Daily energy expenditure, activity patterns, and energy costs of the

European Journal of Clinical Nutrition (2002) 56, 819–829 ß 2002 Nature Publishing Group All rights reserved 0954–3007/02 $25.00 www.nature.com/ejcn

ORIGINAL COMMUNICATION Daily energy expenditure, activity patterns, and energy costs of the various activities in French 12 – 16-y-old adolescents in free living conditions M Vermorel1*, J Vernet1, A Bitar2, N Fellmann3 and J Coudert3 1

Energy and Lipid Metabolism Research Unit, INRA, St Gene`s Champanelle, France; 2Department of Biology and Animal Physiology, Faculty of Sciences, El Jadida, Morocco; and 3Physiology and Sports Biology Laboratory, Medical Faculty, ClermontFerrand, France Background: Changes in lifestyle and increases in sedentary activities during recent decades have been shown to contribute to the prevalence of overweight in adolescents. Objectives: To determine the inter-individual variability and the day-to-day variations in daily energy expenditure (DEE) and activity pattern, and the energy costs and EE of the various activities of adolescents in free-living conditions. Design: Sixty adolescents (four groups of 14 – 16 boys or girls aged 12 – 16 y) participated in this cross-sectional study during spring or autumn. Activity patterns and EE were determined during five consecutive days, using both a diary and the heart rate recording method validated by whole-body calorimetry and laboratory tests. Results: Mean DEE increased significantly with age in boys, but not in girls. However, the physical activity level did not vary significantly with sex and age. Mean DEE was significantly higher in spring than in autumn in the 12.6-y-old subjects. It was also 21% higher during the free days than during the schooldays in the active subjects, but 7% lower in the sedentary subjects. The energy cost of 22 activities was determined. Time and energy devoted to moderate and sport activities exhibited great interindividual variability. They were lower in girls than in boys and decreased with age. The increase in EE resulting from moderate and sport activities instead of sedentary activities ranged from 0.2 to 2.7 MJ=day over the week. Conclusion: The great variability in DEE of adolescents resulted mainly from differences in the nature, duration and intensity of physical activities during the free days. European Journal of Clinical Nutrition (2002) 56, 819 – 829. doi:10.1038=sj.ejcn.1601395 Keywords: adolescents; body composition; energy expenditure; free-living conditions; physical activity; season

Introduction Epidemiological studies have shown alterations of the lifestyle and food habits of adolescents in industrialized

*Correspondence: M Vermorel, Energy and Lipid Metabolism Research Unit, INRA Theix, 63122 Saint-Gene`s Champanelle, France. E-mail: [email protected] Guarantors: M Vermorel and J Coudert. Contributors: MV was responsible for all stages of the study, including the study design, measurements, interpretation of results and writing the paper. JV analysed the data and was responsible for statistical analysis. AB and NF were responsible for the measurements of physical capacities and energy expenditure during the laboratory tests, and participated in the discussion. JC, medical doctor, supervised the study, participated in the discussion of the study design, recruited the volunteers and participated in the discussion of the results. Received 5 July 2001; revised 23 November 2001; accepted 28 November 2001

countries during recent decades (Rolland-Cachera et al, 2000; Samuelson, 2000). Job-related physical activity and spontaneous physical activities have indeed declined because of the increase in the school-leaving age, travel to school by bus or car, urbanization of country people, lack of space for outdoor activities, especially in big cities etc, to the benefit of sedentary activities such as TV and video watching and computer games. However, sports activities sometimes increased, especially in adolescents performing competition sports. Thus, physical activity can be very variable among adolescents (Torun et al, 1996; Bratteby et al, 1997a). In the meantime, improvement of standard of living and availability of energy-dense foods resulted in frequent imbalance between energy intake and energy expenditure (EE), which explains the prevalence of overweight and obesity in children and adolescents (Martinez, 2000).

Energy expenditure of adolescents M Vermorel et al

820 Daily energy expenditure (DEE) of adolescents in freeliving conditions has been assessed during the last decade using the doubly labelled water (DLW) method (Bandini et al, 1990; Davies et al, 1991; Livingstone et al, 1992; Wong, 1994; Bratteby et al, 1998; Roemmich et al, 1998). However, this valuable method does not supply information on day-to-day variations in DEE of subjects, and EE devoted to physical activities. In addition, to our knowledge information on the time spent on different categories of physical activities has only been obtained from four studies carried out with 10 – 15-y-old children from activity diaries (Armstrong et al, 1990; Bratteby et al, 1997a; Henry et al, 1999; Ekelund et al, 2000). Furthermore, approximate costs of activities in adolescents have been derived from measurements in adults, and adapted for a diary of activities (Bouchard et al, 1983) or the DLW method (Bratteby et al, 1997b). Unfortunately, the estimated energy costs of high or very high intensity manual work and sport differed greatly between the two sets of data. Therefore, a group of experts concluded that ‘there is a need to obtain more information on the energy cost of activities and tasks in children and adolescents’ (Torun et al, 1996). The variations and determinants of EE of 62 12 – 16-y-old boys and girls have been studied in our laboratories with a standardized activity program using whole-body indirect calorimeters (Bitar et al, 1999). The objectives of the present cross-sectional study were to determine: (1) DEE and its main components in the same adolescents in free living conditions; (2) day-to-day variations in DEE and activity pattern; and (3) time devoted to, EE and energy costs of the various activities.

