Skip to main content
NCBI home page
HHS Author Manuscripts logoLink to HHS Author Manuscripts
. Author manuscript; available in PMC: 2024 Mar 9.
Published in final edited form as: Am J Prev Med. 2020 Oct 1;59(5):e197–e206. doi: 10.1016/j.amepre.2020.06.013

Ultra-Processed Foods and Excess Heart Age Among U.S. Adults

Quanhe Yang 1, Zefeng Zhang 1, Euridice Martinez Steele 2,3, Latetia V Moore 4, Sandra L Jackson 1
PMCID: PMC10924299  NIHMSID: NIHMS1968205  PMID: 33012621

Abstract

Introduction:

A high percentage of total calories from ultra-processed foods has been associated with several cardiovascular disease risk factors. No study has examined the association between ultra-processed foods and heart age. This study examines the association between ultra-processed foods and excess heart age (difference between estimated heart age and chronological age) among U.S. adults.

Methods:

The National Health and Nutrition Examination Survey (2009–2016) data for participants aged 30–74 years without cardiovascular disease or stroke (n=12,640) was used. Ultra-processed food was assigned based on NOVA classification of food processing, with ultra-processed food being the highest level. This study estimated the usual percentage of calories from ultra-processed foods and used sex-specific Framingham heart age algorithms to calculate heart age. The multivariable linear or logistic regression was used to examine the association between ultra-processed foods and excess heart age or likelihood of excess heart age being ≥10 years. Data analyses were conducted in 2020.

Results:

The median usual percentage of calories from ultra-processed foods was 54.5% (IQR=45.8%–63.1%). Adjusted excess heart age increased from 7.0 years (95% CI=6.4, 7.6) in the lowest quintile (Q1) to 9.9 years (95% CI=9.2, 10.5) in the highest quintile (Q5) (p<0.001). Compared with Q1, AORs for excess heart age of ≥10 years were 1.16 (95% CI=1.08, 1.25) in Q2, 1.29 (95% CI=1.14, 1.46) in Q3, 1.43 (95% CI=1.20, 1.71) in Q4, and 1.66 (95% CI=1.29, 2.14) in Q5 (p<0.001). The pattern of association was largely consistent across subgroups.

Conclusions:

U.S. adults consumed more than half of total daily calories from ultra-processed foods. A higher percentage of calories from ultra-processed foods was associated with higher excess heart age and likelihood of excess heart age of ≥10 years.

INTRODUCTION

Ultra-processed foods (UPFs) are industrial food products that contain multiple ingredients and are manufactured through multiple sequences of processes to create the final products.1 UPFs are usually branded; ready-to-eat; high in added sugars, salt, and saturated fats; and often contain many food additives.2,3 The consumption of UPFs has increased significantly during the past few decades, especially among developed countries.1 UPFs represent an important part of food consumption in the U.S., providing more than half of total calories for U.S. adults, and these calories are typically nutritionally poor and inconsistent with diets recommended by U.S. nutrition guidelines.4-8 A recent inpatient crossover randomized trial concluded that a diet with a large proportion of UPFs increased energy intake and caused weight gain among participants.9 Many observational studies suggested that high consumption of UPFs is associated with several major cardiovascular disease (CVD) risk factors, such as obesity,4,10-12 hypertension,13 dyslipidemia,14 and metabolic syndrome.15 Several cohort studies showed that high consumption of UPFs was associated with increased risk of CVD16 and early death from all-cause mortality.2,17-19

Reducing UPF intake may help to reduce risk factors and improve prevention of heart disease and stroke, the leading causes of death and serious disability in the U. S.20 However, prevention of CVD through recommended lifestyle change is challenging, especially when few Americans meet guidelines for healthy diets.20,21 During efforts to prevent and manage CVD, multivariable prediction models were developed that use the individual’s CVD risk profile to estimate the absolute risk of developing a CVD event during the next 10 years.22-24 Model-predicted absolute CVD risk can be difficult for patients to understand and may provide false assurances for people with high lifetime, but low short-term, CVD risk. Therefore, the effectiveness of predicted CVD risk in promoting lifestyle changes or adherence to recommended treatment may be limited.25-27 To simplify risk communication, the Framingham Heart Study (FHS) introduced the concept of heart age (i.e., predicted age of a person’s vascular system based on an individual’s risk profile, which would be equal to the person’s chronological age if their risk factor profile were ideal).28 The difference between predicted heart age and one’s chronological age was defined as excess heart age (EHA), which provides a simplified way to describe the risk of developing CVD.25,28 This study assesses the association between percentage of total calories from UPFs and EHA using data from nationally representative samples of U.S. adults.

METHODS

Study Population

The National Health and Nutrition Examination Survey (NHANES) is a cross-sectional assessment of the U.S. population’s health with information gathered through interviews, medical examinations, and laboratory tests. The survey provides demographic and laboratory data for a nationally representative sample of non-institutionalized U.S. residents.29 This study pooled NHANES 2009–2016 data for stable estimates by demographic subgroups. Adults aged 30–74 years were included in accordance with the age range used in the FHS heart age calculation.28 Among 16,835 adults aged 30–74 years, this study excluded 108 pregnant women; 1,329 participants who reported a history of heart attack, stroke, or congestive heart failure; 1,679 with incomplete or unreliable data on first-day 24-hour dietary recall; 975 who had missing information on risk factors used in heart age calculation; and 104 who had missing values on covariates. This left 12,640 participants for analysis.

Measures

This study used 24-hour dietary recall to estimate intake of UPFs.30 All NHANES participants who received physical examinations provided first-day 24-hour dietary recall through in-person interviews at the mobile examination center, and 88.7% provided a second recall via telephone interview 3–10 days later. NHANES estimated participants’ nutrient intake from foods by using U.S. Department of Agriculture Food and Nutrient Databases for Dietary Studies for each 2-year NHANES cycle.

