Assessing and promoting physical activity among youth have gained ample attention in recent years. This is mainly due to the high prevalence of pediatric overweight and obesity all over the world. Trends suggest declines in children's physical activity level, in particular in active commuting, organized sports, and physical education (10,13). However, there is conflicting information on the proportion of children meeting the current health-related physical activity guideline to accumulate a minimum of 60 min of moderate-to-vigorous physical activity (MVPA) per day (5,15,21). Self-reported data of 115,981 Europeans aged 11, 13, or 15 yr suggest that about one third of young people meet the 60-min guideline (3), whereas in another European study using accelerometers, the proportion of children meeting the guideline was much higher and ranged from 62% in 15-yr-old girls to over 97% in 9-yr-olds (18). In the Netherlands, this proportion was found to range from 20% to 90% in different studies using self-reports among school-aged children (9,27). This conflicting information hampers national policy making.
The proportion of children meeting the 60-min physical activity guideline is likely to vary with the assessment method. Numerous methods are available to assess physical activity. In most epidemiologic studies, self-reports are used. Self-reports are easily administered, low-cost measurements but tend to overestimate the time spent in vigorous physical activities and to underestimate the time spent in unstructured daily physical activities such as walking and playing outdoors (3,14). At present, accelerometers are being used with increasing regularity. Accelerometers are lightweight, unobtrusive, and relatively inexpensive compared with other objective methods, such as direct observation or doubly labeled water. However, they cannot always be worn and underestimate certain activities (e.g., stair climbing, weight lifting, cycling, and rowing). Furthermore, regarding measuring intensity, there is a ceiling effect with running at high speeds (6).
Data processing decisions, such as the definition of moderate-intensity activity, are also likely to influence adherence to the 60-min guideline. Moderate-intensity activity is described by the American College of Sports Medicine as "generally equivalent to a brisk walk, or activity that noticeably accelerates the heart rate." This relatively loose definition leaves room for different interpretations. For children in most studies, moderate intensity is defined as "at least 3 METs," whereas in the Netherlands, a threshold of 5 METs is used (12). In a study of Pate et al. (17), the proportion of 12-yr-old girls meeting the 60-min guideline ranged from 1% to 88% using three different intensity thresholds (i.e., 3.0, 3.8, and 4.6 METs). Another aspect that might influence adherence to the guideline is whether intermittent, cumulative physical activity or sustained periods of physical activity are included in the analysis. In adults, accumulation of physical activity in intermittent bouts of at least 10 min is stated to be as effective in effecting chronic disease risk factors as longer bouts. For children, the guideline is less specific and bout durations between 1 and 30 min are used in the literature. Olds et al. (16) showed that the definition of days on which a guideline must be met also affects adherence to the guideline. In their study among 13- to 19-yr-olds, the proportion of children meeting the guideline differed from 20% for "all days" to 68% for "on average across all days."
The purpose of the present study was to investigate the effect of data processing decisions in terms of intensity threshold, bout duration, and days on the proportion of 6- to 11-yr-old children meeting the 60-min physical activity guideline using a subjective and an objective assessment method.
Spatial Planning and Children's Exercise Study Description and Subjects
This study was part of the Spatial Planning and Children's Exercise study (26). The study involved a convenience sample of 6- to 11-yr-old children recruited from 20 elementary schools in 10 disadvantaged neighborhoods of six Dutch cities (>70,000 inhabitants). The study consisted of measurements of physical activity, energy consumption, and anthropometry at the individual level and an observation checklist for the built environment at the neighborhood level. All measurements were conducted between October 2004 and January 2005. The study was approved by the ethics committee of the Leiden University Medical Center. Informed consent was obtained from the parents of 1228 children after they were given written information about the purpose and nature of the study.
Physical activity diary
Physical activity was assessed subjectively by a 7-d physical activity diary that was completed by one of the parents together with their child. During seven consecutive days for all waking hours, all physical activities were noted at the end of each day, including the duration (in minutes) and the corresponding physical activity category (i.e., active commuting, activities during school time, organized sports, playing outdoors, and activities at home). Because recall of physical activity is a complex cognitive task, the information processing guidelines of Baranowski (4) were followed. Every day was segmented into the following: morning, 6:00 a.m. to 12 noon; afternoon, 12 noon to 6:00 p.m.; and evening, 6:00 p.m. to 11:00 p.m. Furthermore, memory cues were provided, that is, a bookmarker with examples of commonly performed activities for each category and guiding questions (e.g., "At what time did your child get up this morning/go to bed this evening?," "At what time does school begin/end?," and "How long does recess/lunch break last?"). In addition, an example of a filled-out day was provided. All instructions and guiding questions were addressed to the parents. They were not instructed about reporting a minimum duration of the activities. The diary was pilot tested before the study.
