence
to a specific recommended diet or dietary pattern using an a priori diet quality index, as
described in the Dietary Patterns Methods Project (13), can lead to a better understanding of
the impact of a person’s overall diet on survival for women with breast cancer.
Diet quality indices use evidence of proposed benefits for foods or nutrients, categorized
a priori as “healthy” or “unhealthy,” to rate the quality of a person’s diet or adherence
to recommended dietary patterns. Here, adherence to recommended dietary patterns was
evaluated using four diet quality indices, namely the Healthy Eating Index (HEI-2015)
Author Manuscript
(14), the Alternative Healthy Eating Index (AHEI) (15), the alternative Mediterranean Diet
(aMED) score (16,17), and the Dietary Approaches to Stop Hypertension (DASH) score
(18). These indices were chosen based on findings from the Dietary Patterns Methods
Project initiated by the National Cancer Institute (19). We examined associations between
the four diet quality indices and all-cause mortality in a large sample of women with breast
cancer. We also evaluated whether the association between adherence to these diet quality
indices and all-cause mortality differed by pre- vs. post-diagnosis diet, menopausal status,
Cancer Epidemiol Biomarkers Prev. Author manuscript; available in PMC 2023 November 01.
tumor hormone receptor status, hormonal cancer treatment, race and ethnicity, physical
Author Manuscript
activity, and body mass index (BMI).
Materials and Methods

Study Sample
The Breast Cancer Family Registry (BCFR) was established in 1995 and recruited and
prospectively follows over 15,000 families with a history of breast cancer from the
U.S., Canada, and Australia. The study’s design, implementation, and follow-up have
been described in detail previously (20). Because the Australian site used a different
diet questionnaire, the current analysis was limited to women enrolled at the five North
American BCFR sites. Population-based families were recruited through incident breast
cancer cases (i.e., probands) ascertained by regional cancer registries in Northern California
and the province of Ontario, Canada, based on site-specific criteria (sex, race and ethnicity,
Author Manuscript
age at diagnosis, and cancer family history). Clinic-based families were recruited through
affected or unaffected probands identified through cancer clinics and community outreach in
New York City, Philadelphia, Utah, and Ontario. At enrollment, structured questionnaires
were administered to collect information about personal and family medical history,
demographics, lifestyle, and diet, and treatment for those with a personal history of breast
cancer. Information on clinical and tumor characteristics was obtained from cancer registries
and pathology reports. The study was approved by the Institutional Review Board at each
participating site. All participants provided written informed consent.
The current analysis was limited to women diagnosed with a first primary, invasive breast
cancer diagnosed between 1993 and 2011 who completed a Food Frequency Questionnaire
(FFQ) (n=6,861). Women who were not followed for a full year (n=237) or died within the
Author Manuscript
first year after the baseline questionnaire was completed (n=167) were excluded to minimize
the impact of reverse causation.
Death Ascertainment
Deaths up to 2018 were identified through regular follow-up with participating family
members via telephone or mailed questionnaires, and linkage to cancer and death registries.
Information on underlying cause of death was available only for participants from the
Northern California site. The International Classification of Diseases, Tenth Revision
(ICD-10) was used to classify deaths from breast cancer (C509).
Dietary Assessment
The FFQ developed for the Multiethnic Cohort (MEC) was used to collect information
Author Manuscript
about usual dietary intake of BCFR participants. The 108-item FFQ assessed frequency of
consumption (never or hardly ever, once a month, 2–3 times a month, once a week, 2–3
times a week, 4–6 times a week, once a day, and ≥2 times a day) and usual serving size
(3 typical serving sizes). The FFQ was validated with repeated 24-hour diet recalls for a
racially and ethnically diverse population in the MEC (21). For probands from the two
population-based sites, the FFQ asked about dietary intake during the year before diagnosis.
For all other participants, the FFQ asked about dietary intake during the year before FFQ
Author Manuscript
completion.
Of the 6,457 women who completed the FFQ, 300 women were excluded including 37
probands from the population-based sites who completed the FFQ more than 5 years after
their diagnosis; 35 women who were diagnosed with breast cancer more than 5 years after
they completed the FFQ; 54 women with unreliable dietary data defined as total energy
intake exceeding 3 standard deviations above or below the mean value of the natural log-
transformed energy intake in the study sample; and 174 women with insufficient dietary data
to calculate scores for the dietary indices (Figure 1). Thus, the study sample included 6,157
women. Of these, 4,461 were probands from the two population-based sites who reported
their pre-diagnosis dietary intake, and 96 women who were unaffected at enrollment and
developed breast cancer within 5 years of completing the FFQ. Together, they comprise
4,557 women with FFQ data that reflect dietary intake before their breast cancer diagnosis.
Author Manuscript
An additional 1,600 women had a prior diagnosis of breast cancer when they enrolled in the
BCFR, and they reported their dietary intake for the year prior to enrollment. Thus, their
FFQ data reflect dietary intake after their breast cancer diagnosis.
Diet Quality Indices

The MyPyramid Equivalents Database, 2.0 (MPED 2.0) was used to estimate intake of major
food groupings based on dietary data collected using the FFQ (22,23). Each individual’s
intake was scored using four diet quality indices (Supplemental Table S1): the Healthy
Eating Index-2015 (HEI-2015) (14), the Alternative Healthy Eating Index (AHEI) (15),
the alternative Mediterranean Diet (aMED) score (16,17), and Dietary Approaches to Stop
Hypertension (DASH) score (13,18,19). The HEI-2015 reflects a diet that abides by the
2015–2020 Dietary Guidelines for Americans (range of scores: 0–100) (14). The AHEI
Author Manuscript
measures adherence to a diet associated with reduced risk of cardiovascular diseases

