Effects of product reformulation on sugar intake and health_a review

Prévia do material em texto

Nutrition Science$Policy
Effects of product reformulation on sugar intake and health—a
systematic review and meta-analysis
Kawther M. Hashem, Feng J. He, and Graham A. MacGregor
Context: Obesity, type 2 diabetes, and dental caries are all major public health
problems in the United Kingdom and contribute substantially to healthcare costs.
Objective: A systematic review and meta-analysis was conducted to determine the
effect of product reformulation measures on sugar intake and health outcomes.
Data sources: Using a combination of terms, the following databases were
searched—The Cochrane Library, EMBASE, MEDLINE (Ovid), and Scopus.
Additionally, multiple gray literature searches were undertaken. Data extraction:
A total of 16 studies met the inclusion criteria. There were 4 randomized controlled
trials, 6 studies that modeled reformulation in a country, 5 studies that modeled a
different approach of reformulation, and 1 study was both a modelling study of a
different approach to reformulation and a retrospective observational study. The stud-
ies were assessed for risk of bias and overall quality of evidence was rated using the
Grades of Recommendation, Assessment, Development and Evaluation Working
Group (GRADE) framework. Results: Results from randomized controlled trials
suggest that consumption of reformulated products can reduce sugar intake and
body weight. The pooled estimates were �11.18% (95% confidence interval [CI],
�19.95 to �2.41; P < 0.00001) for changes in percentage of sugar intake,
�91.00 g/day (95%CI, �148.72 to �33.28; P< 0.00001) for changes in sugar in-
take in grams per day, and �1.04 kg (95%CI, �2.16 to �0.08; P¼ 0.0002) for
changes in body weight. However, the quality of the evidence was very low. Results
from the other studies suggested that reformulation can reduce sugar intake and im-
prove health. Much of the evidence draws on modeling studies. Conclusions: This
systematic review and meta-analysis suggests that product reformulation to reduce
sugar content could reduce sugar intake in individuals and thus improve popula-
tion health. These findings provide an important starting point for ongoing work
on sugar reformulation.
INTRODUCTION
Obesity, type 2 diabetes, and dental caries are all major
public health problems in the United Kingdom1–7 and
contribute substantially to healthcare costs.8 It is
recognized that consumption of excessive free sugars
(“sugar”) is associated with these conditions.9–14
In July 2015, the Scientific Advisory Committee on
Nutrition (SACN) recommended that the average in-
take of sugar across the UK population should not
Affiliation: K.M. Hashem, F.J. He, and G.A. MacGregor are with the Wolfson Institute of Preventive Medicine, Barts and The London School
of Medicine & Dentistry, Queen Mary University of London, United Kingdom.
Correspondence: K.M. Hashem, Wolfson Institute of Preventive Medicine, Barts and The London School of Medicine & Dentistry, Queen
Mary University of London, Charterhouse Square, London, EC1M 6BQ, UK. E-mail: k.hashem@qmul.ac.uk.
Key words: reformulation, systematic review, sugar.
VC The Author(s) 2019. Published by Oxford University Press on behalf of the International Life Sciences Institute.
All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
doi: 10.1093/nutrit/nuy015
Nutrition ReviewsVR Vol. 77(3):181–196 181
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
exceed 5% of total energy intake.10 This is in line with
the World Health Organization’s (WHO) new guide-
lines on sugar intake.15,16 As of 2014, average intakes of
sugar in the United Kingdom exceeded recommenda-
tions in all age groups.17 Children have a 3-fold higher
intake of sugar than that recommended; the average in-
take was 54 g (13% of total energy) and 73 g (15%) per
day in those aged 4–10 years and those aged 11–
18 years, respectively.17
A number of population-level interventions have
been proposed, including pricing interventions, food
procurement interventions, restrictions on marketing
to children, on-pack nutrition information, information
campaigns, and product reformulation.18 These inter-
ventions can potentially reduce population sugar intake,
however, in recent years product reformulation has
gained prominence as a potential cost-effective inter-
vention, particularly in the United Kingdom.8,19
Product reformulation in whole jurisdictions (eg,
countries) are efforts to lower the “unhealthy” compo-
nents (eg, saturated fat, trans fats, sugar, salt) of products
at the time of production without worsening the profile
of other ingredients (eg, increasing calorie content).20
The reformulated products should replace existing prod-
uct (eg, the same brand of drink with less sugar). This
approach does not rely on substantial behavioral change
on the consumer’s part.
Reformulation has been proven to be effective at
reducing salt intake in the United Kingdom. The key to
the success of the UK salt reduction program has been
the setting of incremental targets for each food group
with a specified timeframe to complete reformulation,
using maximum and average or sales-weighted average
targets.21 Because the reduction in salt has been gradual
and progressive, the UK population has adjusted to the
taste of lower salt concentrations, without adding salt at
the table.22 The average salt intake for adults fell from
9.5 g in 2000–2001 to 8.1 g in 2011; this was accompa-
nied by a fall in population blood pressure and mortal-
ity from stroke and coronary heart disease.23,24
Given the progress made with the salt reduction
program in the United Kingdom, it has been proposed
that sugar consumption can be reduced through a simi-
lar systematic, unobtrusive, and gradual reformulation
program for manufacturers,19 particularly because most
of the sugar in the diet comes from manufactured prod-
ucts, with the food and drink industry representing an
important leverage point for reformulation.
Because sugar reformulation is a relatively new area
that has undergone a recent surge in research, it was
deemed necessary and most appropriate to collate all of
the evidence available using a systematic methodology
to demonstrate the state of the evidence as well as the
gaps that can be addressed by researchers and policy
makers. Therefore, the aim of this study was to system-
atically review the evidence on the effect of product
reformulation measures to reduce sugar intake and im-
prove health outcomes.
METHODS
The study design adhered to the Preferred Reporting
Items for Systematic Review and Meta-analysis
Protocols (PRISMA) 2015 statement.25 The protocol
has been previously published.26 The guiding question
of this review was, “What is the effect of reformulation
on sugar intake and health?”
Inclusion and exclusion criteria
The criteria for study eligibility in this review, including
population, intervention, comparison, outcome, type of
study, and setting (PICOTS) elements, are described in
Table 1. Studies published between 1990 and early 2016
were included. This is because publications about reformu-
lation, particularly salt reformulation, began to appear in
the literature in the early 2000s; therefore, starting the
search from 1990 guaranteed the capture of any papers
published before 2000. Studies were excluded if they were
non-English language studies, had no relevant outcome
measures, or involved interventions with short durations
(< 8 wk).
Search strategy
Electronic databases were systematically searched using
a combination of terms tailored to each database.
An initial limited search of EMBASE was under-
taken; this was followed by an analysis of the text words
contained in the titles and abstracts. A second search
using all identified keywords and index terms was un-
dertaken across all included databases. The databases
searched included the Cochrane Library, EMBASE,
MEDLINE (Ovid), and Scopus. Thereference lists of
key articles that matched inclusion criteria were
searched by hand to identify any further relevant refer-
ences, which were subject to the same screening and se-
lection process.
Gray literature searches were also undertaken using
the broad search terms “sugar” and “food” or “drink”
and “reduction” or “reformulation.” These searches in-
cluded key government and organization websites, as
well as general searches in Google.
Screening and data extraction
All papers identified from the initial electronic search
process were imported into an Endnote library, and
182 Nutrition ReviewsVR Vol. 77(3):181–196
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
duplicates were removed. The eligibility criteria were
applied to the results, and all identified references were
screened using a 2-stage approach. The first stage in-
volved 2 reviewers independently reviewing the title
and the abstract. In the second stage, the full-text ver-
sions of the shortlisted papers were extracted and
assessed by the lead author. The number of excluded
studies (including reasons for exclusion) was recorded
at each stage. A standardized data extraction form was
developed and piloted prior to its use.
Meta-analysis of randomized controlled trials
For each trial, the mean effect on sugar intake and body
weight was calculated. For crossover trials, the mean ef-
fect was the difference in outcomes between the end of
the reduced sugar period and the end of the sugar (ie,
control) period. For parallel trials, the mean effect was
the difference between the 2 groups in the change in
outcomes from baseline to the end of follow-up.
The variance of the mean effect for outcomes in
each trial was calculated. This was derived from stan-
dard deviations of paired differences between baseline
and the end of follow-up for each group in a parallel
trial or between the 2 diet periods in a crossover trial.
To assess the mean effect sizes, the data were pooled
using the inverse variance method in a random-effects
meta-analysis. The I2 test was used to examine heterogene-
ity, with I2> 50% considered to be important.27
Assessment of risk of bias in included studies
For randomized controlled trials (RCTs), the Cochrane
risk of bias tool28 was used.
Tools for assessing risk of bias in RCTs are well de-
scribed, but much less attention has been given to simi-
lar tools for observational studies.29 A risk-of-bias tool
adapted for nonrandomized controlled studies on
population-level interventions, which has been used by
2 recently published Cochrane systematic reviews on
population-level interventions,18,30 was chosen. It was
deemed the most appropriate tool because it was
adapted from the gold-standard Cochrane risk-of-bias
tool.
Some of the studies included in the systematic re-
view were modeling studies. These studies are based on
many assumptions; therefore, an adapted Cochrane risk
of bias was not appropriate. Hence, only the strengths
and limitations of those studies were discussed.
