ArticlePDF Available

Abstract and Figures

Background Recent population dietary studies indicate that diets rich in ultra-processed foods, increasingly frequent worldwide, are grossly nutritionally unbalanced, suggesting that the dietary contribution of these foods largely determines the overall nutritional quality of contemporaneous diets. Yet, these studies have focused on individual nutrients (one at a time) rather than the overall nutritional quality of the diets. Here we investigate the relationship between the energy contribution of ultra-processed foods in the US diet and its content of critical nutrients, individually and overall. Methods We evaluated dietary intakes of 9,317 participants from 2009 to 2010 NHANES aged 1+ years. Food items were classified into unprocessed or minimally processed foods, processed culinary ingredients, processed foods, and ultra-processed foods. First, we examined the average dietary content of macronutrients, micronutrients, and fiber across quintiles of the energy contribution of ultra-processed foods. Then, we used Principal Component Analysis (PCA) to identify a nutrient-balanced dietary pattern to enable the assessment of the overall nutritional quality of the diet. Linear regression was used to explore the association between the dietary share of ultra-processed foods and the balanced-pattern PCA factor score. The scores were thereafter categorized into tertiles, and their distribution was examined across ultra-processed food quintiles. All models incorporated survey sample weights and were adjusted for age, sex, race/ethnicity, family income, and educational attainment. ResultsThe average content of protein, fiber, vitamins A, C, D, and E, zinc, potassium, phosphorus, magnesium, and calcium in the US diet decreased significantly across quintiles of the energy contribution of ultra-processed foods, while carbohydrate, added sugar, and saturated fat contents increased. An inverse dose–response association was found between ultra-processed food quintiles and overall dietary quality measured through a nutrient-balanced-pattern PCA-derived factor score characterized by being richer in fiber, potassium, magnesium and vitamin C, and having less saturated fat and added sugars. Conclusions This study suggests that decreasing the dietary share of ultra-processed foods is a rational and effective way to improve the nutritional quality of US diets.
Content may be subject to copyright.
R E S E A R C H Open Access
The share of ultra-processed foods and the
overall nutritional quality of diets in the US:
evidence from a nationally representative
cross-sectional study
Euridice Martínez Steele
1,2
, Barry M. Popkin
3
, Boyd Swinburn
4
and Carlos A. Monteiro
1,2*
Abstract
Background: Recent population dietary studies indicate that diets rich in ultra-processed foods, increasingly
frequent worldwide, are grossly nutritionally unbalanced, suggesting that the dietary contribution of these foods
largely determines the overall nutritional quality of contemporaneous diets. Yet, these studies have focused on
individual nutrients (one at a time) rather than the overall nutritional quality of the diets. Here we investigate the
relationship between the energy contribution of ultra-processed foods in the US diet and its content of critical
nutrients, individually and overall.
Methods: We evaluated dietary intakes of 9,317 participants from 2009 to 2010 NHANES aged 1+ years. Food items
were classified into unprocessed or minimally processed foods, processed culinary ingredients, processed foods,
and ultra-processed foods. First, we examined the average dietary content of macronutrients, micronutrients, and
fiber across quintiles of the energy contribution of ultra-processed foods. Then, we used Principal Component
Analysis (PCA) to identify a nutrient-balanced dietary pattern to enable the assessment of the overall nutritional
quality of the diet. Linear regression was used to explore the association between the dietary share of ultra-
processed foods and the balanced-pattern PCA factor score. The scores were thereafter categorized into tertiles,
and their distribution was examined across ultra-processed food quintiles. All models incorporated survey sample
weights and were adjusted for age, sex, race/ethnicity, family income, and educational attainment.
Results: The average content of protein, fiber, vitamins A, C, D, and E, zinc, potassium, phosphorus, magnesium,
and calcium in the US diet decreased significantly across quintiles of the energy contribution of ultra-processed
foods, while carbohydrate, added sugar, and saturated fat contents increased. An inverse doseresponse association
was found between ultra-processed food quintiles and overall dietary quality measured through a nutrient-
balanced-pattern PCA-derived factor score characterized by being richer in fiber, potassium, magnesium and vitamin
C, and having less saturated fat and added sugars.
Conclusions: This study suggests that decreasing the dietary share of ultra-processed foods is a rational and
effective way to improve the nutritional quality of US diets.
Keywords: NHANES, Ultra-processed, Dietary nutrient profile, PCA, Dietary patterns, Diet quality, Macronutrients,
Micronutrients
* Correspondence: carlosam@usp.br
1
Department of Nutrition, School of Public Health, University of São Paulo,
Av. Dr. Arnaldo, 715, 01246-907 São Paulo, Brazil
2
Center for Epidemiological Studies in Health and Nutrition, University of São
Paulo, São Paulo, Brazil
Full list of author information is available at the end of the article
© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Martínez Steele et al. Population Health Metrics (2017) 15:6
DOI 10.1186/s12963-017-0119-3
Background
Ultra-processed foods are formulations manufactured
using several ingredients and a series of processes (hence
ultra-processed). Most of their ingredients are lower-
cost industrial sources of dietary energy and nutrients,
and additives used for the purpose of imitating sensorial
qualities of minimally processed foods or of culinary
preparations of these foods, or to disguise undesirable
sensory qualities of the final product. They are made to
be hyper-palatable and attractive by the use of many
additives, with long shelf life, and are able to be con-
sumed anywhere, anytime. Ultra-processed foods include
but are not limited to soft drinks, sweet or savory
snacks, reconstituted meat products, and pre-prepared
frozen dishes [16].
In studies carried out in nationally representative sam-
ples of the Brazilian population it has been shown that
the group of ultra-processed foods have higher content
of free sugars, total fats, saturated fats, and trans fats,
and lower content of protein, fiber, and most micronutri-
ents than the rest of the diet, and that high consumption
of ultra-processed foods renders grossly nutritionally un-
balanced diets [79]. In Canada, similar results have been
documented regarding free sugars, total fats, protein, and
fiber [10]. In the US, using 20092010 National Health
and Nutrition Examination Survey (NHANES) day 1 data,
a positive association was found between the dietary
contribution of ultra-processed foods and the dietary con-
tent of added sugars [11]. Another US study found that
highly processed barcoded consumer packaged foods and
beverages, mostly ultra-processed products, are higher in
saturated fat, sugar, and sodium contents compared to
less-processed foods [12].
Based on the detrimental effects of ultra-processed
foods on the dietary content of critical nutrients and
taking into account their increasing predominance in
global food supplies [3, 6, 1316], the dietary share of
ultra-processed foods, expressed as a percentage of
total energy intake, has been proposed [1, 4, 17] and further
recognized by the United Nations Food and Agriculture
Organization [5], the Pan-American Health Organization
[6], and INFORMAS (International Network for Food and
Obesity/non-communicable diseases Research, Moni-
toring and Action Support) [18] as a potentially mean-
ingful determinant of the overall nutritional quality of
contemporaneous diets.
In order to further evaluate the influence of the dietary
share of ultra-processed foods on the nutritional dietary
quality we need to study its relationship with the overall
nutrient profile of diets. As several authors have pointed
out [1922], studying nutrients one at a time has a
number of drawbacks, which may be overcome by focus-
ing on dietary patterns [19, 2330]. Yet, to date, popula-
tion studies assessing the impact of ultra-processed food
consumption on the nutritional quality of diets have
focused on the dietary content of individual nutrients.
Dietary patterns can be derived using two approaches:
a priori or a posteriori [31]. A priori techniques use
scoring systems or overall measures of dietary quality
based on nutritional variables, generally foods and/or
nutrients, in order to assess the degree to which a par-
ticipant complies with a predefined theoretical dietary
pattern, created based on current nutrition knowledge.
Empirically derived dietary patterns, on the other hand,
are patterns derived a posteriori based on observed
dietary intake of the various foods and/or nutrients.
While a posteriori derived patterns may not necessarily
represent optimal dietary patterns, as they are outcome-
independent, a priori techniques are limited by the current
knowledge which may generate uncertainty regarding
which nutrients and cutoff points to use when generating
scores [19].
The objective of this study was to examine the relation-
ship between dietary contribution of ultra-processed foods
and the nutritional quality of the US diet through the evalu-
ation of dietary contents of critical nutrients individually
and also overall, using dietary pattern analysis.
Methods
Data source, population and sampling
We utilized nationally representative data from the
20092010 National Health and Nutrition Examination
Survey (NHANES), a continuous, nationally representative,
cross-sectional survey of non-institutionalized, civilian US
residents [32].
The survey included an interview conducted in the
home and a subsequent health examination performed
at a mobile examination center (MEC). All NHANES
examinees were eligible for two 24-h dietary recall inter-
views. The first dietary recall interview was collected
in-person in the MEC while the second was collected
by telephone three to ten days later. Dietary inter-
views were conducted by trained interviewers using
the validated [3335] US Department of Agriculture
Automated Multiple-Pass Method.
Among the 13,272 people screened in NHANES in
20092010, 10,537 (79.4%) participated in the household
interview and 10,253 (77.3%) also participated in the
MEC health examination. Of these, 9,754 individuals
provided one day of complete dietary intakes, and 8406
provided two daysworth.
We evaluated 9,317 survey participants aged 1 year and
above who had at least one day of 24-h dietary recall data
and had not been breast-fed on either of the two days. Data
for two recall days were used when available, and one day
otherwise. These 9,317 individuals had similar sociodemo-
graphic characteristics (gender, age, race/ethnicity, family
income, and educational attainment) to the full sample of
Martínez Steele et al. Population Health Metrics (2017) 15:6 Page 2 of 11
10,109 interviewed participants aged 1 year and above
(Additional file 1: Table S1).
Food classification according to processing
We classified all recorded food items (N=280,132 Food
Codes) according to Nova, a food classification based
on the extent and purpose of industrial food processing
[4, 17]. Nova includes four groups: unprocessed or
minimally processed foods(suchasfresh,dry,or
frozen fruits or vegetables; packaged grains and pulses;
grits, flakes, or flours made from corn, wheat, or
cassava; pasta, fresh or dry, made from flour and water;
eggs; fresh or frozen meat and fish and fresh or pas-
teurized milk); processed culinary ingredients(includ-
ing salt, vinegar, oils, fats, sugar, and other substances
extracted from foods and used in kitchens to season
and cook unprocessed or minimally processed foods
and to make culinary preparations), processed foods
(including pickled vegetables, fruit preserves, salted
meat products, canned fish in water or oil, cheeses,
artisan-style breads (no additives), and other ready-to-
consume products manufactured with the addition of
salt, vinegar, sugar, oil, or other substances of culinary
use to unprocessed or minimally processed foods), and
ultra-processed foods.