and mortality (range: 0–110) (15). The aMED score captures the traditional diet in the
Mediterranean region characterized by a high intake of olive oil, legumes, nuts, fruits,
and vegetables, and a low intake of saturated fats, meat, and poultry (range: 0–9) (16,17).
The DASH score is based on a diet that has been shown to decrease blood pressure in
randomized-controlled trials (range: 8–40) (18,24,25). For all four indices, a higher score
corresponds to better adherence to that particular index. The AHEI and DASH indices used
in this analysis are modified versions of the original indices: the AHEI did not include trans
fat and the DASH used dairy instead of low-fat dairy as a component due to lack of intake
data on trans fat and low-fat dairy. Thus, the AHEI had a modified scoring range in this
study (AHEI range: 0–100).
Author Manuscript
Statistical Analysis
Multivariable Cox proportional hazards regression models with robust variance estimates
were used to evaluate the association of adherence to the four diet quality indices with
all-cause mortality. Total scores for each index were categorized into quartiles, with the
lowest quartile (worst adherence) as the reference group. For pre-diagnostic diet, days
since diagnosis was used at the time scale. For post-diagnostic diet, delayed entry models
were used with the follow-up time being calculated from the date of completion of the
dietary assessment. Cases were censored at the date of death or date of last contact.
Author Manuscript
Two multivariable models were constructed: model 1 was adjusted for age at diagnosis,
study site, and total energy intake, and model 2 was additionally adjusted for covariates
that are associated with both diet quality and all-mortality among breast cancer survivors
based on the literature (7,26,27) and also changed the parameter estimates by more
than 10%, including self-reported race and ethnicity (non-Hispanic White, non-Hispanic
Black, Hispanic, and Asian/Pacific Islander, other), education (high school or less, some
college or bachelor’s degree, or graduate degree), menopausal status (premenopausal,
postmenopausal), ER status (positive, negative, unknown/missing), PR status (positive,
negative, unknown/missing), tumor stage (I, II, III, IV), self-reported treatment type [surgery
(yes/no), radiation (yes/no), chemotherapy (yes/no), hormone therapy (yes/no), and other
treatment (yes/no)], and lifestyle factors including recent moderate-to-vigorous physical
activity in the three years prior to enrollment (meets recommendations: ≥150 minutes/week
Author Manuscript
vs. does not meet recommendations: <150 minutes/week), and cigarette smoking status at
enrollment (never, ever), pack-years of cigarette smoking (continuous). For women who
reported pre-diagnosis dietary intake, the lifestyle factors pertained to the year prior to breast
cancer diagnosis; for women who reported post-diagnosis dietary intake, the lifestyle factors
pertained to those in the year prior to enrollment (hereafter referred to as ‘at enrollment’).
For women with missing physical activity data (n=243), the missing values were replaced by
the median minutes/week of moderate-to-vigorous physical activity among women who had
data on physical activity. Women missing data on any other covariate (n=187) were excluded
from multivariable Model 2. BMI was not included as a covariate because it may act as a
potential mediator of the association between diet quality and survival.
To evaluate potential effect modification, stratified analyses were conducted by BMI