Assessment of quality of evidence
The overall quality of evidence was rated using the
Grades of Recommendation, Assessment, Development
and Evaluation Working Group (GRADE) frame-
work,31 which is Cochrane’s recommended approach
for grading the quality of evidence and the strength of
recommendations. When using GRADE, evidence is
rated not by study but across studies for specific out-
comes. It is based on 5 considerations: risk of bias, im-
precision, inconsistency, indirectness, and publication
bias. There are four possible GRADE ratings: high,
moderate, low, and very low.32
Randomized controlled trials studies start at a
GRADE rating of high, which may be decreased. The
grade may be decreased by 1 or (if very serious) 2 levels
in the following circumstances: serious or very serious
limitations to study quality; important inconsistency;
some or major uncertainty about directness; imprecise
or sparse data; or high probability of reporting bias.31
Observational studies start at a GRADE rating of
low, which may be increased or decreased. The grade
may be increased in the following circumstances: strong
evidence of association based on consistent evidence
from > 2 or more observational studies with no plausi-
ble confounders (þ1); very strong evidence of associa-
tion based on direct evidence with no major threats to
validity (þ2); evidence of a dose-response gradient
(þ1); or all plausible confounders with reduced effect
(þ1).31
GRADE rating was not applied to modeling stud-
ies, but the outcomes assessed by the modeling studies
were added in a narrative form to the summary of find-
ings table.
RESULTS
A total of 16 studies met the inclusion criteria. There
were 4 RCTs, 6 studies that modeled reformulation in a
country, 5 studies that modeled a different approach of
reformulation, and 1 study both modeled a different ap-
proach to reformulation and was also a retrospective
observational study (Figure 1).
Table 1 PICOS criteria for the inclusion and exclusion of
studies
Parameter Criteria
Population Studies involving populations of any age
Intervention Interventions and studies investigating the ef-
fect of product reformulation measures on
sugar intake and health outcomes
Comparison Control/sugar group, unexposed population,
or preintervention group without reformu-
lated products and intervention group/pop-
ulation with reformulated products
Outcome Sugar intake and health outcomes (eg, body
weight, dental health, and type 2 diabetes)
Type of study Both quantitative and qualitative study types
Setting Any setting—randomized trials (� 8 wk dura-
tion) or population-based models
Nutrition ReviewsVR Vol. 77(3):181–196 183
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
The results are summarized in Table 2.33–48 As re-
quired by the GRADE method, the evidence was graded
as very low for the RCTs due to the lack of blinding, in-
consistency, indirectness, and imprecision (as a result
of low participants in pooled RCTs). For the 1 observa-
tional study, the grade was kept as low because there
was only 1 observational study and the effect was small.
The outcomes assessed with modeling studies were not
graded.
Randomized controlled trials
Four RCT publications assessed the effect of sugar-
reformulated products over a period of 8–10 weeks
(Table 3).33–36 Two of the publications were based on
the same trial. Of the 3 trials, 2 used paralleled compari-
sons,34–36 and 1 used crossover design.33
Figures 2–433–36 show the change in sugar intake
(%), sugar intake (g/d) and body weight in individual
trials included in the meta-analysis and the mean effect
size. The pooled estimates of reduction were �11.18%
(95% confidence interval [CI], �19.95 to �2.41, P <
0.00001) in percentage of sugar intake, �91.00 g/day
(95%CI, �148.72 to �33.28; P < 0.00001) in sugar
intake in grams per day, and �1.04 kg (95%CI, �2.16 to
�0.08; P ¼ 0.002) in body weight.
Modeling of reformulation
Six modeling studies assessed the impact of reducing
sugar intake through reformulation on either popula-
tion sugar intake38–41 (Figure 5)38–41,43–48 and/or health
outcomes37,38,42 (Table 2). The studies focused on 2
countries: the United Kingdom37–39 and France.40–42
Three studies were based on the UK (England)
population (Table 437–39 and Figure 5). All of the stud-
ies assumed a reduction in sugar intake and/or an im-
provement in health outcomes.
Three studies were based on the French population
(Table 540–42 and Figure 5). Two studies assessed the
impact on sugar intake,40,41 and 1 assessed the impact
on both intake and health outcomes.42
Different reformulation approaches
The search retrieved several different approaches to
reformulation. The different approaches are described
as 1) cap and trade (1 simulation), 2) Choices program
Id
en�fi
ca
�o
n
Sc
re
en
in
g 
El
ig
ib
ili
ty
 
In
clu
de
d 
Records iden�fied through 
databases searching 
(n = 5009)
Addi�onal records iden�fied 
through other sources 
(n = 95 )
Records a�er duplicates removed 
(n = 4279 )
Records screened 
(n = 4279 )
Records excluded: �tles and 
abstracts not relevant
(n = 4104 )
Full-text ar�cles assessed for 
eligibility 
(n = 175 )
Studies included in qualita�ve 
synthesis 
(n = 16)
Full-text ar�cles excluded: no 
relevant data available
(n = 159)
Studies included in meta-analysis 
(n = 3)
Figure 1 PRISMA flow diagram used for the selection of studies.
184 Nutrition ReviewsVR Vol. 77(3):181–196
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
Table 2 Summary of findings
Effects of product reformulation on sugar intake and health outcomes
Patient or population: people of all ages
Setting: any setting—randomized trials (� 8 wk duration) or population-based observational or models
Intervention: sugar-reformulation
Comparison: change in sugar intake or health outcomes from pre- to post-reformulation
Outcomes Effect (mean) No. of participants
(studies)
Quality of the
evidence (GRADE)a
Sugar, % Change in sugar intake (%): �11.18 (95%CI,
�19.95 to �2.41)
123 (3 RCTs) ����
Very lowb,c,d,e
Sugar intake, g/d Change in sugar intake (g): �91.00 (95%CI,
�148.72 to �33.28)
91 (2 RCTs) ����
Very lowb,c,d,e
One observational study found that during a
7-y period (2005–2012) sugar intake fell by
2.06% (reduction in mean intake of 0.25 g)
in adult consumers of reformulated
products
In teenagers, the reduction was 1.36%
(�0.36 g) in consumers of reformulated
products
In children, the reduction was 1.21% (�0.32
g) in consumers of reformulated products
In preschoolers, the reduction was 1.07%
(�0.10 g) in consumers of reformulated
products
Eight modeling studies suggested a reduc-
tion in sugar intake, which ranged from 0.2
to 62.1 g/d
(1 observational
and 8 modeling
studies)
����
Low
—
Body weight, kg Change in weight (kg): �1.04 (95%CI, �2.16
to �0.08)
123 (3 RCTs) ����
Very lowb,c,d,e
Overweight and
obesity
One modeling study found reformulation of
soft drinks reduces overweight by 1.0%
(from 35.5% to 34.5%; 0.5 million adults)
and obesity by 2.1% points (from 27.8% to
25.7%; 1 million adults). The same model-
ling study found if fruit juices were ex-
cluded, the reduction will be lower,
overweight will be reduced by 0.7% (0.3
million) and obesity by 1.7% (0.8 million)
Another study found that reformulation
through a cap-and-trade intervention
would reduce obesity by 1.7% (95%CI, 0.9–
2.4) over 20 y
Under the “optimistic” scenario simulated, in
which consumers continued consuming
their preferred product after its reformula-
tion (rather than switching to another
product with the nearest sugar content),
obesity was reduced by 3.2% (95%CI, 2.4–
4.0)
(2 modeling studies) —
Glycemia and
insulinemia
Sugar group resulted in significant increases
of postprandial glycemia and insulinemia
compared to reformulated group
23 (1 RCT) ����
Very lowb,c,d,e
Type 2 diabetes One modeling study found that reformula-
tion of soft drinks could prevent about 274
000–309 000 cases of obesity-related type
2 diabetes over the 2 decades after the
predicted reduction in bodyweight is
achieved. The same modelling study found,
if fruit juices were excluded, 221 000–250
000 cases will be prevented
(2 modeling studies) —
(continued)
Nutrition ReviewsVR Vol. 77(3):181–196 185
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
(4 simulations), and 3) industry-led reformulation (1
that was both a simulation and observational). The
modeling studies that reported reduction in sugar in-
take in grams are illustrated in Figure 5.
Cap and trade. Basu and Lewis43 suggested a cap-and-
trade approach to reformulation. They provided initial
permits to all manufacturers (year 2015) at no cost. The
permits to each manufacturer equalled the mean base-
line sugar emissions level. They then gradually lowered
the cap in each subsequent year; they simulated a 20%
reduction over 20 years (at a rate of 1% annually from
the starting level).
In the less optimistic scenario, Basu and Lewis,43
estimated that calorie intake may fall by 31 kcal/person/
day (95%CI, 29–33) at the 20-year mark, which would
equate to 7.8 g of sugar. In a sensitivity analysis, where
manufacturers passed on the cost of sugar emissions
permits to consumers rather than absorbing the costs
themselves, calorie intake may fall by 36 kcal/person/
day (95%CI, 33–39), which would equate to 9 g of sugar.