The Nova group of ultra-processed foods, of particular
interest in this study, includes soft drinks, sweet or
savory packaged snacks, confectionery and industrialized
desserts, mass-produced packaged breads and buns,
poultry and fish nuggets and other reconstituted meat
products, instant noodles and soups, and many other
ready-to-consume formulations of several ingredients.
Besides salt, sugar, oils, and fats, ultra-processed foods
ingredients include food substances not commonly used
in culinary preparations, and this is what distinguishes
them from processed foods. These ingredients include
modified starches, hydrogenated oils, protein isolates, and
additives whose purpose is to imitate sensorial qualities of
unprocessed or minimally processed foods and their culin-
ary preparations, or to disguise undesirable qualities of the
final product, such as colorants, flavorings, non-sugar
sweeteners, emulsifiers, humectants, sequestrants, and
firming, bulking, de-foaming, anti-caking, and glazing
agents. Unprocessed or minimally processed foods repre-
sent a small proportion of, or are even absent from, the
list of ingredients of ultra-processed products. A detailed
definition of each Nova food group and examples of food
items classified in each group are shown elsewhere [11].
The rationale underlying the classification is also
explained elsewhere [13, 36, 37].
For all food items (Food Codes) judged to be a hand-
made recipe (prepared from fresh or minimally proc-
essed foods and processed culinary ingredients), the
classification was applied to the underlying ingredients
(Standard Reference Codes -SR Codes-) obtained from
the USDA Food and Nutrient Database for Dietary Studies
(FNDDS) 5.0 [38]. More details in this regard have been
previously published [11].
Assessing energy and nutrient contents
For this study, we used Food Code nutrient values as
provided by NHANES.
For handmade recipes, we calculated the underlying
ingredient (SR Code) nutrient values using variables
from both FNDDS 5.0 [38] and USDA National Nutrient
Database for Standard Reference, Release 24 (SR24) [39].
The following nutrients were considered in this study:
protein, carbohydrates, added sugars, fats, saturated fats,
sodium, vitamins A (as retinol activity equivalents), C, D,
and E (as alpha-tocopherol), iron, zinc, potassium, phos-
phorus, magnesium, calcium, and fiber. These included
most underconsumed (vitamins A, C, D, and E, calcium,
magnesium, potassium, and fiber) and all overconsumed
(sodium, added sugar, and saturated fat) nutrients in the
US population [40].
Data on added sugars per Food Code and per SR Code
were obtained by merging the Food Patterns Equivalents
Database (FPED) 20092010 and Food Patterns Equiva-
lents Ingredients Database (FPID) 20092010 [41].
We used the following conversion factors: 4 kcal/g for
carbohydrates and protein, 9 kcal/g for fat and 7 kcal/g for
alcohol. Total energy intake was calculated as the sum of
calories from carbohydrates, proteins, fat, and alcohol.
Data analysis
We utilized all available dietary intake data for each par-
ticipant, using means of both recall days when available
(86% of participants) and one day otherwise.
Food items were sorted into mutually exclusive food
subgroups within each of the four Nova groups, as
shown in Table 1. First, we evaluated the contributions
of each food group and subgroup to total energy intake
and across quintiles of the dietary energy contribution of
ultra-processed foods (henceforth dietary share of ultra-
processed foods). The group of unprocessed or minim-
ally processed foods was also combined with the group
of processed culinary ingredients, as foods belonging to
these two groups are usually combined together in culin-
ary preparations and therefore consumed together.
We then compared the average dietary content of mac-
ronutrients (expressed as percent of total energy) and of
micronutrients and fiber (both expressed as g/1,000 kcal)
across quintiles of dietary share of ultra-processed foods.
Principal Component Analysis (PCA) is one of the
methods that can be used to empirically derive dietary
patterns. This is a mathematical technique that allows
reducing the complexity of interrelationships among ob-
served variables into a smaller number of uncorrelated
Martínez Steele et al. Population Health Metrics (2017) 15:6 Page 3 of 11
Table 1 Distribution (%) of the total daily per capita energy intake (kcal) according to NOVA food groups by quintiles of the dietary
share of ultra-processed foods, US population aged 1+ years (NHANES 20092010) (N=9,317)
Quintile of dietary share of ultra-processed foods
(% of total energy intake)
a
All quintiles
(n=9,317)
(2,069.9 kcal)
Q1
(n=1,941)
(1,970.9 kcal)
Q2
(n=1,903)
(2,017.6 kcal)
Q3
(n=1,791)
(2,061.8 kcal)
Q4
(n=1,785)
(2,151.5 kcal)
Q5
(n=1,897)
(2,147.7 kcal)
Unprocessed or minimally processed foods 30.2 48.3 36.7 29.4 23.3 13.2*
Meat (includes poultry) 8.0 11.6 9.6 8 6.7 4*
Fruit and freshly squeezed fruit juices 5.5 8.8 6.8 5.4 4.3 2.5*
Milk and plain yogurt 5.1 6.4 6.1 5.3 4.8 2.9*
Grains 2.9 6.3 3.4 2.3 1.6 0.7*
Roots and tubers 1.7 2.6 2.3 1.7 1.2 0.7*
Eggs 1.5 2.1 1.8 1.4 1.2 0.7*
Pasta 1.4 2.4 1.6 1.4 1.1 0.5*
Legumes 0.9 1.8 1.1 0.8 0.5 0.2*
Fish and seafood 0.8 1.5 1 0.7 0.4 0.2*
Vegetables 0.9 1.5 1 0.8 0.6 0.4*
Other unprocessed or minimally processed foods
b
1.7 3.2 2 1.5 1 0.5*
Processed culinary ingredients 2.9 4.9 3.4 2.9 2.2 1.2*
Sugar
c
1.1 1.6 1.3 1.1 0.9 0.6*
Plant oils 1.2 2.5 1.4 1.2 0.7 0.3*
Animal fats
d
0.5 0.7 0.6 0.6 0.5 0.2*
Other processed culinary ingredients
e
0.05 0.1 0.04 0.05 0.03 0.01
Unprocessed or minimally processed foods + Processed
culinary ingredients
33.1 53.2 40.1 32.4 25.4 14.5*
Processed foods 9.3 14.1 11.2 9.2 7.2 4.8*
Cheese 3.6 4.1 4.1 3.9 3.4 2.5*
Ham and other salted, smoked, or canned meat or fish 1.2 1.5 1.4 1.4 1.1 0.8
Vegetables and other plant foods preserved in brine 0.7 0.9 0.8 0.7 0.6 0.5*
Other processed foods
f
3.7 7.6 4.8 3.2 2.1 1*
Ultra-processed foods 57.5 32.6 48.6 58.4 67.3 80.7*
Breads 9.5 7.2 9.9 10.3 10.6 9.4*
Soft and fruit drinks
g
6.9 3 4.7 6.7 8.2 11.8*
Cakes, cookies, and pies 5.5 2.6 4.6 5.5 6.8 7.9*
Salty snacks 4.4 2.4 3.7 4.3 5.4 6.2*
Frozen and shelf-stable plate meals 3.9 1.3 2.2 3.7 5.2 7.3*
Pizza (ready-to-eat/heat) 3.3 0.5 1.4 2.6 4.1 7.8*
Breakfast cereals 3.1 2.2 3.2 3.6 3.5 3.1
Sauces, dressings, and gravies 2.5 2.4 2.7 2.7 2.8 2.1
Reconstituted meat or fish products 2.3 0.9 2.1 2.4 2.9 2.9*
Ice cream and ice pops 2.3 1.1 1.9 2.4 2.9 3*
Sweet snacks 2.3 1.1 2.1 2.4 2.7 3.4*
Milk-based drinks 1.9 1.1 1.7 1.9 2.1 2.6*
Desserts
h
1.8 1.3 1.9 2.1 2.1 1.8*
French fries and other potato products 1.7 0.4 1.1 1.7 1.9 3.5*
Sandwiches and hamburgers on bun (ready-to-eat/heat) 1.4 0.2 0.5 1.2 1.5 3.5*
Martínez Steele et al. Population Health Metrics (2017) 15:6 Page 4 of 11
linear combinations of them referred to as components
and which maximize the explained variance [19, 42].
Using PCA, through the correlation matrix applied to
the dietary content of macronutrients, micronutrients, and
fiber, we identified four nutrient dietary patterns in the
sample (Vitamin E was excluded because it loaded on all
main extracted components). The four patterns were se-
lected based on the Kaiser criterion (eigenvalue > 1.0), scree
plot, and PCA components interpretability. The compo-
nents were rotated using the varimax procedure and a
factor score was calculated for each of the four patterns.
PCA was conducted in the whole sample and stratify-
ing by age (15, 611, 1219, 2039, 4059, 60+ years),
sex, race/ethnicity (Mexican-American, Other Hispanic,
Non-Hispanic White, Non-Hispanic Black, Other Race),
ratio of family income to poverty line (0.001.30, >1.30
3.50, and >3.50) [32] and educational attainment of
respondents aged 20+ years or of household reference
person otherwise (<12, 12 years, and >12 years). Final
PCA results are presented for all strata combined be-
cause, despite some variations, comparable patterns were
observed across sociodemographic strata.
We used Gaussian regression to estimate the associ-
ation between the dietary share of ultra-processed
foods and the four component factor scores. To relax
the linearity assumption of the association, the dietary
contribution of ultra-processed foods variable was
transformed using restricted cubic splines with five
knots. The model was also fit using z-standardized
scores. The factor scores were then regressed on the
quintiles of the dietary share of ultra-processed foods.
Finally, factor scores were categorized into tertiles to
express low,middle, and high adherence to the dietary
pattern in order to examine the category distribution
across quintiles of the dietary share of ultra-processed
foods.
All regression models were adjusted for age, sex, race/
ethnicity, family income [32], and educational attain-
ment. As 908 participants had missing values on family
income and/or educational attainment, adjusted analyses
included 8,409 individuals.
NHANES survey sample weights were used in all
analyses except the PCA correlation matrix, to account
for differential probabilities of selection for the individ-
ual domains, nonresponse to survey instruments, and
differences between the final sample and the total US
population. The Taylor series linearization variance
approximation procedure was used to account for com-
plex sample design and sample weights [32]. Tests of
linear trend were performed to evaluate the effect of
quintiles as a single continuous variable.
