Temporal trends in physical fitness among preschoolers from Macao Special Administrative Region between 2002 and 2020

Abstract

Objective

This study examined temporal trends in physical fitness among preschool children aged 3–5 years from the Macao Special Administrative Region, China between 2002 and 2020.

Methods

Representative repeated cross-sectional physical fitness data were collected in 2002, 2005, 2010, 2015, and 2020 (n = 4514). Body size (height, weight, and chest, waist and hip circumferences) and physical fitness (2x10-m shuttle run, standing long jump, walking balance, two-leg continuous jump, overhead throw, and sit-and-reach) were objectively measured. Trends in means were calculated using general linear models. Models were adjusted for gender, age, height, and weight. Trends in distributional characteristics were calculated as the ratio of the coefficients of variation and described visually.

Results

We found significant but small increases in height, weight, and chest circumference (standardised effect size [ES] = 0.27–0.42), a significant moderate increase in hip circumference (ES = 0.59), and a negligible trend in waist circumference. Physical fitness trends were conflicting, with negligible to small declines in throwing (ES = −0.14) and balance (ES = −0.32) performance, and negligible to small improvements in other measures (ES = 0.19–0.34). We found negligible trends in distributional variability and differing trends in distributional asymmetry.

Conclusion

Overall, these findings suggest modest growth and development among Macao preschoolers over the past two decades. Our findings also highlight the importance of ongoing monitoring to support physical fitness and overall health in early childhood. Continuous screening and monitoring are crucial for identifying trends and informing future health initiatives.

1. Introduction

Physical fitness, encompassing cardiorespiratory, musculoskeletal (power, strength, endurance, and flexibility), and motor fitness (including performance- or skill-related measures like speed, agility, balance, and coordination), is a crucial indicator of overall health and physical competence throughout childhood and adolescence.^1,2 Accurate assessment requires standardised and validated tests but obtaining reliable data from preschool children is challenging due to their variable motor skills and short attention spans.^3,4 Despite these challenges, assessments such as the FITness testing in PREschool children battery have focused on preschoolers.^1,5 Early fitness assessment may prevent future health issues, as low fitness levels in childhood are associated with cardiovascular disease, obesity, and mental disorders later in life.6, 7, 8 However, there is a significant gap in research on intergenerational changes in physical fitness levels among preschoolers aged 3–5 years.^1,3 Understanding fitness levels in this critical developmental stage is essential for promoting long-term health and well-being.

Temporal trends in physical fitness are important because they can reveal changes in health patterns over time. Fitness can be measured in various ways, and among preschoolers, their motor skills or muscle fitness are typically assessed. Recent studies show a concerning decline in the motor fitness of preschool children, possibly linked to reduced physical activity levels and increased screen time.9, 10, 11 This trend could negatively impact their overall health and development, potentially leading to overweight and obesity issues.^9,12 However, a trends study of Polish preschoolers between 2008 and 2018 provided some contrasting findings, with improvements in overall handgrip strength and standing broad jump performance among girls, despite a decline in overhead medicine ball throw and sit-and-reach flexibility.³ Similarly, a study of German preschoolers showed stable trends between 2003 and 2017, with only the stand-and-reach flexibility of girls declining.¹³ While these results suggest that aspects of motor fitness have trended differently over time, there is little information on temporal trends in preschoolers’ fitness levels. Some studies have included preschoolers as part of a broader analysis of children and adolescents,^13,14 but they specifically analysed trends among preschoolers, making it difficult to discern true trends in this age group. Additionally, most research has focused on European preschool children, leaving a notable lack of studies in Asia, where social and cultural differences may significantly influence trends in physical fitness and health outcomes.¹⁵ Therefore, it is crucial to conduct more research to fill this knowledge gap and to gain insights into the trends in physical fitness levels of Asian preschoolers.