The greater decline resulted from consumers switching
Table 2 Continued
Effects of product reformulation on sugar intake and health outcomes
Patient or population: people of all ages
Setting: any setting—randomized trials (� 8 wk duration) or population-based observational or models
Intervention: sugar-reformulation
Comparison: change in sugar intake or health outcomes from pre- to post-reformulation
Outcomes Effect (mean) No. of participants
(studies)
Quality of the
evidence (GRADE)a
Another study found that reformulation
through a cap-and-trade intervention
could reduce type 2 diabetes by 21.7
cases/100 000 people (95%CI, 12.9–30.6)
over 20 y
Under the “optimistic” scenario simulated, in
which consumers continued consuming
their preferred product after its reformula-
tion, type 2 diabetes could be reduced by
37.7/100 000 people (95%CI, 22.4–53.1)
Dental caries One modeling study found a 20% reduction
in sugar intake from possible reformula-
tion, which increased the proportion of 5-
y-olds with no obvious decay by 7%, in-
creased caries-free 15-y-olds by 6%, re-
duced mean dmft/DMFT by 13% among
5-y-olds and 16% among 15-y-olds. The
study also modeled a 40% reduction in
sugar intake and suggested reformulation
as an approach. It found that no dmft/
DMFT increased by 15% for 5-y-olds and
13% for 15-y-olds and mean dmft/DMFT
values decreased by 24% for 5-y-olds and
29% for 15-y-olds
(1 modeling study) —
Deaths averted One modeling study found a reduction in
sugar through reformulation would result
in 746 deaths per year being averted in
the French population
(1 modeling study) —
Abbreviations: CI, confidence interval; DMFT, decayed, missing, or filled teeth.
aGRADE Working Group grades of evidence: High quality: Very confident that the true effect lies close to that of the estimate of the
effect. Moderate quality: Moderately confident in the effect estimate The true effect is likely to be close to the estimate of the effect,
but there is a possibility that it is substantially different. Low quality: Confidence in the effect estimate is limited. The true effect may
be substantially different from the estimate of the effect. Very low quality: Very little confidence in the effect estimate. The true effect
is likely to be substantially different from the estimate of effect.
bOverall risk of bias in the randomized controlled trials was high due to lack of effective blinding.
cThere was serious concern over the heterogeneity of the randomized controlled trials because the v2 P value was significant and the
I2 > 50%.
dRandomized controlled trials are not an entirely suitable assessment of a whole population reformulation intervention because the
duration of trials were short.
eThe number of participants was low.
186 Nutrition ReviewsVR Vol. 77(3):181–196
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
to lower-sugar alternatives when high-sugar products
became more expensive. This approach was estimated
to reduce the prevalence of obesity by 1.7 percentage
points (95%CI, 0.9–2.4; a 4.6% decline) and the inci-
dence of type 2 diabetes by 21.7 cases/100 000 people
(95%CI, 12.9–30.6; a 4.2% decline) over 20 years.
Racial/ethnic minorities may experience the largestdeclines in obesity prevalence and type 2 diabetes inci-
dence. Under the “optimistic” scenario simulated, in
which consumers continued consuming their preferred
Table 3 Summary of the randomized controlled trials included in the systematic review
References Study type (size), population (age),
duration (wk), and BMI
Comparator Outcome measurement
Gatenby et al. (1997)33 RCT (49 [13 control, 17 fat-reformu-
lated, 19 sugar-reformulated])
Adults (18–50 y)
10 wk
BMI
Control (24.5 6 3.3 kg/m2)
Fat reformulated (22.7 6 3.1 kg/m2)
Sugar reformulated group (22.4 6
2.9 kg/m2)
1. Control: maintain usual diet
2. Fat reformulated: used re-
duced-fat products ad libi-
tum in place of habitually
consumed products with tra-
ditional composition
3. Sugar reformulated: used re-
duced-sugar products ad libi-
tum in place of habitually
consumed products with tra-
ditional composition
1. 4 d weighed food diaries
2. Anthropometry
Raben et al. (2002)34 RCT (41 [21 sugar group and
20 reformulated group])
Adults (20–50 y [mean 33 y in sugar,
37 y in reformulated group])
10 wk
BMI
Sugar (28.0 6 0.5 kg/m2)
Reformulated (27.6 6 0.5 kg/m2)
1. Sugar group: received prod-
ucts containing sugar (�70%
from drinks and �30% from
foods to reach a sugar intake
of �2 g/kg body weight;
foods/drinks included soft
drinks, fruit juices, yogurt, ice
cream)
2. Reformulated group: re-
ceived similar products con-
taining reformulated
products (noncaloric sweet-
eners) in similar amounts to
sugar group
1. 7 d dietary records for en-
ergy and nutrient intakes
2. 7 d diaries for monitoring
hunger, fullness, and palat-
ability of the products
3. Anthropometry
4. 24 hour urine collection
5. Diurnal appetite scores
Raben et al. (2011)35 RCT (23 [12 sugar-group and 11
reformulated group])
Adults (20–50 y [mean 35.3 y in
sugar group, 35.5 y in reformu-
lated group])
10 wk
BMI
Sugar (28.7 6 0.7 kg/m2)
Reformulated (27.6 6 0.8 kg/m2)
1. Sugar group: received prod-
ucts containing sugar (�70%
from drinks and �30% from
foods to reach a sugar intake
of �2 g/kg body weight;
foods/drinks included soft
drinks, fruit juices, yogurt, ice
cream)
2. Reformulated group: re-
ceived similar products con-
taining reformulated
products (noncaloric sweet-
eners) in similar amounts to
sugar group
1. 7 d dietary records for en-
ergy and nutrient intakes
2. 7 d diaries for monitoring
hunger, fullness, and palat-
ability of the products
3. Anthropometry
4. Blood pressure
5. Blood sampling and bio-
chemical analysis
Markey et al. (2015)36 RCT, double-blind, crossover (50)
Adults (mean, 31.6 y)
8 wk
BMI (24.0 6 3.3 kg/m2)
1. Regular sugar-containing
products
2. Sugar-reduced products
(reformulated)
Participants were provided with
a choice of 7 drinks, including
juice drinks and soft drinks, and
19 foods, including pasta sauce,
baked beans, muesli, puddings,
and sweet confectionary
products
1. 4 d weighed food diaries
2. A visual analogue scale
questionnaire assessed the
products’ visual appeal,
smell, taste, aftertaste, and
palatability
3. 7 d monitor of physical
activity
4. Anthropometry
5. Blood pressure and arterial
stiffness
6. Blood sampling and bio-
chemical analysis
Abbreviations: BMI, body mass index; RCT, randomized controlled trial.
Nutrition ReviewsVR Vol. 77(3):181–196 187
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
product after its reformulation (rather than switching
to another product with the nearest sugar content
when their preferred product was reformulated), obe-
sity prevalence could be reduced by 3.2 percentage
points (95%CI, 2.4–4.0), and type 2 diabetes incidence
could be reduced by 37.7/100 000 people (95%CI,
22.4–53.1).
Choices program. The Choices program’s 2 main aims
were to encourage the food and drink industry to
reformulate and/or develop new healthier products
and to help consumers make healthier choices by rec-
ognizing the Choices logo.49 The criteria were devel-
oped and set for the Dutch market by an
independent scientific committee advising the
Choices Foundation Board.50 Manufacturers used the
Choices logo if their products met the Choices crite-
ria and were approved. The committee periodically
evaluates the product criteria so as to keep up-to-date
with the latest scientific and reformulation develop-
ments. After each review, a transition period is given
during which manufacturers can reformulate to align
their products with the new criteria and continue to
use the logo.49
Four studies modeled the potential impact of the
Choices program on the Dutch populations’ sugar in-
take, among other nutrients (Table 6).44–47 One of the
studies also modeled the potential impact of the
Choices program on diets in Greece, Spain, the United
States, Israel, China, and South Africa.46 None of them
evaluated the potential impact of nutrient intake
changes on health outcomes. All of the studies showed
the Choices program could encourage reformulation
and reduce sugar intake from current levels. Figure 5
shows the reduction in sugar intake in the models that
reported sugar intake in grams per day.
Industry-led reformulation. Another approach to refor-
mulation is industry-led reformulation. Only 1 study
evaluated the effect of industry-led sugar-reformulation.48
The Food and Drink Industry Ireland (FDII) evaluated
the effect of industry-led reformulation between 2005
and 2012 on the Irish population by 1) calculating the
Figure 2 Change in sugar intake percentage (%).
Figure 3 Change in sugar intake (g/d).
Figure 4 Change in weight (kg).
188 Nutrition ReviewsVR Vol. 77(3):181–196
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
levels of sugar sold via reformulated products pre- and
post-reformulation and 2) calculating daily sugar
intakes of Irish subpopulations at pre- and post-
reformulation. The nutrient composition levels for
sugar, as well as energy, total fat, saturated fat, and/or
sodium, were submitted for nearly 600 products from
the 14 FDII members.
This analysis generated 2 scenarios estimating the
impact of reformulation on the 4 subpopulations.
Scenario A assumes all companies across the food and
drink sector with similar products reformulated in a
similar way to the 14 FDII member companies in-
volved (optimistic); this was the modeling part of the
study. Scenario B assumes only the 14 FDII members
conducted their reported reformulation, thereby esti-
mating the minimum impact of reformulation in
Ireland (conservative); this was the observational part
of the study.
