THE ASSOCIATION BETWEEN PARTICIPATION IN ORGANISED PHYSICAL ACTIVITY AND LEVEL OF PHYSICAL ACTIVITY AND INACTIVITY IN ADOLESCENT GIRLS

BACKGROUND: An increase of leisure time and vigorous physical activity (PA) in girls can be supported by their regular participation in organised PA, as it brings numerous health benefi ts and is an important part of healthy lifestyle. OBJECTIVE: This study aimed to examine whether there are diff erences in physical activity and inactivity structure and levels in girls who do not participate in any organised PA and girls who regularly participate in physical education or training classes. METHODS: The sample comprised 497 girls (age 17.94 ± 0.52 years) from Poland participating never (group 1), or once (group 2), twice (group 3), and three or four times (group 4) a week in any organised PA. The Polish version of the International Physical Activity Questionnaire (IPAQ), accelerometer ActiGraph GT1M and activity and inactivity logs were used to assess the level and structure of physical activity and inactivity. RESULTS: Girls participating in organised PA three or more times a week showed signifi cantly the highest values of steps/day, moderate and vigorous PA, and leisure time PA, but also inactivity in comparison to girls without participation in organised PA. CONCLUSIONS: This study provides the evidence supporting the positive link between a number/week participation in organised PA and PA level. However, the fi ndings from this study also advocate the need of increasing PA and reducing inactivity in girls in school or after-school periods, regardless of the participation in organised PA.

An increase of leisure time and vigorous PA in girls can be supported by their regular participation in orga nised PA, as it brings numerous health benefi ts (Pangrazi, 2000) and is an important part of healthy lifestyle; and school has an essential role in it (Park & Kim, 2008;Trost & Loprinzi, 2008).The possibility to participate in regular organised PA is within either phy sical education (PE) classes during school hours or within after-school programmes.Due to the regular participation in PE classes at school children and youth might increase their overall daily PA (Fairclough & Stratton, 2005a;2005b), whether it is the number of steps a day (Jago, Baranowski, Zakeri, & Harris, 2005;Morgan, Beighle, & Pangrazi, 2007;Mota et al., 2008;Peiró-Velert, Devís-De vís, Beltrán-Carrillo, & Fox, 2008;Tudor-Locke et al., 2006) or moderate to vi gorous PA (Gordon-Larsen, McMurray, & Popkin, 2000; Morgan

BACKGROUND
The health benefits of regular physical activity (PA) in youth include the prevention of many risk factors (obesity, high levels of body fat and cholesterol, hyper tension) that could set up themselves throughout childhood and project into adulthood (Steinberger & Daniels, 2003;Sung et al., 2006).However, PA drops exponentially during the adolescent period (Allison, Adlaf, Dwyer, Lysy, & Irving, 2007) and many adolescents place their health at risk due to insuffi cient PA (Lowry, Galuska, Fulton, Wechsler, & Kann, 2002).Most studies show that girls are less physically active than boys (Gavarry, Giacomoni, Bernard, Seymat, & Falgairette, 2003;Norman et al., 2006;Pearson, Atkin, Biddle, Gorely, & Edwardson, 2009;Sanchez et al., 2007) and at the age of 9-12 show less moderate to vi gorous physical activity (MVPA) than boys (Kahn et al., 2008).Furthermore, concerning the daily structure of PA in young adolescents (Loucaides & Jago, 2008), it is apparent that girls show signifi cantly lower total daily PA than boys before school, at school, in a 20 min.recess and after school.Majority of adolescent females do not meet health recommendations for physical activity and generally are a population group at risk as far as et al., 2007;Mota et al., 2008;Myers, Strik miller, Webber, & Berenson, 1996).Moreover, PA programmes wit hin after-school period have greater potential for improving physical activity levels and other health-related aspects and are emerging as potentially useful and feasible for PA promotion (Beets, Beighle, Erwin, & Huberty, 2009).
Nevertheless, studies have examined the eff ect of the participation in organised PA using either subjective or objective methods of assessment.E.g.Flohr et al. (2006) evaluated the pedometer-assessed PA patterns in children and diff erences between participants in afterschool organised PA and non-participants; using a questionnaire, Santos et al. (2009) assessed the relationship between participation in organized and non-organized PA, commuting to school and BMI; and Fairclough and Stratton (2005b) used heart rate telemetry to assess PA levels during high school PE classes.The authors are not aware of any research that investigated the frequency of the participation in organised PA using the combination of subjective and objective methods e.g.questionnaire, accelerometers and daily PA and physical inactivity selfrecord logs, thus allowing the examination of its eff ects in a more comprehensive way.
Therefore, the aim of the study was to examine, using objective and subjective methods, whether there are differences in PA and physical inactivity (PI) structure and levels in girls who do not participate in any organised PA and girls who regularly participate in PE or train ing classes once, twice or three or more times a week.