Macao is a Special Administrative Region (SAR) of the People's Republic of China, which uniquely blends Eastern and Western cultures, making it an intriguing setting to examine trends in the physical fitness of preschoolers. The region has a strong commitment to education, offering 15 years of free preschool to secondary school education, ensuring broad access to early childhood education and development programs for all children.¹⁶ With a remarkable life expectancy surpassing 84 years in 2019,¹⁷ Macao has prioritised its healthcare and social systems. Recognising the importance of physical health, standardised and systematic fitness testing of people of all ages has been implemented since the turn of the century, when the Government implemented a fitness surveillance system aligned with China's national fitness monitoring.¹⁸ Using population-representative data, this study examined the temporal trends in the physical fitness of Macao preschool children aged 3–5 years between 2002 and 2020. The findings could have important implications for identifying key interventions and shaping future public health and physical education policies in Macao, as well as offer valuable lessons for other regions in the Western-Pacific region.

2. Methods

2.1. Data source

The Sports Bureau of Macao SAR conducts the Physical Fitness Report of Macao SAR Residents, a citywide, repeated cross-sectional fitness survey. This survey's objective is to assess the physical fitness levels of Macao residents aged 3–79 years. Since 2005, data have been collected once every 5 years, encompassing a wider range of tests, building upon the initial implementation of the first large-scale Research Report on Physical Fitness of 3–6 Years Old Children in Macao in 2002. The monitoring system serves as a vital means to monitor the physical fitness status of residents. It provides scientific data for government to formulate social development policies and promote physical activity.¹⁹ The sampling and testing procedures are detailed elsewhere, with sex- and age-specific percentile scores available for all fitness tests in each report.20, 21, 22 Our research specifically utilised data on preschoolers, which were derived from these comprehensive reports. Preschoolers from six kindergartens in Macao were recruited using a combination of random stratified and cluster sampling methods. Prior to data collection, written parental consent was provided, confirming participants were apparently healthy and suitable for fitness testing. Across five data collection periods (December 2002, January to April 2005, 2010, and 2015, and September to November 2020), standardised test protocols were used to assess the body size and physical fitness of the participants. Additionally, an on-site medical professional handled any emergencies. Physical fitness data were collected for a total of 4519 preschoolers (boys, 59 % or n = 2680; girls, 41 % or n = 1839) aged 3–5 years between 2002 and 2020. The frequency breakdown is reported in Table 1.

Table 1.

Frequency of sample characteristics by gender and age.

Age		Study
	Gender	2002	2005	2010	2015	2020
3	Male	106	159	193	206	232
	Female	103	96	102	128	143
4	Male	115	191	185	180	194
	Female	115	113	117	131	129
5	Male	114	165	189	221	230
	Female	116	132	107	142	165

2.2. Physical fitness indicators

Parents reported demographic information including the child's age at their last birthday and their sex as male or female. For the body size measurements, height was recorded in the nearest 0.1 cm using a stadiometer, weight was measured in the nearest 0.1 kg using a digital weighing scale, body mass index was calculated as weight divided by height in metres squared and reported to the nearest 0.1 kg/m², and circumferences of the chest, waist, and hip were measured in the nearest 0.1 cm using a measuring tape.

Physical fitness was assessed using six tests. Speed-agility was assessed in a single trial by instructing participants to run 10 m from the starting line to a turning line (by touching the wall) and back. The time taken recorded in the nearest 0.1 s. Additionally, participants performed two-leg continuous jumps over ten 5 cm soft packs placed at 0.5 m intervals, with the better of two trials recorded in the nearest 0.1 s using a stopwatch. Lower body muscle power was measured by having participants perform the standing long jump using an electronic mat, with the better of two trials recorded to the nearest 1 cm. Upper body muscle power was assessed by having participants throw a standardised tennis ball as far as possible. The throw distance was measured using a tape marked at every 0.5 m interval, with the better of two trials recorded. If the ball landed beyond 20 m, then the actual distance was measured using a measuring tape. If the ball landed outside of a 6 m wide boundary, the participant repeated the throw. Balance was evaluated by having participants walk a 3 m long balance beam as fast as possible, with the better of two trials recorded in the nearest 0.1 s using a stopwatch. Flexibility was measured by having participants perform the sit and reach test, with the better of two trials recorded in the nearest 0.1 cm using an electronic apparatus.

2.3. Statistical analyses

All data were analysed using R for Windows 4.3.3. General linear models (GENMOD) were used to assess trends, with the testing year as the independent variable and the body size and physical fitness measures as the dependent variables. Trend models were adjusted for gender and age, with the trends in physical fitness additionally adjusted for body size (height and weight). To reflect the true underlying confidence intervals (CI), we applied an approach described in detail elsewhere to adjust the 95 % CI²³.

