Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Apr 15.
Published in final edited form as: Am J Cardiol. 2022 Jan 25;169:71–77. doi: 10.1016/j.amjcard.2021.12.043

Relation of DASH Dietary Pattern to Heart Failure Risk and to Socioeconomic Status (From the Southern Community Cohort Study)

Rachel S Chang a, Meng Xu b,c, Sarah H Brown a, Sarah S Cohen d, Danxia Yu e, Elvis A Akwo b,f, Debra Dixon b,g, Loren Lipworth b,e, Deepak K Gupta b,g
PMCID: PMC9007893  NIHMSID: NIHMS1769881  PMID: 35090697

Abstract

The Dietary Approaches to Stop Hypertension (DASH) dietary pattern has been associated with lower risk of incident heart failure (HF); however, prior studies were conducted in mostly middle-income white populations. The association between DASH and incident HF risk in lower income and Black individuals is less well understood. We analyzed 25,300 white and Black adults without history of HF at enrollment (2002 – 2009) in the Southern Community Cohort Study receiving Centers for Medicare/Medicaid Services (CMS). Alignment with DASH was assessed at enrollment using a validated food frequency questionnaire. Incident HF was ascertained from CMS claims through 2016. The association between DASH diet alignment and incident HF was examined in multivariable-adjusted Cox proportional hazards regression models, including an interaction term testing effect modification by income. The cohort was predominantly middle-aged (median 54 years), Black (68%), female (63%), and low-income (88% < $25,000/year/household). Socioeconomic factors, including education and annual income, were larger contributors to the variance in DASH score than were cardiovascular comorbidities. The association between DASH dietary alignment and HF risk was not significant overall (HR 1.00; 95% CI 0.96–1.04) or in race-sex groups. However, the association between alignment with DASH diet and HF risk significantly varied by income (interaction-p=0.030), with neutral and inverse associations in lower (<$25,000/year) and higher ($≥25,000) income individuals, respectively. In conclusion, income modified the association between healthier dietary pattern and risk of incident HF. In lower income individuals, greater alignment with the DASH diet was not associated with lower HF risk.

Keywords: DASH diet, heart failure, income

Introduction

Despite advances in the prevention and treatment of cardiovascular disease, heart failure (HF) is increasing in prevalence, highly morbid, and lethal.1,2 Increasing attention to the prevention of HF has revealed a substantial proportion of incident HF is attributable to modifiable risk factors, such as hypertension, diabetes, and obesity.3 Diet is a universally applicable, modifiable lifestyle contributor to cardiovascular health, and thus a potential target for intervention regarding HF. The Dietary Approaches to Stop Hypertension (DASH) diet focuses on consumption of fruits, vegetables, whole grains, and lean meats. In a randomized controlled trial, the DASH diet lowered blood pressure to a greater extent than a typical Western diet.4 Few studies, however, have investigated the association between DASH diet alignment, defined as the extent to which an individual’s diet corresponds with established DASH diet components, and incident HF risk. The Southern Community Cohort Study (SCCS), a prospective cohort study based in the southeastern US, affords the opportunity to address these evidence gaps in a low-income, predominantly Black population, in whom traditional cardiovascular risk factors are common.

Methods

The SCCS is an ongoing, prospective cohort study designed to examine the incidence of cancer and other chronic diseases, and to elucidate causes of health disparities in a predominantly low-income and Black population in the southeastern United States. Detailed methods on its study design have been previously published.5 Briefly, a total of 84,797 individuals aged 40–79 years (65% Black, 65% female) were recruited from 2002–2009 in 12 southeastern states. Most participants (86%) enrolled through Community Health Centers (CHC), which provide health and preventive care mainly to the underserved; the remaining 14% enrolled through general population sampling. At enrollment, participants completed an extensive questionnaire which ascertained sociodemographic and anthropometric characteristics, personal medical history, and lifestyle behaviors including smoking, diet and physical activity.5,6 The questionnaire was administered through standardized computer-assisted personal interviews for community health center participants, and via self-administered mailed questionnaires for population participants. SCCS protocols were approved by the Vanderbilt University Medical Center and Meharry Medical College Institutional Review Boards (IRB) and participants provided written informed consent. Meng Xu, Loren Lipworth, and Deepak K. Gupta had access to all data in the study and take full responsibility for its integrity and the data analysis.