The results showed that during the 7 years (2005–
2012) covered by the research, sugar content fell by
Tedstone et al (2015) (Adults 19-64 years) 
Ma et al (2016) (sugar-sweetened drinks)
France
United Kingdom
Ba�s� et al (2013) (Adults) 
Combris et al (2011) (Children – Breakfast cereals)*
Cap and trade
Basu et al (2014)
Basu et al (2014) (with price increase)
Choices program 
Temme et al (2010)
Roodenburg et al (2011) (Netherlands)
Roodenburg et al (2011) (Greece)
Roodenburg et al (2011) (USA)
Roodenburg et al (2011) (Israel)
Roodenburg et al (2011) (South Africa)
Industry-led 
Roodenburg et al (2011) (Spain)
FDII (2016) (Adults)
Study, year
Tedstone et al (2015) (Children 4-10 years) 
Tedstone et al (2015) (Children 11-18 years) 
FDII (2016) (Teenagers 13-17 years)
FDII (2016) (Children 5-12 years)
FDII (2016) (Preschoolers 1-4 years)
Combris et al (2011) (Children – Biscuits/pastries)*
Combris et al (2011) (Adults – Bread-based products)*
Ma et al (2016) (sugar-sweetened drinks, ex. juices)
Roodenburg et al (2013) - (Netherlands - Adults 19-30 y)*
26.0
9.6
0.4
1.4
9.0
7.8
50.1
58.9
38.1
62.1
17.0
2.9
36.1
1.0
Mean/Median (g)
19.0
17.0 
2.0 
3.5 
1.0
0.6
0.2
40.7
7.8
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
0 10 2030 40 50 60
Sugar reduc�on (g/day)
Tedstone et al (2015) (Adults 65 years +) 14.0 
Figure 5 Sugar reduction (g/d) in modeling studies. Different modelling approaches were included; therefore, the uncertainty measures
are not comparable and are not plotted on the graph.
*Scenario 3/optimistic scenario.
Nutrition ReviewsVR Vol. 77(3):181–196 189
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
14%. In individuals who consumed reformulated prod-
ucts, the mean daily sugar intake was reduced by 2.06%
(observed) to 7.70% (modeled) in adults, 1.36% (ob-
served) to 7.44% (modeled) in teenagers, 1.21% (ob-
served) to 13.56% (modeled) in children, and 1.07%
(observed) to 9.84% (modeled) in preschoolers. Table 1
shows the modelled reduction in sugar (g) intake in the
Irish population’s daily total diet.
Risk of bias of included studies
The overall risk of bias in the RCTs was high due to in-
effective blinding. However, insufficient information
was provided to show whether there was sufficient al-
location concealment in the trials. (Table S1 in the
Supporting Information online).
Observational studies
The Cochrane risk-of-bias tool adapted for nonrandom-
ized controlled studies on population-level interventions
showed the overall risk of bias was high for the single
study48 in this systematic review that included a retro-
spective observational study component (Tables S2 and
S3 in the Supporting Information online).
DISCUSSION
This review assessed the impact of reformulation on
sugar intake and health outcomes. Although the num-
ber of studies was limited and the types of studies and
the types of interventions are extensively heterogeneous,
there is some indication that sugar reformulation can
reduce sugar intake and hence improve health out-
comes. Because the types of studies and the types of
interventions were heterogeneous, the groups of studies
are each discussed separately.
Randomized controlled trials
Meta-analysis of RCTs showed sugar intake and body
weight were reduced after reformulation. However,
there was extensive heterogeneity among the trials and
a number of factors may have contributed to the results.
Sugar and energy intake. The amount of energy and
sugar consumed in the sugar (or control) groups versus
the sugar-reformulated groups in the trials varied. The
2 trials that showed no effect on health outcomes
(weight) had a lower net difference in energy intake by
the end of trial compared with the 2 trials that had an
effect on energy intake and, therefore, health outcomes.
In the Markey et al.36 and Gatenby et al.33 trials, the net
differences between the 2 groups were 181 kcal/day and
52 kcal/day, respectively, whereas in the trial conducted
by Raben et al.,34 the net differences between the 2
groups was 492 kcal/day.
Administration. The way the products were adminis-
tered to the participants was very different across trials.
This may have affected the potential impact on the par-
ticipants. In the Markey et al.36 trial, participants were
provided with sufficient study products from a choice
of 7 drinks and 19 foods. The participants were also
asked to replace habitually used sugar and condiments
with those provided by the investigators ad libitum
throughout the study periods. Similarly, in the trial by
Raben et al.34,35 the participants received products con-
taining sugar and similar reformulated products con-
taining noncaloric sweeteners (in similar amounts to
sugar group). In contrast, in the trial by Gatenby
et al.,33 the participants were told to seek commercially
available reformulated products and record what they
consumed. The participants knew very well they were
consuming reduced-sugar reformulated products over
their regular products. The difference in the way the
products were administered to participants may have
led to differences in energy compensatory behaviors.
Other differences in the trials, such as blinding, and
underreporting in outcome measures (eg, sugar intake),
may have influenced the outcomes. Although there were
attempts in 2 trials34,36 to blind participants to whether
they were on a sugar/control product diet versus a refor-
mulated product diet, the blinding may have been inef-
fective. In the Markey et al.36 study, 83% of the
participants correctly identified the sugar products and
the reformulated products.
Moreover, in the Raben et al.34,35 trial, the partici-
pants were told that they received products containing
Table 4 Methods used in modeling studies of the impact
of sugar reformulation on the UK population
References Method
Bedi et al. (2005)37 Modeled the effect of a 20% re-
duction in sugar intake and
suggested reformulation as
approach
Modeled the effect of a 40% re-
duction in sugar intake and
suggested reformulation as
approach
Tedstone et al. (2015)39 Modeled 50% reformulation in
all of the main contributors of
sugar intake
Ma et al. (2016)38 Modeled 40% reformulation in
soft drinks
Modeled 40% reformulation in
soft drinks (excluding fruit
juice)
190 Nutrition ReviewsVR Vol. 77(3):181–196
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
https://academic.oup.com/nutritionreviews/article-lookup/doi/10.1093/nutrit/nuy015#supplementary-data
https://academic.oup.com/nutritionreviews/article-lookup/doi/10.1093/nutrit/nuy015#supplementary-data
https://academic.oup.com/nutritionreviews/article-lookup/doi/10.1093/nutrit/nuy015#supplementary-data
https://academic.oup.com/nutritionreviews/article-lookup/doi/10.1093/nutrit/nuy015#supplementary-data
https://academic.oup.com/nutritionreviews/article-lookup/doi/10.1093/nutrit/nuy015#supplementary-data
noncaloric sweeteners. At the end of the study, partici-
pants in the reformulated group believed this, but par-
ticipants in the sugar group guessed to some extent the
true content of their products. This could have influ-
enced the participant’s eating behavior so that partici-
pants in the sugar group would have eaten less of their
own products and participants in the reformulated
group would have eaten more of their own products.
Therefore, the net difference in calorie intake between
sugar and reformulated groups may have been smaller
than intended.
All of the trials relied on food diaries as a key out-
come measure because there is no biomarker to assess
sugar intake. Participants tend to underestimate and
underreport their intake of high fat, salt, and sugar
products; therefore, the reported intakes may have
underestimated the real sugar intake and ultimately the
total energy intake.51–54 This outcome measure is also
known to place a burden on participants, and indeed
there was evidence of reduced reporting toward the end
of the Gatenby et al.33 trial.
Further to the concerns above, GRADE allowed for
the assessment of the RCTs’ quality of evidence. Due to
the lack of blinding and allocation, inconsistency, and
imprecision (as a result of low participants in pooled
RCTs), the quality of evidence was very low.
In summary, a limited number of RCTs addressed
the use of reformulation to reduce sugar intake and as-
sess the impact on health outcomes. The mean effect
suggests reformulation has an impact on sugar intake
and body weight. However, long-term RCTs on dietary
manipulation and impact on health outcomes may not
be practical to perform due to expense, adherence, and
legality.