Trends in means were presented as absolute trends (i.e., the regression coefficient) and standardised effect sizes (Cohen's d; the regression coefficient divided by the pooled standard deviation), while trends in coefficient of variation (CV; the standard deviation divided by the mean) reflected trends in variability. Additionally, trends in percentiles were calculated to examine distributional asymmetry. Positive trends indicated temporal increases in means, while negative trends indicated temporal declines in means. We used thresholds of 0.2, 0.5, and 0.8 to interpret the magnitude of standardised trends in means, with values below 0.2 considered negligible.²⁴ Trends in means were visualised by expressing the linear standardised rates of change in physical fitness to the year 2002 = 0. The ratio of CVs was calculated by dividing the 2015–2020 CVs by the 2002–2010 CVs. A ratio below 0.9 indicated a significant decrease in variability, while a ratio above 1.1 indicated a significant increase in variability. Ratios between 0.9 and 1.1 suggested that the trend in variability was negligible.²⁵ Trends in distributional asymmetry were visualised using LOWESS (Locally Weighted Scatterplot Smoother) curves (span = 0.75), by plotting the covariate-adjusted trends in physical fitness from the 10th to the 90th percentiles.²⁶

3. Results

3.1. Trends in body size

Table 2 presents the trends for body size measurements among Macao preschoolers from 2002 to 2020. There were significant increases in height (ES = 0.42), weight (ES = 0.30), and chest circumference (ES = 0.38). A significant moderate increase was observed for hip circumference (ES = 0.59). However, the trend in waist circumference was negligible and the trend in body mass index was not statistically significant. There were negligible trends in distributional variability. Furthermore, there were small gender-related temporal differences for height, waist and hip circumferences, with negligible differences for weight and chest circumference (Supplementary Table 1). Increases in weight, chest circumference, and boys’ waist circumference were similar across the distribution, while some measures exhibited more nuanced patterns (Supplementary Fig. 1).

Table 2.

Temporal trends in mean body size for 3- to 5-year-old Macao preschool children between 2002 and 2020 after adjusting for trends in gender and age.

			Change in means (95 % CI)		Ratio of CVs (95 % CI)
Test	n	Mean ± SD	Absolute	Standardised ES
Height (cm)	4514	105.4 ± 4.5	1.9 (1.5, 2.3)	0.42 (0.34, 0.51) ∗	1.01 (0.94, 1.08)
Weight (kg)	4519	17.2 ± 2.5	0.8 (0.5, 1.0)	0.30 (0.21, 0.38) ∗	1.00 (0.84, 1.16)
Body mass index (kg/m2)	3845	15.4 ± 1.5	−0.1 (−0.2, 0.0)	−0.04 (−0.13, 0.05)	1.02 (0.92, 1.09)
Chest circumference (cm)	4512	53.3 ± 3.0	1.2 (0.9, 1.4)	0.38 (0.29, 0.47) ∗	1.00 (0.85, 1.15)
Waist circumference (cm)	3843	49.8 ± 3.9	0.7 (0.4, 1.1)	0.18 (0.09, 0.27) ∗	1.02 (0.62, 1.41)
Hip circumference (cm)	3844	55.9 ± 4.0	2.4 (2.0, 2.8)	0.59 (0.50, 0.68) ∗	1.01 (0.68, 1.33)

Note. ∗p < 0.01. All absolute changes in mean values. The trend for body mass index trend was from 2005 to 2020. Positive trends indicate temporal increases in means, while negative trends indicate temporal declines in means. Abbreviations. SD: standard deviation; CI: confidence intervals; ES: effect size; CV: coefficient of variation.