This study used the NOVA foods classification system that considers the nature, extent, and purpose of processing when categorizing foods and beverages into 4 groups: (1) unprocessed or minimally processed foods, (2) processed culinary ingredients, (3) processed foods, and (4) UPFs.1 This study focused on UPFs, the highest level of food processing. Examples of UPFs include sugarsweetened beverages, breads, packaged salty snacks, packaged cakes, and processed meats. Briefly, the U.S. Department of Agriculture Food and Nutrient Databases for Dietary Studies converts foods and beverages (consumed among participants) into gram amounts and determines their nutrient values by using 8-digit food codes. The NOVA system was applied to this classification to classify all foods and beverages into 4 groups. The classification procedures of the NOVA system have been described in detail elsewhere.4,6

Dietary data from a single 24-hour recall may not represent a participant’s usual intake because of day-to-day variations in diet, and use of such data may bias estimate of the association between nutrient intake and health outcomes because of the measurement errors.31 Therefore, the methods developed by the National Cancer Institute were used to estimate the usual percentage of total calories from UPFs.31,32 These methods require that some of the participants have multiple days of nutrient values to estimate the within- and between-individual variations; 88.7% of participants in this study had second-day recalls.31 The models for estimating usual UPFs included the following: age, sex, race/ethnicity (white, non-Hispanic; black, non-Hispanic; Mexican American; and others), educational level (less than high school, high school graduate, and more than high school), leisure-time physical activity (inactive, insufficient, and met recommended physical activity),33 poverty–income ratio (PIR, the ratio of household income to poverty threshold after accounting for inflation and family size [<1.3, 1.3–3.49, ≥3.5, missing, n=1,034]),34 total calorie intake (first-day recall), and day of the week recall (weekday vs weekend [Friday–Sunday]).

FHS provided the sex-specific laboratory-based and nonlaboratory-based Framingham Risk Score (FRS) to estimate 10-year risk of developing CVD for each participant.28 This study used the sex-specific nonlaboratory-based FRS, and the parameters from FRS models were used to calculate predicted heart age.28 Heart age is the age of a person’s heart based on their risk factor profile for heart attack and stroke. For example, a 50-year-old woman who smokes and has uncontrolled high blood pressure could have a heart age of 75 years and EHA of 25 years, whereas the same woman without smoking and with optimal blood pressure and normal weight could have a heart age of 47 years, that is, 3 years younger than her chronologic age. EHA represents an alternative way to express a person’s risk for developing CVD that may simplify CVD risk communication.25 Because most U.S. adults have 1 or more CVD risk factors, the average adult has a positive EHA (i. e., heart age is greater than chronological age).20,35 The nonlaboratory-based FRS model included the following 7 variables: age, sex, systolic blood pressure, hypertension treatment status (yes/no), smoking status (current/no), diabetes (yes/no),36 and BMI. The average systolic blood pressure with up to 3 measurements (98.8% had 3 measurements) was used. Hypertension treatment, smoking status, and diabetes were self-reported. BMI was calculated as measured weight in kilograms divided by height in meters squared. For sensitivity analysis, this study calculated the heart age by using the laboratory-based FRS (replacing BMI with total cholesterol and high-density lipoprotein cholesterol) (Appendix Tables 1 and 2, available online).

Statistical Analysis

This study estimated the weighted prevalence and means (95% CIs) of selected covariates and CVD risk factors used in FRS by sex and tested for significance between men and women based on t-tests for continuous variables and Wald F-tests for categorical variables. Multivariable linear regression was used to examine the association between EHA and UPFs. It is not recommended to classify the predicted usual percentage of calories from UPFs into categories (e.g., quintiles, because of potential misclassifications at the quintile boundaries).37 To present the results in quintile format, this study first used the restricted cubic spline in multivariable linear regression models with 4 knots (20th, 40th, 60th, and 80th percentiles) to examine the departure from a linear relationship between UPFs and EHA,38 and there was no evidence of departure from a linear relationship (p=0.16 for nonlinearity). Second, this study calculated the 10th, 30th, 50th, 70th, and 90th percentile distribution of the percentage of calories from UPFs as the middle value of each quintile,32 and then estimated the adjusted mean EHA associated with these UPF percentiles by multiplying the regression coefficient (β-coefficient) by the middle value of each quintile. This study estimated mean EHA adjusted for age, age squared, sex, and race/ethnicity and fully adjusted models, including additional covariates of education, physical activity, PIR, and total calorie intake. Multivariable logistic regression was used to examine the association between usual percentage of calories from UPFs and likelihood for an EHA of ≥10 years (1/0). Similar to the linear regression models, this study first examined the linear relationship between UPFs and EHA of ≥10 years (p=0.28 for nonlinearity) and then estimated the AORs by comparing the middle values of each quintile with the first quintile as reference (Q5, Q4, Q3, and Q2 vs Q1). This study also presented the stratified analyses by age group (<60 years vs ≥60 years), sex, race/ethnicity, educational level, physical activity, and PIR. This study tested for interactions between UPFs and selected covariates by including cross-product terms in the multivariable regression models based on Wald F-tests and presented false discovery rate adjusted p-values to account for multiple comparisons.39 All analyses were conducted in 2020 using SAS, version 9.4 and SUDAAN, version 11, which accounted for the survey’s complex sampling design.

RESULTS

Participant information is provided in Table 1. The mean age was 50 (range=30—74) years; the percentage of participants aged 30–44 years was higher among men than among women (p<0.001); more men than women met the recommended amount of physical activity (p<0.001); more women had a PIR <1.3 (p<0.001) and had more than high school education (p=0.026); and the percentage of black, non-Hispanic women was higher than that of men (p=0.002). For CVD risk factors, the mean systolic blood pressure, prevalence of diabetes, and current smoking were significantly higher among men than women (p<0.05), but the percentage of hypertension treatment was higher among women than men (p<0.001). The mean usual percentage of calories from UPFs was 54.5% (IQR=45.8%–63.1%) and was similar between men and women.

Table 1.

