JSSM 2003 - 2, SUPPLEMENTUM 2, Table 1

Table 1. Stability and changes of physical activity. Longitudinal studies tracking physical activity since the 1990s.
Correlations (Pearson or rank order) <30 are considered low (+/-), 0.30-0.60 moderate (+) and >60 high (++) (Malina, 1996).

Author	Country	Year	Longitudinal follow up	Material	Methods	Results	Tracking
Kelder et al.	USA	1994	7 years Minnesota heart health programme	N= 2376Baseline 6^th grade.	Self-reported physical activity.	Students at baseline reporting high physical activity remained high, and those reporting low PA remained low.	+
Telama et al.	Finland	1994	9 years Cardiovascular risks of young Finns-cohort	N=35963-18 years.	Self-reported physical activity.	Physical activity peaked at 12 years and then reduced considerably, but the intensity and strain increased at the same time. Physical activity is a weak predictor of physical activity 9 years later.	+/-
Telama et al.	Finland	1997	12 years Cardiovascular risks of young Finns-cohort	Age 9 (n=610), 12 (n=624),15 (n=572), 18 (n=503).	Predicted P.A.12 years later.Questionnaire, sum index of 5 variables.	Tracking significant, but low.9-year internal 0.18-0.4712-year internal 0.00-0.27Participation in competitive sport and PE number were the best predictors.	+/-
Yang et al.	Finland	1999	12 year sCardiovascular risks of young Finns-cohort	N= 2411Baseline 9,12,15,18 year olds.	Self-reported physical activity.	Early physical activity was the best predictor of adult physical activity with the exception of 21-year- old women.	+
Telama et al.	Finland	1996	12 yearsCardiovascular risks of young Finns-cohort	N= 3596Baseline 3,6,9,12 year old.	Self-reported physical activity.	All tracking correlations significant but low, varying 0.50-0.80 among boys and 0.40-0.61 among girls. Highest tracking correlation was frequency of participation in sports clubs.	+/-
Raitakari et al.	Finland	1994	6 years Cardiovascular risks of young Finns-cohort	N=961	Self-reported physical activity.Sum index from frequency +intensity +duration.	Significant tracking. 3-year sum index correlation ranged 0.35-0.54 among boys and 0.33-0.39 among girls. Physical inactivity showed better tracking than activity.	+
Sallis et al.	USA	1999	20 months	N= 362 boys, 370 girlsBaseline 4^th and 5^th grade	Physical activity computed from child reports, parents’ reports and objective activity monitoring.	Significant decline in children’s physical activity during 4^th and 5^th grade. The rate of decline ranged from 3% to 6% for boys and from 7% to 12% for girls.
Pate et al.	USA	1996	3 years Youth risk survey	N=22 boys25 girlsaged 3-4	Heart rate measuring: HR over 50% rest pulse between 3-6 p.m.	Spearman rank order correlation 0.57-0.66 (p< 0.001). Physical activity behaviour tended to track during early childhood.	+
Jantz et al.	USA	2000	5 years Muscatine study	N=126Mean age 10.8 boys and 10.3 girls	Physical fitness measured by oxygen uptake and maximal isometric contraction,Physical activity measured by questionnaire.	Tracking of physical fitness and activity variables was moderate to high. Sedentary behaviour tracked better in boys, and vigorous activity in girls.	+
Anderssen	Den-mark	1996	7 years	Aged 18-30 at baseline.N=2328 menand 2787 women	Questionnaire four times in the follow-up period.	Moderate tracking.Intraclass correlation 0.57 (0.42-0.57).PA declined sharply during early adulthood.	+
Van Mechelen et al.	Nether-lands	2000	15 years Amsterdam Growth Study	98 female83 male aged 13, 14, 15, 16, 21, 27	Structured interview. Physical activity over 4METs.	Significant decline of habitual physical activity among males and females between 13 and 27 years.
Kemperet al.	Nether-lands	2001	20 years Amsterdam Growth Study	400 boys and girls	Physical activity measured by cross check interview.	Stability coefficients 0.35 –0.29.	+/-
Dovey et al.	New Zealand	1998	12 months before turning 15 and 18	775 boys and girls	Interview.	Total participation at the age of 18 was 63% of that reported at 15 years. Boys spent significantly more time in physical activity than girls.
Armstrong et al.	UK	2000	3 years	202 boys and girls	Annual measurement: HR, body mass, skinfold thickness.	PA decreased from 11 to 13 years, specially among girls.

Abbrevations: PA= physical activity, PE= physical activity.

Material