The Caloric Calculator

Average Caloric Impact of Childhood Obesity Interventions

Y. Claire Wang, MD, ScD, Amber Hsiao, MPH, C. Tracy Orleans, PhD, Steven L. Gortmaker, PhD
This activity is available for CME credit. See page A4 for information.

Background: The childhood obesity epidemic reects the daily accumulation of an energy
gap excess calories consumed over calories expended. Population-level interventions to
reverse the epidemic can be assessed by the degree to which they increase energy expenditure
and/or reduce caloric intake. However, no common metric exists for such comparative
assessment.

Purpose: To develop a common metric, the Average Caloric Impact (ACI), for estimating and
comparing population-level effect sizes of a range of childhood obesity interventions.

Methods: An iterative, collaborative process was used to review literature from 1996 to 2012
and select illustrative interventions showing effects on youth diet and/or activity levels, energy
balance, and weight. The ACIs of physical activity interventions were estimated based on
program reach, frequency, duration, and intensity and mean body weight of the targeted age and
gender group from the 20092010 National Health and Nutrition Examination Survey. ACIs of
dietary interventions were based on reach and changes in foods and/or beverages consumed.

Results: Fifteen interventions informed by 29 studies were included, ranging from individual
behavioral to population-level policies. A web tool, the Caloric Calculator, was developed to
allow researchers and policymakers to estimate the ACIs of interventions on target populations
with reference to energy gap reductions required to reach the nations Healthy People
childhood obesity goals.

Conclusions: The Caloric Calculator and ACIs provide researchers and policymakers with a
common metric for estimating the potential effect sizes of various interventions for reducing
childhood obesity, providing a platform for evidence-based dialogues on new program or policy
approaches as data emerge.
(Am J Prev Med 2013;45(2):e3e13) & 2013 American Journal of Preventive Medicine

Background

he obesity epidemic costs the U.S. $147$210

billion in annual healthcare costs.1 Although the
trends have shown some signs of leveling, more
than one third of U.S. adults and nearly 17% of children
and adolescents are obese.2 As a result, it was predicted
that one in three children born in 2000 would be
diagnosed with type 2 diabetes in his or her lifetime.3
From the Department of Health Policy & Management (Wang, Hsiao),
Mailman School of Public Health, Columbia University, New York, New
York; the Robert Wood Johnson Foundation (Orleans), Princeton, New
Jersey; and Department of Society, Health, and Human Development
(Gortmaker), Harvard School of Public Health, Cambridge, Massachusetts
Address correspondence to: Y. Claire Wang, MD, ScD, Department
of Health Policy and Management, Mailman School of Public
Health, Columbia University, 600 W 168th St, Rm 602 New York NY
10032. E-mail: Ycw2102@columbia.edu.
0749-3797/$36.00
http://dx.doi.org/10.1016/j.amepre.2013.03.012

The rise in childhood obesity since the early 1970s

reects the accumulation of the small daily energy
gapthe excess of calories consumed over calories
expended.4,5 Previous analyses estimated that an average
surplus of 110165 kcal/day in energy accounted for the
excess weight gain seen in U.S. children and youth over
a 10-year period.4 Thus, effective interventions would
have to bring about a net reduction in this energy gap to
reverse the epidemic. A recent study estimated that
among U.S. children aged 219 years, a net reduction
of 64 kcal/day per capita in energy surplus would be
needed to achieve the Healthy People 2020 childhood
obesity goals, with a range from 22 kcal/day for those
aged 25 years, to 77 kcal/day for those aged 611
years,
98 kcal/day for those aged 1219 years, and much
higher levels among those who are already overweight or
obese.5
The evidence base for population-level interventions
to reduce childhood obesity levels has grown rapidly,

ranging
& 2013 American Journal of Preventive Medicine

Published by Elsevier Inc.

Am J Prev Med 2013;45(2):e3e13 e3

e4

Wang et al / Am J Prev Med 2013;45(2):e3e13

from strategies to change individual behaviors to those

that seek to alter policies, environments, and social
norms. In most cases, however, these policies or
programs are evaluated independently. No common
metric exists to allow comparative assessments of
effects across interven- tions with varied congurations
for a target population.6,7
In the current paper, the Average Caloric Impact
(ACI) is proposed as a metric to gauge the populationlevel average effect on daily calories expended/consumed.
This metric was applied to an illustrative set of interventions evaluated in the literature. Greater emphasis
was placed on population-, school-, or state-level programs than on medical treatments of overweight/obese
youth. The results are presented using a user-friendly
web tool, the Caloric Calculator.