Characteristics of Participants Aged 30–74 Years by Sex, NHANES 2009–2016

Characteristics Total
n=12,640		 Men
n=6,059		 Women
n=6,581		 p-valuea
Demographics and covariates
 Age, years, mean (95% CI) 49.7 (49.2, 50.1) 49.0 (48.6, 49.5) 50.2 (49.7, 50.7) <0.001
 Age group in years, % (95% CI)
  30–44 36.9 (35.1, 38.7) 38.7 (36.7, 40.7) 35.3 (33.2, 37.4) <0.001
  45–59 39.9 (38.3, 41.4) 39.8 (37.8, 41.8) 39.9 (38.0, 41.8)
  60–74 23.3 (22.0, 24.4) 21.5 (19.9, 23.1) 24.8 (23.4, 26.2)
 Race or ethnicity, % (95% CI)
  White, non-Hispanic 68.8 (64.6, 71.7) 68.9 (65.4, 72.4) 67.4 (63.6, 71.3) 0.002
  Black, non-Hispanic 10.3 (8.7, 12.0) 9.5 (8.0, 11.0) 11.1 (9.2, 12.9)
  Mexican American 8.4 (6.4, 10.3) 8.7 (6.6, 10.7) 8.1 (6.1, 10.1)
  Other 13.2 (11.6, 14.8) 13.0 (11.2, 14.7) 13.4 (11.8, 15.1)
 Education, % (95% CI)
  Less than high school 14.7 (13.1, 16.4) 15.0 (13.2, 16.9) 14.4 (12.7, 16.2) 0.026
  High school graduate 20.6 (19.2, 22.0) 22.0 (20.2, 23.7) 19.3 (17.7, 20.9)
  More than high school 64.7 (62.3, 67.1) 63.0 (60.4, 65.6) 66.3 (63.6, 68.9)
 Physical activity, % (95% CI)b
  Met recommendation 32.6 (30.8, 34.3) 34.1 (32.0, 36.2) 31.2 (29.1, 33.2) <0.001
  Insufficient 15.6 (14.5, 16.8) 13.9 (12.2, 15.6) 17.2 (15.9, 18.5)
  Inactive 51.8 (49.8, 53.7) 52.0 (49.4, 54.6) 51.6 (49.4, 53.8)
 PIR, % (95% CI)c
  <1.3 18.1 (16.2, 20.0) 16.8 (14.8, 18.8) 19.3 (17.3, 21.3) <0.001
  1.3–3.49 30.8 (28.9, 32.7) 30.3 (28.3, 32.3) 31.3 (29.2, 33.4)
  ≥3.5 44.8 (42.1, 47.6) 46.8 (44.0, 49.6) 43.1 (40.1, 46.10)
CVD risk factors
 SBP, mmHg, mean (95% CI) 122.3 (121.7, 122.9) 124.3 (123.6, 124.9) 120.4 (119.8, 121.1) <0.001
 BMI, kg/m2, mean (95% CI) 29.4 (29.2, 29.6) 29.3 (29.0, 29.5) 29.5 (29.3, 29.8) 0.066
 Hypertension, % (95% CI)
  No 67.6 (66.2, 69.0) 67.5 (65.6, 69.5) 67.7 (66.2, 69.2) 0.898
  Yes 32.4 (31.0, 33.8) 32.5 (30.5, 34.3) 32.3 (30.8, 33.8)
  Participants taking medications for hypertension 71.0 (68.8, 73.1) 65.2 (62.1, 68.2) 76.3 (73.6, 79.1) <0.001
(n=4,474) (n=2,084) (n=2,390)
 Diabetes, % (95% CI)d
  No 86.9 (86.0, 87.8) 86.0 (84.6, 87.3) 87.7 (86.7, 88.7) 0.024
  Yes 13.1 (12.2, 14.0) 14.0 (12.7, 15.4) 12.3 (11.3, 13.3)
 Smoke, % (95% CI)
  Nonsmoker 80.8 (79.6, 82.0) 79.1 (77.6, 80.5) 82.4 (80.8, 84.0) 0.001
  Current 19.2 (18.0, 20.4) 20.9 (19.5, 22.4) 17.6 (16.0, 19.2)
 Calories from UPFs, median % (IQR)e 54.5 (45.8–63.1) 55.0 (48.4–61.7) 54.8 (47.8–61.4) 0.325f
Open in a new tab

Note: Boldface indicates statistical significance (p<0.05).

a

p-value for testing significance between men and women based on t-test for continuous variable and Ward-F-test for categorical variables.

b

Definition of physical activity: inactive, defined as <10 minutes per week; insufficient, defined as some activity but not enough to meet the recommended amount; met recommendation, defined as ≥150 minutes per week moderate or ≥75 minutes per week vigorous or an equivalent combination. Minutes of vigorous-intensity activity were given twice the credit of minutes of moderate-intensity activity to calculate the equivalent combination.

c

There were 1,034 participants with missing PIR.

d

Definition of diabetes: self-reported or healthcare provider diagnosis or fasting glucose ≥126 mg/dL or HbA1c concentration ≥6.5%.

e

Estimated usual percentage of calories from UPFs.

f

p-value for testing significance in median calories from UPFs between men and women based on Wilcoxon test.

CVD, cardiovascular disease; NHANES, National Health and Nutrition Examination Survey; PIR, poverty–income ratio; SBP, systolic blood pressure; UPF, ultra-processed food.

Table 2 presents the results for the linear regression models. Adjusted EHA increased from 7.0 (95% CI=6.4, 7.6) in the lowest quintile of UPF (Q1) to 9.9 (95% CI=9.2, 10.5) in Q5 (p<0.001). The association appeared to be stronger among women, increasing from 5.2 (95% CI=4.2, 6.2) in Q1 to 9.4 (95% CI=8.6, 10.3) in Q5 (p<0.001), but increased from 8.7 (95% CI=7.8, 9.6) in Q1 to 10.2 (95% CI=9.3, 11.1) in Q5 among men (p=0.064; false discovery rate adjusted p=0.067 for interaction between men and women). The pattern of association was largely consistent by age group, race/ethnicity, education, and PIR (p>0.05 for all interactions) (Figure 1A).

Table 2.