To minimize chance findings from multiple compari-
sons, statistical hypotheses were tested using a two-
tailed p0.001 level of significance. Data were analyzed
using Stata version 12.1.
Results
Distribution of total energy intake according to food groups
and across quintiles of dietary share of ultra-processed foods
The average US daily energy intake in 20092010 was
2,069.9 kcal, 57.5% of calories coming from ultra-
processed foods, 30.2% from unprocessed or minimally
processed foods, 9.3% from processed foods and 2.9%
from processed culinary ingredients (Table 1). The
energy contribution of most subgroups belonging to
ultra-processed foods increased monotonically from the
first to the last quintile of the dietary share of ultra-
processed foods, with a few exceptions that showed a
slight decrease between the fourth and fifth quintiles.
An opposite trend was observed among subgroups from
all three remaining groups.
Table 1 Distribution (%) of the total daily per capita energy intake (kcal) according to NOVA food groups by quintiles of the dietary
share of ultra-processed foods, US population aged 1+ years (NHANES 20092010) (N=9,317) (Continued)
Instant and canned soups 0.9 0.7 0.8 0.9 0.9 1
Other ultra-processed foods
i
3.8 3.9 4 3.9 3.7 3.2
Total 100.0 100.0 100.0 100.0 100.0 100.0
a
Mean (range) dietary share of ultra-processed foods per quintile: 1st =32.6 (0 to 42.6); 2nd = 48.6 (42.6 to 54.0); 3rd = 58.4 (54.0 to 62.8); 4th = 67.3 (62.8 to 72.3);
5th = 80.7 (72.3 to 100)
b
Including nuts and seeds (unsalted); yeast; dried fruits (without added sugars) and vegetables; non pre-sweetened, non-whitened, non-flavored coffee and tea;
coconut water and meat; homemade soup and sauces; flours; tapioca
c
Including honey, molasses, maple syrup (100%)
d
Including butter, lard, and cream
e
Including starches; coconut and milk cream; unsweetened baking chocolate, cocoa powder, and gelatin powder; vinegar; baking powder and baking soda
f
Including salted or sugared nuts and seeds; peanut, sesame, cashew, and almond butter or spread; beer and wine
g
Including energy drinks, sports drinks, nonalcoholic wine
h
Including ready-to-eat and dry-mix desserts such as pudding
i
Including soy products such as meatless patties and fish sticks; baby food and baby formula; dips, spreads, mustard, and catsup; margarine; sugar substitutes,
sweeteners, and all syrups (excluding 100% maple syrup); distilled alcoholic drinks
*Significant linear trend across all quintiles (p< 0.001), both in unadjusted and models adjusted for sex, age group (1-5, 611, 1219, 2039, 4059, 60+ years),
race/ethnicity (Mexican-American, Other Hispanic, Non-Hispanic White, Non-Hispanic Black and Other Race Including Multi-Racial), ratio of family income to
poverty (SNAP 0.001.30, >1.303.50, and >3.50 and over), and educational attainment (<12, 12 years, and >12 years)
Martínez Steele et al. Population Health Metrics (2017) 15:6 Page 5 of 11
Nutrient dietary contents according to dietary share of
ultra-processed foods
The average dietary protein content decreased signifi-
cantly and monotonically across quintiles of the dietary
share of ultra-processed foods (from 17.9% of total energy
intake in the lowest quintile to 13.1% in the highest). The
content of alcohol evolved in a similar way (from 4.1% to
0.9% of total energy intake). In contrast, across the same
quintiles, there were significant increases in the content of
carbohydrates (from 46.5% to 53.4%), added sugars (7.7%
to 19.2%), and saturated fats (10.1% to 10.9%) (Table 2).
The average dietary content of fiber and of all micro-
nutrients except iron and sodium decreased significantly
and monotonically across quintiles of the dietary share
of ultra-processed foods: fiber (from 9.6 in the lowest
quintile to 6.7 g/1,000 kcal in the highest), vitamin A
(377.5 to 272.3 μg/1,000 kcal), vitamin C (58.2 to
32.4 mg/1,000 kcal), vitamin D (3.3 to 2.0 μg/1,000 kcal)
and vitamin E (4.1 to 3.3 mg/1,000 kcal), zinc (6.3 to
4.9 mg/1,000 kcal), potassium (1.6 to 1.0 g/1,000 kcal),
phosphorus (728.9 to 605.9 mg/1,000 kcal), magnesium
(173.3 to 117.3 mg/1,000 kcal), and calcium (531.1 to
464.7 mg/1,000 kcal). The sodium dietary content de-
creased non-significantly across quintiles of the dietary
share of ultra-processed foods (from 1.74 to 1.63 g/
1,000 kcal), while the iron content increased between
the first and third quintiles and decreased thereafter.
Nutrient dietary patterns obtained through PCA
Through PCA, four of 15 components had an eigen-
value >1.0 and explained 67% of the variance, and all
four were retained. The rotated factor loadings of these
four components are displayed in Table 3 (factor loadings
above 0.20 and below -0.20 have been highlighted).
The first component was characterized by being richer in
fiber, potassium, magnesium, and vitamin C, and having less
saturated fat and added sugars (variables with factor load-
ings above 0.20 or below -0.20). The factor loading for so-
dium was close to zero in this first component (0.04). This
component, called nutrient balanced pattern,wasselected
as an instrument to measure the quality of the diet overall.
Each of the three remaining components mixed
healthy and unhealthy features regarding dietary nutrient
contents. The second component indicated higher con-
tent in both saturated fat and micronutrients such as
calcium, vitamin D, phosphorus, and vitamin A and
lower content in sodium. The third showed higher
content in protein, saturated fat, and sodium and
Table 2 Indicators of the dietary content of macronutrients and micronutrients according to the dietary share of ultra-processed
foods, US population aged 1+ years (NHANES 20092010) (N=9,317)
Quintiles of dietary share of ultra-processed foods (% of total energy intake) [n]
a
Q1 [n=1,941] Q2 [n=1,903] Q3 [n=1,791] Q4 [n=1,785] Q5 [n=1,897]
Macronutrient Indicators (mean %
of total energy intake)
Protein 17.9 16.7 15.8 14.7 13.1*
Total carbohydrates 46.5 48.6 49.9 51.3 53.4*
Added sugars 7.7 11 13.4 15.7 19.2*
Total fats 31.4 32.2 32.5 32.6 32.5
Saturated fats 10.1 10.7 10.9 10.9 10.9*
Alcohol 4.1 2.4 1.8 1.4 0.9*
Micronutrient Indicators (mean density) Fiber (g/1,000 kcal) 9.6 8.9 8.2 7.4 6.7*
Sodium (g/1,000 kcal) 1.74 1.69 1.69 1.66 1.63
Vitamin A (μg/1,000 kcal) 377.5 358.5 347.4 306.2 272.3*
Vitamin C (mg/1,000 kcal) 58.2 51.4 42.9 40.3 32.4*
Vitamin D (μg/1,000 kcal) 3.3 3.2 2.9 2.5 2.0*
Vitamin E (mg/1,000 kcal) 4.1 3.8 3.6 3.5 3.3*
Iron (mg/1,000 kcal) 7.4 7.7 7.8 7.5 7.4
Zinc (mg/1,000 kcal) 6.3 6 5.8 5.4 4.9*
Potassium (g/1,000 kcal) 1.6 1.4 1.3 1.2 1.0*
Phosphorus (mg/1,000 kcal) 728.9 715.9 691.7 653.9 605.9*
Magnesium (mg/1,000 kcal) 173.3 156.6 144.3 130.6 117.3*
Calcium (mg/1,000 kcal) 531.1 539.6 532.2 507 464.7*
a
Mean (range) dietary share of ultra-processed foods per quintile: 1st =32.6 (0 to 42.6); 2nd = 48.6 (42.6 to 54.0); 3rd = 58.4 (54.0 to 62.8); 4th = 67.3 (62.8 to 72.3);
5th = 80.7 (72.3 to 100)
*Significant linear trend across all quintiles (p0.001), both in unadjusted and models adjusted for sex, age group (1-5, 611, 1219, 2039, 4059, 60+ years),
race/ethnicity (Mexican-American, Other Hispanic, Non-Hispanic White, Non-Hispanic Black and Other Race Including Multi-Racial), ratio of family income to
poverty (SNAP 0.001.30, >1.303.50, and >3.50 and over), and educational attainment (<12, 12 years, and >12 years).
Martínez Steele et al. Population Health Metrics (2017) 15:6 Page 6 of 11
phosphorus, and lower content in carbohydrates and
added sugars. The fourth presented higher content in iron,
zinc, vitamin A and sodium, and lower in vitamin C.
Comparable PCA patterns were observed across socio-
demographic strata. This was especially true for the
Nutrient balanced pattern as illustrated for race/ethni-
city strata in Additional file 1: Table S2.
Association between the dietary share of ultra-processed
foods and the nutrient balanced pattern
In unadjusted restricted cubic splines Gaussian regres-
sion analysis, a strong linear association was identified
between the dietary share of ultra-processed foods and
the nutrient balanced pattern factor score (coefficient
for linear term = -0.03, 95% CI: -0.04 to -0.02) (Fig. 1).
There was little evidence of nonlinearity in the restricted
cubic spline model (Wald test for linear term p< 0.001;
Wald test for all non-linear terms p= 0.16). The strength of
the association remained nearly the same after adjusting for
sex, age, race/ethnicity, family income, and educational
attainment (coefficient for linear term = -0.04, 95% CI: -0.05
to -0.03). According to the adjusted model, one standard
deviation increase in the dietary share of ultra-processed
foods leads to a 0.38 standard deviation decrease in the
nutrient balanced pattern factor score.