Simultaneously with the diary, physical activity was assessed objectively in a subsample using the ActiGraph AM-7164 accelerometer (ActiGraph, Pensacola, FL). The ActiGraph has proved to be valid and reliable for use in children (8). It is a lightweight uniaxial accelerometer, which is designed to measure normal human movement by using an internal piezoelectric cantilever beam that creates a charge proportional to the magnitude of the movement. Movement values (counts·min−1) are accumulated and stored over a user-specified period. In this study, the time sampling interval was set at 1 min. The device was attached securely to the child's right hip by an elastic waist belt. Children were asked to wear the accelerometer during waking hours for eight consecutive days. They were instructed to remove the device during swimming and bathing.
Other variables that were collected included age, sex, body height and body weight, parental education level, and country of origin of the child and both parents. Body height and body weight (while wearing indoor clothes without shoes) were measured by two trained research assistants with a portable stadiometer (Stanley 04-116, Stanley-Mabo Ltd., Poissy, France) and a digital scale (Seca 812, Vogel & Halke Gmbtt & Co., Germany). Body mass index was calculated (kg·m−2) and categorized into normal weight, overweight, and obesity according to age- and sex-specific cutoffs for children (7).
Fifty-one percent (n = 625) of the initial sample returned the physical activity diary. Children failing to complete the diary for at least 4 d (including at least one weekend day; n = 104) were excluded from analyses (23). Each activity was assigned a MET value using the compendium of physical activities (1).
Ten percent (n = 62) of the diary sample simultaneously wore an ActiGraph accelerometer. Children were included in the analyses if the accelerometer was worn for at least 4 d during at least 500 min·d−1 (23). The first day of monitoring was excluded as there was a significant difference between the first and the following monitoring day (day 1 = 766 ± 338 counts·min−1; day 2 = 641 ± 183 counts·min−1; t = 3.448; P = 0.001; both weekdays). Substantial periods of zero activity counts (≥10 min) were excluded under the assumption that the accelerometer had been removed. Accelerometer data were processed using Actisoft 3.2 (MTI Health Services, Fort Walton Beach, FL) and MAHUffe 184.108.40.206 (Institute of Metabolic Science, Medical Research Council Epidemiology Unit, Cambridge, UK) software programs. Age-specific count cutoffs were used to calculate the number of minutes spent in moderate-to-vigorous physical activity (MVPA) of at least 3 and 5 METs (11,24).
All statistical analyses were performed using SPSS 14.0 (SPSS Inc., Chicago, IL). Descriptive statistics were used to characterize the samples. For both assessment methods, the amount of time spent above moderate-intensity thresholds of 3 and 5 METs, respectively, was calculated. Time was calculated in three ways: 1) by summing every minute per day that the specific intensity threshold was met (called 1-min bouts); 2) by only summing five or more consecutive minutes above the specific threshold (5-min bouts); and 3) by only summing 10 or more consecutive minutes above the specific threshold (10-min bouts). No interruptions below the specific threshold were allowed in identifying bouts. For the diary, it was assumed that the complete duration of the activity was spent above the assigned MET value. Mean differences between intensity thresholds and bout durations were tested with paired-samples t-tests. Between-sex comparisons were made using independent-samples t-tests. For each intensity threshold and bout duration, the proportion of children meeting the 60-min guideline was calculated. A distinction was made between meeting the 60-min threshold on each assessment day and meeting this threshold on average across all assessment days. Differences between proportions of children meeting the guideline were analyzed using chi-square analysis. Values were considered statistically significant at P < 0.05.