Adjusted Mean EHA According to Usual Percentage of Calories From UPFs—NHANES 2009–2016

Characteristic Mid-value of quintiles of usual percentage of calories from UPFs among U.S. adults p-valuea
Q1
38.1%		 Q2
47.8%		 Q3
54.5%		 Q4
61.2%		 Q5
70.9%
Range/usual percent, % 0 to <43.7 43.7 to <51.3 51.3 to <57.7 57.7 to <65.3 ≥65.3
Total
 Mean EHA adjusted for age, age squared, sex, and race or ethnicity only 5.4
(4.7, 6.0)		 7.0
(6.6, 7.4)		 8.1
(7.8, 8.5)		 9.3
(8.9, 9.7)		 10.9
(10.3, 11.5)		 <0.001
 Fully adjusted EHAb 7.0
(6.4, 7.6)		 7.8
(7.5, 8.2)		 8.4
(8.1, 8.7)		 9.0
(8.6, 9.4)		 9.9
(9.2, 10.5)		 <0.001
Menc
 Mean EHA adjusted for age, age suared, sex, and race or ethnicity only 7.4
(6.6, 8.2)		 8.4
(7.9, 8.9)		 9.1
(8.7, 9.5)		 9.8
(9.3, 10.3)		 10.8
(10.0, 11.7)		 <0.001
 Fully adjusted EHAb 8.7
(7.8, 9.6)		 9.1
(8.6, 9.7)		 9.4
(9.0, 9.9)		 9.7
(9.2, 10.3)		 10.2
(9.3, 11.1)		 0.064
Womenc
 Mean EHA adjusted for age, age squared, sex, and race or ethnicity only 3.2
(2.4, 4.1)		 5.5
(4.9, 6.0)		 7.0
(6.5, 7.5)		 8.5
(8.0, 9.1)		 10.8
(10.0, 11.7)		 <0.001
 Fully adjusted EHAb 5.2
(4.2, 6.2)		 6.4
(5.8, 7.1)		 7.3
(6.8, 7.8)		 8.2
(7.7, 8.7)		 9.4
(8.6, 10.3)		 <0.001
Open in a new tab

Note: Boldface indicates statistical significance (p<0.05). Data presented as mean (95% CI) unless otherwise noted.

a

p-value for testing significant association between the usual percentage of calories from UPFs and EHA based on t-test (β-coefficient); all tests are 2-tailed.

b

Adjusted for age, age squared, sex, race or ethnicity, educational attainment, physical activity level, poverty–income ratio, and total calorie.

c

FDR-adjusted p=0.067 for interaction between usual percentage of calories from UPFs and sex for association with EHA based on Wald F-test; all tests are 2-tailed.

EHA, excess heart age; FDR, false discovery rate; NHANES, National Health and Nutrition Examination Survey; Q, quintile; UPF, ultra-processed food.

Figure 1.

Figure 1.

Open in a new tab

Adjusted mean EHA (panel A) and AOR of risk for EHA ≥10 years (panel B) according to usual percentage of calories from UPFs by selected subgroups—NHANES 2009–2016. Q1 usual percentage of calories from UPFs serves as reference group. *FDR adjusted p-value for interaction by subgroups. EHA, excess heart age; FDR, false discovery rate; NHANES, National Health and Nutrition Examination Survey; Q, quintile; UPF, ultra-processed food.

Table 3 presents the results for the logistic regression models. Compared with those with the lowest intake of UPFs (Q1, 38.1% of calories from UPFs), AORs for an EHA of ≥10 years increased 29% for those with average consumption (Q3, 54.5% calories from UPFs) and 66% among those with the highest intake (Q5, 71% calories from UPFs) (p<0.001). Among men, having an EHA of ≥10 years increased 19% in Q3 to 41% in Q5 (p=0.088), and among women the corresponding numbers were 42% and 101%, respectively (p<0.001; false discovery rate adjusted p=0.785 for interaction between men and women). These associations were similar by age group, race/ethnicity, education, and PIR (p>0.05 for all interactions) (Figure 1B).

Table 3.

AOR of EHA ≥10 Years According to Usual Percentage of Calories From UPFs—NHANES 2009–2016

Characteristic Mid-value of quintiles of usual percentage of calories from UPFs among U.S. adults p-valuea
Q1
38.1%		 Q2
47.8%		 Q3
54.5%		 Q4
61.2%		 Q5
70.9%
Range/usual percent, % 0 to <43.7 43.7 to <51.3 51.3 to <57.7 57.7 to <65.3 ≥65.3
Total
 Mean EHA adjusted for age, age squared, sex, and race or ethnicity only 1.00 1.34
(1.25, 1.43)		 1.63
(1.45, 1.83)		 1.99
(1.69, 2.35)		 2.66
(2.10, 3.36)		 <0.001
 Fully adjusted EHAb 1.00 1.16
(1.08, 1.25)		 1.29
(1.13, 1.46)		 1.43
(1.19, 1.70)		 1.66
(1.29, 2.13)		 <0.001
Menc
 Mean EHA adjusted for age, age squared, sex, and race or ethnicity only 1.00 1.25
(1.12, 1.40)		 1.47
(1.22, 1.77)		 1.72
(1.32, 2.24)		 2.16
(1.48, 3.14)		 <0.001
 Fully adjusted EHAb 1.00 1.11
(0.98, 1.24)		 1.19
(0.97, 1.45)		 1.27
(0.96, 1.69)		 1.41
(0.94, 2.11)		 0.088
Womenc
 Mean EHA adjusted for age, age squared, sex, and race or ethnicity only 1.00 1.42
(1.30, 1.55)		 1.81
(1.56, 2.11)		 2.32
(1.87, 2.86)		 3.30
(2.44, 4.46)		 <0.001
 Fully adjusted EHAb 1.00 1.23
(1.12, 1.35)		 1.42
(1.21, 1.66)		 1.64
(1.31, 2.04)		 2.01
(1.47, 2.75)		 <0.001
Open in a new tab

Note: Boldface indicates statistical significance (p<0.05). Data presented as AOR (95% CI) unless otherwise noted.

a

p-value for testing significant association between the usual percentage of calories from UPFs and an EHA of ≥10 years based on t-test (β-coefficient); all tests are 2-tailed.

b

Adjusted for age, age squared, sex, race or ethnicity, educational attainment, physical activity level, poverty–income ratio, and total calorie.

c

FDR-adjusted p=0.785 for interaction between usual percentage of calories from UPFs and sex on risk of an EHA of ≥10 years based on Wald F-test; all tests are 2-tailed.

EHA, excess heart age; FDR, false discovery rate; NHANES, National Health and Nutrition Examination Survey; Q, quintile; UPF, ultra-processed food.