categories. BMI was calculated by dividing self-reported weight (kg) by the square of
Author Manuscript
self-reported height (m) at enrollment, and categorized as having a healthy weight: 18.5-<25
kg/m2, overweight: 25-<30 kg/m2, or obesity: ≥30 kg/m2. Underweight women (BMI <18.5
kg/m2 were excluded from stratified analyses due to small sample size (n=143)]. Additional
stratified analyses were conducted by physical activity (meets recommendations vs. does not
meet recommendations), joint tumor estrogen receptor (ER) and progesterone receptor (PR)
status [hormone receptor positive (ER+ or PR+) tumors vs. hormone receptor negative (ER-
and PR-) tumors], receipt of hormonal breast cancer treatment (yes/no), menopausal status at
enrollment (pre- vs. post-menopausal), and whether pre- or post-diagnosis diet was assessed.
Women who still had periods at enrollment, were perimenopausal, pregnant or breastfeeding
were categorized as pre-menopausal. Women who started using menopausal hormone
therapy before they stopped menstruating were classified as unknown menopausal status
unless their age was ≥55 years, in which case they were categorized as post-menopausal.
Author Manuscript
Women whose periods had stopped naturally or medically due to bilateral oophorectomy,
hysterectomy, radiation, chemotherapy or other reasons or were aged 55 years or older were
categorized as post-menopausal. Statistical interaction was tested by comparing the log-
likelihood statistics of models that included interaction terms and models without interaction
terms. Multilevel model analysis was not conducted as the main analysis because only 10%
of the women had familial relation to at least one other woman in the sample. We conducted
a sensitivity analysis by restricting the analytic sample to unrelated women (n=5,553).
All analyses were conducted using SAS version 9.4 (SAS Institute, Inc., Cary, NC, USA).
Author Manuscript
All reported p-values were two-sided, and results were considered statistically significant at
p < 0.05.
Data Availability
The data that support the findings of this study are available from the corresponding author
upon reasonable request and with appropriate IRB approval.
Results
The analysis was based on 6,157 women diagnosed with a first primary, invasive breast
cancer who completed the FFQ. The mean ± standard deviation (SD) age at enrollment
was 53 ± 24 years (Table 1). Nearly half of the women were from racially and ethnically
minoritized populations (19% Hispanics, 13% non-Hispanic Blacks, and 12% Asians/Pacific
Author Manuscript
Islanders) and 42% were post-menopausal at enrollment. For all four indices, women in
the highest quartile (Q4) were more likely to be post-menopausal, have higher levels of
education and lower BMI, be physically active, and were less likely to have obesity or
a history of cigarette smoking, and smoked fewer pack-years. Total energy intake was
higher among women in the highest quartile of the AHEI, aMED, and DASH indices, but
energy intake was lower among women in the highest quartile of the HEI-2015 due to
energy-adjusted intake categories. The mean ± SD diet quality scores were 70.2 ± 9.4 for
HEI-2015, 48.2 ± 9.0 for AHEI, 4.0 ± 1.8 for aMED, and 24.0 ± 4.3 for DASH. All indices
showed a moderate-to-high Pearson correlation with each other, ranging from 0.45 to 0.78
(p<0.001 for all) (Supplemental Table S2).
During a median follow-up time of 11.3 years, 1,265 deaths occurred. Associations between
Author Manuscript
all-cause mortality and adherence to the four indices are shown in Table 2. Participants in
the highest compared to the lowest quartile (Q4 vs. Q1) of adherence had lower risk of all-
cause mortality for all 4 indices, although the 95% CI for HEI-2015 included the null value:
HEI-2015 (HR=0.88, 95% CI: 0.74–1.04, Ptrend=0.12), AHEI (HR = 0.82, 95% CI: 0.69–
0.97, Ptrend=0.02), aMED (HR = 0.74, 95% CI: 0.59–0.92, Ptrend=0.02), and DASH (HR =
0.78, 95% CI: 0.65–0.94, Ptrend<0.01). No significant differences were found for pre- vs.
post-diagnosis diet, although the association was slightly stronger for post-diagnosis diet for
aMED (post-diagnosis diet: HR=0.63, 95% CI: 0.43–0.94, Ptrend=0.04; pre-diagnosis diet:
HR=0.81, 95% CI: 0.62–1.06, Ptrend=0.23; Pinteraction=0.99), and for pre-diagnostic diet for
DASH (pre-diagnosis diet: HR=0.76, 95% CI: 0.61–0.96, Ptrend=0.02; post-diagnosis diet:
HR=0.93, 95% CI: 0.66–1.31; Ptrend=0.59; Pinteraction=0.40). Results were similar, though
attenuated slightly, in sensitivity analyses restricted to unrelated women (Supplementary
Author Manuscript
Table S3). In secondary analyses of a subset of women with breast cancer-specific mortality
data, no clear associations were observed, but there was a suggestion of reduced risk of
breast cancer-specific mortality with better adherence to the HEI-2015 index (Q4 vs. Q1:
HR=0.73, 95% CI: 0.52–1.02, Ptrend=0.03) (Supplemental Table S4).
For all four indices, the association between higher adherence (Q4 vs. Q1) and lower risk of
all-cause mortality was found only in women in the healthy weight range (BMI=18.5-<25
kg/m2): HEI-2015 (HR=0.67, 95% CI: 0.52–0.87, Ptrend=0.006), AHEI (HR=0.68, 95% CI:
0.52–0.88, Ptrend=0.005), aMED (HR=0.62, 95% CI: 0.44–0.87, Ptrend=0.001), and DASH
Author Manuscript
(HR=0.69, 95% CI: 0.52–0.90, Ptrend=0.005) (Figure 2 and Supplemental Tables S5 and
S6). No clear associations were observed among women with overweight and obesity for
any of the indices (Ptrend≥0.05), with interactions between diet quality and BMI category
observed for the HEI-2015 (Pinteraction=0.005) and DASH (Pinteraction=0.03) indices. For a
subset of women with information available on tumor hormone receptor status (n=4,644),
stratified analyses indicated that higher adherence (Q4 vs. Q1) to 3 of the 4 indices was
associated with reduced all-cause mortality only among women with hormone receptor
positive tumors: AHEI (HR=0.76, 95% CI: 0.61–0.95, Ptrend=0.03), aMED (HR=0.67, 95%
CI: 0.50–0.90, Ptrend=0.02), and DASH (HR=0.69, 95% CI: 0.54–0.88, Ptrend=0.003) (Figure
3), with an interaction between diet quality and hormone receptor status observed for the
DASH index (Pinteraction=0.02). No clear associations were observed among women with
hormone receptor negative tumors (Ptrend≥0.05). An interaction was also observed between
Author Manuscript
the DASH index and race and ethnicity (Pinteraction=0.005), where the association between
adherence to the DASH index and all-cause mortality was confined to non-Hispanic White
women (Q4 vs. Q1: HR=0.75, 95% CI=0.60–0.95, Ptrend=0.002). No clear associations
were observed among non-Hispanic Black, Asian/Pacific Islander, and Hispanic women
(Ptrend≥0.05) (Supplemental Table S7). No interactions were detected for other indices, or by
menopausal status, and patient or treatment characteristics (Supplemental Table S5 and S6).
Restricting the analysis to women reporting post-diagnosis diet greater than 12 months after
breast cancer diagnosis did not alter the results (Supplemental Table S8).
DISCUSSION
In this study of North American women with breast cancer from diverse racial and ethnic
backgrounds, greater adherence to HEI-2015, AHEI, aMED, and DASH diet quality indices
Author Manuscript
was associated with lower risk of all-cause mortality. The associations were largely confined
to women with a BMI < 25 kg/m2 and may be limited to women diagnosed with hormone
receptor positive tumors.
The four diet quality indices examined in this study have been associated with lower risk of
all-cause mortality in a variety of populations (19,28,29), although results are inconsistent
for women with breast cancer (5,6,13,27,30). In the present study, higher scores of the
HEI-2015, AHEI, aMED, and DASH indices were associated with reduced risk of all-cause
mortality among women with breast cancer. A previous study of 4,103 women diagnosed
with stage I-III breast cancer enrolled in the Nurses’ Health Study observed that higher
scores of the AHEI (Q5 vs. Q1: RR=0.57, 95% CI: 0.42–0.77; Ptrend =0.03) and DASH
(Q5 vs. Q1: RR=0.72, 95% CI: 0.53–0.99; Ptrend =0.03) indices were associated with lower
Author Manuscript
risk of non-breast cancer mortality, but not with breast cancer-specific mortality (5). An
earlier investigation from the same cohort found no association between aMED score and
all-cause mortality or non-breast cancer mortality (6,31). In the present analysis, lower
risk of all-cause mortality was associated with higher scores of the HEI-2015 only among
women with a BMI < 25 kg/m2. Two previous studies also reported lower risk of all-cause
mortality for the HEI, including a study of 670 women diagnosed with local or regional
breast cancer in the Healthy Eating, Activity, and Lifestyle (HEAL) study (Q4 vs. Q1: HR
0.40, 95% CI: 0.17–0.94) (30,32,33), and a study of 2,317 women diagnosed with invasive
breast cancer in the Women’s Health Initiative (WHI) (Q4 vs. Q1: HR=0.74, 95% CI: 0.55–
Author Manuscript
0.99) (26). However, these two studies reported inconsistent results for breast cancer-specific
mortality (26,30); one study reported lower breast cancer-specific mortality with higher
DASH score [32], and the other study reported a null association between HEI and breast
cancer-specific mortality [28], although the number of deaths was small in both studies. In
the present study with more breast-cancer specific deaths than the two studies, there was
some suggestion that higher adherence to HEI-2015 was associated with a lower risk of
breast cancer-specific mortality, but these results should still be interpreted with caution due
to the limited sample size.
There is consistent evidence to suggest that obesity negatively impacts survival among
women with breast cancer (34). The present study indicates that, among women with a
BMI <25 kg/m2, higher adherence to all four indices was associated with decreased risk of
all-cause mortality. This suggests that eating a healthy diet may have additional benefits of
Author Manuscript
lower mortality among women who maintain a healthy weight. In addition, the association
between the AHEI, aMED, and DASH indices and lower all-cause mortality was largely
confined to women who were diagnosed with hormone receptor positive breast cancer,
consistent with another study that observed lower all-cause mortality risk associated with
better adherence to HEI-2005 among women with hormone receptor positive tumors (26).
In two other studies, associations with AHEI, DASH, and HEI-2005 indices did not differ
by tumor hormone receptor status (5,30). There are some suggestions that having better
diet quality may be associated with lower mortality from causes other than breast cancer
(35). Given that women with hormone receptor positive breast cancer have lower all-cause
mortality than those with ER-PR- tumors and thus are more likely to die from causes other
than breast cancer, having better diet quality may have a larger impact on survival among
women with hormone receptor positive tumors. Nevertheless, the present findings should be
Author Manuscript
interpreted with caution because only 1,201 women were diagnosed with hormone receptor
negative breast cancer, and the number of deaths in this subsample was small (n=208). More
studies are needed to further elucidate whether the association between diet quality and
mortality outcomes differs by tumor hormone receptor status.
The majority of studies conducted to date have included primarily non-Hispanic White,
post-menopausal women. Our sample included a large proportion (44%) of Hispanic, Asian/
Pacific Islander, and non-Hispanic Black women. The results from this study were largely
consistent across the different racial and ethnic groups, but there was some suggestion
that the association between higher adherence to the DASH index and reduced all-cause
mortality may be confined to non-Hispanic White women. A larger sample of women from
different racial and ethnic groups is necessary to confirm this finding. Our results also
Author Manuscript
suggest that the association between diet quality estimated through all four indices and lower
mortality risk may be confined to post-menopausal women. This is the first study to examine
the association between diet quality and all-cause mortality risk among pre-menopausal
women with breast cancer (n=3,569). Thus, further research is necessary to understand
whether higher diet quality may improve survival for pre-menopausal women with breast
cancer.
The present study has several strengths, including the large sample size of women with
Author Manuscript
breast cancer from a racially and ethnically diverse population, and data collection of
multiple patient, tumor, and treatment characteristics that allowed us to investigate potential
effect modification by these factors. Bias due to differential follow-up was minimized by
the use of linkages to cancer registries and death registry records whenever possible. The
association of dietary intake patterns with all-cause mortality was examined using four
indices that evaluate the adherence to four recommended diet or dietary patterns.
There are some limitations to consider. The use of a single FFQ is subject to larger
measurement errors due to self-report than it would be if multiple FFQ measures were
available (36). Thus, changes in dietary intakes during follow-up were not captured which
could have introduced exposure misclassification. However, the FFQ used in the BCFR has
been shown to have reasonable validity when compared against repeated 24-hour diet recalls
in another multiethnic cohort (21). To further improve validity, women with unreliable
Author Manuscript
dietary reporting and those who completed the FFQ more than 5 years after diagnosis were
excluded to reduce errors associated with recalling more distant diet. All analyses were
adjusted for total energy intake to reduce confounding and improve validity by removing
correlated errors. Tumor hormone receptor status was not available for approximately 25%
of the women. The study lacked information on co-morbidities that could influence all-cause
mortality. Our evaluation of breast cancer-specific mortality was based on a smaller sample
size, as it was limited to one study site. There is also potential for bias in the standard errors
of our hazard ratios due to familial correlations among some of the women in our sample,
though sensitivity analyses suggest that this bias is likely to be low.
In summary, in this large cohort of women diagnosed with a first primary invasive
breast cancer, adherence to the HEI-2015, AHEI, aMED, and DASH diet quality indices
Author Manuscript
characterized by a high intake of whole fruits, vegetables, dairy, whole grains, nuts, seafood,
and legumes, along with low intake of red and processed meat, sugar-sweetened beverages,
and sodium, was associated with lower risk of death after breast cancer diagnosis. These
findings are generally consistent with earlier studies that suggest adherence to an overall
healthy dietary pattern is associated with reduced mortality among breast cancer survivors.
Supplementary Material
Refer to Web version on PubMed Central for supplementary material.
Funding Support:
The Breast Cancer Family Registry (BCFR) is supported by grant U01 CA164920 from the U.S. National Cancer
Author Manuscript
Institute. The content of this manuscript does not necessarily reflect the views or policies of the National Cancer
Institute or any of the collaborating centers in the BCFR, nor does mention of trade names, commercial products,
or organizations imply endorsement by the U.S. Government or the BCFR. DEH was partially supported by NIH
5T32HL069772-15 and NIH 2T32CA009001-39.
References
1. American Cancer S. Cancer Treatment & Survivorship Facts & Figures 2019–2021. 2019.
2. Bradshaw PT, Stevens J, Khankari N, Teitelbaum SL, Neugut AI, Gammon MD. Cardiovascular
Disease Mortality Among Breast Cancer Survivors. Epidemiology 2016;27(1):6–13 doi 10.1097/
Author Manuscript
EDE.0000000000000394. [PubMed: 26414938]