Across quintiles of the dietary share of ultra-processed
foods, the adjusted mean nutrient balanced pattern
factor score decreased monotonically, from 1.1 in the
lowest quintile to -0.9 in the highest (Tabl e 4 ).Across
the same quintiles, the proportion of individuals with
high adherence to the nutrient balanced pattern
decreased monotonically from 58.4% in the lowest
quintile of the dietary share of ultra-processed foods
to 11.0% in the highest. Inversely, the proportion of
individuals with low adherence increased from 13.3%
Table 3 Rotated factor loadings for the first four components from principal component analysis using nutrients, US population
aged 1+ years (NHANES 20092010) (N=9,317)
PC1 PC2 PC3 PC4
Indicator
a
(% expl.
b
= 20.4) (% expl. = 18.0) (% expl. = 17.7) (% expl. = 10.9)
Fiber density (g/1,000 kcal) 0.47
c
-0.12 0.00 0.09
Sodium density (g/1,000 kcal) 0.04 -0.22 0.39 0.20
Potassium density (mg/1,000 kcal) 0.44 0.15 0.10 -0.08
Iron density (mg/1,000 kcal) 0.02 0.00 -0.09 0.68
Zinc density (mg/1,000 kcal) -0.08 0.06 0.16 0.53
Phosphorus density (mg/1,000 kcal) 0.09 0.38 0.21 0.08
Magnesium density (mg/1,000 kcal) 0.44 0.05 0.11 0.05
Calcium density (mg/1,000 kcal) 0.02 0.55 -0.07 0.02
Vitamin A density (μg/1,000 kcal) 0.06 0.24 -0.09 0.24
Vitamin C density (mg/1,000 kcal) 0.40 0.07 -0.15 -0.21
Vitamin D density (μg/1,000 kcal) -0.02 0.55 -0.08 0.00
Protein (% of total energy) 0.05 0.03 0.45 0.14
Carbohydrate (% of total energy) 0.17 0.00 -0.54 0.17
Added sugars (% of total energy) -0.24 -0.03 -0.41 0.14
Saturated fat (% of total energy) -0.34 0.30 0.22 -0.16
a
For details on indicators, see Methods section
b
Proportion of the variance explained by each factor after orthogonal varimax rotation (Kaiser on)
c
Items with a factor loading above 0.20 or below -0.20 have been highlighted using boldface
-2 -1 0 1 2 3
PCA factor 1 score
27.3 46.8 58.1 68.4 85.2
% of total energy intak e from ultra-processed foods
95% CI predicted values
Fig. 1 Nutrient balanced patternfactor score regressed on the dietary
share of ultra-processed foods evaluated by restricted cubic splines, US
population aged 1+ years (NHANES 20092010) (N=9,317). Legend:
The values shown on the x-axis correspond to the 5th, 27.5th, 50th,
72.5th, and 95th percentiles for percentage of total energy from
ultra-processed foods (knots). Coefficient for linear term = -0.03,
95% CI: -0.04 to -0.02 (beta = -0.35). There was little evidence of
nonlinearity in the restricted cubic spline model (Wald test for
linear term p< 0.001; Wald test for all non-linear terms p= 0.16)
Martínez Steele et al. Population Health Metrics (2017) 15:6 Page 7 of 11
in the lowest quintile to 61.7% in the highest (overall
Chi square test p< 0.001).
The dietary share of ultra-processed foods also pre-
sented an inverse association with the remaining three
components (Additional file 1: Figure S1). The mean
factor scores of these three remaining components also
decreased across the dietary share of ultra-processed
foods (Additional file 1: Table S3).
Discussion
In this analysis of US nationally representative data, we
show that a significant linear inverse relationship exists
between the dietary contribution of ultra-processed
foods and the dietary content of protein, fiber, vitamins
A, C, D, and E, zinc, potassium, phosphorus, magne-
sium, and calcium. On the other hand, carbohydrate,
saturated fat, and added sugar contents increased signifi-
cantly with the dietary contribution of ultra-processed
foods. Only diets in the lowest quintile of ultra-
processed consumption had the average added sugar
content below the upper limit recommended by the
20152020 Dietary Guidelines for Americans [40], while
the average saturated fat content exceeded the same
limit in all quintiles, with the lowest quintile moving
closest to the recommendation.
We also found an inverse doseresponse association
between ultra-processed food dietary contribution and
the overall dietary quality measured through a nutrient
balanced pattern PCA-derived factor score characterized
by being richer in fiber, potassium, magnesium, and vita-
min C, and having less saturated fat and added sugars.
Furthermore, we found substantially higher adherence to
the nutrient balanced pattern in lower quintiles of ultra-
processed food dietary contribution than in higher ones.
These results are relevant because both individual edu-
cation interventions and food environment regulatory
policies have the potential to modify the dietary content
of ultra-processed foods. To our knowledge, this is the
first study to evaluate the association between the diet-
ary contribution of ultra-processed foods and the overall
nutritional quality of diets in the US.
The non-significant but somewhat unexpected sodium
content decrease across quintiles of the dietary share of
ultra-processed foods may be partly explained by the fact
that in the US processed foods include basically salty
products”–such as cheese, ham, or vegetables in brine
while most ultra-processed foods are either sweet prod-
ucts(soft, fruit, and milk drinks, cakes, cookies, breakfast
cereals, ice cream, sweet snacks, industrialized desserts) or
products containing both salt and sugar (breads, sauces,
canned soups, dressings, gravies, dips, spreads, mustard,
catsup). Still, the sodium dietary content was above the
Tolerable Upper Intake Level for any sex-age group [40]
regardless of the share of ultra-processed foods.
The not uncommon iron fortification of ultra-processed
foods or their ingredients may explain why the iron
content does not show the reverse gradient across quin-
tiles of ultra-processed food consumption seen among
other micronutrients.
Few studies have assessed the impact of levels of food
processing on the nutrient contents of the US diet. One
study [43] that applied a food-industry-supported classifi-
cation system [44] to NHANES 20032008 food intake
data found that, together, mixtures of combined ingredi-
entsand ready-to-eat,which are mostly ultra-processed
foods, contributed to 51% of total energy intake in the US
diet but to only 37% of the protein intake and to 73% of
the added sugar intake. These two food groups also con-
tributed to 37% of the fiber intake and to between 30%
and 60% of the intake of micronutrients [43]. Analyses of
the same data restricted to children and adolescents [45]
and to adults [46] showed similar results. Unfortunately,
these studies on data from NHANES 20032008 failed to
explore whether the dietary content of critical nutrients
Table 4 Nutrient balanced patternfactor score means and adherence according to the dietary share of ultra-processed foods, US
population aged 1+ years (NHANES 20092010)
Dietary share of ultra-processed foods (% of total energy intake) Nutrient balanced patternfactor score Adherence to Nutrient balanced pattern
b
Quintiles Mean (range) Mean Low (%) Middle (%) High (%)
unadj. (R2 = 0.18) adj.
a
(R2 = 0.24)
Q1 (n=1,941) 32.6 (0 to 42.6) 1.2* 1.1* 13.3 28.3 58.4
Q2 (n=1,903) 48.6 (42.6 to 54.0) 0.6* 0.5* 19.6 35.0 45.5
Q3 (n=1,791) 58.4 (54.0 to 62.8) 0.04 0.002 30.0 37.3 32.7
Q4 (n=1,785) 67.3 (62.8 to 72.3) -0.5* -0.4* 42.2 38.8 19.0
Q5 (n=1,897) 80.7 (72.3 to 100) -1.0*
¥
-0.9*
¥
61.7 27.4 11.0
a
Adjusted for sex, age group (1-5, 611, 1219, 2039, 4059, 60+ years), race/ethnicity (Mexican-American, Other Hispanic, Non-Hispanic White, Non-Hispanic
Black and Other Race Including Multi-Racial), ratio of family income to poverty (SNAP 0.001.30, >1.303.50, and >3.50 and over), and educational attainment
(<12, 12 years, and >12 years)
b
Nutrient balanced pattern(PC1) factor score tertiles: T1 (-4.7 to -0.9 points); T2 (-0.9 to 0.6 points); T3 (0.6 to 9.9 points)
*Statistically significant p0.001
¥
Significant linear trend across all quintiles (p0.001), both in unadjusted and models adjusted for sex, age group, race/ethnicity, ratio of family income to
poverty, and educational attainment
Martínez Steele et al. Population Health Metrics (2017) 15:6 Page 8 of 11
actually differed between high and low consumers of
mixtures of combined ingredientsplus ready-to-eat.
Another study evaluated US household barcoded
purchasing data from 2000 to 2012 using a classification
system guided by the one used in our study [12]. In
2012, the mean per capita purchase of highly processed
foods,a category similar to ultra-processed foods, had
higher adjusted median saturated fat, total sugar, and
sodium content than less processed foods.This report
did not capture non-barcoded items such as unpackaged
fresh fruit, vegetables, and meat, or highly processed
foods such as ready-to-eat store-prepared items, and did
not explore whether the dietary content of critical nutri-
ents actually differed between high and low consumers
of highly processed foods".
Consistent with our results, an investigation in Canada
using 2001 household purchasing data found a decrease in
protein content and fiber density across quintiles of the
energy share of ultra-processed foods, together with an
increase in the content of free sugars and total fats [10].
A study carried out in Brazil using 20082009 national
food intake data found that protein, fiber, sodium, and
potassium decreased significantly across quintiles of the
dietary contribution of ultra-processed foods, while free
sugars, total fats, and saturated fats increased [8]. After
adjusting for family income, there was a significant drop in
the dietary content of vitamin D, vitamin E, phosphorus,
magnesium, and zinc, and an increase in calcium [9].
Our study has several strengths. We studied a large,
nationally representative sample of the US population,
increasing generalizability. Our investigation was based
on total effective individual consumption data, rather
than on household purchasing data [7, 10, 47], which do
not evaluate the fraction of wasted food or purchases at
restaurants.
Potential limitations should be considered. As with
most population measures, dietary data obtained by 24-
h recalls are imperfect. However, 24-h recalls are the
least-biased self-report instrument available. Also, stan-
dardized methods and approach of NHANES have been
shown to produce accurate intake estimates [3335],
and will therefore be suitable for assessing food group
contributions and nutrient densities in the overall diet.
Although NHANES collects limited information indica-
tive of food processing (i.e., place of meals, product
brands), these data are not consistently determined for
all food items and this may lead to groups classification
errors. Also, as some authors have highlighted, the num-
ber of food items reported in NHANES is smaller than
the number available in the marketplace, and national
food composition data are not updated as required to
include all brand-specific products and to examine diet-
ary profiles sensitive to brand preferences [48]. The PCA
method also has limitations such as subjective decisions
regarding the number of extracted components, method
of rotation, naming of components, and cutoffs for factor
loadings [23, 31, 49].
Conclusions
This study suggests that decreasing the dietary share of
ultra-processed foods is a rational and effective way to
substantially improve dietary quality in the US.