Table 5 Modeling studies of sugar reformulation on the French population
References Method Consumption levels Health outcomes
Battisti et al. (2013)40 Modeled the impact of a refor-
mulation intervention—the
French Nutrition and Health
Programme’s commitments
(PNNS 2)—on nutrient in-
take, including sugar intake
Sugar consumption decreased by
0.4 g (0.4%) in both sexes
For men, the decrease in sugar
intakes was mainly due to refor-
mulations of dairy products and
soft drinks
For men, 1.6% of the objectiveset
for sugars (25% decrease of
added sugars consumption)
was achieved
Combris et al. (2011)41 Modeled different food refor-
mulation scenarios for
breakfast cereals, biscuits
and pastries, and bread-
based products to assess im-
pact on population intake of
sugar, fat, and salt
In breakfast cereals scenarios 1, 2,
and 3, the amount of sugars
supplied to the market would be
reduced by 993–3830 tons. At
the individual intake level for
the average child consumer of
breakfast cereals, this represents
a 2.6%–9.1% decrease in the in-
take of sugars provided by these
products. The daily intake of
sugars provided by breakfast
cereals decreased by 0.4–1.4 g
Biscuits and pastries scenarios 1, 2,
and 3 decreased the amount of
sugars supplied to the market by
1111–8888 tons. In the average
child consumer of biscuits and
pastries, this change represents
a decrease in sugar intake by
0.5%–4.2%. In the children, the
3 scenarios decrease the daily
intake of sugar by 0.03–0.22 g
For adults who are high consum-
ers of bread-based products, the
simulation led to a 0.13–0.63 g
decrease in the daily intake of
sugar
Leroy et al. (2015)42 Modeled reformulation scenar-
ios on sugar intake and
health outcomes
The number of deaths
avoided by reducing sugar
was 746 deaths per year
Nutrition ReviewsVR Vol. 77(3):181–196 191
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
Table 6 Summary of the Choices program modeling studies included in the systematic review
References Country Method Outcome measures (sugar intake)
Roodenburg et al. (2009)44 Netherlands Three scenarios were calculated
and compared using Dutch
National Food Consumption
Survey 2003:
Scenario 1—usual nutrient intakes
“as measured”
Scenario 2—the same as scenario
1, except that all foods that did
not comply with the Choices cri-
teria were replaced, where pos-
sible, by similar foods that did
comply with the Choices criteria
Scenario 3—the same as scenario
2, but corrected for the differ-
ence in energy density between
the original and the replacement
food*
The distribution curve of scenario 2 and
3 shifted to the left of the curve for
scenario 1
Approximately 37% reduction in median
total sugar intakes were observed if the
Dutch population were to eat only
foods that comply with Choices (with
adjustment for energy density) as com-
pared with the “as measured” intakes
Temme et al. (2010)45 Netherlands Three replacement scenarios were
compared using the Dutch con-
sumption survey
The foods not complying with
Choices criteria were replaced
by:
Scenario 1— foods complying
with the Choices criteria and the
real 2007 market shares
Scenario 2—the same as scenario
1 but with 100% market shares
Scenario 3—the existing similar
foods with a composition that al-
ready complied with the Choices
criteria with 100% market shares
Scenarios 1 and 2 included food
groups for which market share
information was available, such
as nonalcoholic drinks, proc-
essed fruit and vegetables,
cheese, dairy and dairy products,
soya foods, prepared meals,
soup, fats (baking and spread-
ing), fat-based sauces, and wa-
ter-based sauces
Scenario 3 included all food
groups
Scenario 1—Sugar intake reduced by 1%
Scenario 2—Sugar intake reduced by 6%
(8.9 g median)
Scenario 3—All noncomplying foods
replaced by complying foods with ac-
tual compositions, estimated intake of
sugar would be 88.5 g/d (95%CI, 85.8–
91.6), a reduction of 36% (50 g) com-
pared with the reference intake
The main contributors to this reduction
are nonalcoholic drinks, responsible for
half of the reduction, and sweets, sweet
snacks, and dairy products (5%–6%
lower intake compared with baseline)
The distribution curve of scenario 3
shifted to the left of the curve for
scenarios 1 and 2
Roodenburg et al. (2011)46 Netherlands, Greece,
Spain, United States,
Israel, China, South
Africa
Average intakes of energy, trans
fatty acids, saturated fats, so-
dium, added sugar, and fiber
were derived from dietary in-
take studies and food consump-
tion surveys. For each of the key
nutrients, these average intakes
were translated into 3 typical
daily menus per country. These
3 menus were compared with
average intakes from 3 Choices
daily menus. Foods from the
typical menus that did not com-
ply with the Choices criteria
were replaced with foods that
did comply and are available on
the market
Added sugar intake reduced by
2.9�62.1 g/d with the Choices menus
For Spain, China, and Israel, typical added
sugar intakes were already below
recommendation
None of the foods in the Chinese and the
South African typical menus were
replaced because added sugar levels
were low (China) or no alternatives
were available on the market (South
Africa). For the other countries, sweet
snacks and cereals were replaced and
also sugar in coffee (in Greece, where
sweeteners are commonly used)
(continued)
192 Nutrition ReviewsVR Vol. 77(3):181–196
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
Modeling of reformulation
All of the modeling studies suggest reformulation can
reduce sugar intake38–42 and improve health out-
comes.37,38,42 The results provide useful insight, reveal-
ing the extent to which reformulation could contribute
to improving health, even in the absence of changes in
individual’s behavior. However, the results must be
considered with caution given the uncertainties sur-
rounding modeling studies.
There are known limitations to national dietary
survey data, which are used extensively in modeling
studies. Many studies show that dietary surveys relying
on self-reported intake data are prone to errors of re-
call.51–57 Therefore, any studies of associations with
health based on self-reported dietary intake data may
potentially underestimate the effect of reformulation
but also underestimate the interventions needed to deal
with excessive sugar intake.
Moreover, there are concerns related to the type of
food composition data used. For example, generic food
composition data do not include many branded and
new products, meaning that several products are aver-
aged to 1 nutrient composition, which may not repre-
sent the products consumed. Indeed, studies have
shown that very similar products can have a wide varia-
tion in nutrient levels, particularly in sugar content.58
Also, nutrient data may not have been correctly
measured or calculated. For instance, the definition of
free sugars excludes other sugars such as lactose and in-
trinsic sugars in fruit and vegetables. However, there
are no verified tests to distinguish the different types of
sugars, and current labeling requirements for total sug-
ars do not distinguish the different types. This can add
a layer of complication when estimating the effect of
reformulation. Also, this can act as a disincentive to
manufacturers attempting to reformulate when their
efforts may be confounded by the presence of milk,
fruit, and vegetables.
Nevertheless, generic food composition databases
are the best open source of nutrition information avail-
able for a wide range of products.
Some modeling studies may have underestimated
the potential impact of reformulation. In the Ma et al.38
study several separate analyses were conducted the
results suggested that the main findings were conserva-
tive. The authors also estimated sugar-sweetened drink
intake more accurately by combining sales and intake
data to overcome the prevalent issue of diet under-
reporting, which has rarely been considered in similar
studies.59,60 Although they tried to correct for under-
reporting, the results are still likely to have underesti-
mated the effect of reformulation. However, some mod-
els may have overestimated the impact because in
foods, unlike drinks, reformulation in practice can be
more complex than simulated.61
At the same time, most of the modeling studies
assessed the initial impact of reformulationand not
Table 6 Continued
References Country Method Outcome measures (sugar intake)
Roodenburg et al. (2013)47 Netherlands Scenarios were calculated and
compared using data from the
2003 Dutch food consumption
survey in young adults (aged
19–30 y) and the Dutch food
composition table
Scenario 1—the “actual intakes”
Scenario 2—same as scenario 1
except all foods that did not
comply were replaced by similar
foods that did comply with the
Choices criteria*
Scenario 3—same as scenario 2
adjusted for the difference in en-
ergy density between the origi-
nal and replacement food
Scenarios 4 and 5 were calculated
where snacks were not or were
partially replaced
Scenario 1—The intake guidelines (15%
of energy intake in Netherlands) for to-
tal sugar were not reached by 95% of
the population
Scenario 2—Median sugar intake was re-
duced from 142.0 to 91.7g/d, reduction
of 50.3g/d. The proportion of the popu-
lation not meeting the recommended
sugar intake guideline was reduced to
71%
Scenario 3—Median sugar intake was re-
duced by 40.7g/d. The percentage not
meeting the recommended sugar in-
take guideline were 80% of the
population
Scenarios 4 and 5—Median sugar intake
was reduced by 42.7 and 38.9 g/d
The snacks scenarios indicated that not
replacing snacks with Choices compli-
ant products resulted in substantially
higher intakes of sugar, indicating
snacks are an important source of sugar
Abbreviations: BMI, body mass index; CI, confidence interval.
*It was hypothesized that consumers may compensate for the reduction in energy intake by eating more foods with lower energy den-
sity. Therefore, when foods were replaced with foods with lower energy, a multiplication factor was applied, so that the total amount
of energy consumed was the same as the energy delivered by the replacement food.
Nutrition ReviewsVR Vol. 77(3):181–196 193
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
subsequent reformulation. Therefore, the long-term im-
pact of reformulation may have been underestimated.
Furthermore, it is likely that once the products with the
lowest level of sugar are reformulated, some products
that were initially more nutritious will also be reformu-
lated to maintain product differentiation.
Modeling studies did not consider practical con-
cerns with reformulation, such as whether the interven-
tion will be implemented by a government jurisdiction
on a voluntary or mandatory basis. There are several
examples of programs implemented by governments
that sought to reduce the population’s sugar intake as
part of a wider intervention to improve population
diets, including programs in England,62 the
Netherlands,63 France,64 and Australia.65 Most of these
programs are voluntary, and therefore their effective-
ness has been questioned.40,65–68 The concern seems to
revolve around lack of compliance from manufacturers;
progress maybe slowed down because there are no con-
sequences for non-compliance. Nevertheless, the results
from the modeling studies are promising.
Different approaches of reformulation
All of the modeling studies of different reformulation
approaches show a reduction in sugar intake and/or
improvement in health outcomes. However, all of the
concerns raised previously in relation to the limitations
of modeling studies would apply to the Basu and
Lewis43 modeling study and the Choices program
studies.44–47
Moreover, the Choices program and health logo/
label initiatives in general are often implemented on a
voluntary basis. Therefore, not all manufacturers will
reformulate their products to comply with the criteria.69
Furthermore, studies that assess only products with
such logos often do not collect data on how many less
healthy products or those not meeting the logo criteria
were introduced during the same time frame, limiting
their ability to evaluate the overall supply chain.70
Logos/labels are intended to enable consumers to
make informed choices, but this can only work if con-
sumers understand the information they are provided.
Also, the need to see reformulation permeate the entire
supply chain argues against the use of logos/labels,
which are used by manufacturers to promote their
products.71 Nevertheless, in the absence of population-
level reformulation interventions, logos/labels can en-
courage some reformulation if the criteria are strict.
Another approach is industry-led reformulation,
which has been shown to reduce sugar intake. Many
manufacturers around the world claim to be reformu-
lating.72 However, when their reformulation plans are
closely assessed, it is found that what is meant by
“reformulation” is actually creating new reduced-sugar
products, which is new product development. New
product development is not reformulation; in such
cases, the manufacturers are not reformulating a prod-
uct to replace an existing one.
Strengths and limitations
This review is the first to investigate the impact of sugar
reformulation on intake and health outcomes and grade
the quality of the evidence. The review captured a wide
range of studies because the inclusion criteria was
broad.
However, several limitations must be acknowl-
edged. The limitations of the findings are those inherent
to the primary research on which they are based, nota-
bly inadequacy of dietary intake and food composition
data, variation in the interventions, and the limited
number of trials and studies.
