Total daily energy expenditure and pattern of physical activity ... - Nature

European Journal of Clinical Nutrition (2000) 54, 195±202 ß 2000 Macmillan Publishers Ltd All rights reserved 0954±3007/00 $15.00 www.nature.com/ejcn

Total daily energy expenditure and pattern of physical activity measured by minute-by-minute heart rate monitoring in 14 ± 15 year old Swedish adolescents U Ekelund*, M SjoÈstroÈm, A Yngve and A Nilsson Unit for Preventive Nutrition, Department of Medical Nutrition, Karolinska Institutet, Stockholm, Sweden; and Department of Physical È rebro University, O È rebro, Sweden Education and Health, O

Objective: To assess total daily energy expenditure (TDEE) and patterns of physical activity among Swedish male and female adolescents and to relate the amount and intensity of physical activity to existing recommendations (energy expenditure equal to or above 12.4 kJ=kg=day or accumulation of 30 min=day in moderate physical activity equal to 4.5 times sedentary energy expenditure or more). Design: TDEE, physical activity level (PAL TDEE=BMR), energy expenditure (EE) and time spent in different intensities of physical activity were assessed by using minute-by-minute heart rate monitoring in combination with laboratory measured sedentary energy expenditure (SEE) and peak oxygen uptake. È rebro University, and Department of Clinical Setting: Department of Physical Education and Health, O È rebro Medical Centre Hospital, Sweden. Physiology, O È rebro, randomly selected Subjects: Eighty-two 14 ± 15 y old adolescents (42 boys, 40 girls) from the city of O through a two-stage sampling procedure. Results: TDEE was 12.8 MJ=day and 10.0 MJ=day for boys and girls respectively (P < 0.001) and PAL was 1.74 and 1.67 (NS). Forty-four percent and 47%, respectively, of TDEE referred to EE in physical activity, of which 70% for both genders referred to light physical activity (corresponding to < 4.5 times SEE). Eleven boys and 14 girls had an EE lower than 12.4 kJ=kg=day and=or did not accumulate 30 min=day in physical activity 4.5 SEE. Those (n 20) with the highest PAL values ( > 2.01 and 1.81, respectively) spent 149 min=day at a 4.5 SEE intensity level compared to 40 min=day for those (n 30) with the lowest PAL values ( < 1.55 and 1.45, respectively). Conclusions: Swedish adolescent boys and girls are similarly physically active. The major amount of time devoted to physical activity refers to light physical activity. At least thirty percent of adolescents seem not to achieve appropriate levels of physical activity considered to be bene®cial for health. È rebro County Council, The Public Health Committee of Stockholm County Council, SweducaSponsorship: O tion Foundation. Descriptors: adolescents; energy expenditure; heart rate; physical activity European Journal of Clinical Nutrition (2000) 54, 195±202

Introduction There is a general concern about low levels of physical activity in youth in industrialized countries. Habitual physical activity declines with increasing age (Verschuur & Kemper, 1985; Livingstone et al, 1992). Moreover, it seems that boys are more active than girls, especially regarding vigorous physical activity, but the reported data are not consistent (Riddoch et al, 1991; Janz et al, 1992; Livingstone et al, 1992; Gavarry et al, 1998). EngstroÈm (1989) reported, based on questionnaire data, a gradual decline from 1968 to 1984 in participation in organized *Correspondence: U Ekelund, Department of Physical Education and È rebro University, S-70182 O È rebro, Sweden. Health, O E-mail: [email protected] Guarantors: U Ekelund and M SjoÈstroÈm. Contributors: UE initiated the study and was responsible for the design in collaboration with AY and MS. AN collected and organized the data. UE and AN did the data analysis. AY and MS participated in the interpretation and discussion of the results. UE was the main writer of the manuscript. Received 11 May 1999; revised 7 September 1999; accepted 14 September 1999

sports and leisure time spontaneous physical activity in 15 y old Swedish adolescents. On the other hand, Bratteby et al (1997) concluded, by using activity diary data, that Swedish 15 y olds have a high energy expenditure and a high physical activity level. To what extent Swedish adolescents achieve appropriate levels of physical activity in relation to existing physical activity recommendations is not known. Different activity recommendations for young people have been proposed (Blair et al, 1989; Corbin et al, 1994; Sallis & Patrick, 1994; Biddle et al, 1998). The recommendation from Blair et al (1989) is an energy expenditure recommendation based on epidemiological studies that link physical activity to health in adults. They estimated energy expenditure in physical activity of 3 kcal=kg=day (12.4 kJ=kg=day) to be an appropriate target. Corbin et al (1994) proposed a `lifetime physical activity' guideline. It recommended, as a minimum, that children and adolescents should accumulate 30 min of moderate physical activity daily (equal to 3 ± 4 kcal=kg=day). Based on review of the scienti®c literature on various health effects of physical activity in adolescents, an international consensus conference generated two guidelines (Sallis & Patrick, 1994). The