Additional file
Additional file 1: Table S1. Characteristics of study participants and of
the full sample of interviewed participants aged 1 year and above, US
population aged 1+ years (NHANES 20092010). Table S2. Rotated factor
loadings for the first four components from principal component analysis
using nutrients, across race/ethnicity strata, US population aged 1+ years
(NHANES 20092010) (N=9,317). Table S3. PC2-PC4 score means and
adherence according to the dietary share of ultra-processed foods, US
population aged 1+ years (NHANES 20092010). Figure S1. PC2-PC4
factor scores regressed on the dietary share of ultra-processed foods
evaluated by restricted cubic splines, US population aged 1+ years
(NHANES 20092010) (N=9,317). (DOCX 1047 kb)
Abbreviations
FNDDS: USDA food and nutrient database for dietary studies; FPED: Food
patterns equivalents database; FPID: Food patterns equivalents
ingredients database; INFORMAS: International network for food and
obesity/non-communicable diseases research, monitoring and action
support; NHANES: National health and nutrition examination survey;
PCA: Principal component analysis; SR Codes: Standard reference codes;
USDA: National nutrient database for standard reference, release 24 (SR24)
Acknowledgments
Not applicable.
Funding
This research received funding from Conselho Nacional de Desenvolvimento
Científico e Tecnológico, Edital MCTI/CNPq/Universal (Processo CNPq nº
443477/2014-0) and from Fundação de Amparo à Pesquisa do Estado de São
Paulo (Processo FAPESP nº 2015/14900-9).
Availability of data and materials
Publicly available datasets have been used for this study.
Authorscontributions
CAM and EMS designed research; CAM and EMS analyzed data and
performed statistical analysis; CAM, EMS, BP, and BS wrote the paper and
CAM and EMS had primary responsibility for final content. All authors read
and approved the final manuscript.
Competing interests
The authors declare having no competing interests.
Consent for publication
Not applicable.
Ethics approval and consent to participate
Secondary publicly available data were used in this study.
Author details
1
Department of Nutrition, School of Public Health, University of São Paulo,
Av. Dr. Arnaldo, 715, 01246-907 São Paulo, Brazil.
2
Center for Epidemiological
Studies in Health and Nutrition, University of São Paulo, São Paulo, Brazil.
3
Department of Nutrition, University of North Carolina at Chapel Hill, Chapel
Hill, NC, USA.
4
School of Population Health, University of Auckland, Auckland,
New Zealand.
Martínez Steele et al. Population Health Metrics (2017) 15:6 Page 9 of 11
Received: 27 May 2016 Accepted: 10 January 2017
References
1. Monteiro CA. Nutrition and health. The issue is not food, nor nutrients, so
much as processing. Public Health Nutr. 2009;12(5):72931.
2. Ludwig DS. Technology, diet, and the burden of chronic disease. JAMA.
2011;305:13523.
3. Moodie R, Stuckler D, Monteiro C, Sheron N, Neal B, Thamarangsi T, Lincoln
P, Casswell S. Profits and pandemics: prevention of harmful effects of
tobacco, alcohol, and ultra-processed food and drink industries. Lancet.
2013;381(9867):6709. doi:10.1016/S0140-6736(12)62089-3.
4. Monteiro CA, Cannon G, Levy RB, Claro RM, Moubarac J-C. Ultra-processing
and a new classification of foods. In: Neff R, editor. Introduction to U.S. Food
System. Public Health, Environment, and Equity. Jossey Bass A Wiley Brand:
San Francisco; 2015. p. 3389.
5. FAO. Guidelines on the collection of information on food processing
through food consumption surveys. Rome: FAO; 2015.
6. Ultra-processed food and drink products in Latin America: Trends, impact
on obesity, policy implications. Washington D.C.: Panamerican Health
Organization, 2015.
7. Monteiro CA, Levy RB, Claro RM, de Castro IR, Cannon G. Increasing
consumption of ultra-processed foods and likely impact on human health:
evidence from Brazil. Public Health Nutr. 2010a;14(1):513. doi:10.1017/
S1368980010003241.
8. Louzada ML, Martins AP, Canella DS, Baraldi LG, Bertazzi RL, Claro RM,
Moubarac JC, Cannon G, Monteiro CA. Ultra-processed foods and the
nutritional dietary profile in Brazil. Rev Saude Publica. 2015a;49:38. doi:10.
1590/S0034-8910.2015049006132.
9. Louzada ML, Martins AP, Canella DS, Baraldi LG, Bertazzi RL, Claro RM,
Moubarac JC, Cannon G, Monteiro CA. Impact of ultra-processed foods on
micronutrient content in the Brazilian diet. Rev Saude Publica. 2015b;49:45.
doi:10.1590/S0034-8910.2015049006211.
10. Moubarac J-C, Martins APB, Claro RM, Levy RB, Cannon G, Monteiro CA.
Consumption of ultra-processed foods and likely impact on human health.
Evidence from Canada. Public Health Nutr, 2012; doi:10.1017/
S1368980012005009.
11. Martinez Steele E, Baraldi LG, Louzada ML, Moubarac JC, Mozaffarian D,
Monteiro CA. Ultra-processed foods and added sugars in the US diet:
evidence from a nationally representative cross-sectional study. BMJ Open.
2016;6:e009892. doi:10.1136/bmjopen-2015-009892.
12. Poti JM, Mendez MA, Ng SW, Popkin BM. Is the degree of food
processing and convenience linked with the nutritional quality of foods
purchased by US households? Am J Clin Nutr. 2015;101(6):125162. doi:
10.3945/ajcn.114.100925.
13. Stuckler D, McKee M, Ebrahim S, Basu S. Manufacturing epidemics: the role of
global producers in increased consumption of unhealthy commodities
including processed foods, alcohol, and tobacco. PLoS Med. 2012;9, e1001235.
14. Monteiro CA, Cannon G. The impact of transnational Big Foodcompanies
on the South: a view from Brazil. PLoS Med. 2012a;9:e1001252.
15. Monteiro CA, Moubarac JC, Cannon G, Popkin BM. Ultra-processed products
are becoming dominant in the global food system. Obes Rev. 2013;14
Suppl 2:218. doi:10.1111/obr.12107.
16. Juul F, Hemmingsson E. Trends in consumption of ultra-processed foods
and obesity in Sweden between 1960 and 2010. Public Health Nutr. 2015.
doi:10.1017/S1368980015000506.
17. Monteiro CA, Levy RB, Claro RM, Castro IR, Cannon G. A new classification of
foods based on the extent and purpose of their processing. Cad Saúde
Pública. 2010;26(11):203949.
18. Vandevijvere S, Monteiro C, Krebs-Smith SM, Lee A, Swinburn B, Kelly B, Neal
B, Snowdon W, Sacks G. INFORMAS. Monitoring and benchmarking
population diet quality globally: a step-wise approach. Obes Rev. 2013;Suppl
1:13549. doi:10.1111/obr.12082.
19. Hu FB. Dietary pattern analysis: a new direction in nutritional epidemiology.
Curr Opin Lipidol. 2002;13(1):39.
20. Reedy J, Wirfalt E, Flood A, Mitrou PN, Krebs-Smith SM, Kipnis V, Midthune
D, Leitzmann M, Hollenbeck A, Schatzkin A, Subar AF. Comparing 3 dietary
pattern methodscluster analysis, factor analysis, and index analysisWith
colorectal cancer risk: The NIH-AARP Diet and Health Study. Am J Epidemiol.
2010;171(4):47987.
21. Akin JS, Guilkey DK, Popkin BM, Fanelli MT. Cluster analysis of food
consumption patterns of older Americans. J Am Diet Assoc. 1986;5:61624.
22. Patterson RE, Haines PS, Popkin BM. Diet quality index: capturing a
multidimensional behavior. J Am Diet Assoc. 1994;1:5764.
23. Kant AK. Dietary patterns and health outcomes. J Am Diet Assoc. 2004;
104(4):61535.
24. Kant AK. Indexes of overall diet quality: a review. J Am Diet Assoc. 1996;8:78591.
25. Diet and health: implication for reducing chronic disease risk. National
Academy Press, Washington, DC; 1989.
26. Lee CN, Reed DM, MacLean CJ, Yano K, Chiu D. Dietary potassium and
stroke. N Engl J Med. 1988;318:9956.
27. Sacks FM, Obarzanek E, Windhauser MM. for the DASH investigators.
Rational and design of the Dietary Approaches to Stop Hypertension Trial
(DASH): a multicenter controlled-feeding study of dietary patterns to lower
blood pressure. Ann Epidemiol. 1994;5:10818.
28. Farchi G, Mariotti S, Menotti A, Seccareccia F, Torsello S, Fidanza F. Diet and
20-y mortality in two rural population groups of middle-aged men in Italy.
Am J Clin Nutr. 1989;50:1095103.
29. Kant AK, Schatzkin A, Block G, Ziegler RG, Nestle M. Food group intake
patterns and associated nutrient profiles of the US population. J Am Diet
Assoc. 1991;91:15327.
30. Randall E, Marshall JR, Graham S, Brasure J. Patterns in food use and their
associations with nutrient intakes. Am J Clin Nutr. 1990;52:73945.
31. Moeller SM, Reedy J, Millen AE, Dixon LB, Newby PK, Tucker KL, Krebs-Smith
SM, Guenther PM. Dietary patterns: challenges and opportunities in dietary
patterns research an Experimental Biology workshop, April 1, 2006. J Am
Diet Assoc. 2007;107:12339.
32. Johnson CL, Paulose-Ram R, Ogden CL. National Health and Nutrition
Examination Survey: Analytic guidelines, 19992010. National Center for
Health Statistics. Vital Health Stat 2 (161). 2013.
33. Moshfegh AJ, Rhodes DG, Baer DJ, Murayi T, Clemens JC, Rumpler WV, Paul
DR, Sebastian RS, Kuczynski KC, Ingwersen LA, Staples RC, The CLC, USDA.
Automated Multiple-Pass Method reduces bias in the collection of energy
intakes. Am J Clin Nutr. 2008;88:32432.
34. Blanton CA, Moshfegh AJ, Baer DJ, Kretsch MJ. The USDA Automated
Multiple-Pass Method accurately estimates group total energy and nutrient
intake. J Nutr. 2006;136(10):25949.
35. Rumpler WV, Kramer M, Rhodes DG, Moshfegh AJ, Paul DR, Kramer M.
Identifying sources of reporting error using measured food intake. Eur J Clin
Nutr. 2008;62:54452.