For the current analysis, we included SCCS participants who were either ≥ 65 years, or < 65 years at enrollment and: 1) reported being covered by Medicaid or Medicare on baseline questionnaire; or 2) had a Centers for Medicare and Medicaid Services (CMS) claim within 90 days of being enrolled in the SCCS (N = 33,003). Participants with prevalent HF upon enrollment were excluded (N = 4,572). Our analysis was further restricted to self-reported non-Hispanic Black and non-Hispanic white participants (N = 27,119), because the number of participants of other racial groups was too few to allow for stable statistical analysis. Participants with prevalent end-stage renal disease (ESRD) upon enrollment were excluded (N = 51) due to significant dietary restrictions placed on patients with ESRD and potential misclassification of HF in the setting of ESRD. Participants with missing DASH scores (N = 1,768) were excluded, yielding a final analytic cohort of 25,300 participants.

At enrollment into the SCCS, each participant’s dietary pattern was assessed using a validated food frequency questionnaire (FFQ),7 which contained questions on 89 food items designed to reflect a majority of foods common to the southeastern US.8 The FFQ was validated through race-concordant 24-hour dietary recalls in the South against nationwide race-blind 24-hour recall testing (the standard system, which was based on the National Health and Nutrition Examination Survey [NHANES] and the Continuing Survey of Food Intakes by Individuals [CSFII]), in which the Kappa statistic demonstrated substantial or better agreement (κ > 0.73).7 Intakes of total energy and nutrients were calculated through utilizing race- and sex-specific portion size information from NHANES and CSFII as well.7 Participants who had incomplete FFQs (>10 items left blank) or reported implausible total energy intakes (<600 or >8,000 kcal/d) were excluded.

The SCCS FFQ data were linked with the USDA MyPyramid Equivalents Database to generate equivalent intakes of food groups, standardized to intakes per 1000 kilocalories, and to calculate 11 component scores to produce a quantifiable total DASH dietary score based on recommendations in the National Heart, Lung and Blood Institute’s DASH plan.9 The score ranged from 0 to 90 points, with higher scores representing greater alignment with the DASH diet. The 11 DASH diet components included fruits (10 points); vegetables (10 points); total grains (5 points); high-fiber grains (5 points); dairy (5 points); low-fat dairy (5 points); nuts, seeds and legumes (10 points); lean meats, poultry, fish and eggs (10 points); sweets and added sugar (10 points); fats and oils (10 points); and sodium (10 points). Using a similar approach, we also calculated the Healthy Eating Index (HEI)-2010 to assess adherence to the Dietary Guidelines for Americans (DGA) in the SCCS population.10

Participants’ medical history was defined through history of self-reported physician-diagnosed medical conditions, including hypertension, diabetes mellitus, high cholesterol, myocardial infarction or coronary artery bypass surgery (MI/CABG), and stroke or transient ischemic attack (TIA). Physical activity and sedentary time were assessed by a validated physical activity questionnaire from which both total activity metabolic equivalent hours (MET-hrs/day) and total sit hours/day were calculated. As previously reported in the SCCS, outlier analysis for total activity was performed. For MET-hrs/day the Tukey method was used,11 in which out of range values were defined as values > 75th percentile + 1.5 × IQR, and were coded as “missing” for further analysis. Outlier analysis for total sitting hours/day was performed by truncating values > 24 to a maximum of 24 hours. Validation studies have demonstrated the reliability for several self-reported variables in the SCCS through medical records, accelerometers, and biomarkers.5,6

HF events were ascertained via linkage of the SCCS cohort to the CMS Research Identifiable Files. Incident HF was defined as the first claim with an International Classification of Diseases, 9th revision discharge code of 428.x in the Medicare institutional, Part B carrier, outpatient-based claims files, or the Medicaid Analytic Extract Inpatient and Other Services claims files, from the SCCS enrollment date to December 31, 2015.12 Following December 31, 2015, incident HF was defined as the first medical claim with an International Classification of Diseases, 10th revision discharge code of I.50.x, until December 31, 2016. Studies have demonstrated the comparability of ICD-10 codes to ICD-9 codes.13,14 Censoring occurred at the date of HF event, date of death, or end of follow-up period (December 31, 2016), whichever occurred first.15 Prevalent HF was determined through ICD-9 and ICD-10 codes prior to enrollment, and all participants with prevalent HF were excluded from the study.

