Adolescents' physical activity trends over the years: a three-cohort study based on the Health Behaviour in School-aged Children (HBSC) Portuguese survey.

OBJECTIVE: Many young people do not practise enough physical activity (PA) to benefit their health. The three-cohort study aimed to investigate the prevalence of PA and understand the trends between 2002 and 2010, using a representative sample of Portuguese adolescents. DESIGN, SETTING AND PARTICIPANTS: The participants were 8483 adolescents (4067 boys, 4416 girls) who participated in the Health Behaviour in School-aged Children Portuguese survey in 2002, 2006 and 2010. DESIGN, SETTING AND PARTICIPANTS: A questionnaire was used to collect data of PA. Analyses were run separately for boys and girls.
RESULTS: Boys aged 15-17 reported practising 3.5±1.9 times a week in 2002, 3.9±1.9 times a week in 2006 and 3.8±1.9 times a week in 2010, with a significant increase from 2002 to 2006/2010 (p<0.05). PA practice during the past 7 days decreased from 4.2±2.1 to 3.9+1.9 between 2002 and 2006 (p<0.001) and also decreased from 4.4±2.1 to 3.8±1.9 between 2006 and 2010 (p<0.001). Among girls, PA practices during the past 7 days declined significantly from 3.5±1.9 to 2.7±1.7 between 2002 and 2006 (p<0.001), and from 3.3±1.9 to 2.9±1.7 between 2006 and 2010 (p<0.001).
CONCLUSIONS: The study points to age as a factor related to diminished PA participation. The study also revealed that the prevalence of PA has decreased over the years for the same age groups. Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://group.bmj.com/group/rights-licensing/permissions.

Data from a representative sample were collected from three different cohorts (2002, 2006 and 2010), using a standard protocol, which allows for comparisons and for an understanding of the differences of physical activity (PA) over the years in two different age groups of boys and girls.

Measurements of PA were based on self-report. Self-report generally has limited reliability and validity. So the use of accurate objective measures is preferable, but it is not often feasible in epidemiological studies with large samples.

Boys and girls do not practise enough PA to meet the recommended levels of 60 min in moderate to vigorous intensity on all or most days of the week. From a public health perspective, there is a need to augment the time spent on PA.

Soccer is the most practised sports activity among boys and older girls. Thus, when considering an intervention aimed at increasing PA, soccer should be considered as an activity that attracts adolescents.

Introduction

Young people's participation in physical activity (PA) is associated with health benefits, specifically improved bone mineral density, aerobic fitness, muscular strength, endurance, cardiovascular risk factors and mental health.1 There is also a negative relationship between PA and careless nutrition, overweight and obesity among young people.2 3 Additionally, PA incorporated as a personal habit at a young age may persist into adulthood.4 Thus, current guidelines from the WHO recommend that young people should engage in at least 60 min of PA in moderate to vigorous intensity on all or most days of the week.5

In spite of the health benefits, data suggest that many young people are not engaging in PA of sufficient length and intensity to benefit their health.6 7 The Portuguese figures are quite dramatic, revealing that only 36% of children aged 10–11 and 4% of adolescents aged 16–17 were considered sufficiently active.8 Besides the lower levels of PA, the prevalence of active children and adolescents tended to decrease with age, particularly among girls.

In order to intervene and reverse this situation, it is important to understand the trends of PA over the years as well as the PA practices of adolescents. Nonetheless, since most adolescents do not commute actively to school, do not spend much time on PA and spend excessive time in sedentary behaviours,9 it is important to highlight sport participation. Studies have shown that sport participation contributes to adolescent attained PA guidelines,10 and there is evidence that adolescents who participated in sports are more persistent exercisers than non-participants.11 Most studies related to PA and sports participation are cross-sectional8 9 12 and results can only be interpreted in the light of the period in which they were collected. Since little is known about the PA and sport participation of adolescents and how they have evolved over the years, studies that allow researchers to compare patterns and practices of PA and sport participation are needed. Therefore, the present three-cohort study, using a representative sample of Portuguese adolescents, aimed to investigate the prevalence of PA and sport participation among Portuguese adolescents and understand how it has changed between 2002 and 2010.

