Growth, anthropometric indices, physical fitness, and physical activity as health risk indicators in primary school children: Longitudinal trends from the NW-CHILD study

Citation

Muller X (2025) Growth, anthropometric indices, physical fitness, and physical activity as health risk indicators in primary school children: Longitudinal trends from the NW-CHILD study. Doctor of Philosophy. University of Stirling. http://hdl.handle.net/1893/38035

Abstract
Health risks in children caused by challenges such as growth deficits, physical inactivity (PIA), inadequate physical fitness (PF) and obesity can be influenced by several factors, including age, gender, and socioeconomic status (SES). Globally, gender and SES differences have been noted in PF parameters and growth trajectories. In low-middle-income countries (LMICs), those from high SES may be more susceptible to health constraints related to obesity risk, while children from low SES are predisposed to growth deficits. Additionally, different activity patterns are found to emerge among gender and SES groups, which can result in a negative health spiral. Further investigation of these influences on health risks is warranted in the South African child population to ensure timely intervention via screening, monitoring, and identification of preventative strategies. Consequently, this thesis first aimed to conduct a needs analysis for national anthropometric parameters and growth charts. In this regard, the thesis sought to compare global growth standards for height, body mass, and body mass index (BMI) with data from this sample to establish the necessity of recommending national growth charts. Secondly, the study compared physical fitness (PF), trajectories to universal standards, for determining the necessity of national fitness norms. Thirdly, the thesis intended to identify health risks related to physical activity behaviour and obesity. Data from the NW-CHILD longitudinal study were used, where 860, 6 year-old children were randomly recruited, at baseline, by stratified sampling. Participants from 20 schools, representing various school quintiles (quintile 1-3: low SES, quintile 4-5: high SES) as proxies of SES within the North West Province (NWP) of South Africa (SA) formed part of the study. The data were used to identify health risk trajectories over three follow-up measurements (2010, 2013, 2016). The thesis included 349 participants (boys: n = 165 and girls: n = 184) from both low (boys = 83; girls = 118) and high (boys = 82; girls = 66) SES groups, with complete datasets over the three timelines. Participants were 6.8 (± 0.39) years at baseline, 9.90 (± 0.38), during the follow-up and 12.90 (± 0.38) in the final testing year. For objective one, the Statistical Package for the Social Sciences (SPSS) was used to analyse descriptive data of anthropometric health indicators (AHI), including body mass, height, body mass index (BMI), body fat percentage (BFP), skinfold measurements and waist circumference (WC), considering age, gender and SES. Only body mass, height and BMI were comparable to WHO standards. Growth curve models via R software (version 4.4.2) were determined using mixed linear models for all AHI over the seven-year follow-up primary school timespan. Additionally, the LMS method (Lambda, Mu, Sigma) were used to construct percentile cut-points for comparison purposes. For the evaluation of objective two, PF parameters were measured via the Bruininks-Oseretsky Test of Motor Proficiency (2nd Ed) for muscular fitness components (sit-ups, standing long jump (SLJ), push-ups, wall sit and V-ups), while cardiorespiratory fitness (CRF) was assessed by the 20-meter shuttle run (PACER) test. SPSS were used to construct mixed linear regression models with repeated measures to investigate associations between age, gender, and SES as covariates of PF tests. Objective three employed a cross-sectional study design and only included participants (N = 587) in the final year of testing (2016), with a mean age of 12.92 (± 0.42) years. PA behaviour was measured via the Sedentary behaviour questionnaire (SBQ) and the Physical Activity Questionnaire for Older Children (PAQ-C). T-tests and chi-square analyses, via SPSS (version 27), were used to determine significance between gender and SES groups regarding PA behaviour and body composition categories. Spearman’s correlations were used to analyse associations between these variables. Additionally, generalised linear modelling was used to assess associations between PA levels, sedentary behaviour (SB) and overweight and obesity (BMI and BFP) as the main model and were adjusted for gender and SES to identify additional risk (OR). Results pertaining to objective one indicated that those from low SES perform relatively on par or slightly below the WHO reference on the 50th percentile, with children from high SES exceeding the standards. Significant associations with gender, age and SES for various AHI, showed that children from high SES are heavier and taller than those from low SES. Thus, further research is necessary to understand this trend on a national scale. Additionally, the results obtained for objective two reveal that global fitness standards were also viable for use in South African children. However, SES and gender differences again emerged in our sample, with low SES and girls being classified as at-risk for health challenges. The results of research objective three also confirmed that South African children are highly sedentary, especially over weekends, highlighting contributing factors such as screen time (ST) and passive commuting. Findings from the thesis revealed that the WHO references at the 50th percentile are generally suitable for assessing growth in typically developing South African children. However, some unique trends were observed across gender, age, and SES groups, which should be taken into consideration for growth monitoring and assessment. Co-founding factors emerged for PF levels, placing girls, children from low SES and children older than 9 years at risk for poor health. A non-significant trend toward lower odds of obesity was observed for moderate-intensity PA (MPA). Additionally, SB increases obesity risk. Subsequently, MPA interventions, along with a focus on reducing sedentary behaviour (SB), may be viable preventative and intervention strategies, especially for at-risk groups such as girls and children from high SES. The findings of this thesis will cultivate the possible development of national growth standards and fitness percentiles for South African children and adolescents. Additionally, this thesis emphasises the importance of reducing SB in this population. These findings can support the Department of Basic Education (DBE), Kinderkineticists, and all other healthcare providers in enhancing the health and development of children.

Keywords
Physical activity; Physical fitness; Sedentary behaviour; Health risks; Anthropometric health indicators; Primary school children