TDEE of Swedish adolescents U Ekelund et al

196

®rst guideline could be interpreted as an accumulation of 30 min in moderate physical activity daily whereas the second guideline stated that adolescents should engage in at least three sessions per week of activities that last 20 min or more and that require moderate to vigorous effort. Recently, another international consensus conference updated the physical activity recommendations for young people (Biddle et al, 1998). The primary recommendation states that young people should be physically active with at least moderate intensity for at least one hour per day. Objective data assessing both total daily energy expenditure (TDEE) and patterns of physical activity in adolescents are obviously needed. The purpose of the present study was to determine both the amount and the intensity of physical activity among Swedish adolescents by assessing TDEE, physical activity level (PAL TDEE=BMR), energy expenditure in different intensities of physical activity and time spent at different physical activity intensity levels. A number (n 82) of randomly selected 14 ± 15 y old boys and girls was examined by using minute-by-minute heart rate (HR) monitoring as an objective measure. The data were subsequently related to existing physical activity recommendations in adolescents. Subjects and methods Study design and selection of subjects È rebro Youth Activity Study is a cross-sectional study O examining health with special emphasis on physical activÈ rebro is the ity and physical ®tness in youth (Figure 1). O seventh largest municipality in Sweden with about 130,000 inhabitants. It is located about 200 km west of Stockholm and close to the demographic midpoint in Sweden. It has,

È rebro Youth Activity Study. The present Figure 1 The outline of the O study reports data on physical activity in a randomly selected subgroup (n 82). European Journal of Clinical Nutrition

during the last three decades, changed from a typical industrial centre to a municipality predominated by trading, education and small-scale industries. It is organized into 14 smaller districts, each with its own local council. There are a total of 13 comprehensive schools including students from grades 7 ± 9 (14 ± 16 y). Initially, six of the 13 schools were randomly selected and then a number of students from each of these schools were selected at random. The number of students from each school was proportional to the size of the school. The sample of students thus selected (n 241, mean age, 14.8 0.3 y) represented approximately 21% of the eligible population in that age group. The vast majority, 232 students, completed the questionnaire part of the study and 150 (65%) agreed to participate in the complete study. The participation rate varied between 44 and 77% for the six schools. Of those who provided a reason for unwillingness to participate the main answers were `lack of time' and `afraid of blood sampling'. Three subjects reported a chronic disease (cystic ®brosis or asthma) as a reason for not participating. A random subsample (n 82), of the participants was monitored by min-by-min HR recording. In an attempt to detect sample bias, participants (n 150) and nonparticipants (ie subjects who only completed the questionnaire) were compared with regard to gender, age, height, weight, physical education marks, activity habits and outcome of a progressive shuttle run test for assessment of aerobic ®tness. Physical descriptive data (age, height and weight) were obtained from the School Health services. Body mass index (BMI) was calculated from these data and compared between participants and non-participants. Physical education marks were obtained from physical education teachers. Information about activity habits such as time spent in sedentary activities (TV watching, computer games and informatics), participation in physical education and the duration and intensity of leisure time physical activities and organized sports was obtained from a questionnaire. The shuttle run test was performed by instructed physical education teachers. No signi®cant differences were found between participants and non-participants in age, height, weight, BMI, physical education marks or aerobic ®tness. Further, there were no signi®cant differences between participants and non-participants in time spent in sedentary activities or participation in physical education. The only signi®cant difference (P 0.045) found between groups was when calculating an activity index from questions concerning leisure time physical activity and organized sport activities. È rebro County The study protocol was approved by O Council Research Ethics Committee. All students received verbal as well as written information and all parents received written information about the study. The participation was voluntary and the students and their parents provided written informed consent. All data were collected during the school term from February to May. Anthropometry Height was measured to the nearest 0.5 cm using a wallmounted stadiometer and body mass was measured using a standard laboratory scale, to the nearest 0.1 kg in shorts and T-shirt. Body mass index (BMI) was calculated as weight=height2. Relative body fatness was calculated from triceps and subscapula skinfolds (Slaughter et al, 1988). Body fat mass (FM) was obtained by multiplying


197

Table 1 Physical characteristics of the 82 subjects (mean s.d.) Gender (n) Boys (42) (range) Girls (40) (range)

Age (y)

Height (cm)

Weight (kg)

BMI (kg=m2)

FM (%)

FFM (kg)

14.8 0.3 (14.3 ± 15.2) 14.7 0.3 (14.3 ± 15.3)

172 8.9 (152 ± 189) 164 5.8* (152 ± 175)

61.6 13.1 (40.4 ± 96.5) 55.9 8.6* (38.0 ± 75.5)

20.7 3.0 (17.5 ± 30.2) 20.9 3.3 (15.0 ± 30.9)

15.5 8.7 (7 ± 41) 23.6 7.1** (12 ± 41)

52.0 7.1 (34.7 ± 66.0) 42.5 4.6** (30.7 ± 52.1)

BMI, body mass index; FM, fat mass; FFM, fat free mass; FM and FFM calculated from skinfold measurements; *P < 0.05, **P < 0.001, denotes statistical signi®cance between genders.

