Trends in Physical Fitness Among School-Aged Children and Adolescents: A Systematic Review

Introduction and Objective: This systematic review aimed to analyse the international evolution of fitness with its distributional changes in the performance on tests of physical fitness among school-aged children and adolescents. Methods: In accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, the search was undertaken in four international databases (ERIC, PubMed, Scopus, and Web of Science) to identify the studies reporting temporal trends in the physical fitness among school-aged children and adolescents. Results: A total of 485 potential articles were identified, of which 19 articles were relevant for the qualitative synthesis; 1,746,023 children and adolescents from 14 countries (China, Finland, Sweden, Belgium, New Zealand, Denmark, Spain, Norway, Mozambique, Poland, USA, Lithuania, Portugal, Canada), for the period between 1969 and 2017 were included. The subjects were tested using 45 motor tests from eight battery tests. The quality of the study in eight articles was rated as strong, while in 11 articles it was rated as moderate. Discussion: The vast majority of studies show a constant decline in strength and endurance. Three Chinese studies show an increase in strength from 1985 to 1995 and then a decline until 2014. For endurance, similar patterns were found in the two most comprehensive Chinese studies. The decline in flexibility is also evident in European countries. For agility, speed, balance, and coordination, the trend differs among populations.


INTRODUCTION
Physical fitness is a multicomponent construct that is closely related to the ability to perform physical activity (1,2). It is considered to be an important health marker, because high levels of fitness during childhood and adolescence have a positive impact on adult health (3,4). Additionally, higher levels of physical fitness enable participation in a variety of physical activities and decrease the risk of health problems (5)(6)(7). Physical fitness is determined by genetic factors and the level of regular exercise and physical activity (8). The modern era has brought changes in ways of life and work that are associated with lower levels of physical activity (9). From a traditionally active lifestyle in which physical fitness was necessary to manage daily tasks, most people switched to the more sedentary lifestyles. In the previous four decades, studies have indicated an association between the lower physical activity levels (10) and change in body composition and somatotype (11,12). Also, declines in physical fitness are often recorded (13), which are likely influenced by the decreasing trend in physical activity and changes in body composition (14,15). The decline in cardiorespiratory fitness has been extensively documented (16)(17)(18). Moreover, a decline in flexibility (19), repetitive strength and running speed were also recorded (20). Given the reported declining trends in physical activity and consequently physical fitness, some researchers predict the emergence of serious public health concerns (6,21). Contrary to this evidence, in some countries, an increase has been registered on certain components of physical fitness. Examples of this include an increase of muscular fitness in Finland (22) and cardiorespiratory fitness and strength in Canada (23).
These differences in physical fitness among countries have not been examined in detail to date (3). Furthermore, previous studies have only analyzed trends of single physical fitness components (e.g., cardiorespiratory fitness, muscular fitness), which underscores the need for a comprehensive review study that covers more than one physical fitness component and gives a general picture of its trends, which would facilitate conclusion drawing and monitoring. Thus, the objective of this systematic review was to gather the available information on all components of physical fitness among children and adolescents and to analyse the international trends of their performances on the respective physical fitness tests.

METHODS
Data selection, collection, and analyses were performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (24).

Inclusion Criteria
Scientific articles containing data on temporal trends in physical fitness published up to April 2020 were included. Eligibility criteria were the following: (1) cross-sectional, longitudinal, and interventional studies (design criterion); (2) articles published in scientific journals, book chapters, books, conference proceedings, and theses (publish criterion); (3) fitness battery, physical fitness, cardiovascular fitness, speed, flexibility, agility, muscular strength, body composition (outcome measure criterion); (4) children and adolescents aged 10 to 18 years (participant criterion); (5) articles published in English, French, Portuguese, Spanish, or German (language criterion).

Search Strategy
The literature search was undertaken in four international databases: the Education Resources Information Center (ERIC), PubMed, Scopus, and Web of Science. The search was conducted on April 17, 2020. In each database, a search was conducted by title, taking a predefined combination of keywords (through discussion among the research team) into account. The combination of used keywords was the following: "field-based test" OR fit * OR "physical performance" OR "sport performance" OR "physical condition" OR "aerobic capacity" OR "maximum oxygen consumption" OR strength OR flexibility OR motor OR endurance OR speed OR agility OR balance OR "body composition" OR anthropometry OR "body mass index" OR BMI OR skinfolds OR "waist circumference" AND trend * OR tendenc * AND adolescent * OR child * OR young * OR "school age" OR school-age OR youth. The keywords were selected through discussion among the research team, finally defined by the consensus of all authors.