The exclusive reliance on English-language studies
may have not represented all of the evidence available
and contributed to a language bias.
The terms used for the search strategy are one of
the most important aspects of searching through the lit-
erature. It is crucial to capture as many ways to express
an idea or concept as possible to include as many rele-
vant publications as possible. Because the term
“reformulation” is open to interpretation, the terms
used in the search strategy can probably be improved to
retrieve more studies on reformulation. Furthermore,
the search focused on sugar intake, but there are studies
that focused on energy and calorie intake; therefore, not
including such terms may have limited the evidence
retrieved.
The quality of modeling studies was not assessed
because there are no formal assessment tools for them.
Ultimately, the number of studies on sugar refor-
mulation are limited, with high risk of bias and an over-
all grade of low to very low quality; thus, drawing
conclusions on study outcomes/effects was difficult.
CONCLUSION
The findings of this review suggest that sugar reformu-
lation can be beneficial in reducing sugar intake and
improving health outcomes. Much of the evidence
draws on the potential benefits of population-level
interventions through modeling, and these studies sug-
gest there are health gains. This review provides an im-
portant starting point for ongoing work on sugar
reformulation.
194 Nutrition ReviewsVR Vol. 77(3):181–196
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
Acknowledgments
The authors would like to acknowledge Ilham Gokal for
conducting an independent screening of the citations.
Author contributions. K.M.H. conducted the research;
K.M.H. analyzed the data with assistance from F.J.H.;
K.M.H. wrote the first draft of the manuscript. All authors
contributed to the interpretation of the results and revi-
sion of the manuscript and approved the final version.
Funding. This research received no specific grant from
any funding agency in the public, commercial, or not-
for-profit sectors.
Declaration of interest. K.M.H. is an employee of
Consensus Action on Salt, Sugar, & Health (CASSH).
G.A.M. is Chairman of CASSH. CASSH is a nonprofit
charitable organizations. G.A.M. does not receive any
financial support from the organization.
Supporting Information
The following Supporting Information is available
through the onlineversion of this article at the publish-
er’s website:
Table S1 Risk of bias and support for judgments
for the Randomised Controlled Trials included
Table S2 Risk of bias for observational study: bias
domains for nonrandomized controlled studies on
population-level interventions adapted from the
Cochrane Collaboration’s tool for assessing risk of
bias adapted from McLaren et al.18 and Iheozor-
Ejiofor et al30
Table S3 Risk of bias and support for judgments
for the Food and Drink Industry Ireland (FDII)
observational arm of the study48
REFERENCES
1. Public Health England. UK and Ireland Prevalence and Trends. http://webarchive.
nationalarchives.gov.uk/20170110171021/https://www.noo.org.uk/NOO_about_
obesity/adult_obesity/UK_prevalence_and_trends. Accessed July 2, 2015.
2. Health and Social Care Information Centre. Statistics on Obesity, Physical Activity
and Diet: England, 2014. http://www.hscic.gov.uk/catalogue/PUB13648/Obes-
phys-acti-diet-eng-2014-rep.pdf. Accessed July 1, 2015.
3. Public Health England. Adult Obesity and Type 2 Diabetes. 2014. https://www.gov.
uk/government/uploads/system/uploads/attachment_data/file/338934/Adult_
obesity_and_type_2_diabetes_.pdf Accessed July 2, 2015.
4. Health and Social Care Information Centre. National Diabetes Audit 2007–08. 2009.
http://www.hscic.gov.uk/catalogue/PUB02580/nati-diab-audi-07-08-exec-summ.
pdf. Accessed 1 July 2015.
5. Public Health England. National Dental Epidemiology Programme for England: Oral
Health Survey of Five-Year-Old Children 2012. A Report on the Prevalence and
Severity of Dental Decay. http://www.nwph.net/dentalhealth/Oral%20Health
%205yr%20old%20children%202012%20final%20report%20gateway%20ap-
proved.pdf. Accessed December 2, 2015.
6. PHE. Dental Public Health Epidemiology Programme. Oral Health Survey of
Three-year-old Children 2013. A Report on the Prevalence and Severity of Dental
Decay. 2013. http://www.nwph.net/dentalhealth/reports/DPHEP%20for
%20England%20OH%20Survey%203yr%202013%20Report.pdf. Accessed
December 2, 2015.
7. National Health Service. Executive Summary: Adult Dental Health Survey 2009.
2011. http://www.hscic.gov.uk/catalogue/PUB01086/adul-dent-heal-surv-summ-
them-exec-2009-rep2.pdf. Accessed December 2, 2015.
8. Public Health England. Sugar Reduction: Responding to the Challenge. 2014.
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/
324043/Sugar_Reduction_Responding_to_the_Challenge_26_June.pdf. Accessed
June 1, 2014.
9. Scientific Advisory Committee on Nutrition. Carbohydrates and Health. 2015.
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/
445503/SACN_Carbohydrates_and_Health.pdf. Accessed December 1, 2015.
10. Public Health England. Why 5%? 2015. https://www.gov.uk/government/uploads/
system/uploads/attachment_data/file/446010/Why_5__-_The_Science_Behind_
SACN.pdf. Accessed July 1, 2017.
11. Te Morenga L, Mallard S, Mann J. Dietary sugars and body weight: systematic re-
view and meta-analyses of randomised controlled trials and cohort studies. BMJ.
2013;346:e7492.
12. Johnson RK, Appel LJ, Brands M, et al. Dietary sugars intake and cardiovascular
health: a scientific statement from the American Heart Association. Circulation.
2009;120:1011–1020.
13. Xi B, Li SS, Liu ZL, et al. Intake of fruit juice and incidence of type 2 diabetes: a sys-
tematic review and meta-analysis. PLos One. 2014;9:e93471.
14. Vartanian LR, Schwartz MB, Brownell KD. Effects of soft drink consumption on nu-
trition and health: a systematic review and meta-analysis. Am J Public Health.
2007;97:667–675.
15. World Health Organization. Guideline: Sugars Intake for Adults and Children. 2015.
http://apps.who.int/iris/bitstream/10665/149782/1/9789241549028_eng.pdf?
ua¼1. Accessed July 1, 2015.
16. Moynihan PJ, Kelly SA. Effect on caries of restricting sugars intake: systematic re-
view to inform WHO guidelines. J Dent Res. 2014;93:8–18.
17. Public Health England. NDNS: Results from Years 5 and 6 (Combined). 2016.
https://www.gov.uk/government/statistics/ndns-results-from-years-5-and-6-com-
bined. Accessed June 2, 2016.
18. McLaren L, Sumar N, Barberio AM, et al. Population-level interventions in govern-
ment jurisdictions for dietary sodium reduction. Cochrane Database of Syst Rev.
2016;9:CD010166.
19. MacGregor GA, Hashem KM. Action on sugar-lessons from UK salt reduction pro-
gramme. Lancet. 2014;383:929–931.
20. van Raaij J, Hendriksen M, Verhagen H. Potential for improvement of population
diet through reformulation of commonly eaten foods. Public Health Nutr.
2009;12:325–330.
21. He FJ, Brinsden HC, MacGregor GA. Salt reduction in the United Kingdom: a suc-
cessful experiment in public health. J Hum Hypertens. 2014;28:345–352.
22. Sutherland J, Edwards P, Shankar B, et al. Fewer adults add salt at the table after
initiation of a national salt campaign in the UK: a repeated cross-sectional analysis.
Br J Nutr. 2013;110:552–558.
23. Department of Health. Department of Health: Assessment of Dietary Sodium Levels
Among Adults (Aged 19-64) in England. 2012. http://transparency.dh.gov.uk/2012/
06/21/sodium-levels-among-adults/. Accessed December 2, 2015.
24. He FJ, Pombo-Rodrigues S, Macgregor GA. Salt reduction in England from 2003 to
2011: its relationship to blood pressure, stroke and ischaemic heart disease mor-
tality. BMJ Open. 2014;4:e004549.
25. Shamseer L, Moher D, Clarke M, et al; the PRISMA-P Group. Preferred reporting
items for systematic review and meta-analysis protocols (PRISMA-P) 2015: elabo-
ration and explanation. BMJ 2015;350:g7647.
26. Hashem KM, He FJ, MacGregor GA. Systematic review of the literature on the ef-
fectiveness of product reformulation measures to reduce the sugar content of
food and drink on the population’s sugar consumption and health: a study proto-
col. BMJ Open. 2016;6:e011052.
27. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analy-
ses. BMJ. 2003;327:557–560.
28. Higgins JPT, Altman DG, Gøtzsche PC, et al. The Cochrane Collaboration’s tool for
assessing risk of bias in randomised trials. BMJ. 2011;343:d5928.
29. Sanderson S, Tatt ID, Higgins JP. Tools for assessing quality and susceptibility to
bias in observational studies in epidemiology: a systematic review and annotated
bibliography. Int J Epidemiol. 2007;36:666–676.
30. Iheozor-Ejiofor ZWH, Walsh T, O’Malley L, et al. Water fluoridation for the preven-
tion of dental caries. Cochrane Database Syst Rev. 2015;6:CD010856.
31. GRADE Working Group. Grading of Recommendations Assessment, Development
and Evaluation. http://www.gradeworkinggroup.org/. Accessed September 7,
2017.
32. BMJ. What is GRADE? http://clinicalevidence.bmj.com/x/set/static/ebm/learn/
665072.html. Last updated September 1, 2012. Accessed September 2, 2017.