3. Bird BRJH, Swain SM. Cardiac Toxicity in Breast Cancer Survivors: Review of Potential Cardiac
Problems. Clin Cancer Res 2008;14(1):14–24 doi 10.1158/1078-0432.CCR-07-1033. [PubMed:
18172247]
4. World Cancer Research Fund International/American Institute for Cancer R. Continuous Update
Project Expert Report 2018. Diet, nutrition, physical activity and breast cancer survivors. 2018.
5. Izano MA, Fung TT, Chiuve SS, Hu FB, Holmes MD. Are diet quality scores after breast cancer
diagnosis associated with improved breast cancer survival? Nutrition and Cancer 2013;65(6):820–6
doi 10.1080/01635581.2013.804939. [PubMed: 23909725]
6. Kim EHJ, Willett WC, Fung T, Rosner B, Holmes MD. Diet quality indices and postmenopausal
breast cancer survival. Nutrition and Cancer 2011;63(3):381–8 doi 10.1080/01635581.2011.535963.
[PubMed: 21462090]
7. Vrieling A, Buck K, Seibold P, Heinz J, Obi N, Flesch-Janys D, et al. Dietary patterns and survival
in German postmenopausal breast cancer survivors. British Journal of Cancer 2013;108(1):188–92
Author Manuscript
doi 10.1038/bjc.2012.521. [PubMed: 23169282]

8. Jochems SHJ, Osch FHMV, Bryan RT, Wesselius A, Schooten vFJ, Cheng KK, et al. Impact
of dietary patterns and the main food groups on mortality and recurrence in cancer survivors: a
systematic review of current epidemiological literature. BMJ Open 2018;8(2):e014530 doi 10.1136/
bmjopen-2016-014530.
9. Wang F, Cai H, Gu K, Shi L, Yu D, Zhang M, et al. Adherence to Dietary Recommendations
among Long-Term Breast Cancer Survivors and Cancer Outcome Associations. Cancer Epidemiol
Biomarkers Prev 2020;29(2):386–95 doi 10.1158/1055-9965.EPI-19-0872. [PubMed: 31871105]
10. Jaiswal McEligot A, Largent J, Ziogas A, Peel D, Anton-Culver H. Dietary Fat, Fiber, Vegetable,
and Micronutrients Are Associated With Overall Survival in Postmenopausal Women Diagnosed
With Breast Cancer. Nutrition and Cancer 2006;55(2):132–40 doi 10.1207/s15327914nc5502_3.
[PubMed: 17044767]
11. Kant AK. Dietary patterns and health outcomes. Journal of the American Dietetic Association
2004;104(4):615–35 doi 10.1016/j.jada.2004.01.010. [PubMed: 15054348]
12. Health USDo, Human S, Agriculture USDo. 2015 – 2020 Dietary Guidelines for Americans. 8th
Author Manuscript
Edition. 2015.
13. George SM, Ballard-Barbash R, Manson JE, Reedy J, Shikany JM, Subar AF, et al. Comparing
Indices of Diet Quality With Chronic Disease Mortality Risk in Postmenopausal Women in the
Women’s Health Initiative Observational Study: Evidence to Inform National Dietary Guidance.
Am J Epidemiol 2014;180(6):616–25 doi 10.1093/aje/kwu173. [PubMed: 25035143]
14. Panizza CE, Shvetsov YB, Harmon BE, Wilkens LR, Le Marchand L, Haiman C, et al. Testing
the Predictive Validity of the Healthy Eating Index-2015 in the Multiethnic Cohort: Is the Score
Associated with a Reduced Risk of All-Cause and Cause-Specific Mortality? Nutrients 2018;10(4)
doi 10.3390/nu10040452.
15. Chiuve SE, Fung TT, Rimm EB, Hu FB, McCullough ML, Wang M, et al. Alternative Dietary
Indices Both Strongly Predict Risk of Chronic Disease. J Nutr 2012;142(6):1009–18 doi 10.3945/
jn.111.157222. [PubMed: 22513989]
16. Trichopoulou A, Costacou T, Bamia C, Trichopoulos D. Adherence to a Mediterranean Diet and
Survival in a Greek Population. New England Journal of Medicine 2003;348(26):2599–608 doi
Author Manuscript
10.1056/NEJMoa025039. [PubMed: 12826634]

17. Fung TT, McCullough ML, Newby PK, Manson JE, Meigs JB, Rifai N, et al. Diet-quality scores
and plasma concentrations of markers of inflammation and endothelial dysfunction–. Am J Clin
Nutr 2005;82(1):163–73 doi 10.1093/ajcn.82.1.163. [PubMed: 16002815]
18. Fung TT, Chiuve SE, McCullough ML, Rexrode KM, Logroscino G, Hu FB. Adherence to a
DASH-Style Diet and Risk of Coronary Heart Disease and Stroke in Women. Arch Intern Med
2008;168(7):713–20 doi 10.1001/archinte.168.7.713. [PubMed: 18413553]
19. Harmon BE, Boushey CJ, Shvetsov YB, Ettienne R, Reedy J, Wilkens LR, et al. Associations of
key diet-quality indexes with mortality in the Multiethnic Cohort: the Dietary Patterns Methods
Author Manuscript
Project. Am J Clin Nutr 2015;101(3):587–97 doi 10.3945/ajcn.114.090688. [PubMed: 25733644]