Methods

Participants

A proportionate nationally representative sample of 8483 adolescents (4067 boys, 4416 girls), randomly selected from grades 6 and 10, participated in the Health Behaviour in School-aged Children (HBSC) Portuguese survey in 2002 (n=3023), 2006 (n=2556) and 2010 (n=2904).13–15 HBSC Portugal is one of 43 countries and regions across Europe and North America that make up the HBSC Network. HBSC collects data every 4 years on children's and adolescents’ health and well-being, social environments and health behaviours. Data are used at a national and international level to add new vision into young people's health and well-being, understand the social and psychological determinants of health, and incorporate policies to improve young people's lives. The methodological aspects of the HBSC study are well developed and published internationally.16 The study received approval from the Ethical Committee of Porto Medical School and the Portuguese National Data Protection System. All school administrators gave their consent, legal guardians gave written informed consent, and students provided assent.

Measures

For data collection, a questionnaire was used that was designed by the international HBSC members and is described in detail elsewhere.16 17 To assess PA, adolescents were provided with a common definition of PA,18 and then were asked ‘How many days did you practice PA during the past 7 days?’ Answers were given on an 8-point scale (1=none to 8=daily). Seven days of PA recall is shown to be reliable and valid for use in research on PA in adolescents.19 Adolescents were also asked if they practised regular PA during the past 6 months. Answers were yes (≥3 times a weeks) or no (<3 times a week). The item has a high reliability, with intraclass correlation coefficients (1 week interval) of 0.95. Adolescents were then asked about the sports activity they usually practised in organised or non-organised contexts.

Procedures

Questionnaires were administered in schools in January 2002, 2006 and 2010, and were answered anonymously. Participation in the study was voluntary. The administration of the surveys was conducted according to standard guidelines from the HBSC survey protocol,16 and it was carried out by trained schoolteachers during class time.

Analysis

Adolescents were grouped into two age groups (11–13 and 15–17). Descriptive statistics were calculated for all variables (means, SD and percentages) for the entire sample and according to HBSC survey years (2002, 2006 and 2010), age group and sex. Analysis of variance and χ2 test were used to assess the differences among HBSC survey years in each age group. Independent t tests and χ2 were used to compare adolescents from different cohorts. All statistical analyses were performed using IBM SPSS Statistics V.20.0. The level of significance was set at 0.05.

Results

Sample characteristics are displayed in table 1.

Table 1

Characteristics of participants

					Cohorts
	Total		2002		2006		2010
	n	Per cent	n	Per cent	n	Per cent	n	Per cent
Sex
Boys	4067	47.9	1466	48.5	1261	49.3	1340	46.1
Girls	4416	52.1	1557	51.5	1298	50.7	1561	53.9
Age
11–13	4266	50.3	1746	57.8	1275	49.8	1245	42.9
15–17	4217	49.7	1277	42.2	1284	50.2	1656	57.1
School grade
Sixth grade	4348	51.3	1809	59.8	1275	49.8	1264	43.5
Tenth grade	4135	48.7	1214	40.2	1284	50.2	1637	56.5

Descriptive statistics and the differences among HBSC survey years for each age group and sex are presented in table 2.

Table 2

Boys’ and girls’ physical activity participation in 2002, 2006 and 2010, according to age group