Descriptive statistics were calculated as counts (percentages) for categorical variables or medians (25th–75th percentile) for continuous variables and were presented stratified according to quartile of DASH diet score. Comparisons between quartiles were performed using the Pearson chi-squared and Kruskal-Wallis tests for categorical and continuous data, respectively. Correlates of DASH score were examined in multivariable linear regression. The relative contribution to variance in the DASH score for each factor included in the model was ranked according to its respective chi-squared statistic from the model. HF incidence rates were calculated according to the Kaplan-Meier method. The risk of incident HF in association with DASH score was estimated using multivariable-adjusted Cox proportional hazards models. DASH score was modeled as a continuous linear variable as the test for non-linearity was not significant (p = 0.70). The base model adjusted for age, sex, and race, while the fully-adjusted model further accounted for traditional cardiovascular risk factors (body mass index [BMI] in kg/m2, physical activity in total MET-hours/day, total sitting hours/day, smoking status [current, former, or never smoker], and history of hypertension, high cholesterol, diabetes mellitus, MI/CABG, or stroke/TIA); total energy intake (kcal/day); and socioeconomic factors: annual household income [<$15,000, $15,000–24,999, and ≥ $25,000], employment status, highest level of education attained [< high school, high school/vocational training/junior college, and college degree or higher], marital status [married/living as married with partner, separated/divorced, widowed, and single/never married], and enrollment source [CHC vs. general population]. Continuous variables were modeled as restricted cubic splines with 4 knots. Hazard ratios with 95% confidence intervals were calculated per increase from the 25th to 75th percentile in DASH score.

To understand whether the association between DASH diet alignment and incident HF risk varied according to socioeconomic status, we included interaction terms for DASH score x income and education in the multivariable-adjusted Cox regression models. Significant interactions were displayed using marginal effects plots of the 5-year predicted probability of incident HF according to DASH score.

We repeated the primary analysis in each of the 4 race-sex groups. We also performed sensitivity analyses excluding participants with prevalent MI or CABG upon enrollment and excluding participants who developed incident HF within 3 months after SCCS enrollment. To address specificity of the DASH dietary pattern association with incident HF, we repeated the analysis replacing DASH score with the HEI-2010. All analyses were performed using R (The R project, Vienna, Austria). Statistical significance was considered present at p < 0.05.

Results

Enrollment characteristics of the 25,300 SCCS participants are shown in Table 1. The median age was 54 years, and the majority were female, Black, reported a prior diagnosis of hypertension, and had an annual income of less than $15,000. Enrollment characteristics for each race-sex group are shown in Tables S1–S4.

Table 1.

Baseline characteristics of SCCS participants according to quartiles of DASH dietary score.

DASH dietary score
Variable Overall (N = 25,300) Q1 ≤ 40.8 (N = 6,325) Q2 >40.9 to ≤ 47.2 (N = 6,326) Q3 >47.3 to ≤ 54.1 (N = 6,324) Q4 > 54.1 (N = 6,325) P-Value 1
Age (years) 54 (47–65) 50 (44–57) 52 (46–61) 56 (48–65) 63 (53–68) <0.001
Women 62.9% 56.2% 57.4% 64.1% 73.8% <0.001
Black 68.0% 73.1% 70.4% 68.2% 60.3% <0.001
BMI (kg/m2) 29.1 (25.0–34.7) 28.6 (24.2–34.4) 28.9 (24.5–34.4) 29.4 (25.2–34.8) 29.6 (25.7–35.0) <0.001
Hypertension 62.5% 57.0% 60.3% 64.1% 68.8% <0.001
MI or CABG 8.7% 7.4% 8.4% 8.8% 10.2% <0.001
Diabetes mellitus 26.6% 20.3% 22.8% 28.3% 35.2% <0.001
“High cholesterol” 39.9% 32.2% 34.8% 40.8% 51.8% <0.001
Stroke or TIA 9.5% 8.6% 8.8% 10.1% 10.5% <0.001
Smoking status
 Never 34.5% 26.4% 30.1% 36.7% 45.0%
 Former 25.6% 18.2% 22.8% 27.2% 34.0% <0.001
 Current 39.9% 55.4% 47.1% 36.1% 21.1%
Physical activity (MET-hrs/day) 12.8 (6.3–22.9) 12.6 (6.3–23.5) 12.9 (6.3–23.8) 13.1 (6.6–22.7) 12.8 (6.8–21.8) 0.32
Sitting Time (sit-hours/day) 8 (5.5–11.5) 9 (6–12) 8.2 (5.7–12) 8.0 (5.5–11.2) 7.5 (5.0–10.0) <0.001
Total energy intake (kcal/day) 2140 (1482–3114) 2541 (1817–3722) 2387 (1705–3481) 2111 (1503–2995) 1592 (1124–2265) <0.001
Annual household income
 <$15,000 69.3% 76.7% 72.3% 67.4% 60.7%
 $15,000–24,999 18.2% 16.4% 18.0% 18.8% 19.4% <0.001
 $≥25,000 12.5% 6.9% 9.7% 13.8% 19.9%
Employed 15.2% 14.3% 15.0% 14.9% 16.5% <0.001
Education level
 < High School 38.1% 43.8% 39.6% 37.4% 31.5%
 High school / vocational training / some college 53.3% 51.7% 53.7% 53.2% 54.6% <0.001
 College graduate 8.6% 4.4% 6.6% 9.4% 13.9%
Marital status
 Single 21.4% 27.2% 24.5% 20.7% 13.2%
 Widowed 14.5% 9.6% 11.6% 15.4% 21.4%
 Separated or divorced 33.8% 35.7% 35.0% 32.7% 31.9% <0.001
 Married or living together 30.3% 27.5% 28.8% 31.2% 33.5%
Enrollment source
 Community health center 90.0% 93.1% 92.2% 89.2% 85.7% <0.001
Open in a new tab