Data Extraction and Selection
After performing a search in the databases, the necessary data were transferred to a software tool for publishing and managing bibliographies. The process of data extraction was performed based on PRISMA guidelines (24). The articles were downloaded from the databases, after which duplicates, identified by title and author, were removed. Two researchers screened titles and abstracts of the remaining records. Then, the full-text of relevant articles were read and examined according to the inclusion criteria, to decide whether or not to include them in the systematic review. The following information was extracted from each study: author's name and year of publication, study design, country, sample characteristics (number of participants, gender, and age), the instrument/battery for assessing physical fitness, main results, and study quality.

Study Quality and Risk of Bias
The methodological quality of the studies was assessed using the Quality Assessment Tool for Quantitative Studies (25), which is a 19-item checklist, assessing eight methodological domains: selection bias, study design, confounders, blinding, data collection methods, withdrawals and dropouts, intervention integrity, and analyses. Each section was graded as being of strong, moderate, or weak methodological quality. A global rating is determined based on the scores of each component. Two researchers rated the studies in each domain, as well as the overall quality of each study. Discrepancies were resolved by consensus. Figure 1 shows the flow chart of records selection. A total of 485 potential articles were identified through the electronic database search (three from ERIC; 145 from PubMed; 166 from Scopus; 171 from Web of Science). After exclusion of the duplicates (279), the title and abstract of 206 were assessed for eligibility. After elimination at the title and abstract level 157 articles, the remaining 49 articles were subsequently read. After reading, another 30 articles were eliminated, leaving 19 relevant articles that satisfied the inclusion criteria and were included in the qualitative synthesis.

Study Characteristics
In Table 1, the studies' characteristics are presented. From 19 studies included in the qualitative synthesis, a total sample of 1,746,023 children and adolescents from 14 countries was represented. All studies had a cross-sectional design with two, three, or more samples from the same number of time-points. Three studies were performed in China and another three in Finland, two in Sweden, and one in each of the following countries: Belgium, New Zealand, Denmark, Spain, Norway, Mozambique, Poland, the United States of America, Lithuania, Portugal, and Canada. The period covered by the studies was between 1969 and 2017. Overall, participants' performance in 45 fitness tests from eight different fitness tests batteries was recorded. Strength was most frequently tested, followed by endurance, flexibility, agility, speed, balance, and coordination.
The quality of the studies in eight articles was rated as strong, while in 11 articles it was rated as moderate.

Strength
Strength was mostly tested topographically (arm and shoulder belt strength in 10 papers for boys and eight for girls; lower limb strength in 10 papers for boys and 11 for girls; abdomen strength in five papers for boys and nine for girls). With the exception of one study, which applied a test assessing the strength of the entire body, eight tests were used to evaluate arm and shoulder strength: handgrip strength (kg), bent-arm hang (s), arm pull (kg), bench-press (n): all four for both boys and girls; pull-ups (n), oblique body pull-ups (n), pull-ups (n/min): three for boys only, and flexed-arm hang (s): only for girls. Three tests were used to evaluate the strength of the lower limbs in both sexes: standing No change in cardiorespiratory fitness over time (1983,1997,2003), and quite high levels were found in these representative cohorts. It founded substantial differences in maximal power output, but no differences when VO2 max was estimated from the equations derived in the validation studies   long jump (cm), vertical jump (cm), and leg lifts. The sit-ups test was used for both sexes to evaluate the strength of the abdomen. In one study, a two-hand lift (N) test, which assessed the strength of the entire body, was used for boys and girls. For the strength of the arms and shoulder belt, a declining trend was found in nine studies for boys and seven for girls. In three Chinese studies (34,36,37), a growth trend from 1985 to 1995 was initially found for all cohorts of boys, followed by a trend of decline until 2014. In the additional six studies for boys (3,22,23,26,27,33) and seven for girls (3,3,23,26,27,30,31,33), a steady decline was observed between 1969 and 2017. For a cohort of Chinese 12-year-olds, the results of upper-body strength showed an increase until 2005, followed by a decline (34). Significant changes were not shown in the cohort of Canadians aged 15 to 19 (23), in Finnish adolescents aged 13 to 16 (22), in boys in Belgium (27), and in Spanish boys and girls on a static strength test (3). Finally, a growth in strength was noted in Mozambique children and adolescents between 1992 and 2012 (30), in Canadian girls aged between 11 and 14 (23), and in Polish girls aged 15.5 years on an absolute strength test (31).
For the explosive strength of the lower extremities, a declining trend was found in six studies for boys and eight for girls. Three Chinese studies for boys and girls (34,36,37) found an initial growth trend from 1985 to 1995 (it peaked in 1995), followed by a downward trend until 2014. In three studies for boys and girls (3,28,33) and two girls only (27,31), a steady decline was noted between 1969 and 2013. No significant changes for explosive strength were noted in the study by Huotari et al. (22) of Finnish boys, the study by Westerstahl et al. (26) for Swedish girls, and in the study with Portuguese adolescents of both sexes (19). Furthermore, no significant changes were noted for the repetitive strength of the subjects of both sexes in Belgium (27). Progress was recorded for a cohort of Swedish adolescents aged 16.4 between 1974 and 1995 (26), which is consistent with the results of studies in China in which a decline in strength occurs after that period. The growth trend was recorded for a cohort of Finnish adolescents aged 13-16 between 1976 and 2001 (22).
The trend of abdomen strength decline in both sexes was found in Swedish adolescents between 1974 and 1995 (26), and in Polish girls (31) between 2006 and 2013. A cohort of Portuguese school-aged boys from 1993 to 2008 and girls from 1993 to 2003 found a trend of decline, followed by a growth trend until 2013 (19). Studies in China have noted a trend of decline in repetitive strength in girls from 1985 to 1995, followed by a growth trend (34,36,37). The growth of repetitive strength in both sexes was recorded in New Zealand from 1991 to 2003 (28) and in Finland between 1996 and 2002 (22). In contrast, between 1992 and 2012, only Lithuanian girls showed an increase, while there were no significant changes among the boys (33).
Westerstahl et al. (26) examined complete body strength and found progress for both sexes of a cohort of Swedish adolescents aged 16.4 years, for the period between 1974 and 1995, which is a similar result to those in Chinese studies indicating fitness growth also up to 1995 and reaching a peak in that period (34,36,37).