Method
s
Selection of Interventions
Using recently published reviews, a set of obesity prevention
interventions targeting U.S. children and adolescents aged
25 years (preschool); 611 years (primary school); 1214 years
(middle school); and/or 1518 years (high school) was selected.
Target populations were dened by grade level based on the
divisions within the typical K12 system. Mean height and weight
for each age group (by gender) were based on the nationally
representative 20092010 National Health and Nutrition Examination Survey (NHANES).
From an initial list of 67 studies published between 2000 and
8
2009, as reviewed by Brennan et al., only seven physical
activity interventions were included that lasted 46 months and
reported outcome measures that were sufcient to have an
inuence on calories. For example, several studies of school lunch
programs or wellness policies were excluded because they
reported consump- tion of only specic nutrients (e.g., % fat),
and/or servings of fruits and vegetables, rather than changes in
total calories consumed or body weight. Similarly, many
evaluations of physical activity programs did not use objective
measures of activity levels (e.g., accelerometers) and thus were
unable to inform changes in energy expenditure.
An iterative and collaborative process was used to identify an
additional 22 studies published between 1996 and 2012; of these,
12 were empirical studies that met the research design and
measurement standards used in the Brennan et al. review. The
remaining studies provided inputs for the model-based estimates.
For dietary interventions selected, the studies assessed changes in
daily caloric intake before and after the intervention
9
(e.g., California schools competitive foods standards ). For studies
reporting changes in consumption of particular foods and/or
beverages, published estimates on the average caloric contribution
of these foods and beverages in the indicated setting (e.g.,
removing sugar-sweetened beverages from schools10) were used.
Strategies were categorized by implementation level (individual,
school, state/national). Because empirical data were lacking for
some strategies (e.g., promoting walking to schools), analytic
models were used to incorporate available evidence to estimate
the likely caloric effect of these strategies, if broadly implemented.

Caloric Impact Calculations

Physical activity interventions.

The physical activity interventions were placed into one of the following categories:
(1) varied school physical education (PE) classes; (2) school PE
interventions designed to increase moderate-to-vigorous physical
activity (MVPA) levels to achieve more active PE; (3) afterschool
physical activity programs; and (4) active commuting (e.g.,
walking) to school. When multiple high-quality studies were
available within a category, the study with the largest effect size
was typically used to represent the best-possible outcome and
population-level implementation.
The effect of the intervention on daily caloric impact
was estimated based on the calculated basal metabolic rate (BMR,
which is a function of age, gender, and body weight), as well
as the frequency (e.g., twice a week); duration (e.g., 30 minutes);
and the intensity of the physical activity (e.g., moderate/vigorous).
BMR for an average-weight child is calculated based on
11
published equa- tions. Intervention intensity was estimated in
METs, representing the amount of energy expended from carrying
out a specic activity
relative to sitting quietly (MET value of 1.0) for a dened period
of time. For instance, walking at a pace of 3 miles per hour
represents an average intensity of 3.3 METs, which burns 3.3
times as many calories than sitting quietly for the same
12
individual.
Pre-intervention activity levels were based on published
baseline measures of study participants and/or national
averages. When MET values were not reported, activity-specic
MET values from
the Ainsworth Compendium for adults12 were combined with
calculated youth-specic BMR estimates, following recommenda13

tions by Ridley et al. Table 1 provides examples of how various

inputs affect the number of calories expended by different
physical
activity
interventions.

Dietary interventions.

Dietary interventions were similarly

reviewed and categorized. For example, a number of interventions
only measured changes in fruit and vegetable intake, and were
excluded because net impact on caloric intake could not be
estimated. One study that empirically measured the caloric impact
of competitive food policies in high schools was included.9
The other ve dietary interventions (e.g., reducing intake of
calories from chips) were estimated based on the authors
calculations.
For policy interventions with limited direct, empirical data
(e.g., removing sugar-sweetened beverages [SSBs] from schools,
14
and a portion-size cap on sugary drinks sold in New York City ),
dietary data from NHANES were used to inform the baseline
consump- tion level among those who would be hypothetically
affected by the policy. For example, NHANES 19992004
showed that SSBs contributed an average of 224 kcal/day to the
overall caloric intake of U.S. children and adolescents, and
7%15% of SSBs were consumed in schools.10 The estimated
caloric impact of replacing all SSB intake from schools (in
session 180 days a year) with water was averaged across the
whole calendar year.

Combined physical activity/dietary interventions.

Sonneville and Gortmaker15 have estimated that every 1-hour increase

in TV watching is associated with a 105.5-kcal increase in net total

energy intake, or a 92-kcal increase in energy intake for video- or

computer-game playing. Their ndings are consistent with a

www.ajpmonline.org

Wang et al / Am J Prev Med 2013;45(2):e3e13

e5

Table 1. Daily caloric effects of physical activity for select groups using Schoeld equations

Population

Average
weight (kg)

Schoeld
equation
a
(BMR)

Inputs
Duration
METs (minutes/day)

Intervention

Schoolb
based?

Caloric effect
c
(kcal/day)

Boys, age in years

25

18

22.706 kg 504.3

Add 30 minutes/
day of walking

2.3

30

No

44

611

34

22.706 kg 504.3

Add 30 minutes/
day of jogging

30

No

186

1214

59

17.686 kg 658.2

Add 15 minutes/
day of PE

2.6

15

Yes

23

1518

77

17.686 kg 658.2

Implement SPARK

3.5

30

Yes

73

25

17

20.315 kg 485.9

Add afterschool
program

3.5

10.5

Yes

11

611

35

13.384 kg 485.9

Make PE more
active

Varies

60

Yes

1214

57

13.384 kg 692.6

Add 30 minutes/
day of PE

2.6

30

Yes

39

1518

65

13.384 kg 692.6

Add 10 minutes/
day of jogging

10

No

Girls, age in years

76

The Schoeld equations are grouped by gender and age groups (broken down as 03 years, 310 years, and 1018 years). Because of this, some
age groups have the same equations.
If the intervention is applied over a full school year, it multiples the caloric impact by 180 days. This is then averaged over 365 days to account for
no change in activity on holidays, weekends, and summer vacation.
c
Daily caloric impact (BMR X METs X duration in minutes) C 1440 minutes/day
d
12
The MET value for Make PE more active is a composite of MET values from ve different activities, based on the Ainsworth Compendium : lying
down, sitting, standing, walking, and running. The change in METs from the intervention depends on user input of baseline versus target % MVPA.
BMR, basal metabolism rate; MVPA, moderate-to-vigorous physical activity; PE, physical education; SPARK, Sports, Play, and Active Recreation
for Kids
b

previously published randomized trial, which found that reducing

16
TV watching among children led to lower caloric intake. It was
hypothesized that children who spend more time watching TV or
playing video games may be more exposed and/or inuenced by
food advertising through characters present in commercials and
1719
interactive games that can shape food preferences and intake.