Data presented as median (25th–75th percentile) for continuous variables or percentages for categorical variables. P-values calculated from Kruskal-Wallis (continuous) or Pearson Chi-squared tests (categorical).

1

BMI = body mass index; CABG = coronary artery bypass graft; TIA = transient ischemic attack. MET = metabolic equivalent hours.

In the overall cohort, median (25th, 75th percentile) DASH dietary score was 47.2 (40.8–54.1) (Table 1). Across increasing quartiles of DASH dietary score, age increased, as did the proportion of women, and the frequency of comorbidities such as hypertension and diabetes. Annual income and educational attainment were also higher with increasing quartile of DASH score. These patterns were largely consistent in each of the 4 race-sex groups (Table S1–S4).

Total energy intake accounted for the greatest proportion of the variance in DASH score, followed by enrollment age, smoking status, sex, education, diabetes, total sit hours/day, and annual income (Figure 1). In contrast, prevalent cardiovascular disease (MI/CABG or stroke/TIA), hypertension, and BMI contributed relatively little to the variance in DASH score.

Figure 1: Relative strength of association of correlates of DASH dietary score in the SCCS.

Figure 1:

Open in a new tab

BMI = body mass index; CABG = coronary artery bypass graft; TIA = transient ischemic attack. MET = metabolic equivalent hours.

Over a median (25th, 75th percentile) follow-up time of 11.0 (8.7, 12.7) years, a total of 7,045 individuals (27.8%) developed HF (incidence rate 27 cases/1000 person-years) (Table 2). In age, sex, and race adjusted model, each 13.3 increase in DASH score (corresponding to the interquartile range) was associated with a significantly lower risk of incident HF (HR 0.962; 95% CI 0.931–0.997; p = 0.032). Further adjustment for traditional cardiovascular risk factors, physical activity, and socioeconomic factors attenuated the association between alignment with DASH score and HF risk (HR 1.00; 95% CI 0.96–1.04; p = 0.86). In analyses stratified by 4 race-sex groups, DASH score was not significantly associated with the risk of HF in the fully adjusted model in any subgroup (Table 2). Exclusion of participants with prevalent MI/CABG at enrollment (N = 2,404) or of participants who developed HF within 3 months of enrollment (N = 297) did not appreciably change the results (Table S5).

Table 2.

Association of Dietary Approaches to Hypertension (DASH) dietary pattern with incident heart failure, overall and in race-sex groups.