In sensitivity analysis using the laboratory-based FRS to estimate heart age, adjusted EHA increased from 3.3 years in Q1 to 5.5 years in Q5 (a 2.2-year difference in EHA) (p=0.001) (Appendix Table 1, available online), and adjusted likelihoods for an EHA of ≥10 years were 15% and 32% higher comparing Q3 and Q5 with Q1 UPF intake (p=0.019) (Appendix Table 2, available online).

DISCUSSION

The main findings from this nationally representative survey are that U.S. adults aged 30–74 years consumed on average more than half of their total daily calories from UPFs, and higher consumption of UPFs was associated with significantly increased EHA (increased EHA represents increased risk of developing CVD). U.S. adults in the highest quintile of UPF consumption had approximately 3 additional years of EHA compared with those in the lowest quintile and 66% increased risk for having an EHA ≥10 years. The pattern of association was largely consistent across age groups, sex, race/ethnicity, education, physical activity, and PIR.

Many studies have suggested that high consumption of UPFs is associated with several major CVD risk factors, including overweight and obesity,4,7,10-12 hypertension,13 increased total cholesterol, low-density lipoprotein cholesterol among children,14 and metabolic syndrome among children and adults.15,40 Numerous studies have reported that dietary factors play significant roles in the incidence and mortality of CVD and other noncommunicable diseases with an estimate of 33% to >40% of incidence or deaths from these conditions associated with unhealthy diets.41,42 Heart-healthy diets consist of fruits and vegetables; whole grains; legumes; nuts; fish; poultry; and limited intake of added sugars, sodium, and saturated fat.8 By contrast, UPFs are typically energy-dense; high in added sugar, sodium, saturated or transfats; and low in dietary fiber and micronutrients.2,6,43,44 In addition, UPFs contain classes of food additives to make the final food products palatable or more appealing and to extend shelf-life. Several animal and observational studies have indicated that cumulative exposure and interactions of multiple food additives may be associated with increased CVD risk factors,16 such as lipid profiles in mice45 and humans,46 inflammation and metabolic syndrome in mice,47 and glucose intolerance and insulin resistance in humans.48 Other studies suggest that the highly refined carbohydrate, added sugar, or fat content of UPFs might produce changes in reward neurocircuitry, leading to addictive-like eating behaviors and overconsumption.49,50 A recent trial revealed that a diet with a large proportion of UPFs significantly increased energy intake and caused weight gain among the adult participants.9 This study showed that high consumption of UPFs was associated with increased EHA, consistent with the findings of other studies on the association between UPFs and CVD risk factors4,7,10-15,40 and increased risk of developing CVD or all-cause mortality.2,16-19 High intake of UPFs appears to be associated with multiple CVD risk factors and pathways to increase CVD risk; however, the exact biological mechanisms remain unclear. The clinical and policy implications regarding UPF consumption depend on better understanding of these pathways.

The major strengths of the study include use of a large, nationally representative sample of the U.S. population with comprehensive measurements of major CVD risk factors and sociodemographic data. This study used a measurement error model to estimate usual percentage of calories from UPFs accounting for within-person day-to-day variation.31 The NOVA is a novel food classification system, and UPF is a promising dietary quality indicator that is relatively simple to understand. In addition, heart age is an alternative and simplified way to express the predicted risk of developing CVD. Risk prediction has played an important role in the prevention of CVD.22 However, identifying effective approaches to communicate CVD risk to patients for lifestyle changes and to support the recommended treatments for CVD prevention remains a challenge.51,52 Studies suggest that heart age might be an effective way to communicate individual-level risk for developing CVD and encourage actions to adopt heart-healthy lifestyles.26,53 Expressing increased heart age in association with high consumption of UPFs may simplify CVD risk communication and motivate more people to adopt heart-healthy diets and lifestyles.

Limitations

First, although NHANES 24-hour dietary recall databases contain some information indicative of food processing, the degree of processing could not be determined consistently for all food items and may result in errors in NOVA classification. This study used a conservative approach, such that the lower level of processing was assigned in case of uncertainty. Therefore, potential misclassifications would lead to underestimation of UPF consumption. Second, studies showed that intake of total calories is under-reported in NHANES,54,55 with different levels of under-reporting by BMI status (i.e., approximately 3%, 15%, and 20% by normal, overweight, and people with obesity, respectively).55 Using the percentage of energy contribution of UPF as the exposure variable may reduce the bias introduced by misreporting as long as the participants nondifferentially misreported calorie intakes from all foods. However, differential under-reporting of calorie intakes by NOVA groups could result in underestimating UPF consumption and attenuating the strength of association. Third, FHS consisted of predominantly white, non-Hispanic adults, and the heart age algorithm may not apply to other racial or ethnic groups. Although the Pooled Cohort Risk Equations included white, non-Hispanic and black, non-Hispanic adults in CVD risk prediction, the parameters for calculating heart age were not available.22 Fourth, FHS developed the general CVD risk score for heart age on the basis of 1967 to 1987 cohorts.28 This study used non-laboratory-based FRS to estimate heart age to be consistent with previous publications.25,56,57 The laboratory-based FRS provided different estimated heart age in this population because of the declining trend of total cholesterol, increasing trend of statin use,58-60 and increasing trend of obesity and diabetes since the 1980s.61-63 In sensitivity analyses, although the pattern of associations remained unchanged, the magnitude of EHA by using laboratory-based FRS was less than the nonlaboratory-based FRS. Fifth, NHANES consists of cross-sectional representative surveys that are subject to the potential for reverse causality between UPF consumption and CVD risk factors, should patients improve their diet after developing symptoms or disease. Sixth, as lifestyle risk factors tend to cluster,64 higher UPF consumption could be a proxy of an overall unhealthy diet or lifestyle, and subsequent residual confounding could overestimate the strength of the association. Furthermore, as a cross-sectional analysis without follow-up CVD outcomes, causal association between UPF and CVD could not be determined. However, the results of this study are consistent with several cohort studies suggesting that high consumption of UPFs was associated with increased incidence of CVD and all-cause mortality.16-19

CONCLUSIONS

U.S. adults aged 30–74 years consumed more than half of total daily calories from UPFs, and a high percentage of calories from UPFs was associated with higher estimated EHA and likelihood of an EHA of ≥10 years. Discussing consumption of UPFs in the context of elevated EHA may be an effective approach for clinicians to communicate CVD risk with patients and to enhance motivation for preventing CVD.