Boys	11–13 years				15–17 years
	2002(n=865)	2006(n=686)	2010(n=625)	p Value	2002(n=601)	2006(n=575)	2010(n=715)	p Value
PA in the past 7 days (M±SD)*	4.2±2.1	4.4±2.1	4.3±2.1	0.243	3.5±1.9	3.9±1.9	3.8±1.9	0.019†
PA in the past 6 months (%)‡				0.012				0.001
Did not practise PA	5.4	5.5	9.0		11.1	12.9	18.0
Practised PA	94.6	94.5	91.0		88.9	87.1	82.0
Sports practised (%)‡
Soccer	72.6	69.8	60.8	<0.001	57.1	50.1	44.3	<0.001
Basketball	22.1	26.4	22.4	0.102	16.6	12.7	13.7	0.130
Volleyball	8.6	12.4	10.6	0.089	9.2	7.8	8.1	0.683
Handball	9.7	11.8	11.2	0.388	8.3	7.1	7.0	0.619
Swimming	18.0	15.3	17.3	0.352	15.8	13.6	15.5	0.500
Athletics	12.7	13.8	13.1	0.806	8.2	7.7	9.2	0.575
Gymnastics	17.5	17.1	22.2	0.027	7.0	3.8	6.7	0.038
Cycling	22.7	24.2	21.4	0.490	18.5	16.7	14.7	0.181
Type of sport activity‡				0.008				0.524
Individual	8.7	8.3	13.0		13.3	13.6	13.4
Team	40.9	38.2	33.8		41.3	38.4	33.0
Individual and collective	38.5	37.8	34.9		24.6	19.7	19.7

*Tested by analysis of variance.

†Data from 2002 are statically different from data for 2006 and 2010. Tested by Tukey's honest significant difference.

‡Tested by χ2 test.

PA, physical activity.

For boys aged 11–13, the number of times they reported practising PA during the past 7 days was between 4.2±2.1 to 4.4±2.1 times. At the age of 15–17, they reported practising 3.5±1.9 times a week in 2002, 3.9±1.9 times a week in 2006 and 3.8±1.9 times a week in 2010, with a significant increase from 2002 to 2006/2010 (p<0.05). In both age groups, the prevalence of PA over the past 6 months decreased slightly between 2006 and 2010, but decreased significantly between 2006 and 2010 (p<0.05 and p<0.01). Soccer was the most practised sports activity, but over the years there was a significant decline in the number of participants (p<0.001 for both age-groups). On the other hand, gymnastics participation decreased from 2002 to 2006 and then increased from 2006 to 2010, reaching the higher value in 2010 for the ages of 11–13 (p<0.05) and 15–17 years (p<0.05). The type of PA practised changed over the years for the younger participants, showing a decrease in the number of team sports participants and an increase in those participating in individual sports (p<0.01). Boys, younger and older, preferred team sports activities more than individual activities. However, over the years, the number of young boys participating in team sports decreased progressively from 2002 to 2010 (p<0.01).

Girls reported practising PA between 3.3±1.9 to 3.5±1.9 times a week at the age of 11–13, and 2.7±1.9 to 2.9±1.7 times a week at the age of 15–17. Soccer, gymnastics and swimming were the most popular sports activities. Over the years, the number of soccer practitioners decreased significantly among younger girls (p<0.001). For the older girls, the number of participants decreased significantly from 2002 to 2006, and then increased significantly from 2006 to 2010 (p<0.001). In respect of gymnastics, the number of participants decreased significantly from 2002 to 2006/2010 for both younger (p<0.01) and older girls (p<0.01). The number of older girls that practised volleyball was inconsistent, decreasing significantly from 2002 to 2006 and increasing significantly from 2006 to 2010 (p<0.05). Girls aged 11–13 preferred individual sports activities rather than team sports activities over the years (p<0.05). On the other hand, among older girls, the number of participants in individual sports activities decreased significantly from 2006 to 2010 (p<0.01).

Table 3 presents PA information from 2 different cohorts, 4 years apart. Among boys, PA practices during the past 7 days decreased from 4.2±2.1 to 3.9±.1.9 between 2002 and 2006 (p<0.001), and again decreased from 4.4±2.1 to 3.8±1.9 between 2006 and 2010 (p<0.001). The prevalence of PA practised over the past 6 months also declined significantly in both cohorts (p<0.001). Looking at the different sports activities, the number of practitioners decreased over the years in almost all activities. It is noteworthy that only volleyball from 2002 to 2006 and swimming from 2006 to 2010 remained relatively stable. In both cohorts, as age increased, the number of participants in team sports activities declined, and, in contrast, the number of participants in individual activities increased significantly (p<0.001).