percentages of body fat by body weight. Fat free mass (FFM) was calculated by subtracting FM from body weight. A summary of the physical characteristics of the subjects is shown in Table 1. Peak oxygen uptake Peak oxygen uptake (PVO2) was measured by using an online, open circuit system through running on a motorized treadmill. The subjects were taught to walk and run on the treadmill before the tests as well as accustomed to the mouthpiece and noseclip. The workload was increased every 3 min, according to Bruce et al's (1972) incremental protocol, until the test was terminated due to exhaustion. Throughout the test, expired air was collected using a twoway non-rebreathing valve (Hans Rudolph Inc., Montana) and a noseclip. Volumes were measured with a Rudolph pneumotachgraph and O2 and CO2 content was measured, at 20 s intervals, by an infrared carbon dioxide analyser and a paramagnetic oxygen analyser (Medical Graphics Inc., Minnesota). Before each test the system was calibrated against gases of known composition and the pneumotachgraph was calibrated against relevant ranges of ¯ow rates by using a 3 ± 1 calibration syringe (Hans Rudolph Inc., Montana). HR was monitored using a bipolar lead (S&B Medico Teknik, Denmark). Four minutes after the test, a ®ngertip blood sample was obtained for measurement of post-exercise blood lactate levels. Lactate concentration in whole blood was immediately analysed using an automated analyser (Analox Instruments Ltd, England). PVO2 was determined by a combination of variables: a respiratory exchange ratio 1.0, HRs within 5 bpm from age-predicted maximum HR, signs of intense effort (hyperpnea, dif®culties in following the treadmill, facial ¯ushing) and a post-exercise blood lactate concentration 7 mmol=l. All subjects ful®lled at least two of these criteria. Subject calibration Calibration procedures were performed in the laboratory after 30 min of supine rest and at least 2 ± 3 h after a meal. Oxygen uptake (VO2) and HR were simultaneously measured under standardized conditions. The following activities were carried out for determination of the individual relationship between VO2 and HR: supine, sitting quietly, standing quietly, and exercising on the treadmill at speeds of 2.7=km at a 10% gradient and 4=km at a 12% gradient. All activities were done in sequence. HR and VO2 were measured at 20 s intervals by using the same system previously described. The gas sampling time was 6 min, initiated by a 3-min acclimatization and equilibrium period. Calibration points were calculated as the mean HR and VO2 for the last 3 min. Sedentary energy expenditure (SEE) was calculated as the average energetic cost of supine, sitting and standing.

Calculation of TDEE, PAL, HR thresholds and energy expenditure and time spent at different levels of physical activity The individual relationship between VO2 and HR was determined by least square regression for both resting activities and physical activity. The critical HR (FLEX HR), above which energy expenditure (EE) can be calculated from HR data was determined as the average of the highest HR, ie standing, during the resting measurements and the lowest HR while exercising on the treadmill. TDEE from HR monitoring was determined as follows. For HR periods during daytime FLEX HR, EE was assumed to equal measured SEE. For HRs > FLEX HR, EE was calculated min-by-min from the individual VO2 ± HR regression lines (VO2 a HRb), where a is the intercept and b the slope of the individual calibration line. EE during sleep was assumed to equal predicted BMR (Scho®eld et al, 1985). TDEE was obtained by summing EE act (HR > FLEX), EE rest (HR FLEX) and EE sleep (BMR). An additional computer program was written to compute EE from HR data. HR thresholds corresponding to 4.5 times SEE, 50% PVO2 and 70% PVO2 were individually calculated from the VO2 ± HR regression equations. The 4.5 SEE threshold was selected for calculation of EE and time spent in physical activity according to published recommendations (Blair et al, 1989; Corbin et al, 1994; Sallis & Patrick, 1994). These authors have suggested a minimum amount of energy expended equal to 12.4 kJ=kg=day, or accumulation of 30 min=day in moderate physical activity (approximately equal to 40 ± 60% PVO2). Recently, the time recommendation was suggested to be changed to accumulation of 60 min=day (Biddle et al, 1998). Subjects were classi®ed into three different activity groups (low, medium and high) based on their PAL values. Low habitual physical activity was de®ned as a PAL value < 1.55 and 1.45, medium physical activity as 1.56 > PAL 2.0 and 1.46 > PAL 1.8 and high physical activity as > 2.01 and > 1.81, for boys and girls respectively (Torun et al, 1996). Activity groups were compared for accumulated time (min=day) and total energy expenditure (kJ=kg=min) at different intensity levels (light, moderate and vigorous). Light intensity was de®ned as HR FLEX > EE HR 4.5 SEE; moderate intensity as HR 4.5 SEE > EE HR 70% PVO2; and vigorous intensity, EE > HR 70% PVO2. Heart rate monitoring HR was monitored over 2 weekdays and 1 weekend day within 2 weeks of calibration procedures by using a Polar Sport Tester PE-4000 Heart Rate Monitor (Polar Electro OY, Finland). HR was recorded on an average of 1 min intervals from the ®rst waking hour until removal close to European Journal of Clinical Nutrition