Base model Fully adjusted model 1
HF events HF incidence Follow-up Time, months (25th, 75th) HF incidence rate (cases/1,000 person-years; 95% CI) Hazard Ratio (95% CI) P Hazard ratio (95% CI) P
Overall
 N = 25,300 		7,045 27.8% 133 (104, 153) 27.1 (26.5, 27.8) 0.963 (0.931–0.997) 0.032 1.00 (0.96–1.04) 0.86
Black women
 N = 10,882 		3,045 28.0% 138 (112–158) 25.7 (24.8, 26.7) 1.00 (0.95–1.05) 0.85 1.00 (0.94–1.06) 0.87
Black men
 N = 6,318 		1,680 26.6% 130 (98, 155) 26.6 (25.4, 27.9) 1.00 (0.93–1.08) 0.92 0.98 (0.91–1.06) 0.62
White women
 N = 5,025 		1,439 28.6% 126 (102–140) 28.9 (27.5, 30.4) 0.90 (0.84–0.97) 0.006 1.00 (0.92–1.08) 0.92
White men
 N = 3,075 		881 28.7% 123 (94, 136) 31 (29, 33) 0.90 (0.82–1.00) 0.06 1.07 (0.95–1.21) 0.25
Open in a new tab

Base model included adjustment for age, race, and sex.

1

Fully adjusted model included age, race, sex, history of myocardial infarction or coronary artery bypass graft, history of diabetes, history of high cholesterol, history of hypertension, history of stroke or TIA, body mass index (BMI), smoking, total physical activity (MET-hrs), total sitting hours, total energy intake, employment status, education, marital status, and enrollment source. Unit of change for hazard ratios for the overall cohort and in each race-sex group was a DASH score increase of 13.3 (interquartile range).

Due to the uniquely low socioeconomic status of our study population underrepresented in previous studies and because socioeconomic factors accounted for a substantial proportion of the variance in DASH score (Figure 1), we examined whether the association between DASH score and HF risk varied according to income (interaction-p = 0.030) and education (interaction-p = 0.66). Among individuals in the lower income groups (household income < $25,000/year), greater alignment with the DASH dietary pattern did not associate with lower risk for incident HF. In contrast, in the highest income group (≥ $25,000 per year), an inverse association between DASH score and HF risk was observed (Figure 2). The income-based variation in the association between DASH score and HF risk was present despite relatively small differences in DASH score, with medians of 46.3, 48.0, and 51.7 in the <$15,000/year; $15,000–24,999/year, and ≥$25,000/year annual income groups, respectively (Table S6). Individuals in the lowest income category comprised nearly 70% of our cohort and were at greater HF risk (28.3 cases/1000 person-years, 95% CI 27.5–29.1) compared with the highest income group (24.1, 95% CI 22.6–25.9). The dietary pattern by income interaction on the association with incident HF risk remained present when HEI-2010 replaced DASH score in the multivariable-adjusted Cox regression (Figure S1).

Figure 2: Five-year predicted probability for incident heart failure (HF) according to DASH score and income in the Southern Community Cohort Study.

Figure 2:

Open in a new tab

The fully adjusted Cox model, adjusted for age, sex, race, hypertension, diabetes, high cholesterol, MI/CABG, stroke/TIA, BMI, smoking, total physical activity, total sitting hours, total energy intake, employment status, education, marital status, and enrollment source. Dashed lines indicate 95% confidence intervals. Multivariable Cox models yielded <$15,000/year: HR 1.01 (95% 0.96–1. 05); $15,000–24,999: HR 1.01 (95% CI 0.92–1.11); and ≥ $25,000: HR 0.92 (95% CI 0.81–1.04) for each 13.3 increase in DASH score (corresponding to the interquartile range).

Discussion

Evidence for a protective association between DASH diet and cardiovascular health has been derived from mostly middle-income or majority white populations.16–19 In a large cohort of predominantly low socioeconomic status and black individuals residing in the southeastern United States, we examined the association between alignment with the DASH dietary pattern and the risk of incident HF. Our principal findings are 1) greater alignment with DASH dietary pattern does not associate with lower risk of incident HF among lower-income individuals, and 2) socioeconomic factors, rather than medical comorbidities such as hypertension and coronary artery disease, are relatively stronger correlates of alignment with DASH dietary pattern.