Online Caloric Calculator Tool

Accompanying the current paper is a web-based tool (www.
caloriccalculator.org) designed to help users visualize and query
the estimated caloric effects of dened interventions within a
dened target population. Programmed in PHP script for HTML,
the tool allows users to choose one or more interventions and
customize their congurations. For example, the user can select as
the target Boys and Middle School (1214) from the dropdown menu, and implement an intervention to increase PE
intensity (e.g., moderate/vigorous) for a duration of time by
specifying the baseline MVPA (default is 37%) and desired postintervention level (e.g., 50% as recommended20).
The resulting caloric effect is benchmarked against two energy
gap goals: to return the prevalence of obesity to (1) the early
1970s and/or (2) the Year-2000 levels. The former more
ambitious goal corresponds to the original goals set in Healthy
People 201021; the

latter provides a rough estimate of the current, more modest

22
Healthy People 2020 goals. The methodology underlying the
calculations of these targets for various population subgroups has
been described previously.5 All interventions listed assume that no
compensatory changes affecting daily energy balance occur,
beyond any effects observed in the empirical studies cited. For
example, the ACI of increasing MVPA from 37% to 50% during
PE classes assumes that students will not consume additional
calories to compensate for additional physical activity, or that
removing a food item from ones diet does not result in increased
consumption of other foods or beverages.

Results
The estimated caloric effect of the 15 interventions in
the tool, by gender and age group, are summarized
in Tables 2 and 3. For instance, for high school boys
and girls, adding 15 minutes of PE time per day for a
full school year was estimated to increase mean energy
expenditure by 25 kcal/day; replacing SSBs with water
in schools for the same group would reduce mean
energy intake by 15 kcal/day. For this group, however, an
average

August 2013

e6

Table 2. Caloric impact of physical activity interventions for average student, by age group
Inputs for caloric calculations
Avg.
caloric
impact
(kcal/day)

Population (age
group, years)

Target
a
METs

Avg.weight
b
(lbs)

Add walking at a
3-mph pace,
15 minutes/day

Both (25)
Both (611)
Both (1214)
Both (1518)

3.3
3.3
3.3
3.3

39
76
127
157

21
30
38
43

Same baseline (1.0, sitting quietly) and target METs for all
ages, based on Ainsworth et al.12 and Ridley et al.13

Add jogging at a
5-mph pace,
15 minutes/day

Both (25)
Both (611)
Both (1214)
Both (1518)

8.0
8.0
8.0
8.0

39
76
127
157

64
90
115
130

Same baseline (1.0, sitting quietly) and target METs for all
12
13
ages, based on Ainsworth et al. and Ridley et al.

Walking to and
from school
(roundtrip)

Both (25)
Both (611)
Both (1214)
Both (1518)

3.3
3.3
3.3
3.3

39
76
127
157

9
12
15
17

Intervention model estimates based on METs from Ainsworth

et al.12 and Ridley et al.,13 and published data on average
distances
from schools and students living within 1 mile of
23,24
school.
Caloric impact estimate uses METs of 1.0 as baseline (i.e.,
sitting in car).
b
Implemented for a full academic year.

Both (25)
Both (611)
Both (1214)
Both (1518)

3.4
3.4
3.6
3.7

39
76
127
157

11
15
21
25

McKenzie et al. estimate 3.4 METs for elementary school PE.

Same value used for preschool.
26
Nader et al. estimate 3.6 METs for middle school PE.
27
Smith et al. estimate 3.7 METs for high school PE.
Implemented for a full academic year.b

Intervention

Assumptions

Empirical estimates
Add school PE
time,
15 minutes/day

25

(continued on next page)

Wang et al / Am J Prev Med 2013;45(2):e3 e13

Modeled estimates

www.ajpmonlin
e.org

August 2013

Table 2. (continued)
Inputs for caloric calculations
Avg.
caloric
impact
(kcal/day)

Population (age
group, years)

Target
a
METs

Avg.weight
b
(lbs)

Make current PE
more active,
30 minutes/day

Both (25)
Both (611)
Both (1214)
Both (1518)

4.5
4.5
4.5
4.5

39
76
127
157

3
4
6
6

MET values used at baseline and target is a composite

of
7
estimated
MET values, based on Wu et al. and Ainsworth
12
et al. (4.5 METs for MVPA, 1.8 METs for non-MVPA).
Because of high variance in METs, baseline activity levels, and
20,26,28,29
population characteristics between CATCH,
25
3035
MSPAN, and TAAG
interventions, same averaged
MVPA% used for all age groups.
Changing the intensity of current PE time (not adding additional
PE time).
Base case increases MVPA from 37% to 50%, based on DHHS
20
national recommendation.
Implemented for a full academic year.b

Implement SPARK
using only PE
specialists to
teach PE, 30
minutes/day

Both (25)
Both (611)
Both (1214)
Both (1518)

7.2
7.2
7.2
7.2

39
76
127
157

34
48
58
64

7.2 METs for PE specialists for SPARK intervention from

36
37
McKenzie et al. and Sallis et al. used in calculation to
demonstrate maximum potential of intervention (compared
to 5.8 METs for trained classroom teachers).
Adding PE time to existing PE time.
Baseline METs assumed to be 3.4 for preschool and
elementary,25 3.6 for middle,26 and 3.7 for high school27.