Supplementary Material

Supplemental Tables

ACKNOWLEDGMENTS

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

Euridice Martinez Steele has a research grant from the State of São Paulo Research Foundation in Brazil (number 2018/17972-9). No other financial disclosures were reported by the authors of this paper.

Footnotes

SUPPLEMENTAL MATERIAL

Supplemental materials associated with this article can be found in the online version at https://doi.org/10.1016/j.amepre.2020.06.013.

REFERENCES

  • 1.Monteiro CA, Cannon G, Moubarac JC, Levy RB, Louzada MLC, Jaime PC. The UN Decade of Nutrition, the NOVA food classification and the trouble with ultra-processing. Public Health Nutr. 2018;21 (1):5–17. 10.1017/S1368980017000234. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Monteiro CA, Cannon G, Lawrence M, da Costa Louzada ML, Pereira Machado P. Ultra-processed foods, diet quality, and health using the NOVA classification system. Rome, Italy: Food and Agriculture Organization of the United Nations. http://www.fao.org/3/ca5644en/ca5644en.pdf. Published 2019. Accessed June 24, 2020. [Google Scholar]
  • 3.Monteiro CA, Cannon G, Levy RB, et al. Ultra-processed foods: what they are and how to identify them. Public Health Nutr. 2019;22 (5):936–941. 10.1017/S1368980018003762. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Juul F, Martinez-Steele E, Parekh N, Monteiro CA, Chang VW. Ultraprocessed food consumption and excess weight among U.S. adults. Br J Nutr. 2018;120(1):90–100. 10.1017/S0007114518001046. [DOI] [PubMed] [Google Scholar]
  • 5.Baraldi LG, Martinez Steele E, Canella DS, Monteiro CA. Consumption of ultra-processed foods and associated sociodemographic factors in the USA between 2007 and 2012: evidence from a nationally representative cross-sectional study. BMJ Open. 2018;8(3):e020574. 10.1136/bmjopen-2017-020574. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Martínez Steele E, Baraldi LG, Louzada ML, Moubarac JC, Mozaffarian D, Monteiro CA. Ultra-processed foods and added sugars in the U.S. diet: evidence from a nationally representative cross-sectional study. BMJ Open. 2016;6(3):e009892. 10.1136/bmjopen-2015-009892. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Monteiro CA, Moubarac JC, Levy RB, Canella DS, Louzada MLDC, Cannon G. Household availability of ultra-processed foods and obesity in nineteen European countries. Public Health Nutr. 2018;21 (1):18–26. 10.1017/S1368980017001379. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.HHS. U.S. Department of Agriculture. 2015–2020 dietary guidelines for Americans. 8th edition Washington, DC: HHS, U.S. Department of Agriculture; 2015. https://health.gov/dietaryguidelines/2015/guide-lines/. Published 2015. Accessed September 18, 2019. [Google Scholar]
  • 9.Hall KD, Ayuketah A, Brychta R, et al. Ultra-processed diets cause excess calorie intake and weight gain: an inpatient randomized controlled trial of ad libitum food intake. Cell Metab. 2019;30(1):67–77. e3. 10.1016/j.cmet.2019.05.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Mendonça RD, Pimenta AM, Gea A, et al. Ultraprocessed food consumption and risk of overweight and obesity: the University of Navarra Follow-Up (SUN) cohort study. Am J Clin Nutr. 2016;104 (5):1433–1440. 10.3945/ajcn.116.135004. [DOI] [PubMed] [Google Scholar]
  • 11.Nardocci M, Leclerc BS, Louzada ML, Monteiro CA, Batal M, Moubarac JC. Consumption of ultra-processed foods and obesity in Canada. Can J Public Health. 2019;110(1):4–14. 10.17269/s41997-018-0130-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Canhada SL, Luft VC, Giatti L, et al. Ultra-processed foods, incident overweight and obesity, and longitudinal changes in weight and waist circumference: the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil). Public Health Nutr. 2020;23(6):1076–1086. 10.1017/S1368980019002854. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Mendonça RD, Lopes AC, Pimenta AM, Gea A, Martinez-Gonzalez MA, Bes-Rastrollo M. Ultra-processed food consumption and the incidence of hypertension in a Mediterranean cohort: the Seguimiento Universidad de Navarra Project. Am J Hypertens. 2017;30(4):358–366. 10.1093/ajh/hpw137. [DOI] [PubMed] [Google Scholar]
  • 14.Rauber F, Campagnolo PD, Hoffman DJ, Vitolo MR. Consumption of ultra-processed food products and its effects on children’s lipid profiles: a longitudinal study. Nutr Metab Cardiovasc Dis. 2015;25 (1):116–122. 10.1016/j.numecd.2014.08.001. [DOI] [PubMed] [Google Scholar]
  • 15.Martínez Steele E, Juul F, Neri D, Rauber F, Monteiro CA. Dietary share of ultra-processed foods and metabolic syndrome in the U.S. adult population. Prev Med. 2019;125:40–48. 10.1016/j.ypmed.2019.05.004. [DOI] [PubMed] [Google Scholar]
  • 16.Srour B, Fezeu LK, Kesse-Guyot E, et al. Ultra-processed food intake and risk of cardiovascular disease: prospective cohort study (Nutri-Net-Santé). BMJ. 2019;365:l1451. 10.1136/bmj.l1451. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Schnabel L, Kesse-Guyot E, Allès B, et al. Association between ultra-processed food consumption and risk of mortality among middle-aged adults in France. JAMA Intern Med. 2019;179(4):490–498. 10.1001/jamainternmed.2018.7289. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Rico-Campà A, Martínez-González MA, Alvarez-Alvarez I, et al. Association between consumption of ultra-processed foods and all cause mortality: SUN prospective cohort study. BMJ. 2019;365:l1949. 10.1136/bmj.l1949. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Blanco-Rojo R, Sandoval-Insausti H, López-Garcia E, et al. Consumption of ultra-processed foods and mortality: a national prospective cohort in Spain. Mayo Clin Proc. 2019;94(11):2178–2188. 10.1016/j.mayocp.2019.03.035. [DOI] [PubMed] [Google Scholar]