198

the subject's bedtime. HR monitors were collected by assistants each morning after monitoring and the data were retrieved via an interface to a personal computer and stored. If data were lost due to malfunction of the recorders or by other reasons a new monitoring attempt was made on the following day. Three days of HR monitoring was achieved in 74 of 82 subjects. Despite repeated attempts, eight subjects provided only two days of monitoring. Unreliable HR values, ie spikes above maximal HR and HRs below 35 bpm, were removed by ®lter (Davidson et al, 1997). Statistics Results in text and tables are expressed as mean s.d. Interindividual differences between slopes and intercepts were examined by calculating the coef®cient of variation (CV). The gender effect on HR thresholds and the components in TDEE were analysed by one-way analysis of variance (ANOVA). The in¯uence of gender and activity group (low, medium and high) on time (min=day) spent at different intensity levels (light, moderate and vigorous) of physical activity was tested by a multivariate analysis of variance (MANOVA). A similar analysis was performed to test the in¯uence of gender and activity group on energy expenditure (kJ=kg=min) at different intensity levels. Differences between activity groups were tested by Tukey's test. Otherwise, statistical analyses were performed using a t-test. For all analyses, a 5% signi®cance level was used. Data were analysed by using SPSS 6.1 statistical package (SPSS Inc., Chicago). Results Individual calibration of VO2 ± HR relationship Intersubject variation in the relationship between VO2 and HR demonstrated the need for individual calibration equations. The intersubject CV for slopes was 32% for boys and 16% for girls. The average slope that re¯ects oxygen transport per heartbeat was steeper in boys compared to

girls (P < 0.001). Gender differences were also suggested for average intercepts (P 0.052). Intersubject CV for intercepts was 60% and 34% for boys and girls respectively. PVO2 and HR data (Table 2) FLEX HR was 24% and 28% higher than resting HR in boys and girls, respectively. The average oxygen consumption corresponding to heart rates 4.5 times SEE was 21.7 4 mlO2=kg=min (41% PVO2) for boys and 19.6 4 mlO2=kg=min (45% PVO2) for girls (P < 0.05). All HR thresholds were signi®cantly different (P < 0.001) from each other in both genders. No gender differences were observed in HR thresholds equivalent to 4.5 SEE, 50% and 70% PVO2. TDEE and its components (Table 3) SEE was 16% and 17% higher than estimated BMR for boys and girls respectively (P < 0.001). On average, 47% and 44% of TDEE refer to EE in physical activity (EE act) above FLEX HR. When EE act was expressed per kilogram of body weight no difference between boys and girls was observed (95.6 59 kJ=kg=day for boys and 82.3 48.7 kJ=kg=day for girls, P 0.15). The major amount, about 67 ± 70%, of EE act was expended in light intensity physical activity (FLEX HR > EE HR 4.5 SEE). Average values for EE related to kilogram of body weight in light intensity physical activity were 67.0 36.9 and 54.6 26.5 kJ=kg=day (P 0.16) for boys and girls, respectively. EE related to kilogram of body weight at a 4.5 SEE intensity level were 28.4 24.9 and 28.1 7.9 kJ=day (P 0.99). Comparison between high, medium and low PAL groups (Figure 2) MANOVA showed an overall activity group effect (P < 0.001). There was a signi®cant activity group effect for all intensity levels (P < 0.001). Signi®cant differences (P < 0.05, Tukey's test) were observed between all activity groups (low, moderate, high) at all intensity levels (light,

Table 2 Peak oxygen uptake (PVO2, ml=kg=min) and heart rates (HR, bpm) (mean s.d.) Gender (n)

PVO2

Boys (42) (range) Girls (40) (range)

53.5 8.0 (35.4 ± 65.4) 43.3 5.5* (31.1 ± 54.0)

Max

Rest

197 9 (181 ± 209) 196 8 (181 ± 209)

78 8 (54 ± 94) 79 9 (62 ± 104)

Flex

4.5 SEE

97 10 (78 ± 111) 101 9 (79 ± 117)

120 15 (91 ± 149) 126 17 (98 ± 149)

50% PVO2

70% PVO2

132 8 (114 ± 153) 134 8 (116 ± 165)

158 8 (144 ± 173) 159 7 (143 ± 174)

Max, maximum HR; Rest, average HR during supine, sitting and standing; Flex, average of the highest HR during resting measurements and lowest HR during exercising; 4.5 SEE, HR corresponding to 4.5 times sedentary energy expenditure; 50% PVO2, HR corresponding to 50% PVO2; 70% PVO2, HR corresponding to 70% PVO2; *P < 0.001, denotes statistically signi®cant difference between genders. Table 3 Estimated basal metabolic rate (BMR), sedentary energy expenditure (SEE), total daily energy expenditure (TDEE) and its components (EE sleep, EE rest, EE act), TDEE in relation to body weight (BW) and physical activity level (PAL) from heart rate monitoring (mean s.d.) Gender (n)

BMR (MJ=day)

SEE (MJ=day)

TDEE (MJ=day)

EE sleep (MJ=day)

EE rest (MJ=day)

EE act (MJ=day)

TDEE=BW (kJ=kg=day)

PAL (TDEE=BMR)

Boys (42) (range) Percentage of TDEE Girls (40) (range) Percentage of TDEE

7.4 0.9 (5.8 ± 9.6) 58 7 6.0 0.4{ (5.1 ± 6.6) 60 4

8.6 1.3 (6.2 ± 11.2) 67 10 7.0 1.4{ (4.5 ± 10.4) 70 14

12.8 2.6 (8.4 ± 19.7) 100 10.0 1.9{ (6.9 ± 14.8) 100

3.3 0.5 (2.3 ± 5.0) 26 4 2.7 0.3{ (2.1 ± 3.4) 27 3

3.5 1.1 (1.6 ± 5.8) 27 9 2.9 1.0** (1.1 ± 5.3) 29 10

6.0 3.2 (1.4 ± 14.0) 47 25 4.4 2.4* (1.0 ± 11.0) 44 24

210 52 (130 ± 292)

1.74 0.37 (1.25 ± 2.44)

182 42** (120 ± 293)

1.67 0.32 (1.21 ± 2.52)

EE sleep, energy expenditure during sleep calculated from BMR; EE rest, energy expenditure for time HR HR ¯ex; EE act, energy expenditure for time HR > HR ¯ex; *P < 0.05, **P < 0.01 and {P < 0.001, denote statistically signi®cant differences between genders. European Journal of Clinical Nutrition


199

Figure 3 The frequency distribution of the PAL values divided by boys (n 42) and girls (n 40).