Add afterschool
physical activity
program

Both (25)
Both (611)
Both (1214)
Both (1518)

4.5
4.5
4.5
4.5

39
76
127
157

11
16
20
22

Gortmaker et al. estimate 4.0 METs in intervention. 4.5

METs is used here as a conservative composite target based
on Wu et al.7
Same baseline (1.0, sitting quietly) and target METs for all
12
13
ages, based on Ainsworth et al. and Ridley et al.
b
Implemented for a full academic year.

Intervention

Assumptions

Wang et al / Am J Prev Med 2013;45(2):e3 e13

38

METs expresses how much energy is needed for physical activities. Caloric impacts expressed in this table are calculated assuming the physical activity is above a baseline of 1.0 METs (except where
noted, as with implementing SPARK), which is the baseline resting metabolic rate when sitting quietly.
Intervention is applied over a full school year (on average, 180 days). The total caloric impact is averaged over 365 days to account for no change in activity on holidays, weekends, and summer
vacation. CATCH, The Child and Adolescent Trial for Cardiovascular Health; MSPAN, The Middle-School Physical Activity and Nutrition intervention; MVPA, moderate-to-vigorous physical activity; PE,
physical
education; SPARK, Sports, Play, and Active Recreation for Kids; TAAG, The Trial of Activity for Adolescent Girls
b

e7

e8

Table 3. Caloric impact of dietary and other interventions for average student by group
Inputs for caloric calculations

Intervention

Population (age group,

years)

Amount

Affected
population,
b
%

Avg. caloric
impact
(kcal/day)

Assumptions

Modeled estimates
All
All

1-oz bag of chips

per day
1 cookie per day

100
100

154
55

Intervention models estimates based on

published caloric values of average bag of
regular potato chips and single Oreo cookie.

Reduce SSB intake

All
All

12-oz can per day

20-oz bottle per day

100
100

136
240

Intervention models estimates based on

published caloric values of average can or bottle
of regular caffeinated cola.

Replace SSBs with

water in schools

Both (25)
Both (611)
Both (1214)
Both (1518)

124
184
301
301

5.5
6.5
10.3
10.3

3
6
15
15

Switch from 1 cup of

sugary cereals to
cereals scored
highest in
nutritional quality

Both (25)
Boys (25)
Girls (25)
Both (611)
Boys (611)
Girls (611)
Both (1214)
Boys (1214)
Girls (1214)
Both (1518)
Boys (1518)
Girls (1518)

0.64 cups
0.64 cups
0.64 cups
0.93 cups
0.93 cups
0.94 cups
1.16 cups
1.32 cups
1.0 cups
1.15 cups
1.25 cups
1.06 cups

48.4
47.3
49.6
39.5
40.2
38.8
34.5
35.5
33.5
26.6
26.1
27.0

7
7
7
6
6
6
5
5
5
4
4
4

Averaged grams/cup
and standardized
serving
39
40
sizes of top ten and bottom ten cereals by
nutrition
score,
as
determined
by
CerealFACTS.
org.41
Affected population and average grams/serving
consumed based on analysis of NHANES 2007
2010 data on 24-hour dietary recall.
Proportion of cups consumed in Amount column
based on standardized 39.2 grams/cup (as
described above), and grams/serving from
NHANES.

Pass NYC's proposed

sugary drink size
limit

Both (25)
Boys (25)
Girls (25)
Both (611)
Boys (611)
Girls (611)
Both (1214)
Boys (1214)
Girls (1214)
Both (1518)
Boys (1518)

24.2
21.1
32.3
67.9
70.0
64.9
93.6
109.3
77.7
111.8
120.3

0.6
0.9
0.4
5.1
6.1
4.2
9.4
10.1
8.7
13.3
15.3

0
0
0
3
4
3
9
11
7
15
18

Amount is average kilocalorie reduction per day if

limited
consumption to 14
16 oz/day as in Elbel
42
et al. and Wang et al.
Affected population and average kilocalorie
reduction based on analysis of NHANES 2007
2010 data on 24-hour dietary recall.
Those consuming 416 oz limit consumption to
maximum of 16 oz/day
No "upsizing" occurs (i.e., individuals purchase
more than one 16-oz beverage to compensate
for size limit).

Affected population and amounts based on

10
published analysis from Wang et al.
c
Implemented for a full academic year.

(continued on next page)

Wang et al / Am J Prev Med 2013;45(2):e3 e13

www.ajpmonlin
e.org

Reduce unhealthy
food intake

August 2013

Table 3. (continued)
Inputs for caloric calculations

Intervention

Population (age group,

years)

Amounta

Affected
population,
%b

Avg. caloric
impact
(kcal/day)

Assumptions

100.1

11.2

11

SSB denition includes sodas, sports drinks, fruit

drinks and punches, low-calorie drinks,
sweetened tea, and other sweetened
beverages consumed in food service
establishments.
Implemented nationally.