Nutrition ReviewsVR Vol. 77(3):181–196 195
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
http://webarchive.nationalarchives.gov.uk/20170110171021/https://www.noo.org.uk/NOO_about_obesity/adult_obesity/UK_prevalence_and_trends
http://webarchive.nationalarchives.gov.uk/20170110171021/https://www.noo.org.uk/NOO_about_obesity/adult_obesity/UK_prevalence_and_trends
http://webarchive.nationalarchives.gov.uk/20170110171021/https://www.noo.org.uk/NOO_about_obesity/adult_obesity/UK_prevalence_and_trends
http://www.hscic.gov.uk/catalogue/PUB13648/Obes-phys-acti-diet-eng-2014-rep.pdf
http://www.hscic.gov.uk/catalogue/PUB13648/Obes-phys-acti-diet-eng-2014-rep.pdf
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/338934/Adult_obesity_and_type_2_diabetes_.pdf
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/338934/Adult_obesity_and_type_2_diabetes_.pdf
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/338934/Adult_obesity_and_type_2_diabetes_.pdf
http://www.hscic.gov.uk/catalogue/PUB02580/nati-diab-audi-07-08-exec-summ.pdf
http://www.hscic.gov.uk/catalogue/PUB02580/nati-diab-audi-07-08-exec-summ.pdfhttp://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/Oral%20Health%205yr%20old%20children%202012%20final%20report%20gateway%20approved.pdf
http://www.nwph.net/dentalhealth/reports/DPHEP%20for%20England%20OH%20Survey%203yr%202013%20Report.pdf
http://www.nwph.net/dentalhealth/reports/DPHEP%20for%20England%20OH%20Survey%203yr%202013%20Report.pdf
http://www.nwph.net/dentalhealth/reports/DPHEP%20for%20England%20OH%20Survey%203yr%202013%20Report.pdf
http://www.nwph.net/dentalhealth/reports/DPHEP%20for%20England%20OH%20Survey%203yr%202013%20Report.pdf
http://www.nwph.net/dentalhealth/reports/DPHEP%20for%20England%20OH%20Survey%203yr%202013%20Report.pdf
http://www.nwph.net/dentalhealth/reports/DPHEP%20for%20England%20OH%20Survey%203yr%202013%20Report.pdf
http://www.nwph.net/dentalhealth/reports/DPHEP%20for%20England%20OH%20Survey%203yr%202013%20Report.pdf
http://www.nwph.net/dentalhealth/reports/DPHEP%20for%20England%20OH%20Survey%203yr%202013%20Report.pdf
http://www.nwph.net/dentalhealth/reports/DPHEP%20for%20England%20OH%20Survey%203yr%202013%20Report.pdf
http://www.nwph.net/dentalhealth/reports/DPHEP%20for%20England%20OH%20Survey%203yr%202013%20Report.pdf
http://www.nwph.net/dentalhealth/reports/DPHEP%20for%20England%20OH%20Survey%203yr%202013%20Report.pdf
http://www.nwph.net/dentalhealth/reports/DPHEP%20for%20England%20OH%20Survey%203yr%202013%20Report.pdf
http://www.nwph.net/dentalhealth/reports/DPHEP%20for%20England%20OH%20Survey%203yr%202013%20Report.pdf
http://www.nwph.net/dentalhealth/reports/DPHEP%20for%20England%20OH%20Survey%203yr%202013%20Report.pdf
http://www.nwph.net/dentalhealth/reports/DPHEP%20for%20England%20OH%20Survey%203yr%202013%20Report.pdf
http://www.hscic.gov.uk/catalogue/PUB01086/adul-dent-heal-surv-summ-them-exec-2009-rep2.pdf
http://www.hscic.gov.uk/catalogue/PUB01086/adul-dent-heal-surv-summ-them-exec-2009-rep2.pdf
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/324043/Sugar_Reduction_Responding_to_the_Challenge_26_June.pdf
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/324043/Sugar_Reduction_Responding_to_the_Challenge_26_June.pdf
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/445503/SACN_Carbohydrates_and_Health.pdf
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/445503/SACN_Carbohydrates_and_Health.pdf
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/446010/Why_5__-_The_Science_Behind_SACN.pdf
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/446010/Why_5__-_The_Science_Behind_SACN.pdf
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/446010/Why_5__-_The_Science_Behind_SACN.pdf
http://apps.who.int/iris/bitstream/10665/149782/1/9789241549028_eng.pdf? ua=1
http://apps.who.int/iris/bitstream/10665/149782/1/9789241549028_eng.pdf? ua=1
http://apps.who.int/iris/bitstream/10665/149782/1/9789241549028_eng.pdf? ua=1
https://www.gov.uk/government/statistics/ndns-results-from-years-5-and-6-combined
https://www.gov.uk/government/statistics/ndns-results-from-years-5-and-6-combined
http://transparency.dh.gov.uk/2012/06/21/sodium-levels-among-adults/
http://transparency.dh.gov.uk/2012/06/21/sodium-levels-among-adults/
http://www.gradeworkinggroup.org/
http://clinicalevidence.bmj.com/x/set/static/ebm/learn/665072.html
http://clinicalevidence.bmj.com/x/set/static/ebm/learn/665072.html
33. Gatenby SJ, Aaron JI, Jack VA, et al. Extended use of foods modified in fat and
sugar content: nutritional implications in a free-living female population. Am J
Clin Nutr. 1997;65:1867–1873.
34. Raben A, Vasilaras TH, Moller AC, et al. Sucrose compared with artificial sweet-
eners: different effects on ad libitum food intake and body weight after 10 wk of
supplementation in overweight subjects. Am J Clin Nutr. 2002;76:721–729.
35. Raben A, Moller BK, Flint A, et al. Increased postprandial glycaemia, insulinemia,
and lipidemia after 10 weeks’ sucrose-rich diet compared to an artificially sweet-
ened diet: a randomised controlled trial. Food Nutr Res. 2011;55. doi:10.3402/
fnr.v55i0.5961.
36. Markey OL, Jeune J, Lovegrove JA. Energy compensation following consumption
of sugar-reduced products: a randomized controlled trial. Eur J Nutr.
2015;55:2137–2149.
37. Bedi R. A futures study of the dental health of children and young people. Health
Educ J. 2005;64:1–111.
38. Ma Y, He FJ, Yin Y, et al. Gradual reduction of sugar in soft drinks without substitu-
tion as a strategy to reduce overweight, obesity, and type 2 diabetes: a modelling
study. Lancet Diabetes Endocrinol. 2016;4:105–114.
39. Tedstone A, Targett V, Allen R. Sugar Reduction: The Evidence for Action Annexe 5:
Food Supply. 2015. https://www.gov.uk/government/uploads/system/uploads/
attachment_data/file/470176/Annexe_5._Food_Supply.pdf. Accessed June 2, 2016
40. Battisti C-Perrin C, Gauvreau-Beziat J, Volatier JL, et al. Assessment of the Potential
Impact of the Nutritional Commitments of Food Operators on French Nutrient
Intakes. IUNS 20th International Congress of Nutrition. 2013.
41. Combris P, Goglia R, Henini M, et al. Improvement of the nutritional quality of
foods as a public health tool. Public Health. 2011;125:717–724.
42. Leroy P, Requillart V, Soler LG, et al. An assessment of the potential health impacts
of food reformulation. Eur J Clin Nutr. 2015;70:694–699.
43. Basu SMP, Lewis KMMS. Reducing added sugars in the food supply through a
cap-and-trade approach. Am J Public Health. 2014;104:2432–2438.
44. Roodenburg AJC, Temme EHM, Davies HO, et al. Potential impact of the choices
programme on nutrient intakes in the Dutch population. Nutr Bull.
2009;34:318–323.
45. Temme EH, van der Voet H, Roodenburg AJ, et al. Impact of foods with health
logo on saturatedfat, sodium and sugar intake of young Dutch adults. Public
Health Nutr. 2010;14:635–644.
46. Roodenburg AJC, Schlatmann A, Dötsch-Klerk M, et al. potential effects of nutrient
profiles on nutrient intakes in the Netherlands, Greece, Spain, USA, Israel, China
and South-Africa. PLos One. 2011;6:e14721.
47. Roodenburg AJC, van Ballegooijen AJ, Dötsch-Klerk M, et al. Modelling of usual
nutrient intakes: potential impact of the choices programme on nutrient intakes
in young Dutch adults. PLoS One. 2013;8:e72378.
48. Food and Drink Industry Ireland. Estimating the Impact of Reformulation by 14 FDII
Members on the Irish Population. http://www.fooddrinkireland.ie/Sectors/FDI/FDI.
nsf/vPages/Publications�fdi-creme-global-reformulation-report-27-01-2016/$file/
TheþFDIþCremeþGlobalþReformulationþProject.pdf. Accessed February 1, 2017.
49. Choices. Foundation Choices. http://www.hetvinkje.nl/organisatie/stichting-ik-kies-
bewust/. Accessed April 2, 2017.
50. Roodenburg AJC, Popkin BM, Seidell JC. Development of international criteria for
a front of package food labelling system: the International Choices Programme.
Eur J Clin Nutr. 2011;65:1190–1200.
51. Hebert JR, Ebbeling CB, Matthews CE, et al. Systematic errors in middle-aged wom-
en’s estimates of energy intake: comparing three self-report measures to total en-
ergy expenditure from doubly labeled water. Ann Epidemiol. 2002;12:577–586.
52. Lara JJ, Scott JA, Lean MEJ. Intentional mis-reporting of food consumption and its
relationship with body mass index and psychological scores in women. J Hum
Nutr Diet. 2004;17:209–218.