Figure 2 (a) Accumulated time (min=day), or (b) energy expenditure (kJ=kg=day) spent at overall, light, moderate and vigorous physical activity for three different activity groups. All activity groups were signi®cantly different (P < 0.05, Tukey's test) from each other at all intensity levels, except between the low and moderate activity groups for moderate intensity of physical activity.

moderate, vigorous) except between the low and moderate activity groups for moderate intensity of physical activity. The high PAL group spent almost 2.5 times as much time in overall physical activity compared to the low PAL group (351 vs 145 min=day). Further, the high PAL group spent twice as much time in light physical activity (197 vs 101 min=day) and more than 3.5 times as much time at a 4.5 SEE intensity level (149 vs 40 min=day) compared to the low PAL group. Energy expenditure in overall physical activity was three times higher in the high PAL group compared to the low PAL group (151 vs 49 kJ=kg=day). The high activity group spent on average 59 kJ=kg=day 4.5 SEE of which 16 kJ=kg=day 70% PVO2, compared to 12 and 2.5 kJ=kg=day for the low activity group. In Figure 3 the frequency distribution of the PAL values is shown.

Patterns of physical activity (Table 4) Boys and girls spent, on average, 555 and 573 min, ie 70.5 and 72.5% of daytime (registered time) in sedentary activities lower than FLEX HR. Time spent at 4.5 SEE intensity level were, on average, 8.6% for both genders, of which 3.3% and 4.0% were spent in physical activity corresponding to 50% PVO2 or higher. One percent and 1.3% were spent in physical activity corresponding to 70% PVO2 or higher, for boys and girls respectively. Physical activity recommendations Thirty percent (11 boys, 14 girls) spent less than 12.4 kJ=kg=day at a 4.5 SEE intensity level. Twenty nine percent (10 boys and 14 girls) did not accumulate 30 min per day of physical activity at this intensity level or higher. Fifty one percent (19 boys and 23 girls) did not accumulate 60 min per day of physical activity at a 4.5 SEE intensity level or higher. Ten percent of the subjects (2 boys and 6 girls) were physical active less than 10 min per day at this intensity level. Discussion Subjects The present anthropometric characteristics (weight, height and BMI) seem to be higher compared to Swedish population reference standards from children born in about 1970 (Lindgren et al, 1995) and similar to those reported by Bratteby et al (1997). Although the data were generated from only one municipality in Sweden, they indicate a European Journal of Clinical Nutrition


200

Table 4 Absolute time (min=day) and relative time (%) spent at different levels of physical activity assessed by heart rate monitoring (mean s.d.) Gender (n) Boys (42) (range) Percentage of reg time Girls (40) (range) Percentage of reg time

Min registered

> HR ¯ex

4.5 SEE

50% PVO2

70% PVO2

788 71 (642 ± 954) 100 793 75 (665 ± 924) 100

233 147 (63 ± 514) 30 18 220 130 (48 ± 526) 28 15

67 71 (4 ± 155) 8.6 9.2 67 76 (2 ± 334) 8.6 9.8

26 17 (2 ± 71) 3.3 2.2 32 27 (2 ± 90) 4.0 3.5

77 (0 ± 24) 1.0 ± 0.8 10 12 (0 ± 39) 13 1.5

HR Flex, average of the highest HR during resting measurements and lowest HR during exercising; 4.5 SEE, HR corresponding to 4.5 times SEE; 50% PVO2, HR corresponding to 50% PVO2; 70% PVO2, HR corresponding to 70% PVO2.