Pass California's
competitive food
nutrition standards
in high schools
nationally

Both (1518)

157.8

100

78

Taber et al.9 estimate 157.9 kcal per weekday

fewer calories consumed in California high
schools, compared to 14 other states with
weaker competitive food laws states.
The intervention only applies to high school
students.
Implemented for a full academic year.c

Reduce TV viewing,
60 minutes/day

All

106

100

106

Reduce video- or
computer-game
playing time, 60
minutes/day

All

92

100

92

Empirical estimates

Sonneville and Gortmaker38 estimate TV watching

and video/computer game playing associated
with 105.5-kcal/hour and 91.8- kcal/hour
increase in total energy intake in boys aged
1315 years and girls aged 1214 years.
16
18
Epstein et al. and Miller et al. report
similar changes in energy intake.

Wang et al / Am J Prev Med 2013;45(2):e3 e13

Girls (1518)

Same calorie change for other age groups

The amount designates the current pre-intervention consumption level of the item by the selected population; amounts are kilocalories unless otherwise specied.
The impact designates the percentage of the selected eligible population that is affected by the intervention.
Intervention is applied over a full school year (on average, 180 days). The total caloric impact is averaged over 365 days to account for no change in activity on holidays, weekends, and summer
vacation.
NHANES, National Health and Nutrition Examination Survey; NYC, New York City; SSB, sugar-sweetened beverage
b
c

e9

e10

Wang et al / Am J Prev Med 2013;45(2):e3e13

per capita reduction of 82 kcal/day in energy surplus

would be needed to meet the Healthy People 2020
obesity prevalence goal of reducing obesity rates from
20.8% to
14.8%. Returning to the early 1970s level of obesity
prevalencethe target set by the more ambitious Healthy
People 2010 goalwould require an average per
capita reduction in energy gap of 217 kcal/day. These
estimates suggest that although any single intervention
may not be sufcient to achieve the Healthy People
goals, substantial progress could be made through a
combination of feasible, sustained policy and
environmental interventions.
Many of the ACI estimates built into the Caloric
Calculator require stipulated assumptions, which are
shown in detail in Tables 2 and 3, as well as within the
web tool. For example, the calculations of energy
expended through increased MVPA during PE involved
the following assumptions: a national baseline of 37%
MVPA during PE time,28 a target level of 50% recommended by the CDC,20 and 180 school days a year for
school-based interventions. The assumed MET level for
non-MVPA PE time was estimated as 1.8 METs, using
an average of lying down, sitting, and standing.12
The time spent on MVPA was estimated to be 4.5
METs based on the average of moderate physical activity
(3 METs) and vigorous physical activity (6 METs).7 For
example, for a typical high school adolescent (average
weight: 157 lbs), increasing MVPA from 37% to 50%
during a daily 30-minute PE class for a school year was
estimated to produce an average increase in energy
expenditure of 6 kcal/dayclearly insufcient on its
own to reverse the childhood obesity epidemic. Further,
even this small effect could potentially be diminished if
compensation occurred for this additional caloric expenditure with increased food or beverage consumption.
It is important to note that all estimates used in
creating the Caloric Calculator were population-based.
In addition, for interventions designed to remove a
particular food or beverage from the diet, caloric benets
were accrued only from the population affected (e.g., the
population affected by the NYC sugary drink portionsize cap was presumed to include those consuming
sugary beverages of 416 ounces per serving, estimated
to include only 12% of adolescents aged 1219 years).14

this point. New evidence from New York City,43 Philadelphia,44 California,9,45 and Mississippi46 demonstrates

Discussio
n
Reversing the nations current childhood obesity epidemic will require multiple individual, behavioral, policy,
environmental, and normative changesthrough public
health and clinical strategiesto reverse the daily accumulation of a positive energy gap that brought us to
www.ajpmonline.org

Wang et al / Am J Prev Med 2013;45(2):e3e13

that broad approaches involving multifaceted policies

and environmental strategies have the power to halt and
reverse the trend.47 However, what has been missing is a
metric for estimating the individual and combined
effects of specic interventions to increase childrens
activity levels and reduce their intake of energydense, low- nutrient foods and beverages.
This paper expands on the previously published
energy gap frameworkwhich estimated the magnitude of energy surplus underlying the obesity epidemic
among U.S. youth4,5to examine the effects of various
interventions, alone or in combination, to favorably tip
the energy balance. The lack of a common metric for
comparing the effectiveness of strategies with differing
behavioral targets (i.e., reducing excess caloric intake
and/or increasing physical activity) has stymied past
efforts to apply analytic tools to rank existing strategies
on their contribution to reversing the childhood obesity
trend. The development and application of the Average
Caloric Impact (ACI) metric and the Caloric Calculator
tool offer an opportunity to ll this gap.
Although the Caloric Calculator begins to address
these issues, there are nuances in the obesity reduction
equation that will require further research and discussion. The evidence used to estimate ACIs is still in
many ways limited and dependent on the rigor of
existing intervention studies and on the availability and
reliability of intervention outcome measures (e.g., the
use of objectively measured, versus self-reported,
outcomes or ecologic associations that can be
examined across stud- ies). In addition, many studies
focus narrowly on specic populations, such as middle
school girls30 or a specic age range.29,36,37
Most challenging at this stage in childhood obesity
prevention research is the lack of high-quality studies
with a sufciently long follow-up. A 2011 Cochrane
review of obesity prevention efforts found that only 14
of the 55 included studies had interventions lasting more
than 12 months, most of which focused only on children
aged 612 years. There is virtually no evidence from
studies aimed at younger children to determine whether
intervention bene- ts can be sustained into later
adolescence or adulthood.6
Therefore, it would be inaccurate to make predictions of
weight change from xed caloric changes using
these estimates, particularly given the multitude of
factors that drive weight change over time48 and the large
changes seen from childhood to adolescence.49
Study populations also have varied widely with
respect to racial/ethnic composition, SES, and
prevalence of obesity at baseline, limiting the
generalizability and comparability of intervention
effects. Thus, the tool represents the authors best
effort to assess the average impact if these programs
August 2013