20. John EM, Hopper JL, Beck JC, Knight JA, Neuhausen SL, Senie RT, et al. The Breast Cancer
Family Registry: an infrastructure for cooperative multinational, interdisciplinary and translational
studies of the genetic epidemiology of breast cancer. Breast Cancer Research 2004;6:R375 doi
10.1186/bcr801. [PubMed: 15217505]
21. Stram DO, Hankin JH, Wilkens LR, Pike MC, Monroe KR, Park S, et al. Calibration of the
Dietary Questionnaire for a Multiethnic Cohort in Hawaii and Los Angeles. Am J Epidemiol
2000;151(4):358–70 doi 10.1093/oxfordjournals.aje.a010214. [PubMed: 10695594]
22. Sharma S, Murphy SP, Wilkens LR, Au D, Shen L, Kolonel LN. Extending a multiethnic food
composition table to include standardized food group servings. Journal of Food Composition and
Analysis 2003;16(4):485–95 doi 10.1016/S0889-1575(03)00015-2.
23. Murphy SP. Unique nutrition support for research at the Cancer Research Center of Hawaii. Hawaii
Med J 2002;61(1):15, 7. [PubMed: 11868199]
24. Sacks FM, Svetkey LP, Vollmer WM, Appel LJ, Bray GA, Harsha D, et al. Effects on Blood
Author Manuscript
Pressure of Reduced Dietary Sodium and the Dietary Approaches to Stop Hypertension (DASH)
Diet. New England Journal of Medicine 2001;344(1):3–10 doi 10.1056/NEJM200101043440101.
[PubMed: 11136953]
25. Günther ALB, Liese AD, Bell RA, Dabelea D, Lawrence JM, Rodriguez BL, et al. Association
Between the Dietary Approaches to Hypertension Diet and Hypertension in Youth With Diabetes
Mellitus. Hypertension 2009;53(1):6–12 doi 10.1161/HYPERTENSIONAHA.108.116665.
[PubMed: 19029488]
26. George SM, Ballard-Barbash R, Shikany JM, Caan BJ, Freudenheim JL, Kroenke CH, et al. Better
Postdiagnosis Diet Quality Is Associated with Reduced Risk of Death among Postmenopausal
Women with Invasive Breast Cancer in the Women’s Health Initiative. Cancer Epidemiol
Biomarkers Prev 2014;23(4):575–83 doi 10.1158/1055-9965.EPI-13-1162. [PubMed: 24493629]
27. McCullough ML, Gapstur SM, Shah R, Campbell PT, Wang Y, Doyle C, et al. Pre- and
postdiagnostic diet in relation to mortality among breast cancer survivors in the CPS-II Nutrition
Cohort. Cancer Causes Control 2016;27(11):1303–14 doi 10.1007/s10552-016-0802-x. [PubMed:
27644127]
Author Manuscript
28. Sotos-Prieto M, Bhupathiraju SN, Mattei J, Fung TT, Li Y, Pan A, et al. Association of Changes
in Diet Quality with Total and Cause-Specific Mortality. N Engl J Med 2017;377(2):143–53 doi
10.1056/NEJMoa1613502. [PubMed: 28700845]
29. Schwingshackl L, Hoffmann G. Diet Quality as Assessed by the Healthy Eating Index, the
Alternate Healthy Eating Index, the Dietary Approaches to Stop Hypertension Score, and Health
Outcomes: A Systematic Review and Meta-Analysis of Cohort Studies. Journal of the Academy
of Nutrition and Dietetics 2015;115(5):780–800.e5 doi 10.1016/j.jand.2014.12.009. [PubMed:
25680825]
30. George SM, Irwin ML, Smith AW, Neuhouser ML, Reedy J, McTiernan A, et al.
Postdiagnosis diet quality, the combination of diet quality and recreational physical activity, and
prognosis after early-stage breast cancer. Cancer Causes Control 2011;22(4):589–98 doi 10.1007/
s10552-011-9732-9. [PubMed: 21340493]
31. Di Maso M, Dal Maso L, Augustin LSA, Puppo A, Falcini F, Stocco C, et al. Adherence to
the Mediterranean Diet and Mortality after Breast Cancer. Nutrients 2020;12(12) doi 10.3390/
nu12123649.
Author Manuscript
32. Deshmukh AA, Shirvani SM, Likhacheva A, Chhatwal J, Chiao EY, Sonawane K. The Association
Between Dietary Quality and Overall and Cancer-Specific Mortality Among Cancer Survivors,
NHANES III. JNCI Cancer Spectr 2018;2(2):pky022 doi 10.1093/jncics/pky022. [PubMed:
29905226]
33. Karavasiloglou N, Pestoni G, Faeh D, Rohrmann S. Post-Diagnostic Diet Quality and Mortality in
Females with Self-Reported History of Breast or Gynecological Cancers: Results from the Third
National Health and Nutrition Examination Survey (NHANES III). Nutrients 2019;11(11) doi
10.3390/nu11112558.
34. Tsilidis KK, Cariolou M, Becerra-Tomás N, Balducci K, Vieira R, Abar L, et al. Postdiagnosis
body fatness, recreational physical activity, dietary factors and breast cancer prognosis:
Author Manuscript
Global Cancer Update Programme (CUP Global) summary of evidence grading. Int J Cancer
2023;152(4):635–44 doi 10.1002/ijc.34320. [PubMed: 36279885]
35. Ergas IJ, Cespedes Feliciano EM, Bradshaw PT, Roh JM, Kwan ML, Cadenhead J, et al. Diet
Quality and Breast Cancer Recurrence and Survival: The Pathways Study. JNCI Cancer Spectr
2021;5(2) doi 10.1093/jncics/pkab019.
36. Willett W. Nutritional Epidemiology. New York, NY: Oxford University Press; 2013. p 70–95.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Figure 1. Study Design and Data Collection Scheme.