trend of increasing height, weight and BMI in Swedish adolescents. Methodology Only a few studies in children and adolescents have calibrated the subjects in relation to their individual VO2 ± HR regression lines (Verschuur & Kemper, 1985; Atomi et al, 1986; Spurr & Reina, 1988; Riddoch et al, 1991; Livingstone et al, 1992; Emons et al, 1992; Ekelund et al, 1999). Despite the fact that HR is affected by emotional status, temperature, day to day variation and type of work performed, validation studies using the DLW method in children and adolescents have shown that minby-min HR monitoring produced similar results for TDEE in groups of children and adolescents with various levels of physical activity (Livingstone et al, 1992; Maffeis et al, 1995; van den Berg-Emons et al, 1996). In the present study BMR was estimated from predictive equations (Scho®eld et al, 1985) and assumed to equal energy expenditure during sleep. It has been shown that overnight energy expenditure and BMR are equivalent and that 1.06BMR is an appropriate factor for approximation of energy expenditure in bed (Seale & Conway, 1999). The approach used in this study, estimated BMR during sleep together with the use of heart rate recording during the awake time of the day, when calculating TDEE was suggested by Spurr et al (1988). Furthermore, the Scho®eld et al (1985) predictive equations produce similar results compared to measured BMR in 15 y old Swedish adolescents (Bratteby et al, 1997). The measured average values for SEE was 16 ± 17% higher than estimated BMR, which is in agreement with others (Livingstone et al, 1992) and close to what was deemed appropriate by Torun et al (1996). This indicates that our estimation of BMR did not signi®cantly affect the calculation of TDEE. In a comparison between min-by-min heart rate monitoring and activity diary, performed in another group of 15 y old boys and girls, we found no signi®cant difference between the methods in TDEE (Ekelund et al, 1999). Individual calibration of the HR ± VO2 relationship is of critical importance for the method. In the present study, the individual calibration was performed once. McCrory et al (1997) examined the between-day and within-day variability in the HR ± VO2 relationship and the effect of this variability on the estimation of TDEE. They found almost identical average TDEE values (CV 1.1%) and concluded that the min-by-min heart rate monitoring method is appropriate for estimation of TDEE in groups, due to the excellent average between-day and within-day repeatability of the HR ± VO2 relation. European Journal of Clinical Nutrition

The intersubject CV for slopes was twice as high in boys compared to girls. This difference may be explained by a greater variation in biological maturity in boys which affects movement economy and blood haemoglobin level and thereby the HR ± VO2 relationship. TDEE and PAL; comparison with other studies (Table 5) The results for TDEE and its components are in agreement to those by Livingstone et al (1992). Spurr & Reina (1988) reported HR TDEE data from normal and marginally malnourished Colombian adolescents similar in age to those in the present study. Absolute values for TDEE were higher in Swedish boys compared to Colombian boys, but energy turnover per kilogram of body weight seemed to be higher in both groups of Colombian boys. The average value for PAL was also higher in Colombian boys (1.94 vs 1.74). For girls, similar results for TDEE related to body weight and in PAL were observed in the present study compared to the study by Spurr & Reina (1988). No difference was observed between gender in PAL values in the present study as well as in the study by Livingstone et al (1992), while the Colombian boys were signi®cantly more physically active than girls (Spurr & Reina, 1988). It could be hypothesized that socio-cultural differences between countries in¯uence activity patterns and thereby contribute to the gender difference. The data indicates that Table 5 Comparison of energy expenditures and activity levels between minute-by-minute heart rate monitoring studies in 14 ± 15 y old adolescents Spurr & Spurr & Livingstone Present Reina (1988)a Reina (1988)b et al (1992) study Boys Subjects (n) Registered time (days) Age (y) Weight (kg) TDEE (MJ=day) TDEE (kJ=kg=day) AEE (MJ=day) PAL Girls Subjects (n) Registered time (days) Age (y) Weight (kg) TDEE (MJ=day) TDEE (kJ=kg=day) AEE (MJ=day) PAL a

20 1±2 14.8 49.9 12.1 244 5.6c 1.94

26 1±2 14.8 38.9 10.6 274 4.9c 1.93

3 2±3 15.4 50.7 11.5 227 4.8 1.71

42 2±3 14.8 61.6 12.8 210 5.4 1.74

19 1±2 14.9 49.3 8.3 174 2.7* 1.61

22 1±2 15.2 42.0 8.2 203 2.8* 1.61

3 2±3 15.6 55.4 9.9 179 4.4 1.88

40 2±3 14.7 55.9 10.0 182 4.0 1.67

Control subjects; bMarginally malnourished subjects; cValues have been calculated by the present authors. AEE, activity energy expenditure (TDEE 7 BMR); PAL, physical activity level (TDEE=BMR).


14 ± 15 y old boys from developing countries are more physically active than their peers from industrialized countries. TDEE and PAL; Swedish adolescents The results for TDEE and PAL are not statistically different from what we found in another study in 15 y old boys and girls (Ekelund et al, 1999). Another Swedish study using a 7 day activity diary has shown higher absolute values for TDEE and for TDEE in relation to body weight (Bratteby et al, 1997). The average values for PAL found by Bratteby et al (1997) were also higher (1.95 and 1.80 for boys and girls, respectively). This could be due to real differences in physical activity between the groups, seasonal differences during data collection or differences in methodology used. The mean values for PAL, 1.74 and 1.67 for boys and girls, respectively, found in the present study are in agreement to those suggested as moderate habitual physical activity (Torun et al, 1996). Comparison between high, medium and low PAL groups In all groups, the major amount (60 ± 75%) of energy expenditure in physical activity (EE act) was due to light physical activity. It was found that, on average, 30 ± 32% of TDEE referred to energy expenditure in light physical activity and no more than 13 ± 15% of TDEE referred to energy expenditure at an intensity level 4.5 REE. The association between overall physical activity and time spent in sedentary activities (HR FLEX HR) indicated that the high PAL groups avoided sedentary activities. This ®nding was supported by the ®ndings from the questionnaire. Boys in the high PAL group spent signi®cantly less time on computer games and informatics compared with the other two groups. Girls in the high PAL group spent signi®cantly less time on TV viewing. The proportion of light and moderate intensity physical activity seems to have a great impact on TDEE and thereby on PAL. Physical activity according to existing recommendations Approximately 70% of the subjects seem to achieve the recommended amount of physical activity according to the energy expenditure recommendation (Blair et al, 1989; Corbin et al, 1994) as well as according to Guideline One in Physical Activity Guidelines for Adolescents (Sallis & Patrick, 1994). On the other hand, it must be considered a matter of concern that at least 30% of the participants in the present study did not reach the desired amount of physical activity. Furthermore, no more than 50% of the boys and girls were physically active for at least one hour at an intensity level equal to 4.5 times SEE (Biddle et al, 1998). The only signi®cant difference (P < 0.05) found, between those who did not reach the recommended amount of physical activity and the rest of the population, was in aerobic ®tness (PVO2). However, it is not known to what extent different variables of physical activity are related to health in youth (Riddoch & Boreham, 1995). If light intensity physical activity (ie > HR FLEX) was included in the analysis, all subjects reached the desired amount of physical activity in relation to existing recommendations. If the total amount of energy expenditure in physical activity is the most important variable in relation to health, the contribution of light intensity physical activity is signi®cant. The average values for PAL from the present study could be compared to the set values in the Nordic Nutrition