Participants
Monitoring of weekly PA was carried out at 15 randomly selected secondary schools with approximately equal material background (sports equipment) for PE and extracurricular PA and in 2 randomly selected classes at each school in Katowice region in Poland in spring (April-May) and in fall (September-October) seasons having the same mean temperatures of ≈ 10 °C (according to measurements taken four times within 24 hour periods), avoiding warm summers and cold winters.A consent from both parents and the girls was required prior the onset of the monitoring.Out of the pupils approached, 29 girls refused to participate.Moreover, we excluded data of 103 participants due to technical causes (e.g. when the accelerometer had been left lying for some time within one week monitoring); and 15 participants due to incorrect completing of the questionnaire.In total, our sample comprised 497 girls (age 17.94 ± 0.52 years) participating never (group 1), once (group 2), twice (group 3), or three or four times (group 4) a week in any organised PA.Organised PA included PE classes at school and any other lessons of exercise and training under the supervision of a teacher or coach.In Poland, there are recommended three classes of PE a week in the curriculum.The groups' cha racteristics are showed in TABLE 1.

Measures and procedure
In order to obtain the most accurate and complex information about the eff ect which the frequency of weekly participation might have on physical activity, the combination of subjective and objective methods was used: To measure the subjective assessment of weekly PA, the Polish version of the International Physical Activity Questionnaire (IPAQ) -long was used.The questionnaire was applied one day prior to weekly monitoring by accelerometer which allows us to eliminate the possible reactivity of accelerometer.The girls completed the questionnaires upon instructions provided by a qualifi ed researcher.The validity and reliability of the IPAQ questionnaire (www.ipaq.ki.se) are proved by previous studies (Cerin, Saelens, Sallis, & Frank, 2006;Craig et al., 2003).Forward and back-translation were applied in the standardization of the Polish version of the questionnaire in compliance with the "Guide to Cultural Adaptation and Translation of the IPAQ Instruments" (http:// www.ipenproject.org/surveytrans.htm).The results of the questionnaire were processing in agreement with the "Guidelines for data processing and analysis of the international physical activity questionnaire" (www.ipaq.ki.se).
Weekly PA monitoring was carried out also objectively using ActiGraph GT1M accelerometers (Manu facturing Technology Inc., FL, USA).According to Welk, Schaben and Morrow study (2004), the ActiGraph accelerometer appeared to have acceptable reliability for most research applications (gene ralizability values above 0.60 and ICC values above 0.80).Girls were instructed on how to affi x the accelerometer by means of an elastic belt clipped at the mid-axillary line of the right hip.Moreover, they were asked to wear the accelerometer for at least 10 waking hours for consecutive seven days, with the exception of times used to perform water activities.Using manufacture supplied software, the time sampling interval of the accelerometers was set at 1 minute, and step mode was activated.
After the seven day period, the recorded data containing activity counts and steps were downloaded with the assistance of the manufacturer's software.Downloaded counts data were analyzed according to scoring for younger adults established by Freedson, Melanson and Sirard (1998).In order to identify the boundary between light (< 3 METs) and mo derate activity (3-6 METs), and moderate and vigorous activity (> 6 METs) the cut-off points of 1951 and 5724 counts/minute, respectively, were used.Daily PA and PI logs were used for subjects to selfrecord times when accelerometer was worn and activity and inactivity patterns so that we could obtain a more accurate picture of the individual's physical activity profi le and were able to make some judgment while assessing the data (Esliger, Copeland, Barnes, & Tremblay, 2005).Specifi cally, the girls re corded the times when they put on the device, ar rived at school, started and fi nished PE class, departed from school, the beginning and end of any other organised PA, and taking off the device in the evening.Further, they were instructed to record the types and time of physical activities performed in at least 10 minutes bouts, e.g.walking, running, cycling, swimming, dancing, playing volleyball, basketball, etc.Moreover, types and time of inactivities were recorded (sitting-lying-standing) while watching TV, working with a computer, studying, etc. at school, attending cultural and sports events, and pas sive transport.Girls were instructed to record the data every evening after taking off the device.
To assess the activity log, physical activities were assigned counts per minute equal to the MET value published in the Compendium of Physical Activities (Ainsworth et al., 2000).