  • 20.Benjamin EJ, Muntner P, Alonso A, et al. Heart disease and stroke statistics-2019 update: a report from the American Heart Association. Circulation. 2019;139(10):e56–e528. 10.1161/CIR.0000000000000659. [DOI] [PubMed] [Google Scholar]
  • 21.Eckel RH, Jakicic JM, Ard JD, et al. 2013 AHA/ACC guideline on lifestyle management to reduce cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation. 2014;129(25 suppl 2):S76–S99. 10.1161/01.cir.0000437740.48606.d1. [DOI] [PubMed] [Google Scholar]
  • 22.Goff DC Jr, Lloyd-Jones DM, Bennett G, et al. 2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation. 2014;129(25 suppl 2):S49–S73. 10.1161/01.cir.0000437741.48606.98. [DOI] [PubMed] [Google Scholar]
  • 23.Siontis GC, Tzoulaki I, Siontis KC, Ioannidis JP. Comparisons of established risk prediction models for cardiovascular disease: systematic review. BMJ. 2012;344:e3318. 10.1136/bmj.e3318. [DOI] [PubMed] [Google Scholar]
  • 24.Wilson PW, D’Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation. 1998;97(18):1837–1847. 10.1161/01.cir.97.18.1837. [DOI] [PubMed] [Google Scholar]
  • 25.Yang Q, Zhong Y, Ritchey M, et al. Vital signs: predicted heart age and racial disparities in heart age among U.S. adults at the state level. MMWR Morb Mortal Wkly Rep. 2015;64(34):950–958. 10.15585/mmwr.mm6434a6. [DOI] [PubMed] [Google Scholar]
  • 26.Soureti A, Hurling R, Murray P, van Mechelen W, Cobain M. Evaluation of a cardiovascular disease risk assessment tool for the promotion of healthier lifestyles. Eur J Cardiovasc Prev Rehabil. 2010;17(5):519–523. 10.1097/HJR.0b013e328337ccd3. [DOI] [PubMed] [Google Scholar]
  • 27.Groenewegen KA, den Ruijter HM, Pasterkamp G, Polak JF, Bots ML, Peters SA. Vascular age to determine cardiovascular disease risk: a systematic review of its concepts, definitions, and clinical applications. Eur J Prev Cardiol. 2016;23(3):264–274. 10.1177/2047487314566999. [DOI] [PubMed] [Google Scholar]
  • 28.D’Agostino RB Sr, Vasan RS, Pencina MJ, et al. General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation. 2008;117(6):743–753. 10.1161/CIRCULATIONAHA.107.699579. [DOI] [PubMed] [Google Scholar]
  • 29.National Health and Nutrition Examination Survey. NHANES questionnaires, datasets, and related documentation. Centers for Disease Control and Prevention, National Center for Health Statistics. https://wwwn.cdc.gov/nchs/nhanes/default.aspx. Updated February 21, 2020. Accessed July 18, 2018. [Google Scholar]
  • 30.Ahluwalia N, Dwyer J, Terry A, Moshfegh A, Johnson C. Update on NHANES dietary data: focus on collection, release, analytical considerations, and uses to inform public policy. Adv Nutr. 2016;7(1):121–134. 10.3945/an.115.009258. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Tooze JA, Midthune D, Dodd KW, et al. A new statistical method for estimating the usual intake of episodically consumed foods with application to their distribution. J Am Diet Assoc. 2006;106(10):1575–1587. 10.1016/j.jada.2006.07.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Usual dietary intakes: the NCI method. National Cancer Institute: Division of Cancer Control & Population Sciences. https://epi.grants.cancer.gov/diet/usualintakes/method.html. Updated March 20, 2020. Accessed February 18, 2019. [Google Scholar]
  • 33.Physical Activity Guidelines Advisory Committee. 2018 Physical Activity Guidelines Advisory Committee scientific report. Washington, DC: U.S. Department of Health and Human Services. https://health.gov/paguidelines/second-edition/report/. Published 2018. Accessed October 30, 2019. [Google Scholar]
  • 34.Ogden CL, Lamb MM, Carroll MD, Flegal KM. Obesity and socioeconomic status in children and adolescents: United States, 2005–2008. NCHS Data Brief. 2010;(51):1–8. [PubMed] [Google Scholar]
  • 35.Yang Q, Cogswell ME, Flanders WD, et al. Trends in cardiovascular health metrics and associations with all-cause and CVD mortality among U.S. adults. JAMA. 2012;307(12):1273–1283. 10.1001/jama.2012.339. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Menke A, Casagrande S, Geiss L, Cowie CC. Prevalence of and trends in diabetes among adults in the United States, 1988–2012. JAMA. 2015;314(10):1021–1029. 10.1001/jama.2015.10029. [DOI] [PubMed] [Google Scholar]
  • 37.Kipnis V, Midthune D, Buckman DW, et al. Modeling data with excess zeros and measurement error: application to evaluating relationships between episodically consumed foods and health outcomes. Biometrics. 2009;65(4):1003–1010. 10.1111/j.1541-0420.2009.01223.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Desquilbet L, Mariotti F. Dose–response analyses using restricted cubic spline functions in public health research. Stat Med. 2010;29(9):1037–1057. 10.1002/sim.3841. [DOI] [PubMed] [Google Scholar]
  • 39.Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Series B Stat Methodol. 1995;57(1):289–300. 10.1111/j.2517-6161.1995.tb02031.x. [DOI] [Google Scholar]
  • 40.Tavares LF, Fonseca SC, Garcia Rosa ML, Yokoo EM. Relationship between ultra-processed foods and metabolic syndrome in adolescents from a Brazilian family doctor program. Public Health Nutr. 2012;15 (1):82–87. 10.1017/S1368980011001571. [DOI] [PubMed] [Google Scholar]
  • 41.Micha R, Peñalvo JL, Cudhea F, Imamura F, Rehm CD, Mozaffarian D. Association between dietary factors and mortality from heart disease, stroke, and type 2 diabetes in the United States. JAMA. 2017;317 (9):912–924. 10.1001/jama.2017.0947. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Anand SS, Hawkes C, de Souza RJ, et al. Food consumption and its impact on cardiovascular disease: importance of solutions focused on the globalized food system: a report from the workshop convened by the World Heart Federation. J Am Coll Cardiol. 2015;66(14):1590–1614. 10.1016/j.jacc.2015.07.050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Martínez Steele E, Popkin BM, Swinburn B, Monteiro CA. The share of ultra-processed foods and the overall nutritional quality of diets in the U.S.: evidence from a nationally representative cross-sectional study. Popul Health Metr. 2017;15(1):6. 10.1186/s12963-017-0119-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Moubarac JC, Batal M, Louzada ML, Martinez Steele E, Monteiro CA. Consumption of ultra-processed foods predicts diet quality in Canada. Appetite. 2017;108:512–520. 10.1016/j.appet.2016.11.006. [DOI] [PubMed] [Google Scholar]