Endurance
Endurance was assessed in boys within 14 research studies and in girls within 15 studies. Eleven tests were used for both sexes: Two studies have shown a trend for forced vital capacity (ml) that shows functional ability, as people with high lung capacity are predisposed to endurance-type sports. The results for this parameter were calculated together for both sexes for children and adolescents

Flexibility
Flexibility was tested within nine research studies for both sexes. For boys, a sit-and-reach test was used, and a sit-and-reach test and standing trunk flexion were used for girls. Tests in Belgium (27), Lithuania (33), Portugal (19), and testing of girls in Poland

Agility
Agility was tested within six studies for boys and within seven studies for girls. Three tests for both sexes were used for assessment: a 10 × 5 m shuttle run (s), a 4 × 10 m shuttle run (s), and a 4 × 9 m run (s). In both sexes, agility declines in Mozambique (30), and New Zealand (28), moreover, for girls in Belgium (27) and Poland (31). The growth of agility for both sexes was found in Spain (3) and Finland (22) and only for boys in Lithuania (33), while the girls showed no significant changes. In Belgium, there is a trend of increasing results for boys between 1969 and 2005, but there were no significant changes between parents and their children (27).

Speed
Speed testing was performed within six studies. For the evaluation of both sexes, the following tests were used: 50 m sprint (s), 40 m sprint (s), and the speed of movement frequency was measured by the Plate tapping test (n/20 s). Tests show a negative trend in the speed of boys and girls in the cohort of Flemish subjects in Belgium (27) and in girls in Poland (31). In a study by Ao et al. (34), Chinese girls' (Han ethnicity) speed decreased from 1985 to 1991 and increased thereafter but only in rural areas, while in boys a constant decrease was recorded. In the Chinese province of Xinjiang, the speed decreased for both sexes from 1985 to 1995, then increased until 2005, after which it remained stable (37). In contrast, the results for the whole of China for both sexes (36) show an increase in speed until 1995, then a decrease until 2005 and retention at a similar level until 2010, followed by a slight increase in 2014. The increase in speed was also visible in urban areas of China (Han ethnicity) for 12-year-olds of both sexes (34), and Portugal (19).

Balance
Balance was tested for both sexes as part of two research studies with the Flamingo balance (n/min) test. Balance tests show that in the cohort of Flemish subjects in Belgium (12-18 years) there are no significant changes when comparing the results of parents and their children later when they reached the same age; however, for the entire female population there is a decrease in balance (27). In Lithuania, greater improvement in balance scores was found in the previous decade and still more in girls 11-18 (33).

Coordination
Only one study shows coordination results for both sexes using three tests: motor coordination track (s), lateral jumping (n/15 s), and Figure 8 (n/min). The only study that examined coordination was conducted by Huotari et al. (35) for a cohort of Finnish adolescents (15-16 years) in the period between 2003 and 2010. The motor coordination track (s) test shows a decrease during the tracked period for both sexes. Lateral jumping test (n/15 s) which evaluates dynamic balance, quickness, and the explosive strength of lower limbs with coordination indicates that there are no significant changes for both sexes, while the Figure 8 test (n/min) test, which evaluates object control skills with coordination, shows that there are no significant changes in boys, while girls have made progress in results.