Statistical analyses
The statistical packages Statistica 9 (StatSoft, 2009) and SPSS 18 (SPSS, Inc., Chicago, IL) were used for data analysis with signifi cant levels set at p < 0.05.Data were normally distributed for all variables.Descriptive statistics were presented as means (M) and standard de viation (SD) in case of step-based data and as median (Mdn) and inter-quartile ranges (IQR) in case of questionnaire and activity and inactivity logs and minutesbased accelerometer data.Kruskal-Wallis test was used to test whether there are diff erences among groups according to frequency of participation in orga nised PA, the intensity of PA, structure of PA (school, transportrelated, domestic and leisure-time domains) and PI (watching TV, working on PC, learning, sitting at school, culture events, passive transport).Mann-Whitney U-test was used to identify the diff erences between two groups.Repeated analysis of variance was used to test whether participation in organised PA is associated with steps/ day of the week.Schéff e post-hoc test was used to identify the diff erences between groups.Coeffi cients ω 2 (Tolson, 1980), Cohenʼs d (Cohen, 1988) and ŋ 2 (Sheskin, 2007) were used as the indicators of eff ect size.

Characteristics of overall physical activity and physical activity in organised lessons
TABLE 2 shows the statistically signifi cant diff erences among groups in daily physical activity (active energy expenditure, steps).There were not found any signifi cant diff erences among groups 2-4 in intensity and volume of PE and training classes.

Daily step values
The girls participating in organised PA 3-4 times a week had signifi cantly higher number of steps/day than girls who did not participate in any organised PA (p = 0.000, d = 0.57).Regardless of the participation in organised PA, the mean values for weekdays were sig nificantly higher than for weekends (F = 104.11,p = 0.000, ω 2 = 0.38) in all girls.TABLE 3 shows that girls who participated in organised PA 3-4 times a week had signifi cantly higher number of steps/day than girls who did not participate at all on both weekdays and weekend days.

Intensity of PA
The signifi cant diff erences in low (< 3 MET), moderate (3-5.9MET) and vigorous (≥ 6 MET) PA on school days among four groups are shown in TABLE 4. On we ekend days, there were found signifi cant diff erences among groups only in vigorous PA (H = 8.302, p = 0.040); whereas the diff erences between groups were not signifi cant in moderate (H = 7.49, p = 0.058) and low PA (H = 5.25, p = 0.154).
According to the questionnaire data, TABLE 5 shows signifi cant diff erences in intensity of PA among groups in vigorous PA.Structure of weekly PA TABLE 6 demonstrates the diff erences in MET-minutes a day betweeen groups within school, transport-related, domestic and leisure-time domains.The girls with the highest participation in organised PA had the lowest value of MET-minute/day in domestic domain, but the highest value of MET-minute/day in their leisure-time.

Structure of weekly PI
The structure of PI and total sums of PI based on inactivity logs and accelerometer data are shown in TABLE 7. Regardless of the structure of PI, signi fi cantly the highest values (min × day -1 ) of PI were found in the girls with the highest participation in organised PA.The signifi cant diff erences in types of PI were found among groups in time spent studying and sitting at school.