This figure provides details about exclusion criteria and identification of women with dietary
assessment pre- versus post- diagnosis. Proband source was the first woman identified in
each family.
Author Manuscript
Author Manuscript
Author Manuscript
Figure 2. Hazard Ratios for All-Cause Mortality in Women Diagnosed with Breast Cancer
Comparing Those in the Highest Quartile for Diet Quality Scores to Those in the Lowest
Quartile, Stratified by BMI, Breast Cancer Family Registry
Among women diagnosed with breast cancer (n=2,693; 532 deaths) who had a healthy
weight women (BMI 18.5-<25 kg/m2), those in the highest quartile of diet quality score
had a significantly lower risk of all-cause mortality compared to women in the lowest
quartile [HR (95% CI) Q4 vs. Q1: HEI-2015, 0.67 (0.52–0.87), Ptrend=0.006; AHEI, 0.68
(0.52–0.88), Ptrend=0.005; aMED, 0.62 (0.44–0.87), Ptrend=0.001; and DASH, 0.69 (0.52–
0.90), Ptrend=0.005]. Similar associations were not observed among women with overweight
(n=1,648; 355 deaths). Among women with obesity (n=1,308; 309 deaths), significant
interactions between diet quality indices and BMI were observed for the HEI-2015
Author Manuscript
(Pinteraction=0.005) and DASH (Pinteraction=0.03) only. All hazard ratios are adjusted for
age at diagnosis, study site, total energy intake, race and ethnicity, education, menopausal
status, ER status, PR status, tumor stage, treatment type, recent recreational physical activity,
cigarette smoking status, and pack-years of cigarette smoking. Vertical bars indicate 95%
confidence intervals.
Author Manuscript
Author Manuscript
Author Manuscript
Figure 3. Hazard Ratios for All-Cause Mortality in Women Diagnosed with Breast Cancer
Comparing Those in the Highest Quartile for Diet Quality Scores with Those in the Lowest
Quartile, Stratified by Tumor Hormone Receptor Status, Breast Cancer Family Registry
Among women diagnosed with hormone receptor positive breast cancer (n=3,469; 747
deaths), those in the highest quartile of diet quality score had a significantly lower risk
of all-cause mortality compared to women in the lowest quartile for three of four indices
[HR (95% CI) Q4 vs. Q1: AHEI, 0.67 (0.50–0.90), Ptrend=0.02; aMED, 0.76 (0.61–0.95),
Ptrend=0.03; and DASH, 0.69 (0.54–0.88), Ptrend=0.003]. Similar associations were not
observed among women diagnosed with estrogen receptor negative tumors (n=1,175; 208
deaths). The only significant interaction was observed for the DASH (p=0.02). All hazard
ratios are adjusted for age at diagnosis, study site, total energy intake, race and ethnicity,
Author Manuscript
education, menopausal status, ER and PR status, tumor stage, treatment type, recent
recreational physical activity, cigarette smoking status, and pack-years of cigarette smoking.
Vertical bars indicate 95% confidence intervals.
Author Manuscript
Author Manuscript Author Manuscript Author Manuscript Author Manuscript
Table 1.
1
Characteristics of Women with Breast Cancer in Quartile 1 (Q1) versus Quartile 4 (Q4) of Adherence to Four Dietary Pattern Indices, The Breast Cancer
Family Registry (n=6,157)
Haslam et al.
HEI-2015 AHEI aMED DASH
Overall Q1 (≤63.8) Q4 (>77.0) Q1 (≤45.8) Q4 (>58.4) Q1 (≤3) Q4 (>6) Q1 (≤21) Q4 (>27)
N 6,157 1,539 1,539 1,539 1,539 2,082 850 1,773 1,311
Age at enrollment, mean (SD) 52.8 (23.5) 51.0 (35.8) 55.5 (26.2) 51.8 (35.9) 53.3 (9.9) 53.5 (37.7) 52.4 (9.6) 51.8 (25.0) 53.1 (10.3)
Race and ethnicity, N (%)
Non-Hispanic White 3,406 (55.3) 815 (53.0) 854 (55.5) 900 (58.5) 813 (52.8) 1252 (60.1) 432 (50.8) 936 (52.8) 763 (58.2)
Hispanic 1,086 (17.6) 290 (18.8) 226 (14.7) 228 (14.8) 292 (19.0) 336 (16.1) 162 (19.1) 236 (13.3) 281 (21.4)
Non-Hispanic Black 813 (13.2) 141 (9.2) 312 (20.3) 225 (14.6) 208 (13.5) 287 (13.8) 101 (12.0) 327 (18.4) 112 (8.5)
Asian/Pacific Islander 743 (12.1) 264 (17.2) 115 (7.5) 157 (10.2) 195 (12.7) 176 (8.5) 134 (15.8) 244 (13.8) 126 (9.6)
Other 109 (1.8) 29 (1.9) 32 (2.1) 29 (1.9) 31 (2.0) 31 (1.5) 20 (2.4) 30 (1.7) 29 (2.2)
Education, N (%)
High School or less 2,397 (38.9) 710 (46.1) 474 (30.8) 733 (47.6) 444 (28.9) 930 (44.7) 234 (27.5) 803 (45.3) 400 (30.5)
Some college or bachelor’s degree 2,858 (46.4) 673 (43.7) 758 (49.3) 665 (43.2) 759 (49.3) 934 (44.9) 417 (49.1) 810 (45.7) 621 (47.4)
Graduate degree 866 (14.1) 146 (9.5) 296 (19.2) 128 (8.3) 326 (21.2) 201 (9.7) 193 (22.7) 150 (8.5) 280 (21.4)
Post-menopausal women, N (%) 2,588 (42.0) 538 (35.0) 764 (49.6) 576 (37.4) 669 (43.5) 855 (41.1) 342 (40.2) 695 (39.2) 563 (42.9)
BMI, kg/m2, mean (SD) 26.6 (5.9) 26.8 (6.3) 26.2 (5.6) 27.0 (6.5) 26.2 (5.8) 27.1 (6.1) 25.6 (5.5) 27.2 (6.4) 25.8 (5.4)
Weight status, N (%)
BMI<18.5 kg/m2 143 (2.3) 55 (3.6) 20 (1.3) 52 (3.4) 30 (2.0) 45 (2.2) 35 (2.9) 50 (2.9) 27 (2.1)
BMI=18.5 – 24.9 kg/m2 2,757 (44.8) 650 (43.7) 746 (48.5) 639 (41.5) 723 (47.0) 874 (42.0) 421 (49.5) 728 (41.1) 653 (49.8)
BMI=25 – 29.9 kg/m2 1,703 (27.7) 401 (26.1) 409 (26.6) 414 (26.9) 439 (28.5) 556 (26.7) 245 (28.8) 476 (26.9) 367 (28.0)
BMI≥30 kg/m2 1,360 (22.1) 383 (24.9) 318 (20.7) 378 (24.6) 303 (19.7) 518 (24.9) 139 (16.4) 454 (25.6) 227 (17.3)
3 2,691 (43.7) 551 (35.8) 731 (47.5) 490 (31.8) 819 (53.2) 704 (33.8) 482 (56.7) 564 (31.8) 723 (55.2)
Recreational physical activity , N (% Active)
Cigarette smokers, N (%) 2,535 (41.2) 660 (42.9) 617 (40.1) 682 (44.3) 651 (42.3) 884 (42.5) 370 (43.5) 820 (46.3) 488 (37.2)
Pack-years (among smokers), mean (SD) 15.4 (14.3) 17.9 (15.1) 14.4 (14.2) 17.7 (15.0) 13.3 (12.9) 17.1 (14.9) 13.0 (13.1) 17.8 (14.9) 12.1 (12.9)
Alcoholic drinks/week, N (%)
None 3,683 (59.8) 1006 (65.4) 850 (55.2) 1056 (68.6) 764 (49.6) 1450 (69.6) 328 (38.6) 1068 (60.2) 790 (60.3)
Page 16
HEI-2015 AHEI aMED DASH
Overall Q1 (≤63.8) Q4 (>77.0) Q1 (≤45.8) Q4 (>58.4) Q1 (≤3) Q4 (>6) Q1 (≤21) Q4 (>27)
N 6,157 1,539 1,539 1,539 1,539 2,082 850 1,773 1,311