The association between DASH dietary pattern alignment and risk of incident HF varies across published studies. Analyses from Sweden and MESA suggested lower risk for incident HF with greater alignment with the DASH dietary pattern.16–18 In contrast, in the CHS and our current analysis in the SCCS, an association between DASH dietary pattern and HF risk was not observed in the overall cohort.19 The Swedish cohorts were largely composed of middle-aged white adults with lower burdens of hypertension (19.7%), diabetes (5–10%), high cholesterol (8–12%), and obesity (average BMI 25 – 26 kg/m2) compared with the SCCS cohort.16,17,20 MESA participants had a lower burden of chronic diseases including hypertension (42.9%), diabetes (11.4%) and other HF risk factors, with relatively higher income (40% > $50,000/year) and education levels (66% > high school) than SCCS participants.18,21 Our results are consistent with the CHS, which included individuals aged ≥ 65 years whose comorbidity profile was more similar to the SCCS (59% with hypertension, 16% diabetes, average BMI 27.4) and who were exclusively Medicare participants with more comparable annual income levels with SCCS participants (61.5% < $25,000/year).19,22 Collectively, the heterogeneity of associations between DASH and HF risk across studies may be due to differences in study populations, both with regards to disease burden and income. It is also possible that once medical conditions become present, they may outweigh dietary contributions to HF risk. For example, primary prevention of HF through diet may be more difficult in a high comorbid population such as the SCCS, perhaps due to cardiovascular structural changes including increased left ventricular mass and hypertrophy in diabetic and hypertensive patients; diabetes and hypertension were present in 26.6% and 62.7% of our SCCS cohort, respectively.23,24

Our results also suggest socioeconomic factors account for the observed differences in the association between dietary patterns and HF risk. Congruent with prior studies, we found a positive correlation between socioeconomic status and alignment to healthier dietary patterns, including DASH.25,26 Moreover, we found that the beneficial association between greater alignment with DASH dietary pattern and lower risk of incident HF is attenuated in lower compared with higher income individuals. This does not mean that lower income individuals should not consume a healthy diet. Rather, our findings indicate that the underlying reasons for this attenuation of the effect of healthier diet warrant further investigation. Potential explanations include that lower income may be a marker of other factors that influence health such as medication adherence, environmental exposures, or safe affordable access to transportation for preventive medical care.27 Additionally, the psychobiological stress experienced by low-income individuals, such as the threat of eviction, lack of housing, or loss of employment, can result in persistent pathophysiologic changes that negate the otherwise beneficial effects of healthier diet.28 In this context, our results emphasize the need to identify and address social determinants which limit the beneficial effect of healthier diet in lower income individuals.

Limitations of our study should be noted. Our cohort was comprised of predominantly low socioeconomic, Black, and female individuals; therefore, our results may not be generalizable to a broader population. Alignment with the DASH dietary pattern was calculated from enrollment FFQs and repeated questionnaires were not administered throughout the follow up period, which could result in misclassification bias. The FFQ used, however, was a validated instrument and studies demonstrate a single FFQ in adults sufficiently reflects an individual’s habitual dietary pattern over time.29,30 Other limitations of the FFQ include the lack of multiple recalls to reliably estimate usual intake, participant burden, and possible literacy demands in estimation of portion size. Variation in food preparations, such as fried versus non-fried meats and preserved versus fresh fruits and vegetables was not distinguished through the FFQ. Cardiovascular comorbidities such as hypertension, prior MI or CABG, stroke or TIA may inform an individual’s dietary pattern but we found that history of these comorbidities contributed little to the variance in DASH score. A sensitivity analysis that excluded participants with prevalent MI/CABG demonstrated that results for the association between DASH score and incident HF risk did not appreciably change. Self-reported covariates may also be subject to misclassification bias; however, the SCCS questionnaire has been previously validated for a number of variables.5–7 Although our large sample size and incident HF events afforded power to adjust for demographic, anthropometric, clinical, and socioeconomic factors, we cannot exclude residual confounding. We acknowledge that prospective studies addressing socioeconomic factors to prevent incident HF are challenging to conduct. Identifying and addressing these factors is important, however, as individuals with the lowest income are at the greatest risk of incident HF.

In conclusion, in a predominantly low-income cohort of adults residing in the southeastern United States, income modified the association between a healthy eating and risk of incident HF. In lower income individuals, greater alignment with DASH dietary pattern was not associated with lower HF risk.

Supplementary Material

1

Acknowledgements

The authors thank all SCCS participants and the SCCS research team.