From the 625 children who returned the physical activity diary, 83% completed the diary for at least 4 d (521 children; 254 boys and 267 girls). On average, these children completed the diary for 7 ± 1 d during 537 ± 197 min·d−1 (i.e., 70% of waking time). From the 62 children who simultaneously wore the ActiGraph, 7 were excluded because they did not complete the diary for at least 4 d, 3 were excluded because they failed to wear the accelerometer for at least 4 d, and 1 was excluded due to monitor failure. These children (n = 51; 16 boys, 35 girls) completed the diary for an average of 7 ± 1 d during 508 ± 224 min·d−1 and wore the accelerometer for an average of 6 ± 1 d during 730 ± 55 min·d−1. The characteristics of the final samples are shown in Table 1. A considerable number of children were overweight or obese, that is, 27% of the diary sample and 22% of the ActiGraph sample.
Table 2 shows children's physical activity level. On average, the diary sample accumulated 1314 ± 489 MET·min·d−1 (boys = 1347 ± 483 MET·min·d−1; girls = 1280 ± 493 MET·min·d−1; NS) (weekdays = 1310 ± 482 MET·min·d−1; weekend days = 1396 ± 646 MET·min·d−1; t = 3.528; P < 0.001). For the ActiGraph sample, mean MET-minutes per day was not significantly different (1205 ± 510 MET·min·d−1). Mean counts per minute of the ActiGraph sample was 584 ± 173 (boys = 689 ± 211 counts·min−1; girls = 536 ± 131 counts·min−1; t = 2.666; P < 0.05) (weekdays = 580 ± 169 counts·min−1; weekend days = 574 ± 207 counts·min−1; NS). Time spent in MVPA differed considerably by guideline operationalization and assessment method (Table 2). It ranged from 53 to 111 min·d−1 using the diary and from 2 to 173 min·d−1 using the ActiGraph. On average, children spent significantly more time in activities of at least 3 METs compared with 5 METs (P < 0.001), independent of bout duration and assessment method. Furthermore, they spent significantly more time in MVPA when shorter bouts were considered compared with longer bouts (diary and ActiGraph; P < 0.001). According to both assessment methods, boys spent significantly more time in MVPA than girls (P < 0.001), with the exception of mean time spent in MVPA of at least 3 METs in bouts of at least 1 min as measured by the ActiGraph (NS).
Effects of intensity threshold and bout duration on MVPA were reflected in the proportion of children meeting the 60-min guideline. This proportion ranged from 3% to 86% using the diary and from 0% to 100% using the ActiGraph (Fig. 1A-D). In the diary sample, the proportion was significantly lower when the 5-MET threshold was used compared with the 3-MET threshold (P < 0.001), independent of bout duration and operationalization of days. A significantly higher proportion of children met the guideline if shorter bouts were considered compared with longer bouts, independent of intensity threshold, operationalization of days, and assessment method. Significantly more children met the guideline on average across all days compared with the guideline on each day (P < 0.001), independent of intensity threshold, bout duration, and assessment method. Between-sex differences in the proportion of children meeting the 60-min guideline were found for some guideline operationalizations. When the mean number of days that the guideline was met was analyzed, boys met the guideline on significantly more days per week than girls (P < 0.05), independent of intensity threshold, bout duration, and assessment method (Table 3), with the exception of the mean number of days the guideline was met using the 3-MET threshold in 1- and 5-min bouts and the 5-MET threshold in 10-min bouts as measured by the ActiGraph.
In this study, the effect of physical activity guideline operationalization in terms of intensity threshold, bout duration, and days was investigated on adherence to the 60-min guideline. The proportion of children meeting the guideline differed considerably by guideline operationalization. It ranged from 3% to 86% using the diary. Similar results were found in a subsample of children who simultaneously wore an accelerometer. Using the ActiGraph, the proportion of children meeting the guideline ranged from 0% to 100%. Overall, a higher proportion of children met the guideline when the 3-MET intensity threshold was used compared with the 5-MET threshold and when a shorter bout duration was used compared with a longer bout duration. More children met the guideline on average across all assessment days compared with the guideline on each assessment day. Furthermore, boys were found to be more active than girls, independent of guideline operationalization and assessment method. This finding is in accordance with other studies in which boys appear to participate in more physical activity, especially more vigorous physical activity, than girls (3). The results of our study are also in line with a study of Sleap and Tolfrey (20) in which 79 children (9-12 yr old) wore a heart rate monitor during 4 d. The interpretation of the children's physical activity level depended on the intensity threshold and whether cumulative or continuous bouts of physical activity were included in the analyses. In a study of Pate et al. (17) with accelerometers, it was also shown that adherence to physical activity guidelines varied by intensity threshold. The proportion of girls that met the 60-min guideline was 1%, 12%, and 88%, respectively, using intensity thresholds of 4.6, 3.8, and 3.0 METs, respectively. Our results are also in line with the study of Olds et al. (16) in which a considerable lower number of children met the guideline "on all 4 d" (20%) compared with "on average across 4 d" (68%).