Recommendations (Nordic Council of Ministers, 1996). In these recommendations, estimated required energy intake is based on PAL values of 1.58 and 1.50 for boys and girls with similar weight as in the present study. These recommendations are based on factorial estimates of energy requirements. However, it has been stated that energy requirements should be based on measurements of energy expenditure (Torun et al, 1996). The ®ndings from the present study raise the question whether the set values for PAL in Nordic Nutrition Recommendations are too low, resulting in low energy intake recommendations.

201

Patterns of physical activity; comparison with others HR monitoring has been used as a way to assess physical activity in children and adolescents (Armstrong et al, 1990; Armstrong & Bray, 1991; Riddoch et al, 1991; Janz et al, 1992; Gavarry et al, 1998). Comparisons between the studies are dif®cult to do due to differences in design, methodology and interpretation of the HR data. However, by using individually calibrated data for the relationship between VO2 ± HR, patterns of physical activity according to speci®c HR thresholds can accurately be assessed (Verschuur & Kemper, 1985; Riddoch et al, 1991; Livingstone et al, 1992). Verschuur & Kemper (1985) reported approximately 60 min per day in physical activity for Dutch 15 y old adolescents at an intensity > 50% PVO2 without any difference between gender. Riddoch et al (1991) reported 7 min (1.4%) and 12 min (2.0%) per day for boys and girls, respectively, spent in physical activity > 50% PVO2 in 14 ± 16 y old Northern Irish school children. Livingstone et al (1992) reported 52 min (5.1%) and 15 min (1.9%) of waking time spent at the same intensity level for 12 ± 15 y old boys and girls, respectively. This could be compared to 26 min (3.3%) and 32 min (4.0%) per day for boys and girls in the present study. It seems that the Dutch teenagers were more physically active while the Swedes were at least as active as their British peers. In fact, the Swedish girls spent twice as much of daytime at this intensity level compared to the British girls. No differences were observed between boys and girls. This is in accordance with the ®ndings from Verschuur & Kemper (1985) and from Riddoch et al (1991), but opposite to the ®ndings from Livingstone et al (1992). Conclusions Swedish boys and girls, 14 ± 15 y old are physically active to a similar extent. The majority of physical activity-related energy expenditure referred to energy expenditure in light intensity physical activity, which seems to have an impact on overall physical activity. At least one-third of the adolescents do not seem to achieve appropriate levels of physical activity according to established recommendations. Acknowledgements ÐWe are grateful to all the subjects who participated È rebro University for in this study, to Associated Professor Olle Carlsson, O valuable statistical assistance and to Rune Hedman and the staff at Clinical È rebro Medical Centre Hospital for their assistance during Physiology, O data collection.

References Armstrong N & Bray S (1991): Physical activity patterns de®ned by continuous heart rate monitoring. Arch. Dis. Childhood 66, 245 ± 247. European Journal of Clinical Nutrition


