SPECIAL COMMUNICATIONS

Rapid Communications

2011 Compendium of Physical Activities: A

Second Update of Codes and MET Values
BARBARA E. AINSWORTH1,2, WILLIAM L. HASKELL3, STEPHEN D. HERRMANN1,2, NATHANAEL MECKES1,2,
DAVID R. BASSETT JR.4, CATRINE TUDOR-LOCKE5, JENNIFER L. GREER1,2, JESSE VEZINA1,2,
MELICIA C. WHITT-GLOVER6, and ARTHUR S. LEON7
1
Exercise and Wellness Program, School of Nutrition and Health Promotion, Arizona State University, Phoenix, AZ;
2
Healthy Lifestyles Research Center, School of Nutrition and Health Promotion, Arizona State University, Phoenix, AZ;
3
Stanford Prevention Research Center, School of Medicine, Stanford University, Palo Alto, CA; 4Department of Kinesiology,
Recreation, and Sports Studies, University of Tennessee, Knoxville, TN; 5Walking Behavior Laboratory, Pennington
Biomedical Research Center, Baton Rouge, LA; 6Gramercy Research Group, Winston-Salem, NC; and 7School of
Kinesiology, University of Minnesota, Minneapolis, MN

ABSTRACT
AINSWORTH, B. E., W. L. HASKELL, S. D. HERRMANN, N. MECKES, D. R. BASSETT JR., C. TUDOR-LOCKE, J. L. GREER,
J. VEZINA, M. C. WHITT-GLOVER, and A. S. LEON. 2011 Compendium of Physical Activities: A Second Update of Codes
and MET Values. Med. Sci. Sports Exerc., Vol. 43, No. 8, pp. 1575–1581, 2011. Purpose: The Compendium of Physical Activities
was developed to enhance the comparability of results across studies using self-report physical activity (PA) and is used to quantify the
energy cost of a wide variety of PA. We provide the second update of the Compendium, called the 2011 Compendium. Methods: The
2011 Compendium retains the previous coding scheme to identify the major category headings and specific PA by their rate of energy
expenditure in MET. Modifications in the 2011 Compendium include cataloging measured MET values and their source references,
when available; addition of new codes and specific activities; an update of the Compendium tracking guide that links information in the
1993, 2000, and 2011 compendia versions; and the creation of a Web site to facilitate easy access and downloading of Compendium
documents. Measured MET values were obtained from a systematic search of databases using defined key words. Results: The 2011
Compendium contains 821 codes for specific activities. Two hundred seventeen new codes were added, 68% (561/821) of which have
measured MET values. Approximately half (317/604) of the codes from the 2000 Compendium were modified to improve the definitions
and/or to consolidate specific activities and to update estimated MET values where measured values did not exist. Updated MET values
accounted for 73% of all code changes. Conclusions: The Compendium is used globally to quantify the energy cost of PA in adults for
surveillance activities, research studies, and, in clinical settings, to write PA recommendations and to assess energy expenditure in
individuals. The 2011 Compendium is an update of a system for quantifying the energy cost of adult human PA and is a living document
that is moving in the direction of being 100% evidence based. Key Words: EXERCISE, EXERTION, ENERGY EXPENDITURE,
METABOLIC, OXYGEN UPTAKE, KILOCALORIE

T
he Compendium of Physical Activities (Compen-
dium) has received widespread acceptance as a re-
Address for correspondence: Barbara E. Ainsworth, Ph.D., MPH, FACSM, source to estimate and classify the energy cost of
Exercise and Wellness Program, School of Nutrition and Health Promotion,
human physical activity (PA). The Compendium provides a
Arizona State University, 425 N. 5th Street, Mail Code 3020, Phoenix, AZ
85004; E-mail: [email protected]. five-digit coding scheme linking categories and types of
Submitted for publication March 2011. PA with their respective MET intensity values. The Com-
Accepted for publication April 2011. pendium was developed to provide a comprehensive list of
Supplemental digital content is available for this article. Direct URL cita- PA and their associated MET values to facilitate the coding
tions appear in the printed text and are provided in the HTML and PDF of self-reported behaviors obtained from PA questionnaires,
versions of this article on the journal’s Web site (www.acsm-msse.org). logs, and records. Its use has since been expanded to include
0195-9131/11/4308-1575/0 estimating the energy cost of individual PA for exercise
MEDICINE & SCIENCE IN SPORTS & EXERCISEÒ and weight management programs (39), a guide in exer-
Copyright Ó 2011 by the American College of Sports Medicine cise rehabilitation settings (5), and use as a reference for
DOI: 10.1249/MSS.0b013e31821ece12 quantifying the types of PA that characterize sedentary

1575

Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 behavior (i.e., 1.0–1.5 METs) (28,29) and light-intensity obtained from published sources and which MET values
SPECIAL COMMUNICATIONS

(1.6–2.9 METs) (28,38), moderate-intensity (3–5.9 METs), were estimated by expert opinion. In 2001, a tracking doc-
and vigorous-intensity (Q6 METs) activities (12). The ument was posted on a Web site to link the 1993 and the
Compendium has been translated into different languages 2000 codes, clearly identifying changes from one version to
(15) and is cited in the 2008 US physical activity guide- the next (http://prevention.sph.sc.edu/tools).
lines report as a reference for quantifying the energy cost of The evolution and widespread availability of portable
PA (31). It also is the focus of a Wikipedia link discuss- metabolic measurement systems in the past 10 yr have fa-
ing MET values (14). MET values from the 2000 Compen- cilitated direct measurement of the oxygen cost and, thus,
dium were used to assign intensities to PA reported in the energy expenditure or MET values of a growing array of PA
American Time Use Survey and thus form the basis for in field settings. At the same time, technology supporting
conclusions about how Americans spend their time (38). many inclusive computer-based literature searches has
The Compendium originated from a need to provide evolved, enabling an exhaustive search of published mea-
consistency in assigning intensity levels of PA captured sured MET values for inclusion in this second revision of the
from questionnaires used in epidemiological studies. The Compendium, called the 2011 Compendium. To emphasize,
idea for a Compendium came from Dr. William Haskell the 2011 Compendium reflects a major revision of past
and was developed during National Heart, Lung, and Blood versions by providing published references for measured
Institute–supported multisite studies awarded to Dr. Arthur MET values representing an increased number of cataloged
S. Leon, Dr. David R. Jacobs Jr., Dr. James Sallis, Dr. Henry PA. We have created the 2011 Compendium (see Supple-
Montoye, and Dr. William Haskell from 1987 to 1989. mental Digital Content, http://links.lww.com/MSS/A82)
Dr. Ainsworth was a postdoctoral associate during the grant which presents the 2011 updated codes and specific activi-
period and was the lead author of the original Compendium ties) that can be used to identify the MET values for asso-
and has maintained subsequent versions (collectively called ciated specific activities. New codes have been added that
the compendia) since 1989. reflect technology-based activities (e.g., computer and video
The first Compendium was published in 1993 (1) and exercise), emergent sports and conditioning activities (e.g.,
contained 476 five-digit codes and their associated MET X Games, Pilates, etc.), and specifically measured activities
values. The first two digits represent a major heading used to (e.g., household tasks such as child care) that have not been
classify PA by its primary purpose or activity domain (e.g., previously published. In addition, some existing codes have
bicycling, home repair, occupation, transportation, etc.). The been modified to reflect current evidence. Modifications to
subsequent three digits represent specific activities within some MET values were necessary to reflect the updated
each major heading. The complete code is then presented evidence by replacing estimated MET values with measured
with an associated MET value. The MET values are defined values. The 2011 Compendium also further clarifies the
as the ratio of the work metabolic rate to a standard resting description of the specific activities contained within each
metabolic rate (RMR) of 1.0 kcalIkgj1Ihj1. One MET is code. The dedicated Taylor codes used for decades to score
considered the RMR or the energy cost of a person at rest the Minnesota Leisure Time Physical Activity Question-
(20). The MET values in the Compendium range from 0.9 naire (37) are retained. Some codes for PA rarely performed
METs for sleeping to 23 METs for running at 14.0 mph. In (e.g., foundry and forestry occupations) have been com-
the 1993 publication, fewer than 15 citations were identified bined to reduce the length of the 2011 Compendium. A final
supporting assigned MET values. Most of the citations were change in 2011 is the development of a Web site to provide
textbooks that published compiled lists of a limited range of an easy way to access the Compendium documents.
PA. Other activities in the 1993 Compendium were identi-
fied from PA records kept by research study participants
CODING SCHEME AND 2011
(34). When published MET values could not be identified
COMPENDIUM CHANGES
for specific activities, experts in the assessment of PA esti-
mated MET values from similar types of activities already Organization. The 2011 Compendium uses the same
listed in the Compendium. organizational structure, described above, as past compen-
In 2000, the Compendium was updated to include two dia to provide consistency in locating specific activities
additional major headings (volunteer and religious activities) and their associated MET values. A brief example of the
and 129 additional codes for specific activities identified
from PA records kept by research study participants were TABLE 1. Example of the 2011 Compendium organizational scheme.a
added (2,3,21,40). In addition, selected codes were dropped Five-Digit MET (Major Heading Code) (Specific Activity Code)
to reduce redundancy of codes and/or specific activities, and Code Intensity Description Description
some MET values were modified to reflect updated infor- 01008 8.5 01 Bicycling (008) BMX
mation about the metabolic cost of specific activities. As in 02045 3.5 02 Conditioning (001) Curvesi exercise
routines in women
1993, efforts were made to identify a measured MET value 03010 5.0 03 Dancing (010) Ballet, modern, or jazz
for the new Compendium activities. However, the previous a
Italicized five-digit codes have estimated MET values; the remaining codes are based
compendia did not identify which MET values were on measured MET values.

1576 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
TABLE 2. Major types of physical activities in the 2011 Compendium. To streamline the occupation major heading, some of

SPECIAL COMMUNICATIONS
1—Bicycling 8—Lawn and garden 15—Sports the activities identified previously with separate five-digit
2—Conditioning exercises 9—Miscellaneous 16—Transportation
3—Dancing 10—Music playing 17—Walking codes were combined into a single code and specific activ-
4—Fishing and hunting 11—Occupation 18—Water activities ity and were provided with a longer description of the types
5—Home activity 12—Running 19—Winter activities
6—Home repair 13—Self-care 21—Religious activities
of activities performed. For example, code 11125, custodial
7—Inactivity 14—Sexual activity 21—Volunteer activities work, is described as light-effort activities such as cleaning
a sink and toilet, dusting, vacuuming, and light cleaning.
Likewise, code 11126, custodial work, is described as
coding scheme is presented in Table 1. References for mea- moderate-effort activities such as using an electric buffer,
sured MET values are available for viewing and down- feathering arena floors, mopping, taking out the trash, and
loading from the Compendium Web site described later in vacuuming. The activities listed in the descriptions for codes
this report. 11125 and 11126 had their own codes in the 2000 Com-
Major headings. The major headings are the same as pendium but are now combined in the 2011 Compendium.
those presented in the 2000 Compendium and are listed in Table 3 provides a brief example of the changes made to the
alphabetical order to facilitate location of specific activities. 2000 Compendium specific activities and their associated
The major headings provide an overall organized presenta- descriptions as presented in the 2011 Compendium. A de-
tion of PA by purpose and include the domains of leisure tailed list of the changes made to the codes is presented on
(including recreational activities), transportation, occupa- the Compendium Web site (https://sites.google.com/site/
tion, home activities (including personal care and other compendiumofphysicalactivities).
care), inactivity, and volunteer activities. A list of the major Intensity of activities. As noted previously, to pro-
headings is provided in Table 2. vide consistency with past compendia, all activities are
Specific activities. Descriptions of the specific activi- assigned an intensity unit on the basis of their rate of en-
ties are designed to provide qualitative information about the ergy expenditure expressed as multiples of 1 MET, de-
intensity (e.g., described as ‘‘vigorous’’), bicycling pedal scribed as the ratio of the metabolic rate for an activity
cadence, walking or running speed (mph), or context for the divided by a standardized expression of the RMR (20). We
specific activities. These detailed descriptions of specific use the definition of a MET as described by Balke (6) to be
activities help Compendium users to identify the most ap- 3.5 mLIkgj1Iminj1, often characterized as the metabolic
propriate code and/or the MET value. Some specific activ- cost of resting quietly. For a reference adult, 1 MET is ap-
ities need little description. For example, code 02101 proximately 3.5 mL oxygenIkgj1 body weightIminj1 or
(stretching), is described only as ‘‘mild.’’ Other specific ac- 1 kcalIkgIhj1 body weight.
tivities and their associated descriptions are very detailed. There has been a discussion about the use of
For example, code 11766 (truck driving) describes the var- 3.5 mLj1Ikgj1Iminj1 as the appropriate denominator to
ious activities performed by a truck driver to include loading
and unloading a truck, tying down a load, standing, walking,
TABLE 3. Example of documentation of changes made to codes in the 2011
and carrying heavy loads. Descriptions for other specific Compendium.a
activities may include types of activities that have the same Code Major Heading Description
MET value but differ in their form. For example, code 01009 Bicycling Changed the description of activities to
03040 (ballroom dancing) identifies slow types of ballroom ‘‘bicycling, mountain, general’’
01015 Bicycling Changed MET level from 8.0 to 7.5
dancing to include the waltz, foxtrot, tango, samba, etc. As 02011 Conditioning exercise Changed MET level from 3.0 to 3.5;
with past Compendia, the specific activities and their asso- changed the description of activities to
ciated descriptions may be very general to reflect the vague read ‘‘bicycling, stationary, 30–50 W,
very light to light effort’’
way that people report PA on questionnaires or records. 02012 Conditioning exercise Changed MET level from 5.5 to 6.8;
For example, if a person writes that they ‘‘played tennis,’’ changed the description of activities to
read ‘‘bicycling, stationary, 90–100 W,
the coder has no idea if the person engaged in singles or moderate to vigorous effort’’
doubles tennis, was playing for leisure or in a competition, 03010 Dancing Changed MET level from 4.8 to 5.0;
changed the description of activities to
or was playing in a game or nongame setting. In this case, read ‘‘ballet, modern, or jazz, general,
the specific activity is coded as 15675, tennis, general. rehearsal, or class’’
Although most specific activities appear under a single ma- 03015 Dancing Changed MET level from 6.5 to 7.3
04001 Fishing and hunting Changed MET level from 3.0 to 3.5
jor heading, depending on the context and purpose of an 04010 Fishing and hunting Changed MET level from 4.0 to 4.3;
activity, a specific activity may be duplicated in more than added ‘‘fishing related’’ to the
description of activities
one major heading. For example, code 17021, walking, 05100 Home activity Changed MET level from 2.0 to 3.3;
carrying a 10-lb child also is listed as 05182, home activi- added ‘‘changing linens’’ to the
ties, carrying a 10-lb child, slow walking. Both specific ac- description of activities
05120 Home activity Changed MET level from 6.0 to 5.8
tivities have the same MET value. This duplication allows a
Selected codes and major headings are displayed for example only. The full list of
the user to attribute the correct context for an activity that changes is available at the 2011 Compendium Web site (https://sites.google.com/site/
may have multiple purposes. compendiumofphysicalactivities).

2011 COMPENDIUM OF PHYSICAL ACTIVITIES Medicine & Science in Sports & Exercised 1577

Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 denote 1 MET for calculating the activity MET listed in the a very close approximation. Thus, there are only very
SPECIAL COMMUNICATIONS

Compendium, especially when assessing the energy cost of small errors involved in conversion of MET (defined as
PA in individuals (9,22,23). The concern is that RMR is 3.5 mLj1Ikgj1Iminj1) to kilocalories.
lower in overweight persons, declines with age, and is lower Calculating the kilocalories expended in PA using Com-
in females than in males (18). Thus, using a fixed RMR of pendium data may not be appropriate for all settings. In-
3.5 mLj1Ikgj1Iminj1 to denote 1 MET may underestimate clusion of the body weight in a PA score can bias the results
the true ‘‘ratio’’ (activity/rest) in certain populations. It has of studies designed to evaluate the effects of PA frequency,
been recommended that a correction factor be used to ad- duration, and intensity on specific health outcomes in clini-
just the MET level on the basis of an estimate of one’s RMR, cal studies or in comparing population PA levels in sur-
which accounts for age, height, weight, and sex. For exam- veillance systems. A person with a higher body weight will
ple, Byrne et al. (9) and Kozey et al. (22) recommend di- expend more kilocalories for a specific activity especially
viding the fixed RMR (3.5 mLj1Ikgj1Iminj1) by an RMR if it is a weight bearing activity (e.g., walking). It is pos-
(mLj1Ikgj1Iminj1) predicted from the Harris–Benedict sible to compute a PA score that is independent of body
(18) equation. This quotient is then multiplied by the Com- weight by dividing the kilocalories expended by body
pendium MET value (9,22). weight. Using the example for 30 min of sweeping with
Standard MET levels in the 2011 Compendium are a di- an energy cost of 133 kcal (3.8 METs  70 kg of body
rect translation of the weight-specific energy costs, com- weight  0.5 h), dividing 133 kcal by 70 kg creates a
puted by taking the energy costs (V̇O2, mLj1Ikgj1Iminj1) summary score of 1.9 kcalIkgj1Ihj1. It also is easy to
and dividing them by 3.5 mLj1Ikgj1Iminj1. As such, they compute MET-h or MET-min scores that are independ-
represent a shorthand way of expressing the weight-specific ent of body weight by multiplying the activity MET
energy costs of activities (19). It should be noted that value by the duration of the activity in hours or minutes,
expressing the values relative to body weight is desirable respectively. In addition, this approach can be used to
because this normalizes the data and minimizes the variation develop an overall activity score combining time spent at
between overweight/obese and normal-weight individuals. different MET levels. PA scores can be expressed as kilo-
To our knowledge, it has not been shown that expressing the calories per kilogram per hour, METs per hour, and METs
energy costs as ‘‘corrected MET’’ would be a superior ap- per minute, and all are used frequently in PA assessment
proach for normalizing the data. (7,30), surveillance studies (4,11), and showing inverse
Most of the source studies cataloged in the 2011 Com- associations between PA and mortality (13,29), body weight
pendium reported the energy costs of activities as either V̇O2 gain (25), and incidence of various chronic diseases (31).
(mLj1Ikgj1Iminj1) or MET using 3.5 as the denominator. Measured MET values. Sixty-eight percent (561 of
Thus, for consistency, we have continued to define 1 MET the 821) of codes included in the 2011 Compendium have
in the same manner. We recognize that some researchers measured MET values derived from published sources. The
may prefer to use corrected MET, and the 2011 Compen- MET values were obtained from laboratory or field
dium can be used to derive these values. To accomplish this, experiments that measured the oxygen cost of the specific
the Compendium MET level for an activity is multiplied by activities listed. To find articles showing the intensities of
the ratio of the standard MET divided by the individual’s the specific activities, we searched databases (PubMed,
predicted RMR (9,22). Thus, for an individual with a pre- Google Scholar, and EBSCOhost to access MEDLINE,
dicted RMR = 2.8 mLj1Ikgj1Iminj1, the corrected MET PsychINFO, SocINDEX, the Cumulative Index to Nursing
would be 3.5/2.8 = 125% of the Compendium values. and Allied Health Literature, and the Education Resources
Calculation of kilocalorie energy expenditure. Information Center), using key words of ‘‘metabolic cost,’’
The methods used to calculate the energy cost of the spe- ‘‘calorimetry,’’ ‘‘energy expenditure,’’ ‘‘oxygen cost,’’ and
cific activities in the 2011 Compendium have been de- ‘‘energy cost.’’ Titles and abstracts were scanned, and
scribed in detail in the 1993 (1) and the 2000 (2) compendia promising references were assembled for closer inspection.
articles by Ainsworth et al. As noted earlier, the energy Additional sources were obtained from reviewing reference
cost of specific PA may differ on the basis of a person’s lists of identified publications. In regard to existing specific
body weight and body fat percentage (8), age (33), physical activities in the Compendium for which we could not iden-
fitness (33), genetics (33), sex (8,18), mechanical effi- tify a measured MET value, estimated MET values were
ciency (27), and the environmental conditions (16) under carried forward from previous compendia. For new specific
which the activity is performed. In general, we recommend activities, if no published data were available, MET values
estimation of the caloric cost of PA with the equation, were estimated from activities that were deemed similar. The
kilocalories = MET  weight in kilograms  duration in 2011 Compendium Web site differentiates the measured and
hours. For example, to compute the kilocalorie energy estimated MET for the specific activities by using a color-
expenditure for a 70-kg person, sweeping at a moderate coded scheme with the five-digit code colored blue if the
effort (3.8 METs) for 30 min would result in 133 kcal MET value is based on measured MET and colored red if
(3.8  70 kg  0.5 h). This method reflects the fact that the MET value is estimated. The printed version of the
3.5 mLj1Ikgj1Iminj1 is equal to 1.0 kcalIkgj1Ihj1, within Compendium presents estimated values in italics. During the

1578 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
literature search, we often identified multiple references for the Web site provides a color coding scheme to iden-

SPECIAL COMMUNICATIONS
the same specific activities. For example, we located 11 tify specific activities with measured and estimated values
references relevant to code 08120, ‘‘mowing lawn, walking, (blue = measured, red = estimated). Users can access de-
power mower, moderate or vigorous effort,’’ with MET tailed reference lists for the specific activities with measured
values ranging from 3.90 to 5.75. In cases such as this, we MET values on the Web site. Users will note that for some
present the average of all the identified measured MET as activities, there are multiple references that present mea-
the representative value (i.e., 5.0 METs for code 08120). sured MET values for the same specific activities. It is then
New codes. As a result of the literature review and possible for users to review the context for the measured
changes in the types of PA identified since the 2000 revision activities (e.g., if activities were measured only in men or in
of the Compendium, several new five-digit codes were women, ages of participants studied, etc.) and evaluate the
added to the 2011 Compendium. Many of the new codes relevance of the MET values for their specific use. The Web
reflect a global approach to including culture-specific ac- site also presents a comprehensive list of the new activities
tivities (e.g., native New Zealand dances and sports activi- and those for which the Compendium MET values or ac-
ties). Other new codes include computerized games and tivity descriptions have changed. Consistent with the 2000
activity promoting videos and/or arcade games (e.g., Wii Fit Compendium, a tracking guide can be viewed to trace the
(Nintendo Co., LTD); Dance Dance Revolution (Konami addition of new codes and changes in existing codes from
Digital Entertainment, Inc.)). Other new activities identify the 1993, 2000, and 2011 compendia. Finally, Web site
the energy cost of approaches to reduce sedentary behaviors, users can download the 2011 Compendium as a PDF file.
such as performing work activities while standing, walking All Web site documents may be downloaded free of charge
for transportation, and fidgeting. The expansion of health by all users.
clubs since 2000 has introduced various types of condition- A primary benefit of using a Web site to present the 2011
ing exercises to include military boot camp–style workouts, Compendium is the ease of providing substantial informa-
energetic dance routines, and different types of yoga exer- tion related to the assessment of PA free to users globally.
cise. A brief example of the new codes and specific activi- The Web site provides resources to estimate the caloric en-
ties included in the 2011 Compendium is presented in ergy expenditure of specific PA and compute corrected
Table 4. A full list of the new codes is available on the 2011 MET values designed to tailor MET values for individual
Compendium Web site. differences in age, sex, body weight, and height and allows
Web site resources. Along with the 2011 Compen- easy search to look up activities and compare changes in the
dium update of the major codes and specific activities, MET values between the different compendia. The Web site
development of the Web site provides for a searchable in- also provides links to programs that use the Compendium to
terface and easy navigation to locate information from the assess the energy cost of PA (e.g., Exercise is Medicinei).
1993, 2000, and 2011 compendia. The Web site search A goal of future compendia is to have measured MET
features allow users to enter key words to locate specific values for all activities. To increase this level from 68% of
activities, their five-digit codes, and associated MET values. measured MET levels, the Web site provides a mechanism
One of the primary goals for the 2011 Compendium update for users to upload findings from published peer-reviewed
was to identify activities that have MET values supported research studies that directly measured the energy cost of
by published literature and to differentiate them from activ- PA. Submission documents are located on the Compendium
ities without measured MET values. As described earlier, Web site. This information will then be readily available to
provide future updates in a timely manner.
TABLE 4. Example of new codes added to the 2011 Compendium.a
Code Major Heading MET Description
DISCUSSION AND LIMITATIONS
01003 Bicycling 14.0 bicycling, mountain, uphill, vigorous
01004 Bicycling 16.0 bicycling, mountain, competitive, The Compendium is an enduring document that has
racing
02001 Conditioning exercise 2.3 activity-promoting video game shown relevance since its initial publication in 1993. The
(e.g., Wii Fit), light effort compendia grew out of a need to provide consistency in
(e.g., balance, yoga)
02003 Conditioning exercise 3.8 activity-promoting video game
assigning an intensity level to activities identified on PA
(e.g., Wii Fit), moderate effort questionnaires used in epidemiological studies. Since the
(e.g., aerobic, resistance routines) introduction of the 1993 Compendium, many studies have
03012 Dancing 6.8 ballet, modern or jazz, performance,
vigorous effort used the coding scheme and standard MET values to assign
03014 Dancing 4.8 tap intensity levels to PA questionnaires globally. The results
04005 Fishing and hunting 4.5 fishing, crab fishing
04007 Fishing and hunting 4.0 fishing, catching fish with hands
from these studies have supported conclusions that regular
05011 Home activity 2.3 cleaning, sweeping, slow, light effort PA is health enhancing and that physical inactivity is a
05012 Home activity 3.8 cleaning, sweeping, slow, major risk factor for chronic diseases and premature mor-
moderate effort
a
tality (31). As discussed in the 2008 PA guidelines (12),
Selected codes of some major headings are displayed for example only. The full list of
new codes is available at the Compendium Web site (https://sites.google.com/site/ regular moderate-intensity PA performed for 150 minIwkj1,
compendiumofphysicalactivities). vigorous-intensity PA performed for 75 minIwkj1, or a

2011 COMPENDIUM OF PHYSICAL ACTIVITIES Medicine & Science in Sports & Exercised 1579

Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 combination of moderate- and vigorous-intensity PA is As noted in the 1993 (1) and the 2000 (2) articles, the
SPECIAL COMMUNICATIONS

sufficient to reduce the risks for premature mortality related Compendium MET values are intended for use in able-
to chronic diseases and disabling conditions. bodied adults who are 18–65 yr old and do not reflect the
The 2011 Compendium has clinical applications for pri- energy cost of children and youth, older adults, and persons
mary and secondary prevention of adverse health condi- with disabilities. Readers interested in identifying the energy
tions. Clinicians, fitness trainers, and health promotion cost of PA in children and youth are referred to Harrell et al.
experts can use the 2011 Compendium to identify and/or (17), Ridley et al. (35), and Ridley and Olds (36) who
track suitable activities for patients and/or clients to perform present results of studies that identify the energy cost of PA
that have sufficient intensity to elicit desired results. For in children and youth. Readers interested in assessing the
example, if an adult wanted to follow the US PA guidelines energy cost of PA in persons with physical disabilities are
to accumulate 150 min of moderate-intensity PA and engage referred to Collins et al. (10) and Lee (26) who present the
in 2 d of strength training activities, an individual could energy cost of PA in persons with spinal cord injuries and
perform the following plan: Monday, walk at 3.0 mph for paraplegia, respectively. Price (32) also provides prediction
30 min (code 17190, 3.5 METs); Tuesday, lift weights for equations that can be used to estimate the energy cost of PA
20 min (code 02054, 3.5 METs); Wednesday, bicycle at in persons with spinal cord injuries. Lante et al. (24) also
9.4 mph for 40 min (code 01019, 5.8 METs); Friday, lift provide a list of the energy costs of activities of daily living
weights for 20 min (code 02054, 3.5 METs); Saturday, at- for adults with intellectual disabilities.
tend a low-impact aerobic dance class for 40 min (code
03020, 5.0 METs); and Sunday, rest. The Compendium is
SUMMARY
very useful to identify PA that may be too intense for clinical
patients’ health/functional status or sufficiently intense to The 2011 Compendium provides 821 codes that reflect 21
elicit the desired health-enhancing benefits. major headings, numerous specific activities and their de-
As a living document, versions of the compendia have tailed descriptions, and associated MET values that can be
been used for nearly 20 yr to provide intensity values to used to identify the energy cost of PA. Building on the 1993
classify PA and estimate their energy cost. Regarded glob- and the 2000 versions of the compendia, the 2011 Com-
ally as the primary source for obtaining MET values to score pendium includes 217 additional specific activities beyond
PA questionnaires, logs, and records and to estimate the the 2000 revision. In addition, the 2011 Compendium
caloric energy costs of PA, the 2011 Compendium provides identifies measured MET values obtained from the pub-
a timely update with the adoption of evidence-based MET lished literature for approximately two-thirds of the of spe-
values. Looking forward, the goal is to have all specific cific activities listed and is supported by a Web site that
activities in the Compendium derived from published direct provides extensive documentation of references for the MET
measurements. Researchers are encouraged to submit stud- values; a tracking guide to link codes, activities, and MET
ies to the Web site that describe measured MET values for values between the 1993, 2000, and 2011 versions; and
current activities with estimated values. Regular updates of equations to compute kilocalories and corrected MET val-
the Compendium will continue to incorporate this emerging ues. Despite its known limitations, the Compendium has
evidence. withstood the test of time to provide a valuable resource to
We have chosen to retain the presentation of the inten- code PA surveys or records and to provide examples of ac-
sity units as standard MET with the RMR represented tivities within a broad intensity range for use in PA
as 3.5 mLj1Ikgj1Iminj1 as in past versions of the com- counseling, research, and clinic settings. It remains a living
pendia. Although the Compendium was not developed to document, and continued improvements are anticipated and
determine the precise energy cost of activities within an in- will be updated at regular intervals.
dividual, we recognize that there are situations when one
may want to use the MET values to estimate an individual This project was supported by a contract from the National Insti-
tutes of Health, National Cancer Institute (HHSN261200900476P).
energy cost of PA as multiples of measured or predicted The authors have no conflicts of interests to report.
RMR. Thus, for someone who has an RMR less than The authors thank Brandon Sawyer, Siddhartha Angadi, Glenn
3.5 mLj1Ikgj1Iminj1, the standard MET overestimates Gaesser, Chelsi Mundy, Tracy Washington, Vanessa Ortiz, Lisa
Smith, and Sean Southland from Arizona State University’s Healthy
the individual’s RMR, resulting in an underestimate of the Lifestyles Research Center and Mary Rosenberg from Stanford
ratio (total EE/resting EE). Readers interested in learning University’s Prevention Research Center for their assistance in re-
more about use of the ‘‘corrected MET’’ to account for dif- vising the Compendium. The authors also thank Dr. Richard Troiano
for his support of this project.
ferences in individual RMR values are referred to Byrne The results of the present study do not constitute endorsement by
et al. (9) and Kozey et al. (22). the American College of Sports Medicine.

REFERENCES
1. Ainsworth BE, Haskell WL, Leon AS, et al. Compendium of 2. Ainsworth BE, Haskell WL, Whitt MC, et al. Compendium of
physical activities: classification of energy costs of human physical physical activities: an update of activity codes and MET intensi-
activities. Med Sci Sports Exerc. 1993;25(1):71–80. ties. Med Sci Sports Exerc. 2000;32(9 suppl):S498–516.

1580 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 3. Ainsworth BE, Irwin ML, Addy CL, Whitt MC, Stolarczyk LM. 22. Kozey S, Lyden K, Staudenmayer J, Freedson P. Errors in MET

SPECIAL COMMUNICATIONS
Moderate physical activity patterns of minority women: the Cross- estimates of physical activities using 3.5 mL  kg(j1)  min(j1)
Cultural Activity Participation Study. J Womens Health Gend as the baseline oxygen consumption. J Phys Act Health. 2010;7:
Based Med. 1999;8:805–13. 508–16.
4. Armstrong T, Bull F, Guthold R. Global Physical Activity Ques- 23. Kozey SL, Lyden K, Howe CA, Staudenmayer JW, Freedson PS.
tionnaire (GPAQ) Analysis Guide [Internet]. [cited Feb 16]. World Accelerometer output and MET values of common physical ac-
Health Organization, Geneva, Switzerland, 2011. Available from: tivities. Med Sci Sports Exerc. 2010;42(9):1776–84.
www.who.int/chp/steps/resources/GPAQ_Analysis_Guide.pdf. 24. Lante K, Reece J, Walkley J. Energy expended by adults with and
5. Arrigo M, Lüscher T. Physical activity in patients with coro- without intellectual disabilities during activities of daily living. Res
nary artery disease before and after cardiac catheterisation: lack of Dev Disabil. 2010;31:1380–9.
improvement at 3 months follow-up without cardiac rehabilitation. 25. Lee IM, Djoussé L, Sesso HD, Wang L, Burning JE. Physical
Kardiov Med. 2007;10:317–25. activity and weight gain prevention. JAMA. 2010;303:1173–9.
6. Balke B. The effect of physical exercise on the metabolic poten- 26. Lee M, Zhu W, Hedrick B, Fernhall B. Determining the metabolic
tial, a crucial measure of physical fitness. In: Staley S, Cureton T, equivalent values of physical activities for persons with paraplegia.
Huelster L, Barry AJ, editors. Exercise and Fitness. Chicago: The Disabil Rehabil. 2010;32:336–43.
Athletic Institute; 1960. pp. 73–81. 27. Mahaudens P, Detrembleur C, Mousny M, Banse X. Gait in ado-
7. Blair SN, Haskell WL, Ho P, et al. Assessment of habitual physical lescent idiopathic scoliosis: energy cost analysis. Eur Spine J.
activity by a seven-day recall in a community survey and con- 2009;18:1160–8.
trolled experiments. Am J Epidemiol. 1985;122:794–804. 28. Pate RR, O’Neill JR, Lobelo F. The evolving definition of ‘‘sed-
8. Browning RC, Baker EA, Herron JA, Kram R. Effects of obesity entary.’’ Exerc Sport Sci Rev. 2008;36(4):173–8.
and sex on the energetic cost and preferred speed of walking. J 29. Patel AV, Bernstein L, Deka A, et al. Leisure time spent sitting in
Appl Physiol. 2006;100:390–8. relation to total mortality in a prospective cohort of US adults. Am
9. Byrne NM, Hills AP, Hunter GR, Weinsier RL, Schutz Y. Meta- J Epidemiol. 2010;172:419–29.
bolic equivalent: one size does not fit all. J Appl Physiol. 2005;99: 30. Pettee Gabriel K, McClain JJ, Lee CD, et al. Evaluation of physical
1112–9. activity measures used in middle-aged women. Med Sci Sports
10. Collins EG, Gater D, Kirati J, Butler J, Hanson K, Langbein WE. Exerc. 2009;41(7):1403–12.
Energy cost of physical activities in persons with spinal cord in- 31. Physical Activity Guideline Advisory Committee. Physical Activ-