Subjects and methods Subjects Sixty adolescents (four groups of 14 – 16-y-old boys or girls aged 12 – 16 y) participated in this study. It was carried out during spring (from 23 April to 14 June) and autumn (from 18 October to 17 December) with seven or eight subjects per sub-group during each period. The volunteers were recruited from two high schools in the suburb of Clermont-Ferrand, France. Before the study began, the purpose and objectives were carefully explained to each subject and his or her parents. Written informed consent was obtained from all adolescents and their parents. The experimental protocol was approved by the University Ethical Committee on Human Research for Medical Sciences. All subjects had a thorough physical examination, and a medical history was taken. Their usual activity was estimated by interview. Anatomical measurements and sexual maturity were determined according to the protocols described previously for the same subjects (Bitar et al, 1999). Body composition was assessed by using both the skinfold-thickness method and bioimpedance analysis. There were no significant differences between the results obtained by the two methods. Peak oxygen uptake (peak VO2) was measured by a direct method (respiratory gas European Journal of Clinical Nutrition

exchange) during exercises on a cycloergometer. The subjects performed several successive 3 min steps against increasing braking forces until exhaustion (Bitar et al, 1999).

Assessment of EE and physical activities EE and physical activities of the volunteers in free-living conditions were assessed using the heart rate (HR)-recording method and an activity diary, respectively. HR was monitered during five consecutive days, from Friday evening to Wednesday evening, using Sport Testers (PE 4000, Polar Electro KY, Kempele, Finland). The system consists of an electro-belt transmitter and wrist microcomputer receiver that stores the pulse in a memory. Pulse was recorded at 1 min intervals up to a maximum recording time of 24 h, when stored information was retrieved via an interface unit and a microcomputer, and the memory was reprogrammed. Volunteers were encouraged to call the investigators at any time during the HR recording period if in any doubt about the recording. Each volunteer was given a detailed explanation and demonstration of the activity diary form and method. He (she) was instructed to write down regularly and carefully all activities (type, intensity, times of beginning and end of each activity) every day during the 5 day recording period. Volunteers were visited at home every evening by the main investigator to reprogram the memory of the Sport Tester and to check the activity diary. When necessary the latter was supplemented by interview of subjects and their parents. The activity records were then compared with the HR graphs to check the agreement of both recordings. In case of disagreement the activity record was immediately reexamined with the subject. The activities were distributed among 22 types ranging from sleep to sport competition.

Prediction equations of EE from HR recordings EE of the same subjects was determined continuously by indirect calorimetry over 24 h during a 36 h stay in two whole-body calorimeters. EE was calculated from oxygen consumption, carbon dioxide production, and 24 h urinary nitrogen (Nu) excretion by using Brouwer’s equation (1965):

EE (kJ) ¼ 16:18 O2 ðlÞ þ 5:02 CO2 ðlÞ 6:00 Nu ðgÞ over 5 min periods during exercise and 15 min periods for the rest of the day (Bitar et al, 1999). HR was recorded continuously by telemetry during the same period according to a standardized activity program including sleep, basal metabolic rate (BMR), meals, seated activities (schoolwork, reading, games and watching television), four 15 min periods of exercise on a cycle ergometer, and the recovery periods (Bitar et al, 1999). In addition, EE of subjects was determined during laboratory tests by open-circuit calorimetry using a face mask and an automated on-line system (CPX ID; Medical Graphics, St Paul, MN USA) for usual activities which could not


821 be performed in the whole-body calorimeters: lying (10 min after adaptation), sitting (5 min), standing quietly (5 min), stepping up and down a wooden stool (one step per s during 4 min), recovery (5 min), walking on a treadmill at 4 – 5 – 6 km=h (5 min at each speed), jogging at own speed (4 min), and recovery (10 min). HR was recorded continuously (Cardiovit CS.6=12; Schiller AG, Baar, Switzerland) during the corresponding periods. Individual general relationships were computed between data for EE and HR recorded both in the whole-body calorimeters during 24 h and during the laboratory tests: two linear relationships with determination of the flex-HR value, an exponential relationship for HRs below the flexHR value and a linear relationship above the flex-HR value, and polynomial relationships (Bitar et al, 1996). The best fit was obtained with polynomial relationships of the third order (Figure 1). The coefficient of determination (r2) averaged 0.900 (s.d. 0.034), and the residual standard deviation (r.s.d.) was 2.11 (s.d. 0.56) kJ=min. In all subjects the values corresponding to the recovery periods following cycling, stepping or jogging exercises were below the regression curves. In addition, individual specific relationships were computed between EE and HR data during sleep, on the one hand, and seated activities, on the other hand.

DEEs and energy costs of activities in free-living conditions DEEs of subjects were computed from HR recorded during 5 days using the individual general relationships. The 5 days included Saturday (school in the morning for 7 or 8 subjects per group), Sunday, Monday (a school day without physical education lessons, PEL), Tuesday (a school day with 2 h PEL), and Wednesday (school in the morning and free activities in

the afternoon). DEEs on Thursday and Friday were estimated in view of assessing mean DEE over a week. Thursday was a school day with 1 h PEL; DEE was assumed to be the mean of DEEs on Monday and Tuesday. Friday was a school day without physical training; DEE was assumed to be similar to that of Monday. EEs during sleep (SEE) and seated activities were computed from HR recordings and the corresponding individual specific relationships to avoid bias caused by the general relationships in some subjects. By contrast, EEs during the other activities were computed from HR recordings using the individual general relationships which took into acount the recovery periods. Mean EE (kJ=min) corresponding to each activity was called the energy cost of this activity. In the case of PEL and physical training, for instance, it included the periods of warming up, exercises, recovery and rest between exercises. The physical activity ratio (PAR) of each activity was obtained by dividing the energy cost by SEE (kJ=min) of each subject as measured in whole-body calorimeters (Bitar et al, 1999) because SEE was more reliable than BMR.