4. Trends in physical fitness

Table 3 and Fig. 1 presents the findings of physical fitness trends among Macao preschoolers from 2002 to 2020. The results showed small significant improvements in speed-agility (two-leg continuous jump: ES = 0.25), lower body muscle power (ES = 0.32), and flexibility (ES = 0.34). Conversely, there were small significant declines in balance (ES = −0.32). The trend in one of the speed agility indicators (2 × 10 m shuttle run) and upper body muscle power was not statistically significant. Negligible gender-related temporal differences were observed (Supplementary Table 2). There were negligible trends in distributional variability and differential trends in distributional asymmetry (Fig. 2). For example, we found larger declines for those with low balance (below the 25th percentile) and smaller declines for those with high balance (above the 75th percentile) compared to those with average balance (at the 50th percentile). In contrast, the trends for speed-agility, lower body muscle power, and flexibility were relatively uniform across the population distribution, while for upper body power, little change was experienced by those with low and average power and small declines were experienced by those with high power.

Table 3.

Temporal trends in mean physical fitness for 3- to 5-year-old Macao preschool children between 2002 and 2020 after adjusting for trends in gender, age, height, and weight.

Test	n	Mean ± SD	Change in means (95 % CI)		Ratio of CVs (95 % CI)
			Absolute	Standardised ES
Speed-agility
2 × 10 m shuttle run (s)	4500	8.5 ± 1.2	0.2 (0.1, 0.4)	0.19 (0.09, 0.29)	1.00 (0.85, 1.16)
Two-leg continuous jump (s)	4030	9.6 ± 3.4	0.9 (0.5, 1.3)	0.25 (0.14, 0.35) ∗	1.04 (0.96, 1.12)
Lower body muscle power
Standing long jump (cm)	4493	71.6 ± 16.3	5.3 (3.6, 7.0)	0.32 (0.22, 0.43) ∗	1.00 (0.62, 1.37)
Upper body muscle power
Overhead throw (m)	4510	3.7 ± 1.1	−0.2 (−0.3, 0.0)	−0.14 (−0.24, −0.03)	1.04 (0.87, 1.20)
Balance
Walking balanced beam (s)	4388	14.0 ± 9.4	−3.2 (−4.3, −2.2)	−0.32 (−0.43, −0.22) ∗	1.03 (0.89, 1.16)
Flexibility
Sit and reach (cm)	4511	8.7 ± 4.2	1.5 (1.0, 1.9)	0.34 (0.24, 0.44) ∗	1.00 (0.76, 1.24)

Note. ∗p < 0.01. All absolute changes in mean values. Positive trends indicated temporal increases in means, while negative trends indicated temporal declines in means. Abbreviation. SD: standard deviation; CI: confidence intervals; ES: effect size; CV: coefficient of variation.

Fig. 1.

Linear standardized rates of mean change in physical fitness among Macao SAR preschoolers aged 3 to 5 from 2002 to 2020.

Notes. Trends are presented as standardized effect sizes (ES). Using a common metric by standardizing trends relative to a reference time point (i.e., 2002 = 0), the curves depict these trends. Positive trends indicate temporal increases in means, while negative trends indicate temporal declines in means.

Fig. 2.

Distributional trends in physical fitness (adjusted for gender and age) among Macao SAR preschoolers aged 3 to 5 from 2002 to 2020.

Notes. Distributional trends are shown as standardized effect sizes (ES). Positive trends indicate temporal increases in means, while negative trends indicate temporal declines in means. The solid lines represent trends in distributional asymmetry, visualised by LOWESS (Locally Weighted Scatterplot Smoother) curves (span = 0.75), which plot the overall trends in adjusted fitness tests from the 10th to the 90th percentiles. Sloped lines indicate asymmetric trends (e.g., lines that sloped upwards from the bottom left to the top right indicated relatively larger improvements in those with better performance than those with lower performance) and flat lines indicated symmetric trends (i.e., uniform trends across all fitness levels).

5. Discussion

This study comprehensively examined trends in physical fitness trends among Macao preschoolers aged 3–5 years from 2002 to 2020. The results indicated increased body size, and modest yet meaningful improvements in several fitness components, such as speed-agility, lower body muscle power, and flexibility, and declines in others. Importantly, these trends were similar for both genders, indicating an equitable improvement in physical fitness. Our findings are of relevance because we provide a crucial glimpse into the body size and physical fitness trajectory of Macao's preschool population, presenting a different temporal picture to that seen among European preschoolers.^3,9,13 Given the general positive trends observed in Macao, researchers and policymakers in other countries may find our findings informative as they work to promote childhood physical fitness through interventions and programs.