202

Armstrong N, Balding J, Gentle P & Kirby B (1990): Patterns of physical activity among 11 to 16 year old British children. Br. Med. J. 301, 203 ± 205. Atomi Y, Iwaoka K, Hatta H, Miyashita M & Yamamoto Y (1986): Daily physical activity levels in preadolescent boys related to VO2max and lactate threshold. Eur. J. Appl. Phys. 55, 156 ± 161. Biddle S, Sallis J & Cavill N (eds) (1998): Young and Active? Young People and Health-Enhancing Physical Activity ± Evidence and Implications. London: Health Education Authority. Blair S, Clark D, Cureton K & Powell K (1989): Exercise and ®tness in childhood: implications for a lifetime of health. In: Perspectives in Exercise Science and Sports Medicine: Youth Exercise and Sport, ed. C Gisol® & D Lamb, pp 401 ± 430. Indianapolis: Benchmark Press. Bratteby LE, Sandhagen B, LoÈtborn M & Samuelson G (1997): 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. Bruce RA, Kusumi F & Hosmer D (1972): Maximal oxygen intake and nomographic assessment of functional aerobic impairment in cardiovascular disease. Am. Heart J. 85, 546 ± 562. Corbin C, Pangrazzi R & Welk G (1994): Toward an understanding of appropriate physical levels for youth. President's Council of Physical Activity and Sport. Physical Activity Fitness Res. Dig. 1, 1 ± 8. Davidson L, McNeill G, Haggarty P, Smith J & Franklin M (1997): Free living energy expenditure of adult men assessed by continuous heartrate monitoring and doubly-labelled water. Br. J. Nutr. 78, 695 ± 708. Ekelund U, Yngve A & SjoÈstroÈm M (1999): Total daily energy expenditure and patterns of physical activity assessed by two different methods in adolescents. Scand. J. Med. Sci. Sports 9, 257 ± 264. Emons HJG, Groenenboom DC, Westerterp K & Saris WHM (1992): Comparison of heart rate monitoring combined with indirect calorimetry and the doubly labelled water (2H218O) method for the measurement of energy expenditure in children. Eur. J. Appl. Phys. 65, 99 ± 103. EngstroÈm LM (1989): Idrottsvanor i foÈraÈndring (Changing sport activities). Stockholm: HLS foÈrlag (in Swedish). Gavarry O, Bernard T, Giacomoni M, Seymat M, Euzet JP & Falgairette G (1998): Continuous heart rate monitoring over 1 week in teenagers aged 11 ± 16 years. Eur. J. Appl. Phys. 77, 125 ± 132. Janz K, Golden J, Hansen J & Mahoney L (1992): Heart rate monitoring of physical activity in children and adolescents: the Muscatine Study. Pediatrics 89, 256 ± 261. Lindgren G, Strandell A, Cole T, Healy M & Tanner J (1995): Swedish population reference standards for height, weight and body mass index at 6 to 16 years (girls) or 19 years (boys). Acta Pediatr. 84, 1019 ± 1028. Livingstone MBE, Coward WA, Prentice AM et al (1992): Daily energy expenditure in free living children: comparison of heart rate monitoring with the doubly labelled water (2H218O) method. Am. J. Clin. Nutr. 56, 343 ± 352.

European Journal of Clinical Nutrition

Maffeis C, Pinelli L, Zaffanello M, Schena F, Iacumin P & Schutz Y (1995): Daily energy expenditure in free living conditions in obese and non-obese children: comparison with doubly-labelled water (2H218O) method and heart-rate monitoring. Int. J. Obes. 19, 671 ± 677. McCrory MA, MoleÂ PA, Nommsen-Rivers LA & Dewey KG (1997): Between-day and within-day variability in the relation between heart rate and oxygen consumption: effect on the estimation of energy expenditure by heart-rate monitoring. Am. J. Clin. Nutr. 66, 18 ± 25. Nordic Council of Ministers (1996): Nordic nutrition recommendations 1996. Scand. J. Nutr. 40, 161 ± 165. Riddoch C & Boreham C (1995): The health-related physical activity of children. Sports Med. 19, 86 ± 102. Riddoch C, Mahoney C, Murphy N, Boreham C & Cran G (1991): The physical activity patterns of Northern Irish schoolchildren ages 11 ± 16 years. Pediatr. Exerc. Sci. 3, 300 ± 309. Sallis JF & Patrick K (1994): Physical activity guidelines for adolescents: consensus statement. Pediatr. Exerc. Sci. 6, 302 ± 314. Scho®eld WN, Scho®eld C & James WPT (1985): Predicting basal metabolic rate. New standards and review of previous work. Hum. Nutr. Clin. Nutr. 39C (Suppl), 5 ± 41. Seale JL & Conway JM (1999): Relationship between overnight energy expenditure and BMR measured in a room-sized calorimeter. Eur. J. Clin. Nutr. 53, 107 ± 111. Slaughter M, Lohman T, Boileau R, Horswill C, Stillman R, Van Loan M & Bemben D (1988): Skinfold equations for estimation of body fatness in children and youth. Hum. Biol. 60, 708 ± 723. Spurr GB & Reina JC (1988): Patterns of daily energy expenditure in normal and marginally undernourished school-aged Colombian children. Eur. J. Clin. Nutr. 42, 819 ± 834. Spurr G, Prentice A, Murgatroyd P, Goldberg GR, Reina JC, Christman NT (1988): Energy expenditure from minute-by-minute heart-rate recording: comparison with indirect calorimetry. Am. J. Clin. Nutr. 48, 552 ± 559. Torun B, Davies PSV, Livingstone MBE, Paolisso M, Sacett R & Spurr GB (1996): Energy requirements and dietary energy recommendations for children and adolescents 1 to 18 years old. Eur. J. Clin. Nutr. 50 (Suppl 1), 37 ± 81. van den Berg-Emons RJG, Saris WHM, Westerterp KR & van Baak MA (1996): Heart rate monitoring to assess energy expenditure in children with reduced physical activity. Med. Sci. Sport Exerc. 28, 496 ± 501. Verschuur R & Kemper HCG (1985): Habitual physical activity in Dutch teenagers measured by heart rate. In: Children and Exercise, ed. RA Binkhorst, HCG Kemper & WHM Saris, pp 194 ± 202. Champaign, IL: Human Kinetics.