Statistical analysis Values are expressed as least-squares means s.d. Data were analysed by ANOVA using PROC GLM of SAS software (1987) with the following models:

y ¼ m þ a sex þ b age þ sex age þ e

ð1Þ

to study the effects of these variates on physical characteristics (Table 1); y ¼ m þ a sex þ b age þ r FFM þ sex age þ e

ð2Þ

using FFM as covariate to study the effects of the variates on mean DEE over a week;

Figure 1 Example of relationship between heart rate (HR) and energy expenditure (EE). Results far below the regression lines corresponded to the recovery periods after the physical exercises.

European Journal of Clinical Nutrition


822

Table 1 Physical characteristics and body composition of adolescentsa 12 – 13 y

14 – 16 y P

Age (y) Body weight (kg) Height (cm) Body mass index (kg=m2) b Fat-free mass (kg) Fat mass (%) 2 Upper arm FFM area (cm ) Peak VO2 (l=min) Adjusted peak VO2c (l=min) a

Boys (n ¼ 15)

Girls (n ¼ 15)

Boys (n ¼ 16)

Girls (n ¼ 14)

Sex

Age

Sexage

12.6** 0.15 45.1** 2.0 153.5** 1.8 19.0 0.6 35.2*** 1.5 21.3** 1.1 28.8** 1.5 1.92** 0.08 2.07** 0.05

12.5** 0.16 43.2** 2.1 153.8** 1.8 18.2 0.6 33.8*** 1.5 21.5** 1.1 25.6** 1.5 1.67*** 0.09 1.90*** 0.06

15.0** 0.15 54.6* 1.9 167.9* 1.7 19.3 0.6 46.2* 1.5 15.2* 1.1 37.1* 1.5 2.58* 0.08 2.26* 0.06

14.9* 0.16 51.3* 2.1 165.6* 1.8 18.7 0.6 39.7** 1.5 22.6* 1.1 29.3** 1.5 1.93** 0.09 1.90*** 0.06

0.42 0.20 0.58 0.27 0.01 0.001 0.001 0.001 0.001

0.001 0.001 0.001 0.51 0.001 0.026 0.001 0.001 0.139

0.92 0.72 0.48 0.86 0.10 0.01 0.13 0.02 0.10

b

c

Least-squares mean s.d. FFM, fat-free mass assessed by the bioimpedance analysis method. Peak VO2 adjusted for FFM. Values in the same row with different symbols are significantly different, P < 0.05.

Table 2 Energy costs of activities (kJ=min) as measured during laboratory testsa 12.6-y-old

Lying Sitting Standing Stepping Walking Jogging

{ {

4 km=h 5 km=h 6 km=h km=h kJ=min

15.0-y-old

P

Boys

Girls

Boys

Girls

5.08** 5.70** 6.11*,** 22.09 14.61*,** 17.37*,** 21.34* 8.1 (0.6) 32.07**

4.52*** 5.16** 5.28*** 20.02 12.93** 15.88** 18.45** 7.9 (0.6) 27.98***

5.63* 6.46* 6.75* 22.28 15.34* 18.94* 21.41* 9.2 (0.4) 37.59*

4.59**,*** 5.23** 5.83**,*** 22.12 14.18*,** 16.82* 19.87*,** 8.0 (0.3) 31.79**

Multiples of s.e.e.

s.e. of LS means

Age

Sex

Mean

s.d.

0.19 0.26 0.29 1.12 0.68 0.75 1.06

0.11 0.13 0.05 0.33 0.15 0.10 0.49

0.001 0.001 0.004 0.31 0.04 0.02 0.04

1.09 1.24 1.32 4.71 3.14 3.79 4.57

0.02 0.02 0.03 0.34 0.13 0.10 0.40

1.44

0.002

0.001

7.09

0.27

Least-squares means. Values in the same row with different symbols are significantly different, P < 0.05. No significant interactions between age and sex for any variables.

a

y ¼ m þ a sex þ b age þ d day þ sex age day þ e

ð3Þ

to study the effects of the variates on DEE, EE and time devoted to the various activities; y ¼ m þ a sex þ b age þ d day þ r FFM þ sex age day þ e ð4Þ using FFM as covariate to study the effects of the variates on DEE, EE and time devoted to the various activities; y ¼ m þ a sex þ b age þ l season þ sex age season þ e ð5Þ

y ¼ m þ a sex þ b age þ l season þ r FFMþ sex age season þ e

ð6Þ

using FFM or body weight (BW) as covariate to study the effects of the variates on DEE, moderate and team sport activities; y ¼ m þ a sex þ b age þ l speed þ r BW þ sex age þ e ð7Þ to study the effect of the variates on EE during jogging. European Journal of Clinical Nutrition

Stage of puberty was not a significant determinant of DEE and SEE as measured by whole-body calorimetry in the same adolescents. Therefore, it was not considered in the models. For all models adjusted means were computed and compared using ‘LSMEANS’ and ‘TDIFF’ options of PROC GLM. Differences were considered as significant for P < 0.05.

Results Physical characteristics of subjects The distribution of subjects according to chronological age and stage of puberty have been published elsewhere (Bitar et al, 1999). Physical characteristics and body composition of subjects depending on sex and age are presented in Table 1. In spite of great individual variability in physical characteristics, none of them were obese since all the body mass index (BMI) values were below the 90th percentiles (RollandCachera et al, 1991). There were no significant differences in height, BW, BMI and body composition between boys and girls at 12.6 y of age. However, body composition was significantly different between boys and girls at 15.0 y of age: fat-free mass (FFM) was 6.5 kg higher, whereas fat mass (FM)


823 was 3.2 kg lower in boys than in girls (P < 0.003). Consequently, the percentage of FM was significantly lower in boys than in girls at 15.0 y of age (P < 0.001). These results were supported by a greater upper arm FFM area in boys (P < 0.001, Table 1). In addition, peak VO2 and adjusted peak VO2 were significantly affected by sex and age. Nevertheless, adjusted peak VO2 was similar in 12.6 and 15.0-y-old girls.