were broadly implemented.

e11

e12

Wang et al / Am J Prev Med 2013;45(2):e3e13

Local contexts and subpopulation characteristics are

likely to modify the actual outcomes. The estimates will
continue to be rened and updated as new data emerge
from periodic scans of newly published data and feedback from collaborators in the eld of childhood obesity
prevention. Going forward, the Calculator will be further
developed to address specic subsets of the population or
allow more user inputs to facilitate broader dissemination and policy discussions. For example, a principal of a
disproportionately low-income school could use the tool
based on the schools demographics, or parents could
use the tool by entering their childs age, gender, and
body weight.
Despite these limitations, there is value in the Caloric
Calculators ability to translate evidence into practice by
generating caloric impact estimates and projecting the
potential cumulative effects of multicomponent interventions addressing one or both sides of the energy
balance equation. The ACI is a summary measure of
several dimensions of the program or policy evaluated:
reach, effectiveness/efcacy, adoption, implementation,
and maintenance.50 These dimensions also convey why
the net caloric impact of the same program will vary
from population to population when implemented in the
real world. As such, the tool is expected to offer a
starting point to support policymakers and practitioners
in using existing evidence to drive decision making in
a more straightforward manner.
The development of a common metric can lay the
groundwork for more evidence-based resource allocation
decisions, both in program implementation and in
further evidence gathering. Future expansion of this
framework may include ner granularity in the population targeted, such as overweight status, race/ethnicity,
and urban/rural locations as well as concerns for equity,
cost effectiveness, and other long-term outcomes.47
Further, the current review underscored the need to
encourage the evaluations of programs and policies to
use and report objective and comparable outcome
measures, such as changes in activity levels (e.g., MET
values); duration (e.g., minutes of MVPA added); net
changes in calories consumed in addition to key nutrients
or diet quality; and measured BMI whenever possible.
Because the Caloric Calculator uses national data with
the aim of estimating mean population-level effect sizes,
the effect of an intervention is averaged across those who
received and beneted from the program and those who
did not. Therefore, an intervention that has a large
effect but reaches only a small number of children may
appear to have less of an impact at the population
level. For example, an active transport program may
target chil- dren who live within 1 mile of their
school, which will reach at most 31% of children in
Grades K8.23 The daily

caloric impact, when averaged across all children, is

therefore a fraction of the net caloric impact for those
who participate in walking to school. Although not
evaluated in the current study, the same consideration
applies to interventions specically targeted at overweight adolescents (who have an average energy gap of
7001000 kcal/day).4
It is important to note that although the analyses
presented in this paper focus on intervention effects
on daily energy gaps and obesity levels in youth, there
are important health and nonhealth benets gained
from improving physical activity and diet that are not
captured by the ACI measure. For instance, there is
growing evidence that physical activity has benecial
effects on mental health outcomes and academic
performance.51 Similarly, an intervention to improve
the nutritional quality of a la carte foods and
beverages improves the overall nutritional prole of
foods consumed at school despite having no
signicant effect on the total number of calories
sold.52,53
Some investments in childhood obesity prevention
have been projected to be cost effective.54 But without
knowing what types of interventions to invest in, efforts
may fail to produce the expected results. There have
been many controversial, yet noteworthy, recent policy
rec- ommendations that will be scaled up to the national
level (e.g., menu labeling). Without experimental
evidence, however, it can be difcult to convince the
public and policymakers of the implications and
demonstrate the possible impact of implementation.
The Caloric Calcu- lator provides a novel tool for
appraising these policies and interventions based on
their potential efcacy, alone or combined, providing an
evidence-based platform to inform practice and policy.
The authors acknowledge the contribution of Dr. Laura K.
Brennan, PhD, MPH, President and CEO of Transtria LLC
(St. Louis, MO), and her team in the evidence-review process.
The authors thank Shawn Nowicki, MPH, and Andrew Wang,
MPH, as graduate student assistants in literature review and
the early development of the tool. The authors also thank
Michael Slaven, MA, who designed and implemented the web
tool, www.caloriccalculator.org, as well as Kevin Hall, PhD,
and Carson Chow, PhD (NIH/National Institute of Diabetes
and Digestive and Kidney Diseases), for their methodologic
advice on the analysis.
This work was supported by the Robert Wood Johnson
Foundation (grant no. 68162). This work is solely the
responsibility of the authors and does not represent the
ofcial views of the Robert Wood Johnson Foundation.
No nancial disclosures were reported by the authors of
this paper.
www.ajpmonline.org