In contrast to previous research on trends in physical development among children and adolescents in Poland, which showed stability in height but increases in weight and body mass index,^12,14 our findings suggest that Macao preschoolers have increased in size, perhaps due to the better living and nutritional conditions.²⁷ These findings are particularly significant, as studies have highlighted the positive relationship between waist circumference and speed-agility, but a negative relationship between cardiorespiratory fitness and lower body muscle power.²⁸ This broadens our understanding of physical fitness in this age group, given previous research shows increases in body mass index and skinfold thickness have coincided with declines in motor fitness in recent decades.^9,28 Notably, this study displayed data for height, weight, and bodily circumferences separately, as past research has highlighted limitations in using body mass index or waist-to-hip ratio as reliable indicators in this age group.²⁸ These findings underscore the necessity for continued monitoring to support healthy growth trajectories among preschool children in Macao.

The standing long jump test which assesses lower body muscle power, has been consistently shown to be a reliable, valid, and health-related physical fitness indicator across different age groups, from preschoolers to adolescents.^1,5 Additionally, this test is important for predicting long-term health outcomes, such as cardiovascular disease risk factors in adulthood.²⁹ In this context, the observed improvement in standing long jump performance among Macao preschoolers is encouraging. This contrasts with research on similar populations in the Czech Republic and Poland, where declines in standing long jump performance were reported alongside increases in body mass index and skinfold thickness.^3,9 A study from Germany found stable standing long jump performance between 2003 and 2017 among preschoolers.¹³ This suggests that the temporal relationship between body size and lower body strength may vary across populations, as well as age and gender groups, underscoring the importance of longitudinal analyses in understanding such complex relationships. Better understanding these relationships may inform the development of targeted interventions to promote healthy physical development and prevent the onset of chronic disease risk factors.

The use of the China-specific fitness test battery allows for a direct comparison of the Macao preschoolers' performance to national norms, enhancing the interpretability of the findings within the local context. However, this also limits the direct comparability of the results to studies using different test batteries. For example, the sit-and-reach test is a measure of lower limb flexibility, but there is limited information on its health-related predictive validity.^1,30 Interestingly, the current study's findings indicate a temporal improvement in flexibility among Macao preschoolers, which differs from the trends observed in Polish preschoolers.^3,12 This suggests that there may be cultural, environmental, or other factors influencing flexibility development in young children across regions and populations. The current study did not include assessments of muscle strength (e.g., handgrip strength) or cardiorespiratory fitness (e.g., the preschool-adapted 20-m shuttle-run), both of which are reliable and valid fitness components that have been used to assess preschoolers and are linked with various health outcomes and future mortality risk.^1,31 The standardised tests included in the current study tended to focus more on motor movement skills, and the inclusion of additional assessments could have provided a more comprehensive understanding of trends in physical fitness levels of Macao preschoolers.

Previous studies have employed LOWESS analysis to examine non-linear changes in biological attributes across a spectrum of body size and physical fitness measures among children and adolescents.^32,33 This analytical approach has also been applied to investigate the relationship between body mass index and health outcomes in preschool populations.^34,35 However, examining trends in the distributional characteristics of physical fitness is rare. Our findings have several important implications for understanding body size and physical fitness trends during early childhood. Trends for most fitness measures were symmetric across the population distribution, particularly the middle of the distribution. Preschool children's fitness levels may be susceptible to environmental, behavioural, or physiological influences that drive more pronounced changes, such as parental education level, access to physical activity resources, or psychomotor development.^26,35 These findings have important implications for a more generalised intervention. Rather than applying a targeted approach, our results indicate that one-size-fits-all strategies may be needed to support the unique needs of preschool children along the body size and fitness continuum. Alternatively, early diagnosis and specialised programming for those at high risk of developmental issues could help ensure these children receive the support they need to reach their full potential.