DISCUSSION
In this cross-sectional study, the diff erences in PA and PI patterns in girls with diff erent frequency in organised PA were investigated.The results of this study are in concordance with previous studies suggesting associations between the number of times/week the youth participate in organised PA and PA levels (Gordon-Larsen et al., 2000;Morgan et al., 2007;Mota et al., 2008;Myers et al., 1996).
The girls attending organised PA, regardless of the frequency a week, achieved more steps/day than those who did not attend the organized PA at all.The similar fi ndings were reported in the USA by Flohr et al. (2006), where seventh-grade after-school program participants achieved 24% more steps on all days as compared to their non-participating counterparts.Similarly in the case study by Pelclová, Ansari, and Vašíčková (2010), Czech adolescent girls participating in afterschool program entered in 21% more steps/day counts than girls who did not participate.Moreover, this was true regardless of the day of the week, month and season of the year.
All participants achieved more steps on weekdays as compared to weekends, with Sunday being the least active day.This "high weekdays-low weekends" PA pat terns were already examined in previous studies in adults (Tudor-Locke et al., 2004), adolescents (Peiró-Velert et al., 2008;Pelclová, Ansari, & Vašíčková, 2010) and as well as in children (Duncan et al., 2008a).This study contributes to these fi ndings by recognizing that these patterns are similar in girls regardless of their frequency of participation in organised PA.
The more organised lessons the girls attended, the more step values they had for all days of the week.How-ever, these girls did not meet the basic recommendation of 10,000 steps/day (Hatano, 1993).Nevertheless, the organised lessons might have potential for signifi cant increase of the number of daily steps.For instance, gra de 6 and 7 pupils can add an average of 1,455 and 2,046 steps respectively in one 45-50 minute PE class (Flohr et al., 2006;Tudor-Locke et al., 2006).Due to regular participation in PE classes at school, children and youth increase their overall daily PA (Fairclough & Stratton, 2005a;2005b), and PA programmes in afterschool time might have greater potential for improving physical activity levels and other health-related as pects (Beets et al., 2009).For example, 104 Czech pre schoolers (aged 5-7 years), 1,174 teenagers (aged 12-17 years) and 787 young adults (aged 18-24 years) had approximately three times lower activity energy ex penditure during school time compared to the their after-school leisure-time periods (Sigmund, Croix, Miklánková, & Frömel, 2007).Similarly, for both genders, the pedometer-measured steps of British children (7-11 year old) suggested that steps accumulated in week day leisure time were greater in the high-active groups than in the mid and low active groups, with relatively smaller diff erences between the activity for steps accumulated at school (Hardman, Horne, & Rowlands, 2009).Likewise in this study, signifi cantly higher values of energy expenditu re were in girls with once, twice or three or more times/week participation compared to girls without any participation in their lei sure-time.However, overall self-reported energy expenditure values in all groups were the highest in school time compared to leisure time, passive transport and home time.Regarding the intensity of PA, frequency of organised PA participation was statistically positively associated mainly with vigorous PA (confi rmed by both accelerometer In addition, this study aimed to investigate the frequency of organised PA attendance and PI eff ect.Unexpectedly, the girls with the most frequent participation in organised PA had the highest self-reported minutes of PI spent at school or while studying.It is possible, that these girls need to compensate this with PA for rest and relaxation.Overall the girls spent on average 71 minutes watching television, 66 minutes working or playing games on computer and 31 minutes by passive transport.These fi ndings were comparable to an American study where 15 year old girls spent 2.35 hours/day while watching TV and videos and playing games (Berkey et al., 2003).Ten and thirty minutes more than in our sample were spent by Australian girls of the same age while watching television and by passive transport respectively, and 39 minutes less while working or playing games on computer (Olds et al., 2009).
Despite the suffi cient sample size, employing both self-reported and accelerometer-based data, and investigating both PA and PI, there are some limitations to the fi ndings from this study.Since the accelerometer is not a suitable device for water activity, swimming had not been taken into consideration while processing the data.Moreover, an analysis of individual sports and activities (e.g.dance, volleyball) in relation to the participation in organised PA, analysis of PA within segmented day (e.g.PA at school and out of school), and the involvement of both genders would help to bring more a comprehensive view into this topic.

CONCLUSION
Despite the limitations discussed above, this study provides the evidence supporting the positive link between a number/week participation in organised PA and PA level.However, the fi ndings from this study also advocate the need of increasing PA and reducing PI in girls in school or after-school periods, regardless of the participation in organised PA.
The main conclusions are: The positive link between a number/week participation in organised PA and PA level was confi rmed using both objective and subjective research methods.The girls participating in organised PA 3-4 times a week had signifi cantly higher number of steps/day, amount of moderate to vigorous PA and PA perfor med at school and in their leisure time than girls who did not participate at all.The girls with the most often participation in organised PA had the highest self-reported minutes of PI spent at school or while studying.The "high weekdays-low weekends" PA patterns are similar in girls regardless of their frequency of participation in organised PA.Regardless of the frequency of participation in organised PA, there is the need of increasing PA and reducing PI in all girls in school or after-school periods.

Fig. 1
Fig. 1 Mean steps a day of the week in groups with diff erent frequency of organised PA participation

TABLE 2
Characteristics of overall physical activity and physical activity in organised lessons according to frequency of organised PA participation (N = 497)

TABLE 3
Mean steps a day on the weekdays and weekend days in groups with diff erent frequency of organised PA participation (N = 497)

TABLE 4
Intensity of PA in school days in groups with diff erent frequency of organised PA participation (accelerometer data)

TABLE 5
Intensity of PA (MET-min × day -1 ) and walk in groups with diff erent frequency of organised PA participation (questionnaire data)

TABLE 7
Structure of PI in groups with diff erent frequency of organised PA participation (inactivity log and accelerometer data)