Energy costs of lying, sitting, standing, stepping, walking and jogging in laboratory tests The energy costs (kJ=min) of these activities (except stepping) were significantly lower in girls than in boys (Table 2). After adjustment for FFM the energy costs of lying and sitting were 8% lower in girls than in boys (P < 0.05). However, after adjustment for BW the energy cost of walking at 4 – 5 and 6 km=h on a treadmill did not vary significantly with age and sex. Similarly, after adjustment for BW and speed the energy cost of jogging did not vary significantly with sex and age. The energy cost of jogging at 7 – 10 km=h could be computed using the following relationship:

EE ¼ 0:475 BW þ 1:97 speed 1:37 sex 5:50

r 2 ¼ 0:675

RSD ¼ 3:57

(EE, kJ=min; BW, kg; speed, km=h; sex, boys ¼ 1; girls ¼ 2). Finally, the energy costs of the various activities, expressed as multiples of s.e.e., did not vary significantly with age and sex. The mean values are presented in Table 2.

Table 3

Variations of DEE in free-living conditions DEE expressed in absolute terms and DEE adjusted for FFM were significantly higher in boys than in girls (P < 0.001; Table 3). However, PAL did not vary significantly with sex and age. Mean DEE over a week increased with age in boys ( þ 10.7%; P < 0.02), but did not vary significantly in girls between 12.6 and 15.0 y of age. DEE during the schoolday with 2 h of PEL was significantly higher than during the schooldays without PEL (P < 0.05). It was also higher than during the weekend in the 15-y-old adolescents (P < 0.05), but not in the 12.6-y-old subjects (Table 3). However, mean DEE covered up great individual differences. Thus, DEE was on average 20.6% (s.d. 9.3) higher during the free days than on Monday in the active boys who practiced moderate physical activities and=or team sport, whereas it was 7.4% (s.d. 9.9) lower in the sedentary boys.

EE, energy cost and time devoted to the various activities Time devoted to each activity was the sum of the durations of the periods devoted to this activity each day or, on average, per day over the week. The definitions are similar for EE and the energy cost of each activity. EE, energy cost and time devoted to each of the 22 defined activities were analysed separately. Energy costs (kJ=min) were similar for some activities such as each of the four meals, washing, dressing, cooking, laying the table, travelling to school (generally by bus), or for school work and home work etc. Consequently the 22 activities were distributed between nine categories depending on type of activity and PAR (Figures 2 and 3).

Mean daily energy expenditures (DEE, MJ) and physical activity level (PAL) of subjects in free-living conditionsa 12.6-y-old

15.0-y-old

P

Boys

Girls

Boys

Girls

s.e. of LSM

Age

Sex

Agesex

DEE b PAL Schoolday without PELc Schoolday with 2 h PEL Wednesday Mean of five schooldays Mean of the weekend

10.78** 1.68 10.27** 11.52** 10.61* 10.62** 10.94**

9.27* 1.61 9.02* 9.81* 9.25* 9.30* 8.84*

11.93*** 1.60 11.43*** 13.39*** 12.47** 12.23*** 11.46*,**

9.48* 1.58 9.20*,** 10.50*,** 9.48* 9.61* 9.17*

0.37 0.036 0.44 0.48 0.46 0.43 0.43

0.07 0.13 0.001 0.005 0.07 0.034 0.253

0.001 0.20 0.001 0.001 0.001 0.001 0.001

0.13 0.42 0.84 0.19 0.16 0.48 0.48

Adjusted for FFM DEE Schoolday without PEL Schoolday with 2 h PEL Wednesday Mean of five schooldays Mean of the weekend

11.49*** 10.91** 12.25** 11.26** 11.28*** 11.62***

10.26** 9.91* 10.86* 10.49** 10.50** 9.89**

10.39** 9.95* 12.07** 10.61** 10.54** 9.78**

9.39* 9.07* 10.47* 9.15* 9.51* 9.07*

0.28 0.35 0.38 0.37 0.33 0.33

0.005 0.001 0.001 0.08 0.001 0.001

0.001 0.001 0.001 0.02 0.001 0.001

0.91 0.99 0.76 0.43 0.69 0.69

a

Least-squares means. Values in the same row with different symbols are significantly different, P < 0.05. PAL, physical activity level ¼ mean energy expenditure over the week=sleeping energy expenditure. c PEL, physical education lessons. b



824

Figure 2

Physical activity ratios (PARs), corresponding to the various activities ðmean and s:d: for the 4 groups of adolescentsÞ; P:A:R: ¼

EE activity ðkj=minÞ Sleeping EE ðkj=minÞ:

1 ¼ Sleep, 2 ¼ Television, 3 ¼ Home work, 4 ¼ Computer, video games, 5 ¼ Parlor games, 6 ¼ School work, 7 ¼ Meals, 8 ¼ Playing music, 9 ¼ Traveling by bus, 10 ¼ Cooking, washing, 11 ¼ Washing, dressing, 12 ¼ Slow walking, 13 ¼ Recreation, 14 ¼ Playing quietly, 15 ¼ Fast walking with satchel, 16 ¼ Recreation sport, 17 ¼ Cycling, 18 ¼ PEL, 19 ¼ Training, competition.

Figure 3 Time devoted to each of the nine categories of activities by boys and girls at 12.6 (light shaded area) and 15.0 (dark shaded area) years of age (means and standard deviations. 1 ¼ Sleep, 2 ¼ Television, 3 ¼ Leisure activities, 4 ¼ School and homework, 5 ¼ Meals, 6 ¼ Miscellaneous activities, 7 ¼ Light physical activities, 8 ¼ Moderate physical activities, 9 ¼ Team sport.



825 Sleep Time spent sleeping was similar in boys and girls, but decreased by 24 min=day with age (P < 0.001; Figure 3). The energy cost of sleeping was, on average, 13% lower in girls than in boys (P < 0.001), and 12.4% higher in the 15.0 than in the 12.6-y-old boys (P < 0.02). Finally, sleep accounted, on average, for 39.1% of time and 25.3% of DEE (Table 4).

age. It was similar for the four groups of subjects during the schooldays (62, s.d. ¼ 23 min=day), but increased three to five times in the older girls, and two to three times in the three other groups of subjects during the weekend. Time devoted to computer games ranged from 0 to 120 min=day. EE corresponding to leisure activities (Table 4) contributed, on average, 7.5% to DEE over the week, for whatever sex and age. The PAR of leisure activities did not vary significantly with sex and age, and averaged 1.65 (s.d. ¼ 0.16, Figure 2).

Watching television Time devoted to watching television averaged 110 (s.d. ¼ 45) min=day over the week, with great inter-individual differences (range 4 – 198 min=day; Figure 3). In addition, it was, on average, 91 min=day longer during the weekend (170, s.d. ¼ 90, range 0 – 382 min) than during the schooldays. EE during watching television was 130% higher during the weekend than during the schooldays, except in the 15.0y-old girls ( þ 60%). However, after adjustment for FFM, EE was no longer significantly affected by sex. Finally, watching television accounted, on average, for 7.6% of time, and 6.9% of DEE (Table 4). The PAR of watching television, did not vary significantly with sex and age, and averaged 1.37 (s.d. ¼ 0.15, Figure 2).

School and home work Time devoted to school and home work was significantly affected by age and weekday, but not by sex (Figure 3). EE corresponding to school and home work (Table 4) contributed to DEE for 21.9% over the week. The PAR did not vary significantly with sex and age, and averaged 1.61.

Meals The four meals consisted of breakfast, generally at 06:30, lunch at 12:00 or 12:30, snack at 16:30 or 17:00 and dinner at 19:30 or 20:00. Time devoted to meals (Figure 3) was significantly affected only by week day. It was, on average, 15 min longer on Sunday, especially for lunch, than during the six other days (101 vs 86 min=day). The energy costs (kJ=min 1) were not significantly different between the four meals. Therefore, the latter were grouped in a single activity, thus reducing the number of activities from 22 to 19. EE during the four meals varied significantly with sex and age,

Leisure activities These activities included reading, playing or listening to music, going to theatre, conversation and parlour and computer games. The time spent at leisure activities (Figure 3) was significantly affected by sex and week day, but not by

Table 4 Mean energy expended (EE) and contribution of each of the nine categories of activities to daily energy expenditure (DEE) depending on age, sex and daya 12 – 13-y-old

15 – 16-y-old

P

Boys

Girls

Boys

Girls

Age

Sex

Day

Inter.

Energy expenditure (MJ=day) Sleep Television Leisure activities School and home work Meals Miscellaneous activities Light physical activities Moderate physical activities Team sport

2.65* 0.10 0.73*,** 0.07 0.78*,** 0.10 2.18** 0.13 0.63** 0.04 0.67* 0.25 0.97 0.18 1.17* 0.27 1.00 0.18

2.34** 0.10 0.69*,** 0.07 0.64*,** 0.10 1.91** 0.13 0.60** 0.04 0.82*,** 0.32 0.82 0.18 0.63**,** 0.27 0.77 0.18

2.86* 0.10 0.77* 0.07 0.69** 0.10 2.94* 0.13 0.76* 0.04 0.95* 0.44 0.83 0.18 0.98*,** 0.26 1.24 0.18

2.38** 0.10 0.55** 0.02 0.93* 0.10 2.08** 0.14 0.65** 0.04 0.75*,** 0.26 0.83 0.18 0.43*** 0.27 0.90 0.18

0.006

0.001

0.001

0.97

0.065 0.001 0.001 0.03 0.36 0.06 0.12

0.24 0.001 0.005 0.66 0.86 0.002 0.001

0.001 0.001 0.01 0.27 0.001 0.001 0.001

0.15 0.002 0.85 0.37 0.04 0.15 0.31

EE % DEE Sleep Television Leisure activities School and home work Meals Miscellaneous activities Light physical activities Moderate physical activities Team sport

25.3*,** 0.5 7.1*,** 0.6 7.4** 0.8 20.8** 1.3 6.1** 0.2 6.2** 0.5 9.4* 0.7 9.8* 0.9 8.2 1.1

25.9* 0.5 8.0* 0.6 7.2** 0.8 21.0** 1.4 6.5*,** 0.2 8.6* 0.5 8.9*,** 0.7 6.7**,*** 0.9 7.6 1.1

24.4** 0.5 6.7*,** 0.6 6.0** 0.8 24.7* 1.3 6.6*,** 0.2 7.7*,** 0.5 7.1** 0.7 8.0*,** 0.9 8.9 1.1

25.5*,** 0.5 5.9** 0.6 10.2* 0.8 21.8*,** 1.4 7.0* 0.2 8.0*,** 0.5 8.7*,** 0.7 4.4*** 0.9 8.5 1.1

0.20 0.04 0.31 0.08 0.08 0.32 0.07 0.03 0.49

0.11 0.87 0.02 0.31 0.10 0.004 0.42 0.004 0.63

0.70 0.16 0.07 0.27 0.95 0.03 0.12 0.83 0.92

a

Least-square mean s.d.. Values in the same row with different symbols are significantly different (P < 0.05).



826 and contributed 6.6%, on average to DEE. Finally, the PAR of meals did not vary significantly with sex, age and weekday, and averaged 1.66.

Miscellaneous activities These activities included washing, dressing, traveling to school by bus, cooking and laying the table. The time spent at these activities did not vary significantly with sex, age and weekday, and averaged 97 min=day (Figure 3). EE associated with these activities averaged 7.8% of DEE (Table 4). The corresponding PARs were similar for the various activities and averaged 1.80 (Figure 2).

Light physical activities These activities consisted of slow walking, shopping, waiting for bus, recreation, playing quietly, and care of animals (dogs, horses etc). Time devoted to these activities (Figure 3) varied significantly with week day and sex. The corresponding EE was lower during the weekend than during the schooldays, except in the 15.0-y-old girls. It contributed 8.5% to DEE, on average, over the week (Table 4). The PAR corresponding to these activities was significantly higher in the 12-y-old boys than in the three other groups of subjects (2.63 vs 2.19, P < 0.001), because the young boys were more active during recreation.

Moderate physical activities These activities included fast walking and recreation sport, mainly basketball, soccer, table tennis, riding, cycling and dancing. Time spent at these activities was lower in girls than in boys, especially in the 15-y-old girls (P < 0.02, Figure 3) by 16 min=day during the school days, and 57 min=day during the weekend. In addition, individual variability was great since time spent at these activities ranged from 5 to 154 min=day. EE and the contribution to DEE were also significantly affected by sex, age and weekday. They were lower in girls than in boys, and decreased with increasing age whatever the weekday (Table 4). Sex, age and weekday did not significantly affect the corresponding PAR. The latter ranged from 2.6 to 5.3 according to activity and exercise intensity.

Team sport Team sport included physical education lessons at school, eurythmics for girls, and physical training or competition in clubs. EE during the 114 (s.d. 7) min PEL on Tuesday was higher in boys than in girls (P < 0.003) and increased with age (P < 0.05): from 2.56 (s.d. 0.48) to 2.78 (s.d. 0.74) MJ in boys, and from 1.63 (s.d. 0.58) to 2.30 (s.d. 0.89) MJ in girls. In addition to PEL at school, time devoted to team sport in clubs on Wednesday afternoon, Saturday and Sunday exhibited a great individual variability and was independent of age European Journal of Clinical Nutrition

and sex: only 37 of the 60 adolescents practiced physical training or competition for 164 (s.d. 116) min (range 15 – 478 min). The corresponding EE ranged from 0.26 to 10.19 MJ, with higher values in boys than in girls, even after adjustment for FFM, whatever age. Practicing team sport in club increased the mean DEE by 15.6%, on average, during these three days, in comparison with the subjects who did not practice. Team sport accounted for 8.3 (s.d. 11.6)% of DEE over the week independently of sex and age (Table 4). The PAR of team sport varied significantly with sex and age (P < 0.01), but not with weekday; however, the interindividual variability was high: from 3.56 to 6.84 (Figure 2).

Sport activities As a whole, time devoted to sport activities (recreation sport at school and at home, PEL, training and competition) was very variable, slightly longer in boys than in girls, and decreased with age, from 84 (s.d. 48) to 59 (s.d. 35) min=day over the week. The increases in EE resulting from sport activities instead of watching television or leisure activities were very variable, and averaged 1.19 (s.d. 0.80), 1.07 (s.d. 0.60) MJ=day in boys, and 0.72 (s.d. 0.37), 0.82 (s.d. 0.68) MJ=day in girls at 12.6 and 15.0 y of age, respectively.

Effect of season DEE and DEE adjusted for FFM were significantly higher in spring than in autumn (P < 0.04). The differences averaged 15% in the 12.6-y-old subjects. Time and EE spent in moderate physical activities and team sport tended to be higher in spring than in autumn (P < 0.08 and P < 0.06, respectively), especially in the 12.6-y-old subjects ( þ 39 min=day and þ 1.00 MJ=day; on average).

Discussion This cross-sectional study enabled direct comparison of DEE and its variation over a week in free-living conditions in four groups of boys and girls aged 12.6 and 15.0 y. In addition, it brought new information on the time devoted to their various activities, the corresponding everyday and mean EEs, and the energy costs of these activities in usual conditions. The results highlighted the great inter-individual variability in time, intensity and EE devoted to physical activities among subjects, although all of them had enough space and facilities to practise outdoor activities and training or competition. Time devoted to the various activities, and the corresponding EE and PAR were determined using both an activity diary and the min-by-min HR recording method, validated by whole-body-calorimetry. Various types of activities including sleep were practised either during laboratory tests or during the stay in the calorimetric chambers. Data corresponding to the recovery periods were also included in the prediction equations. To avoid bias and to improve the


827 prediction accuracy, the individual general relationships were used to estimate DEE and EE during physical activities, whereas specific prediction equations were used to estimate SEE or EE during the sedentary activities. We checked that the DEEs predicted by the two methods were similar. Mean DEE of male adolescents increased significantly between 12.6 and 15.0 y of age, mainly because of increases in FFM since DEE adjusted for FFM decreased with increasing age (Table 3). On the contrary, mean DEE of female adolescents reached a plateau at 12.6 y of age, since it was not significantly different from the values obtained in our laboratory with girls aged 15.0 (present study) or 17.0 y (Ribeyre et al, 2000), in spite of significant increases in BW and FFM. The increases in sleeping, basal and resting metabolic rates were compensated for by decreases in time and EE devoted to moderate physical activities and team sport, especially during the weekend. These results agree with those of Amstrong (1998), showing that males and females generally reduce their level of activity as they grow from childhood through adolescence and into adult life, the rate of decline being greater in females than in males. Mean DEEs of subjects were compared with the values obtained in industrialized countries using either the DLW method, the HR recording method or an activity diary validated by the DLW method. The small differences in BW and FFM were accounted for by expressing DEE per kg BW or per kg FFM, or as PAL, with BMR estimated using the prediction equation of Schofield et al (1985). The mean calculated PAL of the 12.6-y-old adolescents was the same for boys and girls (1.75), and it was similar to the values obtained by Davies et al (1991) in 12-y-old boys and girls (Table 5).