The notable improvements in physical fitness indicators among Macao preschoolers observed in the present study can be attributed to a combination of factors within the education system and home environment. Macao's education system and curriculum guidelines have prioritised the development of physical education and the acquisition of gross motor skills. As the first region in China to incorporate kindergarten into its free education system since 2007, Macao has made the education of its residents a key priority.¹⁶ Additionally, the curriculum has been reformed since 2015 and explicitly emphasises the learning of fundamental movement skills, such as locomotor (i.e., walking, running, jumping, balancing, crawling) and object control skill (i.e., throwing, catching) through various physical activities and games designed by teachers.³⁶ The trends observed in physical fitness among Macao preschoolers can be used to monitor the impact of implemented policies and initiatives. Additionally, the positive influence of parental physical activity modelling may also play a role in these improvements, especially during the COVID-19 pandemic when children had fewer opportunities for physical activity outside the home.^27,37, 38, 39 Future research should continue monitoring the progress of collaborative physical activity promotion efforts between kindergartens and families. This should involve examining recent trends in fitness levels over time, as well as determining the most effective interventions to combat any potential declines in physical fitness among preschoolers.

5.1. Strength and limitation

One of the major strengths of this study is its contribution to the limited body of research on temporal trends in physical fitness among preschoolers worldwide, especially the Western-Pacific region. Our analysis of trends in means and distributional characteristics provides a more holistic understanding of trends in physical fitness among preschoolers than other studies. However, there are several limitations to this study. The use of the China-specific fitness test battery enhanced the local interpretability of the findings but restricted direct comparability with studies using different test batteries. This highlights the need for further research adding standardised assessment items to better understand physical fitness and its implications for Macao preschoolers’ health and development. Additionally, the timing of data collection varied over time, which could potentially have biased our trends. Specifically, the 2002 and 2020 data were collected in the second half of the year and the 2005, 2010, and 2015 data were collected in the first half of the year. The 2002 data collection was the first of its kind, whereas the subsequent years were standardised. The 2020 data collection was delayed to the second half of the year due to disruptions caused by the COVID-19 pandemic, which impacted the usual data collection cycle. This delay was necessary to ensure the safety and health of participants and researchers. Although body mass index is a valuable indicator of body size, data were not available in 2002. This meant that the temporal trend in body mass index could only be calculated for the 2005–2020 period. We could only estimate trends based on descriptive data, without adjustments for other factors like physical activity levels. The observed trends should therefore be interpreted cautiously, as other unmeasured factors may have influenced the results. Another limitation is the potential for selection bias if less fit preschoolers did not participate, leading to an underrepresentation of certain groups.

6. Conclusion

This study provides valuable insights into the trends in body size and physical fitness of Macao preschoolers from 2002 to 2020. The findings indicate small increases in height, weight, and circumferences, improvements in speed-agility, lower body muscle power and flexibility, and declines in upper body muscle power and balance ability. Importantly, no significant gender differences were found, indicating equitable fitness development. These trends suggest increased growth and development and highlight the need for a generalised approach to support fitness across the spectrum. Ongoing monitoring is essential to support physical fitness and overall health in early childhood. Future research should focus on addressing strategies to enhance overall physical fitness levels in early childhood. Continuous screening and monitoring are important for identifying trends and guiding future health initiatives.

Ethical declarations

This repeated cross-sectional study was exempt from ethical approval due to the exclusive use of publicly accessible anonymous register data.

Data availability

The datasets used and analyses during the current study are available from the corresponding authors on reasonable request.

Disclaimer

The content and views expressed in this article are those of the authors and do not necessarily reflect those of the Government of Canada.

Authors' contributions

Conceptualisation: S.M.C. & E.T.C.P., data curation: S.M.C. & H.Y.D., formal analysis: S.M.C. & H.Y.D., investigation: S.M.C. & E.T.C.P., methodology: S.M.C., G.R.T., J.J.L., & E.T.C.P., project administration: S.M.C. & E.T.C.P., resources: S.M.C., software: S.M.C. & J.J.L., supervision: S.M.C. & E.T.C.P., validation: G.R.T., J.J.L., & E.T.C.P., visualisation: S.M.C. & H.Y.D., writing – original draft: S.M.C., and writing – review & editing: S.M.C., G.R.T., J.J.L., C.C.-S., S.M.L. & E.T.C.P.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Cristina Cadenas-Sanchez reports receiving financial support from the Marie Skłodowska-Curie actions, Horizon 2020 Framework Programme, European Union (Project ID: 101028929). For other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: