Background and History

Standardizing Energy Expenditure in Adults

One of the most important resources for physical activity researchers and practitioners is knowledge about the amount of energy that is required to carry out various types of physical activities, from watching TV to running a race. The values used to express energy expenditure are a vital tool for connecting physical activity behavior and health.

In 1993, Ainsworth et al. (2011) published a compendium of energy expenditure (EE) values for adults, and it has been updated twice since then. This Compendium of Physical Activities presents EE in terms of METs, or metabolic equivalents of task. One MET is the rate of energy expenditure while lying supine at rest and is based on an oxygen uptake of 3.5 milliliters per kilogram of body weight per minute (3.5 ml^.kg^-1.min^-1, or 1 kcal/kg.hr^-1). Physical activities that require more intense physical effort have higher METs. For example, leisure biking at a speed of 5.5 mph has an EE of 3.5 METs and competitive mountain bike racing has an EE of 14.0 METs. The Adult Compendium has been widely accepted as a tool to estimate and classify EE for myriad activities.

Other adult EE compendia also have been developed. McCurdy et al. (2000) and Klepeis et al. (2001) have published compendia of EE for diverse activities. They focused on exposure assessment by assessing environmental factors, such as location and presence of smokers, in conjunction with EE estimates for occupational as well as domestic and leisure activities.

A First Attempt to Express Energy Expenditure in Children and Youth

Recognizing that it would be valuable to have similar EE values for children and youth, Ridley et al. developed the first compendium of EE values for youth in 2008 (Ridley et al. 2008). Because of the limited amount of data available on children at the time, many of this compendium’s EE values were based on adult values. This limited the usefulness of the compendium because resting adult METs are known to be less than resting EE values for youth. On average, youth values do not reach adult levels until late in adolescence (McMurray et al. 2015). Nevertheless, this first youth compendium proved to be invaluable in efforts to translate observational and self-reported data concerning physical activity in children into EE units.

Forming NCCOR’s Youth Energy Expenditure Workgroup

The National Collaborative on Childhood Obesity Research (NCCOR) determined that it could potentially make an important contribution to the field of childhood obesity research by supporting an effort to fill the gap between what was known about EE in youth physical activity and the existing Ridley et al. values. In 2012, NCCOR established a Youth Energy Expenditure (YEE) workgroup to develop methods and measures that could be used to create standardized values for the EE of common physical activities in which children and adolescents engage. The workgroup had several main goals, which were designed to improve the utility of youth EE values in research, evaluation, and practice:

• Update the Ridley et al. (2008) literature review to include new estimates of youth EE,
• Update and expand the Ridley et al. values in a new youth compendium, and
• Develop a metric for youth that incorporates the age dependence of resting metabolic rate.

Early Work of NCCOR’s YEE Workgroup

The YEE workgroup’s first task was to convene an Expert Panel that met in April 2012 in Atlanta, GA. The meeting, convened by NCCOR, the Centers for Disease Control and Prevention, and the National Cancer Institute, identified methodological challenges associated with obtaining EE values derived from children and youth. Participants agreed on the need for more EE data on all ages and both sexes of children and youth and began the planning to revise and update the original Ridley compendium.

Following the 2012 meeting, the YEE workgroup carried out two major activities that laid the foundation for work on a revised and updated compendium:

• The workgroup examined the best metric, or unit, to use to describe EE values. It conducted several analyses on a pooled dataset of youth physical activity data from five research groups. These analyses aimed to identify the best EE metric and determine ways to minimize potential confounders, such as age, sex, and physical characteristics. The complete details of this study can be found in a McMurrray et al. paper published in June 2015 in PLoS One. The results indicated that no one metric for EE was superior and fully able to eliminate the relationships of EE with age or sex and physical characteristics, but two metrics—a MET value based on age- and sex-related basal EE (a “youth MET,” or MET_y) and an allometric scaled EE method—were somewhat better in eliminating the influence of sex. The MET_y appeared to be the best for sedentary and light intensity activities, and allometric scaling was the best for moderate-to-vigorous activities. From a research perspective, allometric scaling may be preferred, but for other applications the MET_y may be more intuitive and straightforward. Furthermore, from a population perspective, the MET_y might be acceptable when comparing EE of large groups of children of either sex, but the MET_y must be age- and sex-specific. Thus, the MET_y was chosen as the suitable metric. A paper reporting on additional analyses was published in August 2017 in Pediatric Exercise Science.
• The second activity was a systematic literature review to obtain measured EE values on a range of physical activities in which children and youth participate (Ridley 2013, 2016). Studies were located primarily through electronic databases: CINAHL; Cochrane library; EMBASE; Medline; Proquest; PsychINFO; SCOPUS; SportDiscus; and Web of Science. Numerous key terms were used to describe the youth and the metabolic measurement methodologies used. No language limits were imposed. Studies had to meet the following criteria to be included: 1) full-text article, thesis/dissertation, or data provided from authors from a published conference abstract; 2) data from human children or youth (<18 y) provided; 3) numerical measured EE for individual activities (not total daily or segmented day EE) using indirect calorimetry, direct calorimetry, or CO₂ breathing tests provided; and 4) healthy youth included who were not selected based on specific pathology (excluding weight status). The search resulted in 11,606 citations. Of those citations, a total of 90 eligible studies (excluding walk/run) were included for data extraction in the systematic review. These studies had a total of 487 unique mean energy costs values for 218 different activities. In addition, 75 studies provided 347 mean energy cost values focused on walking and running from speeds of 0.6 to 8 mph (1 to 13 kph).

Having established a metric to use to express EE values and identified available data from the literature, the YEE workgroup decided to advertise for studies on youth activities for which no, or very limited, EE data had been published. This effort culminated in the publication of a special issue of the Journal of Physical Activity and Health (Herrmann and Pfeiffer, 2016). The special issue includes 17 peer-reviewed manuscripts describing nearly 100 separate physical activities. These manuscripts substantially enrich the knowledge about EE associated with child and youth physical activities and their results are included in the compendium.

Developing the Youth Compendium of Activities

The workgroup used these three datasets—the pooled dataset from the five research groups, data from the literature review, and the data from the special supplement—and an imputation process to develop the new Youth Compendium of Physical Activities (Youth Compendium) (Butte 2017). The Youth Compendium includes EE values presented in two ways (smoothed and observed/imputed) for 196 activities, including walking, running, and cycling at various speeds. The values are provided for four age groups—6 to 9, 10 to 12, 13 to 15, and 16 to 18. The smoothed data were estimates calculated from mixed model regression equations. The observed/imputed values were estimates calculated from the mean of the observed and imputed data. The default look-up values are based on the smoothed estimates. The methods section of this website provides additional information on the methods used to develop the Youth Compendium.

The Youth Compendium is a work in progress, and NCCOR plans to update it over time, as additional data become available. The majority of MET_y values were measured in mid-childhood (8 to 12 years); more data are needed in very young children and older adolescents. Sufficient data are available for sedentary and light activities, but additional data on moderate and vigorous activities are needed. In addition, the Youth Compendium is not applicable to children younger than age 6 years or who have illnesses or disabilities, though they may be appropriate for differing weight status. To fill the gaps in the Youth Compendium, future research efforts could solicit data on certain types of activities in specific age groups, similar to the Journal of Physical Activity and Health special issue (Herrmann and Pfeiffer, 2016).

References