The mean DEE of the 15-y-old boys (1.73) was lower than those obtained in the USA (Bandini et al, 1990), in the UK (Davies et al, 1991), and in Sweden (Bratteby et al, 1997a,b,1998; Ekelund et al, 2000). However, the mean DEE expressed per kg BW or per kg FFM was intermediate between these values, and the differences were less than 6%. The mean calculated PAL of boys was similar to that obtained by Ekelund et al (2000), but lower than the values obtained by the other authors (Table 5). The results were similar for the DEE of the 15-y-old girls, but the differences were smaller than for boys when DEEs were expressed per kg BW (Table 5). The mean calculated PAL (1.64) of girls was lower than that of boys, and similar to those obtained by Bandini et al (1990), Davies et al (1991) and Ekelund et al (2000). However, it was lower than the values obtained by Bratteby et al (1997a,b 1998). In the latter studies, adolescents were generally more active than our subjects, since 80 vs 12% of them walked or bicycled to and from school, and 62% of them trained regularly. The greater DEE of adolescents in spring resulted from both a higher sleeping metabolic rate in all subjects (Bitar et al, 1999) and increases in time and EE devoted to moderate outdoor physical activities and team sport in the 12.6-y-old subjects, in agreement with the results obtained in 4 – 10-yold children (Goran et al, 1998). However, the differences were not significant in the 15-y-old subjects because girls had limited moderate physical activities in both seasons, and five of the eight boys studied in autumn practised soccer (training and competition). In spite of the small number of subjects per group, the results showed great inter-individual differences in duration, EE and energy cost of sedentary and physical activities. Boys

Table 5 Comparison of mean daily energy expenditure (DEE) and physical activity level (PAL) of 15-y-old boys and girls in industrialized countries Bandini et al (1990) Method used Boys Number of subjects Height (cm) Body weight (kg) Fat free mass (kg) DEE (MJ) DEE (kJ=kg BW) DEE (kJ=kg FFM) a PAL

Girls Number of subjects Height (cm) Body weight (kg) Fat free mass (kg) DEE (MJ) DEE (kJ=kg BW) DEE (kJ=kg FFM) PALa a

DLW

Davies et al (1991) DLW

Bratteby et al (1997a,b) Activity diary

Bratteby et al (1998) DLW

Ekelund et al (2000) HR record

Present study HR record

14 167 56.4 47.1 13.03 231 277 1.88

12 — 60.1 — 13.47 225 — 1.87

171 174.2 61.1 — 14.16 232 — 1.95

25 174 61.3 51.2 13.82 227 270 1.90

42 172 61.6 52.0 12.8 210 249 1.75

16 167.9 54.6 46.2 11.93 219 258 1.76

14 162 55.7 40.9 9.99 179 244 1.66

11 — 58.0 — 10.12 177 — 1.65

203 166 56.6 — 10.89 192 — 1.79

25 167 58.4 41.9 10.70 183 255 1.73

40 164 55.9 42.5 10.0 179 235 1.66

14 165.6 51.3 39.7 9.48 185 239 1.64

PAL calculated from body weight using the Schofield et al’ (1985) prediction equation.



828 and girls with a high PAL spent, on average, 2.6 and 3.4 times as much time in leisure and team sport (moderate and high physical activities), respectively, as those with a low PAL, and the corresponding EE was 3.4 and 4.9 times greater. Ninety percent of boys had daily physical activities above the minimum known to be associated with diverse health benefits in adults and children: 30 min=day of moderate intensity activities and 12.6 kJ=kg BW=day of exercise EE (Riddoch & Boreham, 1995). However, 28% of the girls had less than the minimum recommended physical activity, and 45% of them had less than 1 h=day of moderate or high physical activity=day on average, over the week. Finally, the energy costs of the 19 activities, expressed as PARs, did not vary significantly with sex and age. The coefficients of variation (CV) averaged 11.0% (s.d. ¼ 1.2) for the sedentary activities, except washing and dressing (CV ¼ 15.9%) and travelling by bus (CV ¼ 16.9%) because some subjects were restless in the bus. However, the CVs were much higher and averaged 20.2% (s.d. ¼ 0.9) for physical activities, including training and competition, except for PEL and eurythmics (CV ¼ 13.2%).

Conclusion Mean DEE, day-to-day variations in DEE and activity pattern of 60 male and female adolescents have been determined, as well as time devoted to, EE and energy costs of the various activities in free-living conditions. The high inter-individual variability in DEE resulted not only from differences in sex, body weight and body composition, but mainly from behavioral differences: time devoted to, and intensity of physical activities vs sedentary activities, especially television and computer games. In sedentary subjects the 3 h of PEL at school per week did not supply generally the recommended minimum moderate physical activity. Team sport in clubs induced a 15.6% increase in DEE during the corresponding days compared to sedentary activities. In addition, regular outdoor moderate physical activities as well as light physical activities such as walking to and from school have a great impact on DEE and thereby against overweight.

Acknowledgements We are grateful to the adolescents who participated enthusiastically in this study, to their parents, and to the high school directors for their cooperation. We thank B Beaune for his expert contribution to the laboratory tests, A Chamoux for lending the sport testers and R Taylor for revising the English.