Sources of Funding: The SCCS is supported by the National Cancer Institute (grants R01 CA092447 and U01 CA202979) and supplemental funding from the American Recovery and Reinvestment Act (3R01 CA 029447-0851). SCCS data collection was performed by the Survey and Biospecimen Shared Resource which is supported in part by the Vanderbilt-Ingram Cancer Center (P30 CA68485). Dr. Dixon was supported by the Training in Cardiovascular Research T32 HL007411, Nashville, TN.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Roger VL. Epidemiology of Heart Failure. Circ Res 2013;113(6):646–659. doi: 10.1161/CIRCRESAHA.113.300268 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Cowie MR, Wood DA, Coats AJ, Thompson SG, Suresh V, Poole-Wilson PA, Sutton GC. Survival of patients with a new diagnosis of heart failure: a population based study. Heart (British Cardiac Society) 2000;83(5):505–510. doi: 10.1136/heart.83.5.505 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Kubicki DM, Xu M, Akwo EA, Dixon D, Munoz D, Blot W, Wang TJ, Lipworth L, Gupta DK. Race and Sex Differences in Modifiable Risk Factors and Incident Heart Failure. JACC: Heart Fail 2020;8(2):122–130. doi: 10.1016/j.jchf.2019.11.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Appel LJ, Moore TJ, Obarzanek E, Vollmer WM, Svetkey LP, Sacks FM, Bray GA, Vogt TM, Cutler JA, Windhauser MM, Lin PH, Karanja N. A clinical trial of the effects of dietary patterns on blood pressure. DASH Collaborative Research Group. N Engl J Med 1997;336(16):1117–1124. doi: 10.1056/NEJM199704173361601 [DOI] [PubMed] [Google Scholar]
  • 5.Signorello LB, Hargreaves MK, Blot WJ. The Southern Community Cohort Study: Investigating Health Disparities. J Health Care Poor Underserved 2010;21(1):26–37. doi: 10.1353/hpu.0.0245 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Buchowski MS, Matthews CE, Cohen SS, Signorello LB, Fowke JH, Hargreaves MK, Schlundt DG, Blot WJ. Evaluation of a questionnaire to assess sedentary and active behaviors in the Southern Community Cohort Study. J Phys Act Health 2012;9(6):765–775. doi: 10.1123/jpah.9.6.765 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Signorello LB, Munro HM, Buchowski MS, Schlundt DG, Cohen SS, Hargreaves MK, Blot WJ. Estimating nutrient intake from a food frequency questionnaire: incorporating the elements of race and geographic region. Am J Epidemiol 2009;170(1):104–111. doi: 10.1093/aje/kwp098 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Buchowski MS, Schlundt DG, Hargreaves MK, Hankin JH, Signorello LB, Blot WJ. Development of a culturally sensitive food frequency questionnaire for use in the Southern Community Cohort Study. Cell Mol Biol (Noisy-le-grand) 2003;49(8):1295–1304. [PubMed] [Google Scholar]
  • 9.Health N, National Heart Lung, Institute B, Services USDH, Lung BNHI. Your Guide to Lowering Your Blood Pressure with DASH: DASH Eating Plan CreateSpace Independent Publishing Platform; 2012. https://books.google.com/books?id=iQOYvwEACAAJ [Google Scholar]
  • 10.Yu D, Sonderman J, Buchowski MS, McLaughlin JK, Shu X, Steinwandel M, Signorello LB, Zhang X, Hargreaves MK, Blot WJ, Zheng W. Healthy Eating and Risks of Total and Cause-Specific Death among Low-Income Populations of African-Americans and Other Adults in the Southeastern United States: A Prospective Cohort Study. PLoS Med 2015;12(5). doi: 10.1371/journal.pmed.1001830 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Tukey John W. Exploratory Data Analysis Vol 2.; 1977. [Google Scholar]
  • 12.Technical Guidance Documentation - Chronic Conditions Data Warehouse. Accessed November 9, 2020. https://www2.ccwdata.org/web/guest/technical-guidance-documentation