36. Moubarac JC, Parra DC, Cannon G, Monteiro CA. Food Classification
Systems Based on Food Processing: significance and implications for
policies and actions: a systematic literature review and assessment. Curr
Obes Rep. 2014;3:25672.
37. Monteiro CA, Cannon G, Levy RB, Claro R, Moubarac JC, Martins AP, Louzada
ML, Baraldi L, Canella D. The Food System. Processing. The big issue for
disease, good health, well-being. World Nutr. 2012;3:52769. www.wphna.
org (accessed Jan 2017).
38. Ahuja JKA, Montville JB, Omolewa-Tomobi G, Heendeniya KY, Martin CL, Steinfeldt
LC, Anand J, Adler ME, LaComb RP, and Moshfegh AJ. 2012. USDA Food and
Nutrient Database for Dietary Studies, 5.0. U.S. Department of Agriculture,
Agricultural Research Service, Food Surveys Research Group, Beltsville, MD.
39. U.S. Department of Agriculture, Agricultural Research Service. USDA National
Nutrient Database for Standard Reference, Release 24. 2011. Nutrient Data
Laboratory Home Page, available at: http://www.ars.usda.gov/ba/bhnrc/
ndl (accessed Jan 2017).
40. U.S. Department of Health and Human Services and U.S. Department of
Agriculture. 20152020 Dietary Guidelines for Americans. 8th Edition.
December 2015. Available at http://health.gov/dietaryguidelines/2015/
guidelines/ (accessed Jan 2017).
41. Bowman SA, Clemens JC, Thoerig RC, Friday JE, Shimizu M, Moshfegh AJ.
Food Surveys Research Group, Beltsville Human Nutrition Research Center,
Agricultural Research Service. U.S.: Department of Agriculture, Beltsville,
Maryland; 2013. Food Patterns Equivalents Database 200910: Methodology
and User Guide [Online], <http://www.ars.usda.gov/ba/bhnrc/fsrg>
(accessed June 2013).
42. Raykov T, Marcoulides GA. An Introduction to Applied Multivariate Analysis,
2008. ISBN-13: 978-0-8058-6375-8 (hardcover) (page 211238).
43. Eicher-Miller HA, Fulgoni III VL, Keast DR. Contributions of Processed Foods
to Dietary Intake in the US from 20032008: A Report of the Food and
Nutrition Science Solutions Joint Task Force of the Academy of Nutrition
Martínez Steele et al. Population Health Metrics (2017) 15:6 Page 10 of 11
and Dietetics, American Society for Nutrition, Institute of Food
Technologists, and International Food Information Council. J Nutr. 2012;142:
2065S72S. doi:10.3945/jn.112.164442.
44. International Food Information Council Foundation. Understanding our
food communications tool kit, 2010. http://www.foodinsight.org/For-
Professionals/Understanding-Our-Food/tabid/1398/Default.aspx (accessed
Jan 2017).
45. Eicher-Miller HA, Fulgoni VL, Keast DR. Processed Food Contributions to
Energy and Nutrient Intake Differ among US Children by Race/Ethnicity.
Nutrients. 2015;7(12):1007688. doi:10.3390/nu7125503.
46. Eicher-Miller HA, Fulgoni VL, Keast DR. Energy and Nutrient Intakes from
Processed Foods Differ by Sex, Income Status, and Race/Ethnicity of US
Adults. J Acad Nutr Diet. 2015;115(6):90718. doi:10.1016/j.jand.2014.11.004.
e6, Epub 2015 Jan 8.
47. Crovetto MM, Uauy R, Martins AP, Moubarac JC, Monteiro C. Household
availability of ready-to-consume food and drink products in Chile: impact
on nutritional quality of the diet. Rev Med Chil. 2014;142(7):8508. doi:10.
4067/S0034-98872014000700005.
48. Slining MM, Yoon EF, Davis J, Hollingsworth B, Miles D, Ng SW. An
Approach to Monitor Food and Nutrition from Factory to Fork. J Acad
Nutr Diet. 2015;115(1):409.
49. Martinez ME, Marshall JR, Sochrost L. Invited commentary: factor analysis
and the search for objectivity. Am J Epidemiol. 1998;148:179.
We accept pre-submission inquiries
Our selector tool helps you to find the most relevant journal
We provide round the clock customer support
Convenient online submission
Thorough peer review
Inclusion in PubMed and all major indexing services
Maximum visibility for your research
Submit your manuscript at
www.biomedcentral.com/submit
Submit your next manuscript to BioMed Central
and we will help you at every step:
Martínez Steele et al. Population Health Metrics (2017) 15:6 Page 11 of 11
... Furthermore, they may contain neo-formed contaminants derived from industrial processing, as well as substances from additives and packaging [6,7]. Considering the association between UPFs and poorer dietary quality, the share of UPFs has been proposed as an effective predictor of population diet quality [8][9][10]. ...
Article
Full-text available
Background: Consumption of ultra-processed foods (UPFs) plays a potential role in the development of obesity and other diet-related noncommunicable diseases (NCDs), but no studies have systematically focused on this. This study aimed to summarize the evidence for the association between UPFs consumption and health outcomes. Methods: A comprehensive search was conducted in PubMed, Embase, and Web of Science to identify all relevant studies. Epidemiological studies were included, and identified studies were evaluated for risk of bias.A narrative review of the synthesized findings was provided to assess the association between UPFs consumption and health outcomes. Results: 20 studies (12 cohort and 8 cross-sectional studies) were included in the analysis, with a total of 334,114 participants and 10 health outcomes. In a narrative review, high UPFs consumption was obviously associated with an increased risk of all-cause mortality, overall cardiovascular diseases, coronary heart diseases, cerebrovascular diseases, hypertension, metabolic syndrome, overweight and obesity, depression, irritable bowel syndrome, overall cancer, postmenopausal breast cancer, gestational obesity, adolescent asthma and wheezing, and frailty. It showed no significant association with cardiovascular disease mortality, prostate and colorectal cancers, gestational diabetes mellitus and gestational overweight. Conclusions: This study indicated a positive association between UPFs consumption and risk of several health outcomes. Large-scale prospective designed studies are needed to confirm our findings.
... According to this classification, UPF are defined as products 'created mostly or entirely from substances extracted from food or derived from food constituents with little or no intact food'. Since Monteiro coined the term UPF, there have been an increasing number of studies that have associated UPF consumption with negative health outcomes in adult subjects (3,39) , including cardiometabolic risk factors (40) , CVD (35) , cancer (16) and many other outcomes (15,25,32) . ...
Article
Full-text available
Increasing evidence suggests that high consumption of ultra-processed foods (UPF) is associated with an increase in non-communicable diseases , overweight and obesity. The present study systematically reviewed all observational studies that investigated the association between UPF consumption and health status. A comprehensive search of MEDLINE, Embase, Scopus, Web of Science and Google Scholar was conducted, and reference lists of included articles were checked. Only cross-sectional and prospective cohort studies were included. At the end of the selection process, twenty-three studies (ten cross-sectional and thirteen prospective cohort studies) were included in the systematic review. As regards the cross-sectional studies, the highest UPF consumption was associated with a significant increase in the risk of overweight/obesity (þ39 %), high waist circumference (þ39 %), low HDL-cholesterol levels (þ102 %) and the metabolic syndrome (þ79 %), while no significant associations with hypertension, hyperglycaemia or hypertriacylglycerolaemia were observed. For prospective cohort studies evaluating a total population of 183 491 participants followed for a period ranging from 3·5 to 19 years, highest UPF consumption was found to be associated with increased risk of all-cause mortality in five studies (risk ratio (RR) 1·25, 95 % CI 1·14, 1·37; P < 0·00001), increased risk of CVD in three studies (RR 1·29, 95 % CI 1·12, 1·48; P = 0·0003), cerebrovascular disease in two studies (RR 1·34, 95 % CI 1·07, 1·68; P = 0·01) and depression in two studies (RR 1·20, 95 % CI 1·03, 1·40; P = 0·02). In conclusion, increased UPF consumption was associated, although in a limited number of studies, with a worse cardiometabolic risk profile and a higher risk of CVD, cerebrovascular disease, depression and all-cause mortality.
... To our knowledge, no study has directly compared the effects of food processing utilizing meal challenges of WF, PF, and GF foods on postprandial substrate utilization. It is well-accepted that diets rich in ultra PF are more obesogenic due to increased energy density, added sugars, and fats, as well as nutrient-depleted of micronutrients, bioactives, and fiber [28,29]. Based on this data, it is hypothesized that PF meals (including GF meals) would increase RER via higher carbohydrate oxidation and lower fat oxidation and therefore alter the TEM response. ...
Article
Full-text available
Consumption of ultra-processed food (PF) is associated with obesity risk compared with whole food (WF) intake. Less is known regarding the intake of gluten-free (GF) food products. The purpose of this study was to directly compare the thermic effect (TEM), substrate utilization, hunger/taste ratings, and glucose response of three different meals containing PF, WF, and GF food products in young healthy women. Eleven volunteers completed all three iso-caloric/macronutrient test meals in a single-blind, randomized crossover design: (1) whole food meal (WF); (2) processed food meal (PF); or (3) gluten-free meal (GF). TEM was significantly lower following GF compared with WF (−20.94 kcal/meal, [95% CI, −35.92 to −5.96], p = 0.008) and PF (mean difference: −14.94 kcal/meal, [95% CI, −29.92 to 0.04], p = 0.04), respectively. WF consumption resulted in significantly higher feelings of fullness compared to GF (mean difference: +14.36%, [95% CI, 3.41 to 25.32%], p = 0.011) and PF (mean difference: +16.81%, [95% CI, 5.62 to 28.01%], p = 0.004), respectively, and enhanced palatability (taste of meal) compared to PF meal (mean ∆: +27.41%, [95% CI, 5.53 to 49.30%], p = 0.048). No differences existed for substrate utilization and blood glucose response among trials. Consumption of a GF meal lowers postprandial thermogenesis compared to WF and PF meals and fullness ratings compared to a WF meal which may impact weight control and obesity risk over the long-term.
... In several instances, foods were classified differently across surveys due to different food systems in country-specific contexts. For example, some FFQs do not separate the bread type and in the US and UK it is mainly UPF [60,104], whereas in France and Spain it is mainly artisanal [52,61]. Additional complexities included application to food databases with a lack of information on the differentiation of canned food into PF or UPF, databases disaggregating foods to nutrient content rather than processing type (e.g., frozen pizza disaggregated to component ingredients could inadvertently be classed as PF or PCI, and not correctly as UPF), and disaggregating handmade dishes (e.g., Bolognese pasta) into major food-items in the recipe (e.g., group 1, pasta) rather than underlying ingredients (pasta, meat, sauce, oil), which is the recommended approach [59,105]. ...