Wang et al / Am J Prev Med 2013;45(2):e3e13

References
1. Trust for America's Health and the Robert Wood Johnson Foundation.
F as in fat: how obesity threatens America's future 2012. healthyamericans.org/report/100/2012.
2. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of obesity and
trends in body mass index among US children and adolescents,
19992010. JAMA 2012;307(5):48390.
3. Narayan KM, Boyle JP, Thompson TJ, Sorensen SW, Williamson DF.
Lifetime risk for diabetes mellitus in the U.S. JAMA 2003;290(14):1884
90.
4. Wang YC, Gortmaker SL, Sobol AM, Kuntz KM. Estimating the
energy gap among U.S. children: a counterfactual approach.
Pediatrics
2006;118(6):e1721e1733.
5. Wang YC, Orleans CT, Gortma ker SL. Reaching the healthy people
goals for reducing childhood obesity: closing the energy gap. Am J
Prev Med 2012;42(5):43744.
6. Waters E, de Silva-Sanigorski A, Hall BJ, et al. Interventions for
preventing obesity in children. Cochrane Database Syst Rev (Online)
2011(12):CD001871.
7. Wu S, Cohen D, Shi Y, Pearson M, Sturm R. Economic analysis of
physical activity interventions. Am J Prev Med 2011;40(2):14958.
8. Brennan L, Castro S, Brownson RC, Claus J, Orleans CT. Accelerating
evidence reviews and broadening evidence standards to identify
effective, promising, and emerging policy and environmental strategies
for prevention of childhood obesity. Annu Rev Public Health 2011;32:
199223.
9. Taber DR, Chriqui JF, Chaloupka FJ. Differences in nutrient intake
associated with state laws regarding fat, sugar, and caloric content of
competitive foods. Arch Pediatr Adolesc Med 2012;166(5):4528.
10. Wang YC, Bleich SN, Gortmaker SL. Increasing caloric contribution from sugar-sweetened beverages and 100% fruit juices among
U.S. children and adolescents, 1988-2004. Pediatrics 2008;121(6):
e1604e1614.
11. Schoeld WN. Predicting basal metabolic rate, new standards and
review of previous work. Hum Nutr Clin Nutr 1985;39(S 1):5
41.
12. Ainsworth BE, Haskell WL, Whitt MC, et al. Compendium of
physical activities: an update of activity codes and MET intensities.
Med Sci Sports Exerc 2000;32(9S):S498S504.
13. Ridley K, Olds TS. Assigning energy costs to activities in children: a
review and synthesis. Med Sci Sports Exerc 2008;40(8):1439
46.
14. Wang Y, Vine S. Caloric impact of a 16-ounce portion size cap on
sugar-sweetened beverages served in restaurants. Am J Clin Nutr 2013
[In Press].
15. Sonneville KR, Gortmaker SL. Total energy intake, adolescent discretionary behaviors and the energy gap. Int J Obes 2008;32(S6):S19S27.
16. Epstein LH, Roemmich JN, Robinson JL, et al. A randomized trial of
the effects of reducing television viewing and computer use on body
mass index in young children. Arch Pediatr Adolesc Med 2008;162(3):
23945.
17. Guran T, Bereket A. International epidemic of childhood obesity and
television viewing. Minerva Pediatr 2011;63(6):48390.
18. Miller SA, Taveras EM, Rifas-Shiman SL, Gillman MW. Association
between television viewing and poor diet quality in young children.
Int
J Pediatr Obes 2008;3(3):16876.
19. Healthy Eating Research. Food and beverage marketing to children
and adolescents: an environment at odds with good health. Robert
Wood Johnson Foundation, 2011.
20. DHHS. Strategies to improve the quality of physical education.
Washington DC: CDC, National Center for Chronic Disease Prevention and Healthy Promotion, Division of Adolescent and School

August 2013

e13

Health, 2010.
21. DHHS, Ofce of Disease Prevention and Health Promotion. Healthy
People 2010: nutrition and overweight. hp2010.nhlbihin.net/2010Objs/
19Nutrition.html.