There are several other factors that may affect adherence to the 60-min physical activity guideline that have not been investigated in this study, such as the MET values that were assigned to each activity (1) and the accelerometer count cutoffs that were used (11,24). The compendium of physical activities is based on adults (1). According to Armstrong and Welsman (3) and Torun (22), the use of adult values may underestimate energy cost by 20% in 10-yr-old children. To our knowledge, there is no comprehensive list of energy costs of children's free-living physical activities. Furthermore, the accelerometer count cutoffs that were used in this study to define moderate intensity are debatable. These count cutoffs were established during running on a treadmill (11,24). Whether this represents children's free-living activities remains uncertain. Anderson et al. (2) showed that accelerometer count cutoffs not only affected time spent in MVPA but also affected agreement between diary and accelerometer estimates of physical activity.
The present study was aimed at investigating the effect of guideline operationalization on the proportion of children meeting the 60-min physical activity guideline. It was not aimed at presenting the absolute proportion of Dutch children meeting the guideline. There are some limitations in this study that warrant careful interpretation of the results. The study was performed in a convenience sample of children living in disadvantaged neighborhoods. Half of the children did not return the diary. However, compliance among the children who did return the diary seems reasonable (i.e., 83% completed the diary for at least 4 d). Accelerometer data were only collected in a subsample of the diary sample. Nowadays, accelerometry is more frequently the method of choice to assess physical activity, but at the time the study was performed this was not commonly used. Financial and time constraints limited our choice of assessment method. Furthermore, agreement between the diary and the ActiGraph was poor. Pearson's correlations between time spent in MVPA according to both assessment methods were low and ranged from r = 0.006 to r = 0.197 (NS) depending on intensity threshold and bout duration. Analogously, the percentage of agreement in meeting the guideline ranged from 31% to 98% with Cohen's κ of 0.013 to 0.179 (NS). Results in Table 2 suggest that in the diary, the duration of short bouts of activities of at least 3 METs was underestimated and the duration of more vigorous activities (≥5 METs) was overestimated compared with ActiGraph recordings. Parents were not instructed about reporting a minimum duration of the activities, but not many activities were reported to last less than 10 min. This might explain the small difference from going from 1- to 5-min bouts to 10-min bouts in the diary sample compared with the ActiGraph sample. Poor agreement between the two methods might also have been caused by removing the ActiGraph during certain activities like swimming and vigorous contact sports (e.g., martial arts and soccer) because some children were afraid of damaging the accelerometer. Furthermore, cycling, an activity that was performed by 61% of the diary sample and 41% of the ActiGraph sample, is underestimated by the ActiGraph. This might have resulted in an underestimation of time spent in activities of at least 5 METs. Besides, a 1-min sampling interval was used in stead of 15 s to be able to monitor during eight consecutive days with the limited accelerometer's memory capacity. With this sampling interval, short bursts of vigorous activity might have been averaged out.
Although there are some methodological limitations, it is concluded that the proportion of children meeting the 60-min physical activity guideline is significantly affected by guideline operationalization and assessment method. The effect size of the intensity threshold, bout duration, and days on the proportion of children meeting the guideline remains to be determined. It is important to reach consensus about how the 60-min physical activity guideline should be operationalized to monitor the extent to which populations of children meet the guideline and to simplify comparison between studies. This will not be easy as there is continuing uncertainty about the optimal intensity, frequency, and duration of physical activity for children (19,25).
This study was supported by the Dutch Ministry of Health, Welfare and Sport and the Dutch Ministry of Housing, Spatial Planning and the Environment. We are grateful to the teachers, principals, parents, children, and research assistants who were involved in this study. The results of the study do not constitute endorsement by ACSM.
1. 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(9 suppl):S498-516.