Article
Full-text available
The nutrition literature and authoritative reports increasingly recognise the concept of ultra-processed foods (UPF), as a descriptor of unhealthy diets. UPFs are now prevalent in diets worldwide. This review aims to identify and appraise the studies on healthy participants that investigated associations between levels of UPF consumption and health outcomes. This involved a systematic search for extant literature; integration and interpretation of findings from diverse study types, populations, health outcomes and dietary assessments; and quality appraisal. Of 43 studies reviewed, 37 found dietary UPF exposure associated with at least one adverse health outcome. Among adults, these included overweight, obesity and cardio-metabolic risks; cancer, type-2 diabetes and cardiovascular diseases; irritable bowel syndrome, depression and frailty conditions; and all-cause mortality. Among children and adolescents, these included cardio-metabolic risks and asthma. No study reported an association between UPF and beneficial health outcomes. Most findings were derived from observational studies and evidence of plausible biological mechanisms to increase confidence in the veracity of these observed associations is steadily evolving. There is now a considerable body of evidence supporting the use of UPFs as a scientific concept to assess the ‘healthiness’ of foods within the context of dietary patterns and to help inform the development of dietary guidelines and nutrition policy actions.
Research Proposal
Full-text available
Ultra-processed foods have become a staple of Western diets due primarily to their low cost, high caloric value, ready-made convenience, and typically high palatability. According to the NOVA classification system, ultra-processed foods are those made from food extracts (e.g., oils, fats, sugar, starch, and proteins), “derived from food constituents (e.g., hydrogenated fats and modified starch),” or synthetic formulas made “from food substrates or other organic sources (e.g., flavor enhancers, colors and several food additives used to make the product hyper-palatable)” (Zinocker et al., 2018). Although consumption of ultra-processed foods has been associated with higher risk of obesity and hypertension, ultra-processed foods are the largest source of calories consumed in America and are increasingly being exported to developing countries (Zinocker et al., 2018; Hall et al., 2019). In this proposal, we will evaluate how consumption of ultra-processed foods alters the gut microbiome and expression of gut satiety hormones in healthy humans. To conduct this study, we will employ techniques such as 16s rRNA sequencing, metabolic profiling, fecal microbiota transplant into germ-free mice, and supplementing ultra-processed diets with satiety-linked hormones and short chain fatty acids (SCFAs). These experiments will potentially elucidate one mechanism by which ultra-processed foods cause increased food intake and weight gain, as an early risk marker for developing obesity and diabetes.
Article
In recent years, there has been an increased interest observed concerning the relationship between the consumption of highly processed foods and health impact [...]
Article
Full-text available
Understanding the drivers and dynamics of global ultra‐processed food (UPF) consumption is essential, given the evidence linking these foods with adverse health outcomes. In this synthesis review, we take two steps. First, we quantify per capita volumes and trends in UPF sales, and ingredients (sweeteners, fats, sodium and cosmetic additives) supplied by these foods, in countries classified by income and region. Second, we review the literature on food systems and political economy factors that likely explain the observed changes. We find evidence for a substantial expansion in the types and quantities of UPFs sold worldwide, representing a transition towards a more processed global diet but with wide variations between regions and countries. As countries grow richer, higher volumes and a wider variety of UPFs are sold. Sales are highest in Australasia, North America, Europe and Latin America but growing rapidly in Asia, the Middle East and Africa. These developments are closely linked with the industrialization of food systems, technological change and globalization, including growth in the market and political activities of transnational food corporations and inadequate policies to protect nutrition in these new contexts. The scale of dietary change underway, especially in highly populated middle‐income countries, raises serious concern for global health.
Article
Objective To assess whether higher adherence to the traditional Mediterranean diet (MedDiet) was associated with lower consumption of ultra-processed foods (UPF) and lower free sugar intake. Design Cross-sectional analysis of baseline information among participants in the SENDO project, a Spanish paediatric cohort. Dietary information was collected through a semi-quantitative FFQ. Food items were classified according to the NOVA classification. Adherence to the MedDiet was evaluated through the KIDMED index. Setting Spain. Participants Three hundred eight-six children (52 % boys) with a mean age of 5·3 years old ( sd 1·0) were included in the analysis. Results 74·4 % of the children had moderate adherence to the MedDiet (mean KIDMED score: 5·9 points; sd 1·7) and overall, 32·2 % of the total energy intake came from UPF. Each two additional points in the KIDMED score was associated with 3·1 % (95 % CI 2·1, 4·0) lower energy intake from UPF. Compared to those with low adherence to the MedDiet, children with medium and high adherence reported 5·0 % (95 % CI 2·2, 7·7) and 8·5 % (95 % CI 5·2, 11·9) lower energy intake from UPF, respectively. We also found that 71·6 % of the variability in free sugar intake was explained by the variability in UPF consumption. Conclusions Adherence to the traditional MedDiet was inversely associated with energy intake from UPF. Furthermore, most of the variability in free sugar intake was explained by the variability of UPF consumption. Public health strategies are needed to strengthen the adherence to the MedDiet in pre-schoolers while regulating the production, marketing and advertising of UPF.
Article
BACKGROUND Various foods are known to have beneficial effects on health when consumed whole, however, there is a trend to prepare foods from processed ingredients, and it remains unclear whether the benefits of the whole food are retained. Therefore, the purpose of this study was to examine whether different processing techniques affect the cholesterol‐lowering and vascular effects of black beans (Phaseolus vulgaris L.). RESULTS Beans were prepared by overnight soaking and boiling, the standard method, as well as by micronization, extrusion, or dehulling and boiling, and then fine milled. Beans prepared by the standard method were also coarse milled. These five materials were incorporated into semi‐purifed diets (30% wt/wt) and fed to spontaneously hypertensive rats for 4 weeks. Body weight, blood pressure and aorta morphology were unaltered by the diets. Fasting total cholesterol was significantly reduced in rats fed micronized beans compared to extruded beans (both fine‐milled) or the bean‐free diet, while boiling combined with coarse milling lowered LDL‐cholesterol. The lack of cholesterol‐lowering in rats fed extruded bean compared to micronized was not explained by amount or composition of dietary fibre or resistant starch. Differences in the polyphenolic profile as determined by HPLC were also unable to explain the variations in cholesterol lowering capacity. CONCLUSION The present study demonstrates that processing of black beans alters the health effects observed with the whole pulse, and suggests products prepared with processed ingredients will need to be tested emperically to establish whether the biological effects are maintained in vivo . This article is protected by copyright. All rights reserved.
Preprint
Full-text available
The use of additives in food products has become an important public health concern. In recent reports, dietary emulsifiers have been shown to affect the gut microbiota, contributing to a pro-inflammatory phenotype and metabolic syndrome. So far, it is not yet known whether similar microbiome shifts are observable for a more diverse set of emulsifier types and to what extent these effects vary with the unique features of an individual's microbiome. To bridge this gap, we investigated the effect of five dietary emulsifiers on the fecal microbiota from 10 human individuals upon a 48 hour exposure. Community structure was assessed with quantative microbial profiling, functionality was evaluated by measuring fermentation metabolites and pro-inflammatory properties were assessed with the phylogenetic prediction algorythm PICRUSt, together with a TLR5 reporter cell assay for flagellin. A comparison was made between two mainstream chemical emulsifiers (carboxymethylcellulose and P80), a natural extract (soy lecithin) and biotechnological emulsifiers (sophorolipids and rhamnolipids). While fecal microbiota responded in a donor-dependent manner to the different emulsifiers, profound differences between emulsifier were observed. Rhamnolipids, sophorolipids and soy lecithin eliminated 91% ± 0%, 89% ± 1% and 87% ± 1% of the viable bacterial population after 48 hours, yet they all selectively increased the proportional abundance of putative pathogens. Moreover, profound shifts in butyrate (-96% ± 6 %, -73% ± 24% and -34 ± 25% respectively) and propionate (+13% ± 24 %, +88% ± 50% and +29% ± 16% respectively) production were observed for these emulsifiers. Phylogenetic prediction indicated higher motility, which was, however, not confirmed by increased flagellin levels using the TLR5 reporter cell assay. We conclude that dietary emulsifiers can severely impact the gut microbiota and this seems to be proportional to their emulsifying strength, rather than emulsifier type or origin. As biotechnological emulsifiers were especially more impactful than chemical emulsifiers, caution is warranted when considering them as more natural alternatives for clean label strategies.
Data
Full-text available
Documentation and dataset available on Worldwide Web Site: Food Surveys Research Group: http://www.ars.usda.gov/Services/docs.htm?docid=12089
Article
Full-text available
Objectives To investigate the contribution of ultra-processed foods to the intake of added sugars in the USA. Ultra-processed foods were defined as industrial formulations which, besides salt, sugar, oils and fats, include substances not used in culinary preparations, in particular additives used to imitate sensorial qualities of minimally processed foods and their culinary preparations. Design Cross-sectional study. Setting National Health and Nutrition Examination Survey 2009–2010. Participants We evaluated 9317 participants aged 1+ years with at least one 24 h dietary recall. Main outcome measures Average dietary content of added sugars and proportion of individuals consuming more than 10% of total energy from added sugars. Data analysis Gaussian and Poisson regressions estimated the association between consumption of ultra-processed foods and intake of added sugars. All models incorporated survey sample weights and adjusted for age, sex, race/ethnicity, family income and educational attainment. Results Ultra-processed foods comprised 57.9% of energy intake, and contributed 89.7% of the energy intake from added sugars. The content of added sugars in ultra-processed foods (21.1% of calories) was eightfold higher than in processed foods (2.4%) and fivefold higher than in unprocessed or minimally processed foods and processed culinary ingredients grouped together (3.7%). Both in unadjusted and adjusted models, each increase of 5 percentage points in proportional energy intake from ultra-processed foods increased the proportional energy intake from added sugars by 1 percentage point. Consumption of added sugars increased linearly across quintiles of ultra-processed food consumption: from 7.5% of total energy in the lowest quintile to 19.5% in the highest. A total of 82.1% of Americans in the highest quintile exceeded the recommended limit of 10% energy from added sugars, compared with 26.4% in the lowest. Conclusions Decreasing the consumption of ultra-processed foods could be an effective way of reducing the excessive intake of added sugars in the USA.