  • 45.Singh K, Ahluwalia P. Effect of monosodium glutamate on lipid peroxidation and certain antioxidant enzymes in cardiac tissue of alcoholic adult male mice. J Cardiovasc Dis Res. 2012;3(1):12–18. 10.4103/0975-3583.91595. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Jang W, Jeoung NH, Cho KH. Modified apolipoprotein (apo) A-I by artificial sweetener causes severe premature cellular senescence and atherosclerosis with impairment of functional and structural properties of apoA-I in lipid-free and lipid-bound state. Mol Cells. 2011;31 (5):461–470. 10.1007/s10059-011-1009-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Chassaing B, Koren O, Goodrich JK, et al. Dietary emulsifiers impact the mouse gut microbiota promoting colitis and metabolic syndrome. Nature. 2015;519(7541):92–96. 10.1038/nature14232. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Pepino MY, Tiemann CD, Patterson BW, Wice BM, Klein S. Sucralose affects glycemic and hormonal responses to an oral glucose load. Diabetes Care. 2013;36(9):2530–2535. 10.2337/dc12-2221. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Carter A, Hendrikse J, Lee N, et al. The neurobiology of “food addiction” and its implications for obesity treatment and policy. Annu Rev Nutr. 2016;36:105–128. 10.1146/annurev-nutr-071715-050909. [DOI] [PubMed] [Google Scholar]
  • 50.Schulte EM, Avena NM, Gearhardt AN. Which foods may be addictive? The roles of processing, fat content, and glycemic load. PLoS One. 2015;10(2):e0117959. 10.1371/journal.pone.0117959. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Usher-Smith JA, Silarova B, Schuit E, Moons KG, Griffin SJ. Impact of provision of cardiovascular disease risk estimates to healthcare professionals and patients: a systematic review. BMJ Open. 2015;5(10):e008717. 10.1136/bmjopen-2015-008717. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Waldron CA, van der Weijden T, Ludt S, Gallacher J, Elwyn G. What are effective strategies to communicate cardiovascular risk information to patients? A systematic review. Patient Educ Couns. 2011;82(2):169–181. 10.1016/j.pec.2010.04.014. [DOI] [PubMed] [Google Scholar]
  • 53.Lopez-Gonzalez AA, Aguilo A, Frontera M, et al. Effectiveness of the Heart Age tool for improving modifiable cardiovascular risk factors in a southern European population: a randomized trial. Eur J Prev Cardiol. 2015;22(3):389–396. 10.1177/2047487313518479. [DOI] [PubMed] [Google Scholar]
  • 54.Archer E, Hand GA, Blair SN. Validity of U.S. nutritional surveillance: National Health and Nutrition Examination Survey caloric energy intake data, 1971–2010. PLoS One. 2013;8(10):e76632. 10.1371/journal.pone.0076632. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Moshfegh AJ, Rhodes DG, Baer DJ, et al. The U.S. Department of Agriculture Automated Multiple-Pass Method reduces bias in the collection of energy intakes. Am J Clin Nutr. 2008;88(2):324–332. 10.1093/ajcn/88.2.324. [DOI] [PubMed] [Google Scholar]
  • 56.Mpofu JJ, Smith RA, Patel D, et al. Disparities in the prevalence of excess heart age among women with a recent live birth. J Womens Health (Larchmt). 2020;29(5):703–712. 10.1089/jwh.2018.7564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Tabaei BP, Chamany S, Perlman S, Thorpe L, Bartley K, Wu WY. Heart age, cardiovascular disease risk, and disparities by sex and race/ethnicity among New York City adults. Public Health Rep. 2019;134 (4):404–416. 10.1177/0033354919849881. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Carroll MD, Kit BK, Lacher DA, Shero ST, Mussolino ME. Trends in lipids and lipoproteins in U.S. adults, 1988–2010. JAMA. 2012;308 (15):1545–1554. 10.1001/jama.2012.13260. [DOI] [PubMed] [Google Scholar]
  • 59.Rosinger A, Carroll MD, Lacher D, Ogden C. Trends in total cholesterol, triglycerides, and low-density lipoprotein in U.S. adults, 1999–2014. JAMA Cardiol. 2017;2(3):339–341. 10.1001/jamacardio.2016.4396. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Salami JA, Warraich H, Valero-Elizondo J, et al. National trends in statin use and expenditures in the U.S. adult population from 2002 to 2013: insights from the Medical Expenditure Panel Survey. JAMA Cardiol. 2017;2(1):56–65. 10.1001/jamacardio.2016.4700. [DOI] [PubMed] [Google Scholar]
  • 61.Flegal KM, Carroll MD, Kuczmarski RJ, Johnson CL. Overweight and obesity in the United States: prevalence and trends, 1960–1994. Int J Obes Relat Metab Disord. 1998;22(1):39–47. 10.1038/sj.ijo.0800541. [DOI] [PubMed] [Google Scholar]
  • 62.Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity among U.S. adults, 1999–2008. JAMA. 2010;303(3):235–241. 10.1001/jama.2009.2014. [DOI] [PubMed] [Google Scholar]
  • 63.Flegal KM, Kruszon-Moran D, Carroll MD, Fryar CD, Ogden CL. Trends in obesity among adults in the United States, 2005 to 2014. JAMA. 2016;315(21):2284–2291. 10.1001/jama.2016.6458. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 64.Schuit AJ, van Loon AJ, Tijhuis M, Ocké M. Clustering of lifestyle risk factors in a general adult population. Prev Med. 2002;35(3):219–224. 10.1006/pmed.2002.1064. [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supplemental Tables
NIHMS1968205-supplement-Supplemental_Tables.docx (48.2KB, docx)

RESOURCES