  • 13.So L, Evans D, Quan H. ICD-10 coding algorithms for defining comorbidities of acute myocardial infarction. BMC Health Serv Res 2006;6:161. doi: 10.1186/1472-6963-6-161 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.McCormick N, Lacaille D, Bhole V, Avina-Zubieta JA. Validity of heart failure diagnoses in administrative databases: a systematic review and meta-analysis. PLoS One 2014;9(8):e104519. doi: 10.1371/journal.pone.0104519 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Akwo EA, Kabagambe EK, Wang TJ, Harrell FE Jr, Blot WJ, Mumma M, Gupta DK, Lipworth L. Heart Failure Incidence and Mortality in the Southern Community Cohort Study. Circ Heart Fail 2017;10(3). doi: 10.1161/CIRCHEARTFAILURE.116.003553 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Levitan EB, Wolk A, Mittleman MA. Consistency with the DASH diet and incidence of heart failure. Arch Intern Med 2009;169(9):851–857. doi: 10.1001/archinternmed.2009.56 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Levitan EB, Wolk A, Mittleman MA. Relation of Consistency with the Dietary Approaches to Stop Hypertension Diet and Incidence of Heart Failure in Men Aged 45 to 79 Years. Am J Cardiol 2009;104(10):1416–1420. doi: 10.1016/j.amjcard.2009.06.061 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Campos CL, Wood A, Burke GL, Bahrami H, Bertoni AG. Dietary Approaches to Stop Hypertension Diet Concordance and Incident Heart Failure: The Multi-Ethnic Study of Atherosclerosis. Am J Prev Med 2019;56(6):819–826. doi: 10.1016/j.amepre.2018.11.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Del Gobbo LC, Kalantarian S, Imamura F, Lemaitre R, Siscovick DS, Psaty BM, Mozaffarian D. Contribution of Major Lifestyle Risk Factors for Incident Heart Failure in Older Adults. JACC Heart Fail 2015;3(7):520–528. doi: 10.1016/j.jchf.2015.02.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Oskarsson V, Sadr-Azodi O, Orsini N, Andrén-Sandberg Å, Wolk A. High dietary glycemic load increases the risk of non-gallstone-related acute pancreatitis: a prospective cohort study. Clin Gastroenterol Hepatol 2014;12(4):676–682. doi: 10.1016/j.cgh.2013.09.058 [DOI] [PubMed] [Google Scholar]
  • 21.De Moraes ACF, Carvalho HB, McClelland RL, Diez-Roux AV, Szklo M. Sex and ethnicity modify the associations between individual and contextual socioeconomic indicators and ideal cardiovascular health: MESA study. J Public Health 2019;41(3):e237–e244. doi: 10.1093/pubmed/fdy145 [DOI] [PubMed] [Google Scholar]
  • 22.Mozaffarian D, Kamineni A, Carnethon M, Djoussé L, Mukamal KJ, Siscovick D. Lifestyle Risk Factors and New-Onset Diabetes Mellitus in Older Adults: The Cardiovascular Health Study. Arch Intern Med 2009;169(8):798–807. doi: 10.1001/archinternmed.2009.21 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Santra S, Basu AK, Roychowdhury P, Banerjee R, Singhania P, Singh S, Datta UK. Comparison of left ventricular mass in normotensive type 2 diabetes mellitus patients with that in the nondiabetic population. J Cardiovasc Dis Res 2011;2(1):50–56. doi: 10.4103/0975-3583.78597 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Drazner MH. The Progression of Hypertensive Heart Disease. Circulation 2011;123(3):327–334. doi: 10.1161/CIRCULATIONAHA.108.845792 [DOI] [PubMed] [Google Scholar]
  • 25.Monsivais P, Rehm CD, Drewnowski A. The DASH Diet and Diet Costs Among Ethnic and Racial Groups in the United States. JAMA Intern Med 2013;173(20). doi: 10.1001/jamainternmed.2013.9479 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Mullie P, Clarys P, Hulens M, Vansant G. Dietary patterns and socioeconomic position. Eur J of Clin Nutr 2010;64(3):231–238. doi: 10.1038/ejcn.2009.145 [DOI] [PubMed] [Google Scholar]
  • 27.Ross JS, Bernheim SM, Bradley EH, Teng HM, Gallo WT. Use of preventive care by the working poor in the United States. Prev Med 2007;44(3):254–259. doi: 10.1016/j.ypmed.2006.11.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Stress and Health Disparities Report. https://www.apa.org. Accessed March 16, 2021. https://www.apa.org/pi/health-disparities/resources/stress-report
  • 29.Goldbohm RA, van ‘t Veer P, van den Brandt PA, van’t Hof MA, Brants HA, Sturmans F, Hermus RJ. Reproducibility of a food frequency questionnaire and stability of dietary habits determined from five annually repeated measurements. Eur J Clin Nutr 1995;49(6):420–429. [PubMed] [Google Scholar]
  • 30.Cui Q, Xia Y, Wu Q, Chang Q, Niu K, Zhao Y. A meta-analysis of the reproducibility of food frequency questionnaires in nutritional epidemiological studies. Int J Behav Nutr Phys Act 2021;18(1):12. doi: 10.1186/s12966-020-01078-4 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1769881-supplement-1.docx (141.7KB, docx)

RESOURCES