Table 3

Boys’ and girls’ physical activity participation over the years

Boys	200211–13 years(n=865)	200615–17 years(n=575)	p Value	200611–13 years(n=686)	201015–17 years(n=715)	p Value
PA in the past 7 days (M±SD*	4.2±2.1	3.9±1.9	0.001	4.4±2.1	3.8±1.9	<0.001
PA in the past 6 months (%)†			<0.001			<0.001
Did not practise PA	5.4	12.6		5.5	18.0
Practised PA	94.6	87.4		94.5	82.0
Sports practised (%)†
Soccer	72.6	51.1	<0.001	69.8	45.1	<0.001
Basketball	22.1	13.9	<0.001	26.4	14.0	<0.001
Volleyball	8.6	8.6	0.082	12.4	9.1	0.010
Handball	9.7	6.9	0.104	11.8	7.2	0.002
Swimming	18.0	13.6	0.029	15.3	15.2	0.941
Athletics	12.7	7.8	0.002	13.8	9.3	0.007
Gymnastics	17.5	4.7	<0.001	17.1	7.1	<0.001
Cycling	22.7	17.0	0.006	24.2	14.8	<0.001
Type of sport activity†			<0.001			<0.001
Individual	8.7	13.6		8.3	13.4
Team	40.9	38.4		38.2	33.0
Individual and collective	38.5	19.7		37.8	19.7

*Tested by student t test.

†Tested by χ2 test

PA, physical activity.

Among girls, PA practices during the past 7 days declined significantly from 3.5±1.9 to 2.7±1.7 between 2002 and 2006 (p<0.001), and from 3.3±1.9 to 2.9±1.7 between 2006 and 2010 (p<0.001). Similar to what was observed among boys, the prevalence of PA over the past 6 months decreased in both cohorts (p<0.001). The practice of PA decreased significantly in all sports activities from 2002 to 2006. Data from 2006 and 2010 show that the practice of PA also declined, with the exception of handball, which remained stable, and volleyball, which in contrast to other sports had a marginally significant increase (p=0.052). Girls team and individual sports participation decreased between 2002 and 2006 (p<0.001) and between 2006 and 2010 (p<0.01).

Discussion

This is the first study among Portuguese adolescents to explore the PA and sport participation of different cohorts, collecting data through the same standard protocol. The goal of the three-cohort study was, using a representative sample of Portuguese adolescents, to investigate the prevalence of PA and sport participation among Portuguese adolescents, and understand how it changed between 2002 and 2010. Overall, the study points to age as a factor related to diminished PA participation. The study also revealed that the prevalence of PA and sport participation has decreased over the years for the same age groups.

Younger and older adolescents reported practising PA less than seven times a week. Although the duration and intensity of PA were not assessed, results indicated that adolescents do not get enough practice to meet the recommended levels of 60 min of moderate to vigorous PA each day.5 This is in line with other recent studies among Portuguese people that described the prevalence of PA.8 12 Since our data are from different HBSC waves, it was possible to observe the evolution of PA in each age group over the 2002–2010 time frames. With the exception of boys aged 15–17, there were no significant changes over the years in regard to PA practised during the past 7 days. Moreover, there was a decrease in the prevalence of PA over the past 6 months, especially among boys. These results can be interpreted in several different ways. First, strategies to increase the number of times that adolescents practise PA have not been successful. Second, adolescents and adults may not be aware of the recommended levels of PA for young people. Consequently, they might assume that they are active enough and therefore may conclude that there is no need to be more physically active.20 Third, schools, even though called to give greater attention to provision of PA for students,21 seem to have failed at promoting PA. Results indicate that there has been no progress made towards attaining this goal.