e14

Wang et al / Am J Prev Med 2013;45(2):e3e13

22. DHHS, Ofce of Disease Prevention and Health Promotion. Healthy

People
2020: nutrition
and
weight
status.
2012.
www.healthypeople.
gov/2020/topicsobjectives2020/objectiveslist.aspx?topicId=29.
23. McDonald NC, Brown AL, Marchetti LM, Pedroso MS. U.S. school
travel, 2009 an assessment of trends. Am J Prev Med
2011;41(2):
146
51.
24. National Center for Safe Routes to School. How children get to school:
school travel patterns from 1969 to 2009. November
2011.
25. McKenzie TL, Sallis JF, Prochaska JJ, Conway TL, Marshall SJ,
Rosengard P. Evaluation of a two-year middle-school physical education intervention: M-SPAN. Med Sci Sports Exerc 2004;36(8):1382
8.
26. Nader PR. Frequency and intensity of activity of third-grade children
in physical education. Arch Pediatr Adolesc Med 2003;157(2):18590.
27. Smith NJ, Lounsbery MA, McKenzie TL. Physical activity in high
school physical education: impact of lesson context and class gender
composition. J Phys Act Health 2013: In Press.
28. McKenzie TL, Nader PR, Strikmiller PK, et al. School physical
education: effect of the Child and Adolescent Trial for
Cardiovascular
Health. Prev Med 1996;25(4):42331.
29. Kelder S, Hoelscher DM, Barroso CS, Walker JL, Cribb P, Hu S. The
CATCH Kids Club: a pilot after-school study for improving elementary students' nutrition and physical activity. Public Health Nutr
2005;8(2):13340.
30. Stevens J, Murray DM, Catellier DJ, et al. Design of the Trial of
Activity in Adolescent Girls (TAAG). Contemp Clin Trials
2005;26(2):22333.
31. Baggett CD, Stevens J, Catellier DJ, et al. Compensation or displacement of physical activity in middle-school girls: the Trial of Activity
for
Adolescent Girls. Int J Obes 2010;34(7):11939.
32. Elder JP, Shuler L, Moe SG, et al. Recruiting a diverse group of
middle school girls into the trial of activity for adolescent girls. J
School Health
2008;78(10):52331.
33. Gittelsohn J, Steckler A, Johnson CC, et al. Formative research in
school and community- based health programs and studies: "state of
the art" and the TAAG approach. Health Educ Behav
2006;33(1):2539.
34. Webber LS, Catellier DJ, Lytle LA, et al. Promoting physical activity in
middle school girls: Trial of Activity for Adolescent Girls. Am J
Prev
Med 2008;34(3):17384.
35. Young DR, Johnson CC, Steckler A, et al. Data to action: using
formative research to develop intervention programs to increase
physical activity in adolescent girls. Health Educ Behav 2006;33(1):
97
111.
36. McKenzie TL, Sallis JF, Kolody B, Faucette FN. Long-term effects of
a physical education curriculum and staff development program:
SPARK. Res Q Exerc Sport 1997;68(4):28091.
37. Sallis JF, McKenzie TL, Alcaraz JE, Kolody B, Faucette N, Hovell MF.
The effects of a 2-year physical education program (SPARK) on
physical activity and tness in elementary school students. Sports,
Play and Active Recreation for Kids. Am J Public Health
1997;87(8):
1328
34.
38. Gortma ker SL, Lee RM, Mozaffarian RS, et al. Effect of an afterschool intervention on increases in children's physical activity. Med
Sci Sports Exerc 2012;44(3):4507.
39. Cereal FACTS. Top 10 cereals by nutrition score. n.d.; cerealfacts.org/
cereal_nutrition_advanced_search.aspx?

lt.
40. Cereal FACTS. Bottom 10 cereals by nutrition score. n.d.;
cerealfacts. org/cereal_nutrition_advanced_search.aspx?lb.
41. Yale Rudd Center for Food Policy & Obesity. Limited progress in the
nutrition quality and marketing of children's cereals, 2012.
42. Elbel B, Cantor J, Mijanovich T. Potential effect of the New York
City policy regarding sugared beverages. N Engl J Med 2012;367(7):
680
1.
43. CDC. Obesity in K-8 studentsNew York City, 2006-07 to
2010-11 school years. MMWR Morbid Mortal Wkly Rep 2011;60
(49):16738.

www.ajpmonline.org

44. Robbins JM, Mallya G, Polansky M, Schwarz DF. Prevalence,

disparities, and trends in obesity and severe obesity among students
in the Philadelphia, Pennsylvania, school district, 2006-2010. Prev
Chronic Dis 2012;9:E145.
45. Babey SH, Wolstein J, Diamant AL, Bloom A, Goldstein HA. Patchwork of progress: changes in overweight and obesity among California's 5th, 7th, and 9th graders, 2005-2010. 2011.
46. Molaison EF, Kolbo JR, Zhang L, et al. Prevalence and trends in
obesity among Mississippi public school students, 2005-2009. J Miss
State Med Assoc 2010;51(3):6772.
47. Gortma ker SL, Swinburn BA, Levy D, et al. Changing the future of
obesity: science, policy, and action. Lancet 2011;378(9793):83847.
48. Hall KD, Sacks G, Chandramohan D, et al. Quantication of the
effect
of
energy
imbalance
on
bodyweight.
Lancet
2011;378(9793):82637.
49. Van Cleave J, Gortmaker SL, Perrin JM. Dynamics of obesity and
chronic
health conditions among children and youth. JAMA 2010;303(7):623
30.

50. Glasgow RE, Vogt TM, Boles SM. Evaluating the public health impact
of health promotion interventions: the RE-AIM framework. Am J
Public Health 1999;89(9):13227.
51. Singh A, Uijtdewilligen L, Twisk JW, van Mechelen W, Chinapaw MJ.
Physical activity and performance at school: a systematic review of the
literature including a methodological quality assessment. Arch
Pediatr Adolesc Med 2012;166(1):4955.
52. Cullen KW, Hartstein J, Reynolds KD, et al. Improving the school
food environment: results from a pilot study in middle schools. J Am
Diet Assoc 2007;107(3):4849.
53. Hartstein J, Cullen KW, Reynolds KD, Harrell J, Resnicow K, Kennel P.
Impact of portion-size control for school a la carte items: changes in
kilocalories and macronutrients purchased by middle school students.
J Am Diet Assoc 2008;108(1):1404.
54. Trasande L. How much should we invest in preventing childhood
obesity? Health Aff 2010;29(3):3728.