53. Rennie KL, Coward A, Jebb SA. Estimating under-reporting of energy intake in die-
tary surveys using an individualised method. BJN. 2007;97:1169–1176.
54. Archer E, Hand GA, Blair SN. Correction: validity of U.S. Nutritional Surveillance: na-
tional Health and Nutrition Examination Survey Caloric Energy Intake Data, 1971–
2010. PLoS One. 2013;8:e76632.
55. Zhang J, Temme EH, Sasaki S, et al. Under- and overreporting of energy intake us-
ing urinary cations as biomarkers: relation to body mass index. Am J Epidemiol.
2000;152:453–462.
56. Heerstrass D, Ock�e M, Bueno-de-Mesquita H, et al. Underreporting of energy, pro-
tein and potassium intake in relation to body mass index. Int J Epidemiol.
1998;27:186–193.
57. Lissner L, Heitmann BL, Lindroos AK. Measuring intake in free-living human sub-
jects: a question of bias. Proc Nutr Soc. 1998;57:333–339.
58. Pombo-Rodrigues S, Hashem KM, He FJ, et al. Salt and sugars content of breakfast
cereals in the UK from 1992 to 2015. Public Health Nutr. 2017;20:1500–1512.
59. Briggs ADM, Mytton OT, Kehlbacher A, et al. Overall and income specific effect on
prevalence of overweight and obesity of 20% sugar sweetened drink tax in UK:
econometric and comparative risk assessment modelling study. BMJ.
2013;347:f6189.
60. Lin BH, Smith TA, Lee JY, et al. Measuring weight outcomes for obesity interven-
tion strategies: the case of a sugar-sweetened beverage tax. Econ Hum Biol.
2011;9:329–341.
61. Food and Drink Federation. Reformulation Guide Spotlight on Sugars For Small to
Medium Sized Companies. 2016. https://www.fdf.org.uk/corporate_pubs/
Reformulation-Guide-Sugars-Aug2016.pdf. Accessed September 2, 2016.
62. Department of Health. Calorie Reduction Pledge Development Tool—Using
Industry Best Practice Examples. Department of Health: London, UK; 2015.
63. Akkoordverbetering. Agree Improving Product Composition—Sector-wide agree-
ments. http://www.akkoordverbeteringproductsamenstelling.nl/Afspraken_en_
resultaten/Sectorbrede_afspraken. Accessed January 3, 2017.
64. Hercberg S, Chat-Yung S, Chaulia M. The French National Nutrition and Health
Program: 2001-2006-2010. Int J Public Health. 2008;53:68–77.
65. Tamara Elliott HT, Sacks G, Dunford E, et al. A systematic interim assessment of the
Australian Government’s Food and Health Dialogue. Med J Aust. 2014;200:92–95.
66. Knai C, Petticrew M, Durand MA, et al. Has a public-private partnership resulted in
action on healthier diets in England? An analysis of the Public Health
Responsibility Deal food pledges. Food Policy. 2015;54:1–10.
67. Durand MA, Petticrew M, Goulding L, et al. An evaluation of the Public Health
Responsibility Deal: informants’ experiences and views of the development,
implementation and achievements of a pledge-based, public-private partner-
ship to improve population health in England. Health Policy. 2015;119:
1506–1514.
68. Milder I, Westenbrink S, Toxopeus I, et al. Monitoring Food Reformulation in the
Netherlands: Sodium, Saturated Fat and Sugar Levels of Foods in 2011 and 2014.
2015. https://www.rivm.nl/Documenten_en_publicaties/Wetenschappelijk/
Rapporten/2015/februari/Monitor_Productsamenstelling_voor_zout_verzadigd_
vet_en_suiker_RIVM_Herformuleringsmonitor_2014. Accessed June 1, 2016.
69. Vyth EL, Steenhuis IH, Mallant SF, et al. A front-of-pack nutrition logo: a quantita-
tive and qualitative process evaluation in the Netherlands. J Health Commun.
2009;14:631–645.
70. Vyth EL, Steenhuis IHM, Roodenburg AJC, et al. Front-of-pack nutrition label stim-
ulates healthier product development: A quantitative analysis. IJBNPA 2010;7. doi:
10.1186/1479-5868-7-65.
71. Bruce Traill W, Bech-Larsen T, Gennaro L, et al. Reformulation for Healthier Food: A
Qualitative Assessment Of Alternative Approaches. http://ageconsearch.umn.edu/
bitstream/123511/2/Traill_
ReformulationforHealthierFoodAQualitativeAssessmentofAlternativeApproches.
pdf. Accessed January 1, 2017.
72. Vlassopoulos A, Masset G, Charles VR, et al. A nutrient profiling system for the
(re)formulation of a global food and beverage portfolio. Eur J Nutr.
2016;56:1105–1122.
196 Nutrition ReviewsVR Vol. 77(3):181–196
D
ow
nloaded from
 https://academ
ic.oup.com
/nutritionreview
s/article/77/3/181/5280774 by guest on 03 N
ovem
ber 2020
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/470176/Annexe_5._Food_Supply.pdf
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/470176/Annexe_5._Food_Supply.pdf
http://www.fooddrinkireland.ie/Sectors/FDI/FDI.nsf/vPages/Publications&sim;fdi-creme-global-reformulation-report-27-01-2016/&dollar;file/The+FDI+Creme+Global+Reformulation+Project.pdf
http://www.fooddrinkireland.ie/Sectors/FDI/FDI.nsf/vPages/Publications&sim;fdi-creme-global-reformulation-report-27-01-2016/&dollar;file/The+FDI+Creme+Global+Reformulation+Project.pdf
http://www.fooddrinkireland.ie/Sectors/FDI/FDI.nsf/vPages/Publications&sim;fdi-creme-global-reformulation-report-27-01-2016/&dollar;file/The+FDI+Creme+Global+Reformulation+Project.pdf
http://www.fooddrinkireland.ie/Sectors/FDI/FDI.nsf/vPages/Publications&sim;fdi-creme-global-reformulation-report-27-01-2016/&dollar;file/The+FDI+Creme+Global+Reformulation+Project.pdf
http://www.fooddrinkireland.ie/Sectors/FDI/FDI.nsf/vPages/Publications&sim;fdi-creme-global-reformulation-report-27-01-2016/&dollar;file/The+FDI+Creme+Global+Reformulation+Project.pdf
http://www.fooddrinkireland.ie/Sectors/FDI/FDI.nsf/vPages/Publications&sim;fdi-creme-global-reformulation-report-27-01-2016/&dollar;file/The+FDI+Creme+Global+Reformulation+Project.pdf
http://www.fooddrinkireland.ie/Sectors/FDI/FDI.nsf/vPages/Publications&sim;fdi-creme-global-reformulation-report-27-01-2016/&dollar;file/The+FDI+Creme+Global+Reformulation+Project.pdf
http://www.fooddrinkireland.ie/Sectors/FDI/FDI.nsf/vPages/Publications&sim;fdi-creme-global-reformulation-report-27-01-2016/&dollar;file/The+FDI+Creme+Global+Reformulation+Project.pdf
http://www.fooddrinkireland.ie/Sectors/FDI/FDI.nsf/vPages/Publications&sim;fdi-creme-global-reformulation-report-27-01-2016/&dollar;file/The+FDI+Creme+Global+Reformulation+Project.pdf
http://www.hetvinkje.nl/organisatie/stichting-ik-kies-bewust/
http://www.hetvinkje.nl/organisatie/stichting-ik-kies-bewust/
https://www.fdf.org.uk/corporate_pubs/Reformulation-Guide-Sugars-Aug2016.pdf
https://www.fdf.org.uk/corporate_pubs/Reformulation-Guide-Sugars-Aug2016.pdf
http://www.akkoordverbeteringproductsamenstelling.nl/Afspraken_en_resultaten/Sectorbrede_afsprakenhttp://www.akkoordverbeteringproductsamenstelling.nl/Afspraken_en_resultaten/Sectorbrede_afspraken
https://www.rivm.nl/Documenten_en_publicaties/Wetenschappelijk/Rapporten/2015/februari/Monitor_Productsamenstelling_voor_zout_verzadigd_vet_en_suiker_RIVM_Herformuleringsmonitor_2014
https://www.rivm.nl/Documenten_en_publicaties/Wetenschappelijk/Rapporten/2015/februari/Monitor_Productsamenstelling_voor_zout_verzadigd_vet_en_suiker_RIVM_Herformuleringsmonitor_2014
https://www.rivm.nl/Documenten_en_publicaties/Wetenschappelijk/Rapporten/2015/februari/Monitor_Productsamenstelling_voor_zout_verzadigd_vet_en_suiker_RIVM_Herformuleringsmonitor_2014
http://ageconsearch.umn.edu/bitstream/123511/2/Traill_ReformulationforHealthierFoodAQualitativeAssessmentofAlternativeApproches.pdf
http://ageconsearch.umn.edu/bitstream/123511/2/Traill_ReformulationforHealthierFoodAQualitativeAssessmentofAlternativeApproches.pdf
http://ageconsearch.umn.edu/bitstream/123511/2/Traill_ReformulationforHealthierFoodAQualitativeAssessmentofAlternativeApproches.pdf
http://ageconsearch.umn.edu/bitstream/123511/2/Traill_ReformulationforHealthierFoodAQualitativeAssessmentofAlternativeApproches.pdf
	nuy015-TF1
	nuy015-TF2
	nuy015-TF3
	nuy015-TF4
	nuy015-TF5
	nuy015-TF6
	nuy015-TF7
	nuy015-TF9