Research Article: Tracking Physical Activity in Different Settings from Late Childhood to Early Adulthood in Germany: The MoMo Longitudinal Study

Regular physical activity is crucial for overall health.
Previous studies on the link between active child- and adulthood were based on limited, non-representative populations.
Research aims to quantify tracking of leisure-time physical activity (PA) in and outside sports clubs over 6 years, from adolescence into young adulthood in Germany.

This study utilizes a subsample from the “Motorik-Modul (MoMo) Longitudinal Study” with a baseline period from 2003–2006 and wave 1 from 2009–2012.
The sample involved N = 947 adolescents and data was collected via the MoMo-physical activity questionnaire (MoMo-PAQ).
Different physical activity indices' stability measured using Spearman’s rank-order correlations and ANOVA with repeated measurement, focusing on age, sex, and socio-economic status (SES) as determinants.

Significant changes were observed in mean leisure-time physical activity outside sports clubs (LTPA), sports club physical activity (SCPA), and overall physical activity (OPA) over time.
No significant change was noted in the overall sports index (OS index).
Low tracking correlations were found for different physical activity indices:
- LTPA: $r = .094$,
- SCPA: $r = .248$,
- OPA: $r = .211$,
- OS index: $r = .266$.
Results varied by sex, age, and SES.

Overall stability of physical activity levels was low but significant, suggesting that physical activity alters over time.
Recommendations for future studies include examining determinants of tracking and variability in physical activity.

The health benefits of physical activity are well established, requiring continuous and regular engagement.
A physically active lifestyle in youth likely carries over into adulthood, making interventions during childhood vital for encouraging long-term activity.
An active upbringing also helps in disease prevention in later life.
Studies have shown that poor fitness and inactivity during childhood can increase risks for health issues like obesity.
Understanding the continuity (or 'tracking') of physical activity from youth into adulthood is critical.

Tracking is defined by Twisk et al. as the stability of a variable over time or the predictability of early measurements for later outcomes. Normalised stability is high when individuals maintain their relative positions within the sample distribution.

Review by Telama et al. indicated that tracking for physical activity typically shows low correlations (e.g., males: $r = .15$ to $r = .44$; females: $r = .09$ to $r = .34$).
Studies conducted across various countries indicated a trend towards low tracking correlations, particularly regarding longitudinal tracking from youth to young adulthood.
Disparities exist in overall tracking due to inconsistent methodologies regarding measurement intervals, participant demographics, and the nature of physical activity (subjective vs. objective measurements).

The KiGGS study is a national longitudinal study of children and adolescents conducted by the Robert Koch Institute, designed to provide representative health status data in Germany.
The MoMo longitudinal study is a module aimed at analysing the interplay between physical fitness, activity, mental, and physical health.

Step 1: Identify study sample points (Total: N = 167).
Step 2: Draw an age-stratified random sample from local registries among 28,400 participants (Response rate: 62.1%).
Step 3: Create a representative subsample for the MoMo study involving N = 7,866 youths; 4,529 of whom participated (Response rate: 57.6%).

Data was collected between 2003 and 2006 (baseline) and again from 2009 to 2012 (wave 1).
Only longitudinal data for participants aged 11 to 17 at baseline were included in the analysis (N = 947; 447 boys, 500 girls).

Utilized the MoMo-PAQ, which assesses physical activity types, frequency, duration, and settings.
Included items for both sports club activity and leisure-time activity, creating indices reflecting active minutes per week.
Club Sports Activity Index is calculated as:
$\text{Club sports index} = \text{duration} \times \text{frequency} \times \text{number of months} / 12$
Leisure-Time Physical Activity Index calculated as:
$\text{Leisure-Time Physical Activity Index} = \text{duration} \times \text{number of months} / 12$

The overall sports index (OS index) combines SCPA and LTPA, while overall physical activity (OPA) measured by a two-item questionnaire, aiming for at least 60 minutes of moderate to vigorous activity daily for ages 5-17.

Included sex, age, and socioeconomic status (SES) as key determinants in this analysis.
Age was categorized into two groups at each measurement point:
- t0: 11-13 years (young) and 14-17 years (old);
- t1: 17-19 years (young) and 20-23 years (old).
SES assessed via parental questionnaires classified into low, middle, and high groups.

Statistical analyses conducted using IBM SPSS 21 with:
- Descriptive analyses for study samples.
- Spearman’s rank-correlations for tracking analysis.
- Repeated measures ANOVA to examine mean stability.
- Adjustments for categorical variables with Kappa values to measure agreement between time points.

Addressing unit-nonresponse through weighting and creating logistic regression models to predict participation probability.
Treated item-nonresponse through regression imputations.

Leisure-time physical activity (LTPA) showed a significant change, predominantly in the younger group.
Sports club physical activity (SCPA) increased significantly across measures, with notable age impacts, predominantly younger participants showing increases.
Overall physical activity (OPA) revealed changes influenced by age and SES.
Overall sports index (OS index) appeared stable, undergoing no significant change.

LTPA showed weak correlations, with an overall value of $r = .094$, while SCPA was rated stronger at $r = .248$.
OPA reflected weak to moderate stability at $r = .211$.
Disparities by gender and SES were observed, especially favoring higher SES for tracking.
Statistical significance for tracking between indices and groups suggested wavering reliability across determinants.
- For example, LTPA’s correlation was significant in girls but non-significant in boys.

An observable trend revealed more transitions from active to inactive status, regardless of setting (LTPA: 54% active to inactive, etc.).
Agreement analysis illustrated varying levels of consistency from t0 to t1 measurements, generally low or moderate correlations.

Results indicate low overall stability for physical activity across the 6-year interval, emphasizing the need for attractive physical activity strategies aimed at engaging young individuals.
The study highlights varying impacts by age, sex, and socioeconomic status, and emphasizes the necessity to tailor interventions by these determinants.
The study's conclusions corroborate with prior research and continue advocating for more robust objective measurement strategies to enhance stability findings.

Strengths include a large, representative sample and consideration of various settings.
Limitations arise from reliance on self-reported data, which may lead to inaccuracies in physical activity assessments, suggesting future evaluations should incorporate objective tracking.