2. Anderson CB, Hagstromer M, Yngve A. Validation of the PDPAR as an adolescent diary: effect of accelerometer
cut points. Med Sci Sports Exerc
3. Armstrong N, Welsman JR. The physical activity patterns of European youth with reference to methods of assessment. Sports Med
4. Baranowski T. Validity and reliability of self report measures of physical activity: an information-processing perspective. Res Q Exerc Sport
5. Biddle SJ, Sallis JF, Cavil N. Young and active? Policy framework for young people and health-enhancing physical activity. In: Biddle SJ, Sallis JF, Cavill N, editors. Young and Active: Young People and Physical Activity
. London: Health Education Authority; 1998. p. 3-16.
6. Brage S, Wedderkopp N, Franks PW, Bo AL, Froberg K. Reexamination of validity and reliability of the CSA monitor in walking and running. Med Sci Sports Exerc
7. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. Br Med J
8. de Vries SI, Bakker I, Hopman-Rock M, Hirasing RA, Mechelen W van. Clinimetric review of motion sensors in children
and adolescents. J Clin Epidemiol
9. de Vries SI, Van Overbeek K, Jongert MWA, Simons M, Chorus AMJ, Bakker I. Evaluation of Scoring for Health (in Dutch)
. Leiden: TNO Kwaliteit van Leven; 2007. p. 33-40.
10. Dollman J, Norton K, Norton L. Evidence for secular trends in children
's physical activity behaviour. Br J Sports Med
11. Freedson PS, Melanson E, Sirard J. Calibration of the Computer Science and Applications, Inc. accelerometer
. Med Sci Sports Exerc
12. Kemper HGC, Ooijendijk WTM, Stiggelbout M. Consensus about the Dutch health-related physical activity guideline (in Dutch). Tijdschr Soc Gezondheidsz
13. McDonald NC. Active transportation to school: trends among U.S. schoolchildren, 1969-2001. Am J Prev Med
14. McMurray RG, Ring KB, Treuth MS, et al. Comparison of two approaches to structured physical activity surveys for adolescents. Med Sci Sports Exerc
15. National Association for Sport and Physical Education. Physical Activity for Children: A Statement of Physical Activity Guidelines for Children Ages 5-12
. 2nd ed. Reston: NASPE Publications; 2004. p. 1-28.
16. Olds T, Ridley K, Wake M, Hesketh K, Waters E, Patton G, Williams J. How should activity guidelines for young people be operationalised? Int J Behav Nutr Phys Act
17. Pate RR, Stevens J, Pratt C, et al. Objectively measured physical activity in sixth-grade girls. Arch Pediatr Adolesc Med
18. Riddoch CJ, Bo AL, Wedderkopp N, et al. Physical activity levels and patterns of 9- and 15-yr-old European children
. Med Sci Sports Exerc
19. Rowland TW. Promoting physical activity for children
's health: rationale and strategies. Sports Med
20. Sleap M, Tolfrey K. Do 9- to 12 yr-old children
meet existing physical activity recommendations for health? Med Sci Sports Exerc
21. Strong WB, Malina RM, Blimkie CJ, et al. Evidence based physical activity for school-age youth. J Pediatr
22. Torun B. Inaccuracy of applying energy expenditure rates of adults to children
. Am J Clin Nutr
23. Trost SG, Pate RR, Freedson PS, Sallis JF, Taylor WC. Using objective physical activity measures with youth: how many days of monitoring are needed? Med Sci Sports Exerc
24. Trost SG, Pate RR, Sallis JF, Freedson PS, Taylor WC, Dowda M, Sirard J. Age and gender differences in objectively measured physical activity in youth. Med Sci Sports Exerc
25. Twisk JW. Physical activity guidelines for children
and adolescents: a critical review. Sports Med
26. Vries de SI, Bakker I, Mechelen van W, Hopman-Rock M. Determinants of activity-friendly neighborhoods of children
: results from the SPACE study. Am J Health Promot
. 2007;21(suppl 4):312-6.
27. Zeijl E, Crone M, Wiefferink K, Reijneveld M. Children in the Netherlands (in Dutch)
. Den Haag/Leiden: SCP/ TNO Kwaliteit van Leven; 2005. p. 52-4.
Keywords:© 2009 American College of Sports Medicine
ACCELEROMETER; CHILDREN; MEASUREMENT; EPIDEMIOLOGY