Article
Full-text available
Background: Processed foodstuff may have a lower nutritional value than natural products. Aim: To analyze the impact of ready-to-consume products on diet quality of Chilean households. Material and Methods: A national representative sample of 10,096 households, based on the 6th Survey on Household Budget and Expenses (VI Encuesta de Presupuestos y Gastos Familiares, 2006-2007), was studied. Foodstuffs were classified as follows: 1) Unprocessed foods or minimally processed foods (G1); 2) Processed culinary ingredients (G2); and 3) Ready-to-consume products (G3). Calorie contribution and energy availability of each household food group, was calculated. The nutritional profile of the national food basket was calculated and compared with two simulated baskets (G3 vs G1+G2), based on international nutritional recommendations. Results: Overall energy availability was of 1,885 kcal per capita/ day; 24% derived from unprocessed foods (G1), 21% from processed culinary ingredients (G2) and 55% from ready-to-consume products (G3), whose proportion increased along with income level. The 2007 national food basket contained an excess of total fat (34% vs 30%), free sugars (16% vs 10%), energy density (2.1 vs 1.3 kcal/ gram) and a low amount of fiber (8.4 vs 12.5 g/1,000 kcal). The basket consisting in ready-to-consume products (G3) had a higher percentage of carbohydrates (61% vs 46%) than the basket consisting in unprocessed foods and ingredients (G1 + G2). It also had a higher percentage of free sugars (17% vs 15%), less dietary fiber (7 vs. 10 g/1,000 kcal) and, above all, a higher energy density (2.6 vs 1.6 kcal/g). Conclusions: The Chilean dietary pattern, based on ready-to-consume products (G3), is affecting the nutritional quality of the diet.
Article
Full-text available
This study determined and compared the mean daily intake of energy and nutrients from processed foods by level of processing (minimally processed; processed for preservation, nutrient enhancement, and freshness; mixtures of combined ingredients; ready-to-eat processed foods; and prepared foods/meals) among non-Hispanic white, non-Hispanic black, and Mexican American US children. Data from participants 2-18 years old (n = 10,298) of the nationally representative cross-sectional National Health and Nutrition Examination Survey 2003-2008 with a complete one day, 24-h dietary recall were used to determine mean intake of energy and nutrients recommended for increase and decrease, as per the 2010 Dietary Guidelines for Americans, among child race/ethnic groups by category of food processing. Regression analysis was used to estimate and compare covariate-adjusted (gender, age, and poverty-income-level) least square means (p < 0.05/3 race/ethnic groups). All children, regardless of race or ethnicity consumed processed foods. Approximately 66% to 84% of total daily energy, saturated fat, cholesterol, fiber, total sugar, added sugars, calcium, vitamin D, potassium, and sodium intake are contributed by one of the five categories of processed foods. Clinicians and policy should primarily advise consideration of the energy and nutrient composition of foods, rather than the processing level, when selecting a healthy diet for children.
Article
Full-text available
OBJECTIVE To evaluate the impact of consuming ultra-processed foods on the micronutrient content of the Brazilian population’s diet. METHODS This cross-sectional study was performed using data on individual food consumption from a module of the 2008-2009 Brazilian Household Budget Survey. A representative sample of the Brazilian population aged 10 years or over was assessed (n = 32,898). Food consumption data were collected through two 24-hour food records. Linear regression models were used to assess the association between the nutrient content of the diet and the quintiles of ultra-processed food consumption – crude and adjusted for family income per capita. RESULTS Mean daily energy intake per capita was 1,866 kcal, with 69.5% coming from natural or minimally processed foods, 9.0% from processed foods and 21.5% from ultra-processed foods. For sixteen out of the seventeen evaluated micronutrients, their content was lower in the fraction of the diet composed of ultra-processed foods compared with the fraction of the diet composed of natural or minimally processed foods. The content of 10 micronutrients in ultra-processed foods did not reach half the content level observed in the natural or minimally processed foods. The higher consumption of ultra-processed foods was inversely and significantly associated with the content of vitamins B12, vitamin D, vitamin E, niacin, pyridoxine, copper, iron, phosphorus, magnesium, selenium and zinc. The reverse situation was only observed for calcium, thiamin and riboflavin. CONCLUSIONS The findings of this study highlight that reducing the consumption of ultra-processed foods is a natural way to promote healthy eating in Brazil and, therefore, is in line with the recommendations made by the Guia Alimentar para a População Brasileira (Dietary Guidelines for the Brazilian Population) to avoid these foods.
Article
Full-text available
OBJECTIVE To assess the impact of consuming ultra-processed foods on the nutritional dietary profile in Brazil. METHODS Cross-sectional study conducted with data from the module on individual food consumption from the 2008-2009 Pesquisa de Orçamentos Familiares (POF – Brazilian Family Budgets Survey). The sample, which represented the section of the Brazilian population aged 10 years or over, involved 32,898 individuals. Food consumption was evaluated by two 24-hour food records. The consumed food items were classified into three groups: natural or minimally processed, including culinary preparations with these foods used as a base; processed; and ultra-processed. RESULTS The average daily energy consumption per capita was 1,866 kcal, with 69.5% being provided by natural or minimally processed foods, 9.0% by processed foods and 21.5% by ultra-processed food. The nutritional profile of the fraction of ultra-processed food consumption showed higher energy density, higher overall fat content, higher saturated and trans fat, higher levels of free sugar and less fiber, protein, sodium and potassium, when compared to the fraction of consumption related to natural or minimally processed foods. Ultra-processed foods presented generally unfavorable characteristics when compared to processed foods. Greater inclusion of ultra-processed foods in the diet resulted in a general deterioration in the dietary nutritional profile. The indicators of the nutritional dietary profile of Brazilians who consumed less ultra-processed foods, with the exception of sodium, are the stratum of the population closer to international recommendations for a healthy diet. CONCLUSIONS The results from this study highlight the damage to health that is arising based on the observed trend in Brazil of replacing traditional meals, based on natural or minimally processed foods, with ultra-processed foods. These results also support the recommendation of avoiding the consumption of these kinds of foods.
Article
Full-text available
(Note. This commentary contains much information not found elsewhere. In one respect it has been updated, with the inclusion of a new Group 3, of processed foods. See the commentary in World Nutrition dated January-March 2016, available above on ResearchGate) Our general theory is that the global food system, and specifically its increasing domination by processed food products as specified and defined here, is the big issue for nutrition, disease, health and well-being. We begin here by explaining the history and development of the theory and its context, and our findings and thinking so far. We also outline implications for assessment of dietary patterns, development of dietary guidelines, promotion of good health and well-being, and prevention and control of obesity and related chronic non-communicable diseases. It is evident, we believe, that the current conceptual framework of nutrition, which places it solely or mainly within the biological sciences, does not adequately respond to the circumstances of our time. Undernutrition, food insecurity and hunger persist in many parts of the world, even within high-income countries, at unacceptable and even scandalous levels. But what is now the pandemic of overweight and obesity, and of other related chronic non-communicable diseases such as diabetes, is out of control. We believe that this global public health crisis will remain uncontrolled until a new and more relevant and appropriate conceptual framework is developed, accepted, and applied. Our purpose here is to develop this new way of thinking, which from our experience and in our judgement so far, is the best fit with the facts.
Article
"Processed foods" are defined as any foods other than raw agricultural commodities and can be categorized by the extent of changes occurring in foods as a result of processing. Conclusions about the association between the degree of food processing and nutritional quality are discrepant. We aimed to determine 2000-2012 trends in the contribution of processed and convenience food categories to purchases by US households and to compare saturated fat, sugar, and sodium content of purchases across levels of processing and convenience. We analyzed purchases of consumer packaged goods for 157,142 households from the 2000-2012 Homescan Panel. We explicitly defined categories for classifying products by degree of industrial processing and separately by convenience of preparation. We classified >1.2 million products through use of barcode-specific descriptions and ingredient lists. Median saturated fat, sugar, and sodium content and the likelihood that purchases exceeded maximum daily intake recommendations for these components were compared across levels of processing or convenience by using quantile and logistic regression. More than three-fourths of energy in purchases by US households came from moderately (15.9%) and highly processed (61.0%) foods and beverages in 2012 (939 kcal/d per capita). Trends between 2000 and 2012 were stable. When classifying foods by convenience, ready-to-eat (68.1%) and ready-to-heat (15.2%) products supplied the majority of energy in purchases. The adjusted proportion of household-level food purchases exceeding 10% kcal from saturated fat, 15% kcal from sugar, and 2400 mg sodium/1000 kcal simultaneously was significantly higher for highly processed (60.4%) and ready-to-eat (27.1%) food purchases compared with purchases of less-processed foods (5.6%) or foods requiring cooking/preparation (4.9%). Highly processed food purchases are a dominant, unshifting part of US purchasing patterns yet may have higher saturated fat, sugar, and sodium content compared with less-processed foods. Wide variation in nutrient content suggests food choices within categories may be important. © 2015 American Society for Nutrition.
Article
To investigate how consumption of ultra-processed foods has changed in Sweden in relation to obesity. Nationwide ecological analysis of changes in processed foods along with corresponding changes in obesity. Trends in per capita food consumption during 1960-2010 were investigated using data from the Swedish Board of Agriculture. Food items were classified as group 1 (unprocessed/minimally processed), group 2 (processed culinary ingredients) or group 3 (3·1, processed food products; and 3·2, ultra-processed products). Obesity prevalence data were pooled from the peer-reviewed literature, Statistics Sweden and the WHO Global Health Observatory. Nationwide analysis in Sweden, 1960-2010. Swedish nationals aged 18 years and older. During the study period consumption of group 1 foods (minimal processing) decreased by 2 %, while consumption of group 2 foods (processed ingredients) decreased by 34 %. Consumption of group 3·1 foods (processed food products) increased by 116 % and group 3·2 foods (ultra-processed products) increased by 142 %. Among ultra-processed products, there were particularly large increases in soda (315 %; 22 v. 92 litres/capita per annum) and snack foods such as crisps and candies (367 %; 7 v. 34 kg/capita per annum). In parallel to these changes in ultra-processed products, rates of adult obesity increased from 5 % in 1980 to over 11 % in 2010. The consumption of ultra-processed products (i.e. foods with low nutritional value but high energy density) has increased dramatically in Sweden since 1960, which mirrors the increased prevalence of obesity. Future research should clarify the potential causal role of ultra-processed products in weight gain and obesity.