jury. Med Sci Sport Exerc. 2010;42(4):691–700. ity Guidelines Advisory Committee Report. Washington (DC): US
11. Craig CL, Marshal AL, Sjöström M, et al. International Physical Department of Health and Human Services; 2008. p. 23.
Activity Questionnaire: 12-country reliability and validity. Med Sci 32. Price M. Energy expenditure and metabolism during exercise in
Sports Exerc. 2003;35(8):1381–95. persons with a spinal cord injury. Sports Med. 2010;40:681–96.
12. Department of Health and Human Services. 2008 Physical Activity 33. Ravussin E, Bogardus C. Relationship of genetics, age, and
Guidelines for Americans [Internet]. [cited 2008 Mar 4]. Available physical fitness to energy expenditure and fuel utilization. Am J
from: http://www.health.gov/paguidelines/pdf/paguide.pdf. Clin Nutr. 1989;49:968–75.
13. Dorn JP, Cerny FJ, Epstein LH, et al. Work and leisure time 34. Richardson MT, Leon AS, Jacobs DJ Jr, Ainsworth BE, Serfass R.
physical activity and mortality in men and women from a general Comprehensive evaluation of the Minnesota Leisure Time Physi-
population sample. Ann Epidemiol. 1999;9:366–73. cal Activity Questionnaire. J Clin Epidemiol. 1994;47:271–81.
14. Dunstan DW, Salmon J, Owen N, et al. Associations of TV 35. Ridley K, Ainsworth BE, Olds TS. Development of a compendium
viewing and physical activity with the metabolic syndrome in of energy expenditures for youth. Int J Behav Nutr Phys Act.
Australian adults. Diabetologia. 2005;48:2254–61. 2008;5:45.
15. Farinatti P. Apresenta0ão de uma versão em Português do Com- 36. Ridley K, Olds TS. Assigning energy costs to activities in
pêndio de Ativadades Fı́sicas: uma contribui0ão aos pesquisadores children: a review and synthesis. Med Sci Sports Exerc. 2008;40:
e profissionais em fisiologia do exercı́cio [in Portuguese]. Rev Br 1439–46.
Fisiol Exer. 2003;2:177–208. 37. Taylor HL, Jacobs DJ Jr, Shucker B, Knudsen J, Leon AS,
16. Gupta JS, Swamy YV, Dimri GP, Pichan G. Physiological re- Debacker G. A questionnaire for the assessment of leisure time
sponses during work in hot humid environments. Indian J Physiol physical activities. J Chronic Dis. 1978;31:741–55.
Pharmacol. 1981;25:339–47. 38. Tudor-Locke C, Washington T, Ainsworth B, Troiano R. Linking
17. Harrell JS, McMurray RG, Baggett CD, Pennell ML, Pearce PF, the American Time Use Survey (ATUS) and the Compendium of
Bangdiwala SI. Energy costs of physical activities in children and Physical Activities: methods and rationale. J Phys Act Health.
adolescents. Med Sci Sports Exerc. 2005;37(2):329–36. 2009;6:347–53.
18. Harris JA, Benedict FG. A biometric study of human basal me- 39. van Mechelen W, Twisk JW, Post GB, Snel J, Kemper HC.
tabolism. Proc Natl Acad Sci U S A. 1918;4:370–3. Physical activity of young people: the Amsterdam Longitudinal
19. Howley E. Letter to the editor. J Phys Act Health. 2011;8:141–2. Growth and Health Study. Med Sci Sports Exerc. 2000;32(9):
20. Howley E. You asked for it: question authority. ACSM’s Health 1610–6.
Fitness J. 2000;4:6. 40. Whitt MC, DuBose KD, Ainsworth BE, Tudor-Locke C. Walking
21. Irwin ML, Ainsworth BE, Conway JM. Estimation of energy ex- patterns in a sample of African American, Native American and
penditure from physical activity measures: determinants of accu- Caucasian women: the Cross-Cultural Activity Participation Study.
racy. Obes Res. 2001;9:517–25. Health Educ Behav. 2004;31:45S–56S.

2011 COMPENDIUM OF PHYSICAL ACTIVITIES Medicine & Science in Sports & Exercised 1581

Copyright © 2011 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.