Haslam et al.
<7 1,645 (26.7) 345 (22.4) 487 (31.6) 313 (20.3) 509 (33.1) 408 (19.6) 372 (43.8) 439 (24.8) 362 (27.6)
≥7 816 (13.3) 184 (12.0) 197 (12.8) 166 (0.8) 264 (17.2) 221 (10.6) 149 (17.5) 264 (14.9) 158 (12.1)
Total energy intake, kcal/d, mean (SD) 2035 (1018) 2230 (1192) 1766 (795) 1827 (862) 2317 (1187) 1549 (669) 2732(1198) 1750 (791) 2350(1099)
Diet quality score, mean (SD)
HEI-2015 70.2 (9.4) 58.0 (4.8) 81.8 (3.7) 62.9 (8.8) 76.5 (7.3) 65.4 (9.2) 76.2 (7.4) 63.3 (8.7) 77.4 (7.0)
AHEI 48.2 (9.0) 41.1 (7.6) 54.4 (8.0) 40.3 (4.6) 64.1 (4.7) 44.9 (7.1) 62.5 (7.3) 39.4 (6.2) 58.2 (6.2)
aMED 4.0 (1.8) 2.9 (1.6) 4.9 (1.6) 2.8 (1.3) 6.1 (1.3) 2.3 (0.8) 7.3 (0.5) 2.6 (1.4) 5.6 (1.3)
DASH 24.0 (4.3) 20.5 (3.5) 27.0 (3.9) 19.7(2.9) 28.1 (3.2) 21.0 (3.4) 28.1 (3.1) 18.9 (2.0) 29.9 (1.8)
4
Cancer treatment , N (%)
Chemotherapy 3,136 (50.9) 843 (54.8) 749 (48.7) 804 (45.2) 788 (51.2) 1034 (49.7) 428 (50.4) 954 (53.8) 662 (50.5)
Hormonal therapy 2,708 (44.0) 663 (43.1) 690 (44.8) 659 (42.8) 700 (45.5) 869 (41.7) 384 (45.2) 764 (43.1) 602 (45.9)
Radiation 3,486 (56.6) 875 (56.9) 869 (56.5) 890 (57.8) 856 (55.6) 1172 (56.3) 477 (56.1) 1049 (59.2) 742 (56.6)
Surgery 5,033 (81.7) 1268 (82.4) 1239 (80.5) 1253 (81.4) 1268 (82.4) 1662 (79.8) 707 (83.2) 1470 (82.9) 1069 (81.5)
Tumor hormone receptor status, N (%)
ER+ and PR− 519 (8.4) 116 (7.5) 128 (8.3) 107 (97.0) 134 (8.7) 161 (7.7) 77 (9.1) 134 (7.6) 107 (8.2)
ER− and PR+ 155 (2.5) 36 (2.3) 48 (3.1) 34 (2.2) †37 (2.4) 43 (2.1) 24 (2.8) †42 (2.4) †28 (2.1)
ER+ and PR+ 2,746 (44.6) 687 (44.6) 710 (46.1) 681 (44.3) 713 (46.3) 927 (44.5) 397 (46.7) 795 (44.8) 620 (47.3)
ER− and PR− 1,201 (19.5) 322 (20.9) 264 (17.2) 333 (21.6) 277 (18.0) 407 (19.6) 150 (17.7) 376 (21.2) 240 (18.3)
Tumor stage, N (%)
Stage I 2,436 (39.6) 577 (37.5) 661 (42.9) 575 (37.4) 660 (42.9) 798 (38.3) 356 (41.9) 691 (39.0) 558 (42.6)
Stage II 2,199 (35.7) 588 (38.2) 494 (32.1) 584 (37.9) 517 (33.6) 747 (35.9) 292 (34.4) 673 (38.0) 436 (33.3)
Stage III 242 (3.9) 70 (4.5) 60 (3.9) 66 (4.3) 51 (3.3) 77 (3.7) 32 (3.8) 76 (4.3) 42 (3.2)
Stage IV 61 (1.0) 15 (1.0) 14 (0.9) 13 (0.8) 13 (0.8) 15 (0.7) 7 (0.8) 18 (1.0) 13 (1.0)
Abbreviations: AHEI, Alternative Healthy Eating Index; aMED, Alternative Mediterranean Diet; BMI, body mass index; DASH, Dietary Approaches to Stop Hypertension; ER, estrogen receptor; PR,
progesterone receptor; HEI-2015, Healthy Eating Index 2015; kcal, kilocalories; SD, standard deviation.
1
P values were computed using ANOVA or Kruskall-Wallis for continuous variables and Chi-Square or Fisher’s Exact tests for categorical variables. All comparisons of Q1 vs. Q4 were statistically
significant at P < 0.05 unless otherwise noted.
Page 17
2
BMI category was defined as underweight: BMI<18.5 kg/m2; healthy weight: BMI =18.5–<25 kg/m2; overweight: BMI=25–<30 kg/m2; and obesity: BMI ≥30 kg/m2.
3
Physically active is defined as recent moderate-to-vigorous physical activity ≥150 minutes/week.
4
The percentage represents the percentage of women who received a particular type of the treatment.
Haslam et al.
Page 18
Table 2.
Hazard Ratios and 95% Confidence Intervals for Risk of All-Cause Mortality in Women Diagnosed with Breast Cancer According to Dietary Quality
Indices, Stratified by Time of Dietary Intake, Breast Cancer Family Registry (n=6,157)
Haslam et al.
Stratified by Time of Dietary Intake

Among All Women
Pre-Diagnosis Diet Post-Diagnosis Diet
# Deaths/PY Model 1 HR Model 2 HR # Deaths/PY Model 1 HR Model 2 HR # Deaths/PY Model 1 HR 2