Adolescent behaviour, such as PA and inactivity over the years, is an issue rarely studied in Portugal. This type of study is important and should be of interest at a national and international level because it helps policymakers, educators and health professionals to understand young people's health in their social context and tracks changes as they develop over the years. Therefore, with data from successive surveys, trends can be identified that make it possible to evaluate more accurately the impact of programmes and strategies implemented in order to promote PA and reduce sedentary behaviour among adolescents. While the HBSC conducts national13–15 and international systematic surveys,6 16 in the USA , the Center for Disease Control and Prevention also conducts a periodic survey of adolescent health risk and health protective behaviours.22 This shows the importance of studying and tracking adolescent behaviour and allows cross-national comparisons.

The trend of PA declining with age was significant among boys and girls in the two different cohorts of adolescents (2002 to 2006 and 2006 to 2010). These results are similar to those found in other studies of adolescents from Portugal23 and other countries.6 7 In Europe, the inverse relationship between PA and age is common in most countries.6 In what is considered the first study to objectively measure physical activity with a representative sample of European children, Riddoch et al24 verified that 9-year-old children were considerably more active than 15-year-old adolescents, presenting similar results in all countries. Virtually all 9-year-old children practised the recommended amount of PA, or more than recommended, while only a small percentage of 15-year-olds did so. This indicated that as age increased, PA decreased.

The decline of PA with age has triggered much discussion in the scientific community. Sallis25 analysed various studies related to the decline of PA, and concluded that the ages between 13 and 18 are at risk. Having also observed the decline in animals, the author was led to suppose that the decline had a biological origin. The consistency of the results of Riddoch et al24 among adolescents from different countries is in some way congruent with Sallis’s25 assumptions, allowing one to conclude that children's and adolescents’ PA is probably determined biologically. However, although the effect of age is significant on PA, it sometimes only explained 1% of the variability,26 which means that the decline in PA with age could be related to other factors such as the great academic pressure adolescents are under,27 longer time in sedentary activities such as screen time,28 and having part-time jobs to earn some money.29 Therefore, qualitative studies to understand clearly the reasons for the decrease in PA among adolescents are needed.

Soccer was the most practised sports activity among boys and older girls over the years and at different ages. This is understandable since it is the most popular sports activity in Portugal, and it is the activity with the most participants.30 31 Nevertheless, the number of participants decreased over the years within both age groups. As the number of participants in others sports activities did not increase significantly, this suggests that the decrease in soccer participants might be related to the prevalence of adolescents who did not practise PA over the past 6 months. This shows the importance of soccer as a sports activity among Portuguese adolescents. Thus, when considering an intervention aimed at increasing PA, soccer should be considered as an activity that attracts adolescents.

The type of activity practised by boys has the tendency to change over the years. While the number of team sports participants decreased between 11–13 and 15–17 years, participation in individual sports increased. On the other hand, for girls, participation in individual and team sports decreased with age. The reason for the choice of individual or team sports may be related to social factors, and some could argue that team sports would be better for the development of social skills. However, studies have shown that there is no difference in social self-esteem between participants in individual or team sports.32 33 For the older boys, the increased participation in individual sports may be related to their level of autonomy.

The present study has some limitations. Measurements of PA were based on self-report. Self-report generally has limited reliability and validity.34 So the use of accurate objective measures is preferable, but it is not often feasible in epidemiological studies with large samples. Moreover, the intensity and duration of PA were not measured. These evaluations would be useful in determining the prevalence of adolescents who attained the recommended levels of PA.

There are also several strengths in the present study. First, data from a representative sample were collected. Second, data from three different cohorts were collected (2002, 2006 and 2010), using a standard protocol, which allows for comparisons and for an understanding of the differences of PA over the years in two different age groups of boys and girls.

Conclusion

The prevalence of physical inactivity increased significantly among boys between ages 11–13 and 15–17, but not among girls. The number of times adolescents practised PA during the past 7 days remained relatively stable, which means that the campaigns and attempts to increase PA among adolescents have not been very successful. With age, there was a significant decrease in most sports activities among boys and girls. This indicated that as adolescents are getting older, they may have different interests leading some to stop practising PA. These findings imply that it is important to more effectively promote PA among adolescents, using different strategies based on age and sex.