References Amstrong N (1998): Young peoples physical-activity patterns as assessed by heart-rate monitoring. J. Sport Sci. 16, S9 – S16. Armstrong N, Balding J, Gentle P & Kirby B (1990): Patterns of physical activity among 11 to 16 y old British children. Br. Med. J. 301, 203 – 205.


Bandini LG, Schoeller DA & Dietz WH (1990): Energy expenditure in obese and nonobese adolescents. Pediatr. Res. 27, 198 – 203. Bitar A, Vermorel M, Fellmann N, Bedu M, Chamoux A, Coudert J (1996): Heart rate recording method validated by whole-body indirect calorimetry in 10-y-old children. J. Appl. Physiol. 81, 1169 – 1173. Bitar A, Fellmann N, Vernet J, Coudert J & Vermorel M (1999): Variations and determinants of energy expenditure as measured by whole-body indirect calorimetry during puberty and adolescence. Am. J. Clin. Nutr. 69, 1209 – 1216. Bouchard C, Tremblay A, Leblanc C, Lortie G, Savard R & Theriault G (1983): A method to assess energy expenditure in children and adults. Am. J. Clin. Nutr. 37, 461 – 467. ¨ tborn M & Samuelson G (1997a): Bratteby LE, Sandhagen BO, Lo Daily energy expenditure and physical activity assessed by an activity diary in 374 randomly selected 15-year-old adolescents. Eur. J. Clin. Nutr. 51, 592 – 600. Bratteby LE, Sandhagen B, Fan H & Samuelson G (1997b): A 7-day activity diary for assessment of daily energy expenditure validated by the doubly-labeled water method in adolescents. Eur. J. Clin. Nutr. 51, 585 – 591. Bratteby LE, Sandhagen B, Fan H, Enghardt H & Samuelson G (1998): Total energy expenditure and physical activity as assessed by the doubly-labeled water method in Swedish adolescents in whom energy intake was underestimated by 7-d diet records. Am. J. Clin. Nutr. 67, 905 – 911. Brouwer E (1965): Report of Sub-Committee on constants and factors. In: Energy Metabolism, ed. KL Blaxter, pp 441 – 443. EAAP publication 11. New York: Academic Press. Davies PSW, Livingstone MBE, Prentice AM, Coward WA, Jagger SE, Stewart C, Strain IJ & Whitehead RG (1991): Total energy expenditure during childhood and adolescence. Proc. Nutr. Soc. 50, 1 – 14A. ¨ stro ¨ m M, Yngve A & Nilsson A (2000): Total daily Ekelund U, Sjo energy expenditure and pattern of physical activity measured by minute-by-minute heart rate monitoring in 14 – 15 y old Swedish adolescents. Eur. J. Clin. Nutr. 54, 195 – 202. Goran MI, Nagy TR, Gower BA, Mazariegos M, Salomons N, Johnson R (1998): Influence of sex, seasonality, ethnicity, and geographic location on the components of total energy expenditure in young children: implications for energy requirements. Am. J. Clin. Nutr. 68, 675 – 682. Henry CJK, Webster Gandy JD & Elia M (1999): Physical activity levels in a sample of Oxford school children aged 10 – 13 y. Eur. J. Clin. Nutr. 53, 840 – 843. Livingstone MBA, Coward WA, White JA, Stewart CM & Kerr MJJ (1992): Daily energy expenditure in free-living children: comparison of the heart-rate monitoring with the doubly labeled water (2H218O) method. Am. J. Clin. Nutr. 56, 343 – 352. Martinez JA (2000): Obesity in young Europeans: genetic and environmental influences. Eur. J. Clin. Nutr. 54(Suppl 1), 56 – 60. Ribeyre J, Fellmann N, Vernet J, Delaitre M, Chamoux A, Coudert J & Vermorel M (2000): Components and variations in daily energy expenditure of athletic and non-athletic adolscents in free-living conditions. Br. J. Nutr. 84, 531 – 539. Ridocch CJ & Boreham (1985): The health-related physical activity of children. Sports Med. 19, 86 – 102. Roemmich JN, Clark PA, Mai V, Berr SS, Weltman A, Veldhuis JD & Rogol AD (1998): Alterations in growth and body composition during puberty: III. Influence of maturation, gender, body composition, fat distribution, aerobic fitness, and energy expenditure on nocturnal growth hormone release. J. Clin. Endocrinol. Metab. 83, 1440 – 1447. Rolland-Cachera MF, Colet J, Sempe´ M, Tichet J, Rossignol C & Charraud A (1991): Body mass index variations: centiles from birth to 87 y. Eur. J. Clin. Nutr. 45, 13 – 21. Rolland-Cachera MF, Bellisle F & Deheeger M (2000): Nutritional status and food intake in adolescents living in Western Europe. Eur. J. Clin. Nutr. 54(Suppl 1), 41 – 46.

Energy expenditure of adolescents M Vermorel et al Samuelson G (2000): Dietary habits and nutritional status in adolescents over Europe. An overview of current studies in the Nordic countries. Eur. J. Clin. Nutr. 54(Suppl 1) 21 – 28. Schofield WN, Schofield C & James WPT (1985): Predicting basal metabolic rate. New standards and review of previous work. Hum. Nutr. Clin. Nutr. 39C(Suppl), 5 – 41. Statistical Analysis System Institute Inc (1987): SAS=STAT guide for personal computers, version 6. Cary, NC: SAS Institute Inc.

Torun B, Davies PSW, Livingstone MBE, Paolisso M, Sackett R & Spurr GB (1996): Energy requirements and dietary energy recommendations for children and adolescents 1 to 18 y old. Eur. J. Clin. Nutr. 50(Suppl 1), 537 – 581. Wong WW (1994): Energy expenditure of female adolescents. J. Am. Coll. Nutr. 13, 332 – 337.

829