1 2 1 2 1 Model 2 HR (95% CI)
(95% CI) (95% CI) (95% CI) (95% CI) (95% CI)
N 6,157 5,970 4,557 4,427 1,600 1,543
3
HEI-2015
Q1 (≤63.8) 332/16,333 1.0 (ref.) 1.00 (ref.) 268/13,427 1.0 (ref.) 1.0 (ref.) 64/2,906 1.0 (ref.) 1.0 (ref.)
Q2 (63.9–70.5) 317/16,493 0.95 (0.80–1.11) 0.99 (0.85–1.17) 236/13,447 0.89 (0.74–1.06) 0.98 (0.81–1.17) 81/3,046 1.25 (0.89–1.75) 1.13 (0.81–1.61)
Q3 (70.6–77.0) 320/16,671 0.94 (0.80–1.09) 0.95 (0.80–1.11) 213/12,993 0.82 (0.69–0.99) 0.88 (0.72–1.06) 107/3,678 1.36 (0.99–1.88) 1.29 (0.92–1.80)
Q4 (≥77.1) 296/16,536 0.87 (0.74–1.02) 0.88 (0.74–1.04) 198/12,007 0.83 (0.69–1.00) 0.87 (0.71–1.06) 98/4,530 1.04 (0.75–1.44) 0.98 (0.70–1.38)
Ptrend = 0.08 Ptrend = 0.12 Ptrend = 0.03 Ptrend = 0.11 Ptrend = 0.95 Ptrend = 0.92
3 Pinteraction =0.10
AHEI
Q1 ≤42.0) 372/16,573 1.0 (ref.) 1.0 (ref.) 281/13,376 1.0 (ref.) 1.0 (ref.) 91/3,196 1.0 (ref.) 1.0 (ref.)
Q2 (42.1–48.1) 327/16,642 0.89 (0.77–1.04) 0.91 (0.77–1.06) 238/13,289 0.88 (0.74–1.04) 0.93 (0.77–1.11) 89/3,353 0.94 (0.69–1.26) 0.87 (0.64–1.20)
Q3 (48.2–54.3) 304/16,425 0.85 (0.73–0.99) 0.88 (0.75–1.03) 219/13,007 0.83 (0.70–0.99) 0.93 (0.77–1.12) 85/3,417 0.90 (0.67–1.23) 0.83 (0.61–1.15)
Q4 (≥54.4) 262/16,393 0.74 (0.63–0.87) 0.82 (0.69–0.97) 177/12,201 0.73 (0.60–0.88) 0.87 (0.71–1.07) 85/4,191 0.73 (0.54–0.98) 0.75 (0.55–1.02)
3 Pinteraction=0.49
aMED
Q1 (≤3) 468/22,089 1.0 (ref.) 1.0 (ref.) 336/17,486 1.0 (ref.) 1.0 (ref.) 131/4,062 1.0 (ref.) 1.0 (ref.)
Q2 (4) 257/12,186 1.01 (0.87–1.18) 1.02 (0.87–1.19) 186/9,603 1.04 (0.87–1.24) 1.06 (0.88–1.28) 71/2,583 0.95 (0.71–1.27) 0.95 (0.70–1.30)
Q3 (5) 415/22,380 0.91 (0.79–1.05) 0.92 (0.80–1.07) 301/17,552 0.95 (0.80–1.12) 0.98 (0.82–1.17) 114/4,828 0.80 (0.60–1.04) 0.86 (0.65–1.15)
Q4 (≥6) 128/9,377 0.67 (0.54–0.82) 0.74 (0.59–0.92) 94/7,232 0.72 (0.56–0.92) 0.81 (0.62–1.06) 34/2,145 0.54 (0.35–0.78) 0.63 (0.43–0.94)
3 Pinteraction =0.99
DASH
Q1 (≤21) 412/19,181 1.0 (ref.) 1.0 (ref.) 330/15,794 1.0 (ref.) 1.0 (ref.) 82/3,387 1.0 (ref.) 1.0 (ref.)
Q2 (22–24) 347/16,944 0.96 (0.83–1.11) 0.98 (0.84–1.14) 248/13,622 0.88 (0.75–1.04) 0.96 (0.80–1.14) 99/3,322 1.23 (0.92–1.65) 1.17 (0.86–1.59)
Page 19
Stratified by Time of Dietary Intake

Among All Women
Pre-Diagnosis Diet Post-Diagnosis Diet
# Deaths/PY Model 1 HR Model 2 HR # Deaths/PY Model 1 HR Model 2 HR # Deaths/PY Model 1 HR 2

1 2 1 2 1 Model 2 HR (95% CI)
(95% CI) (95% CI) (95% CI) (95% CI) (95% CI)
Haslam et al.
N 6,157 5,970 4,557 4,427 1,600 1,543
Q3 (25–27) 299/16,047 0.86 (0.74–1.00) 0.89 (0.76–1.05) 204/12,169 0.81 (0.68–0.96) 0.89 (0.73–1.07) 95/3,878 0.99 (0.73–1.34) 1.07 (0.78–1.48)
Q4 (≥28) 207/13,861 0.70 (0.59–0.83) 0.78 (0.65–0.94) 133/10,290 0.63 (0.51–0.77) 0.76 (0.61–0.96) 74/3,572 0.87 (0.63–1.20) 0.93 (0.66–1.31)
Ptrend < 0.0001 Ptrend = 0.006 Ptrend < 0.0001 Ptrend = 0.02 Ptrend = 0.19 Ptrend = 0.59
Pinteraction =0.40
Abbreviations: AHEI, Alternative Healthy Eating Index; aMED, Alternative Mediterranean Diet; DASH, Dietary Approaches to Stop Hypertension; HEI-2015, Healthy Eating Index 2015; CI, confidence
interval; HR, hazard ratio; PY, person-years.
1
Adjusted for age (continuous), study site, total caloric intake (quartiles).
2
Additionally adjusted for race and ethnicity (non-Hispanic White, Black, Hispanic, Asian/Pacific Islander, other), education (high school or less, some college or bachelor’s degree, graduate degree),
treatment type (yes vs. no for surgery, radiation, chemotherapy, hormone therapy, or other treatment), tumor stage (I, II, III, IV), recent recreational physical activity (<150 minutes/week, ≥150 minutes/
week), cigarette smoking status (never, ever), and pack-years of cigarette smoking (continuous), tumor estrogen receptor status (yes, no, borderline, unknown/missing), tumor progesterone receptor status
(yes, no, borderline, unknown/missing), and menopausal status (pre- vs. post-menopausal).
3
See Supplemental Table S1 for dietary index scoring details.
Page 20

Diet

Uploaded by

Copyright:

Available Formats

Diet

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Diet

Uploaded by

Copyright:

Available Formats

HHS Public Access

Diet Quality and All-Cause Mortality in Women with Breast

breast cancer; breast cancer survivors; diet quality; epidemiology; mortality

Evidence is still limited on whether adherence to a particular recommended diet or dietary

activity, and body mass index (BMI).

Materials and Methods

Diet Quality Indices

measures adherence to a diet associated with reduced risk of cardiovascular diseases

To evaluate potential effect modification, stratified analyses were conducted by BMI

EDE.0000000000000394. [PubMed: 26414938]

doi 10.1038/bjc.2012.521. [PubMed: 23169282]

10.1056/NEJMoa025039. [PubMed: 12826634]

Project. Am J Clin Nutr 2015;101(3):587–97 doi 10.3945/ajcn.114.090688. [PubMed: 25733644]

Figure 1. Study Design and Data Collection Scheme.

HEI-2015 AHEI aMED DASH

Overall Q1 (≤63.8) Q4 (>77.0) Q1 (≤45.8) Q4 (>58.4) Q1 (≤3) Q4 (>6) Q1 (≤21) Q4 (>27)

N 6,157 1,539 1,539 1,539 1,539 2,082 850 1,773 1,311

HEI-2015 AHEI aMED DASH

Overall Q1 (≤63.8) Q4 (>77.0) Q1 (≤45.8) Q4 (>58.4) Q1 (≤3) Q4 (>6) Q1 (≤21) Q4 (>27)

N 6,157 1,539 1,539 1,539 1,539 2,082 850 1,773 1,311

Stratified by Time of Dietary Intake

# Deaths/PY Model 1 HR Model 2 HR # Deaths/PY Model 1 HR Model 2 HR # Deaths/PY Model 1 HR 2

Stratified by Time of Dietary Intake

# Deaths/PY Model 1 HR Model 2 HR # Deaths/PY Model 1 HR Model 2 HR # Deaths/PY Model 1 HR 2

N 6,157 5,970 4,557 4,427 1,600 1,543

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.