DISCUSSION
This study provides a comprehensive overview of longitudinal changes in the physical fitness of children and adolescents and thereby indicates relevant knowledge to develop appropriate public health strategies. The articles that were included in the qualitative synthesis describe the temporal trends for seven physical fitness attributes. Overall, a declining trend was found for one or all three topological areas of strength (9 of 10 studies for boys and in 8 in 11 for girls), endurance (9 in 14 studies for boys and in 8 in 15 for girls), flexibilities (4 in 9 studies), agilities (4 in 6 studies for boys and 4 in 7 for girls), and speed (2 in 6 studied).
Based on the analysis of all three topological areas of strength (arm and shoulder belt strength, lower limb strength, abdomen strength), it can be noted that most studies indicate a declining trend, which is not surprising given the changing lifestyle of children and adolescents, physical inactivity, and increasing screen-time around the world in the previous three decades (38). However, in some studies, in certain cohorts a trend of increase in strength has been noticed whereas in other cohorts a decrease in strength was not evident (19,22,23). It should be emphasized that these are studies with children and adolescents in which (according to the decisions of the governments in these countries) special programmes of additional exercise were applied. Possibly, even in these cohorts there would have been a declining trend in strength levels if there had been only the regular school curriculum, without additional exercise. In several studies, the handgrip strength test shows a growth trend in strength for girls (23,31,36). These results are in line with previous research, since absolute strength is proportional to the size of the muscle cross-sectional area (39) and body height, and which increased over the last 40 years (11,12).
Noteworthy is the fact that the majority of studies point to a trend of declining levels of endurance, which can be explained by the same causes as for the trend of declining strength. A small number of cohorts in certain studies show that there is no decline in endurance levels or even a growth trend. However, these are cohorts for which intervention programmes mirror the longitudinal effects (23,31) or studies covering insufficient periods (3), or studies with non-representative populations (29,34). In summary, the results of these studies must be interpreted with caution, when drawing conclusions about longitudinal trends.
Notably, research conducted in Europe has recorded a trend of declining levels of flexibility. In other parts of the world, growth trends have been reported, for example, in Canada (23), where additional exercise programmes have been conducted. However, the reasons for the growth trend in New Zealand (28) and Mozambique (30) have not yet been determined.
Given the diversity of agility trends in the available studies, it is not possible to discuss a specific direction of movement for this motor ability in children and adolescents. The speed trends are very difficult to interpret because results indicate differences. In particular, the three Chinese studies are quite contradictory. The trend of increasing speed in Portugal (19) probably results from the government's policy on additional exercise, although it is known that speed is highly genetically determined.
Two studies examining the balance are not enough to define the trajectory of the global trend. The results of studies conducted in Belgium (27) and Lithuania (33) are contradictory at first glance, but they are focused on different time periods, and it is possible that in identical periods differences would not exist.
Only one study that have examined coordination (35) does not provide a possibility to discuss global trends.
Generally, the decline of physical fitness of children and adolescents around the world is caused by various factors. In recent years, several studies showed that weight gain is related to physical fitness (14,15,31). As body weight increased, so did BMI, which influenced this trend in China (37), Sweden (26), and New Zealand (28). In addition to these two components, the increase in the thickness of the skin folds caused a decline in physical fitness in Belgium (27). Two Chinese studies showed that the decline in physical fitness was caused by changing lifestyles, characterized by higher media and fast food consumption (34,36). Morales-Demori et al. (32) linked the trend of declining endurance to a sedentary lifestyle, and Venckunas et al. (33) with additional risk factors of smoking, alcohol consumption, non-active lifestyle and longterm television viewing.
This study also has certain limitations. One of the most significant is the insufficient differentiation of the samples by age (e.g., 10-12; 12-14; 14-16; 16-18) which would show the most accurate data. However, the results for the whole population are mostly combined. Also, the authors themselves are self-critical and reported the following limitations in their works: some studies combined results for both sexes; field tests although used worldwide with this age, carry the possibility of individual errors of measurement performers; when testing aerobic fitness, the results will have real values only if all subjects are highly motivated; in some studies urban i rural area are divided, but did not take into account the urbanization of certain rural areas, which likely cause a certain contradiction; results of non-representative samples of some studies can't be generalized for the whole population. Finally, it should be noted, that causes and determinants of physical fitness trends have not been fully identified by researchers. However, weight gain and higher BMI levels are mainly the cause of the declining trend in physical fitness, which facilitates the need to monitor several health-related factors (e.g., food intake) among several settings of the living environment.
Studies from Finland (22), Canada (23), and Portugal (19) show that changing policies, including the development and implementation of health-enhancing programmes, can reduce the negative impact of a sedentary lifestyle, which have taken precedence in our population.
Finally, consistent conclusions about the development of strength and endurance were drawn, because these dimensions were assessed with the largest number of tests, while the other motor skills (flexibility, agility, speed, balance, coordination) were less frequently and insufficiently tested to draw a meaningful conclusion. We encourage researchers to develop a comprehensive and easily applicable test battery enabling the assessment of all motor skills. Doing so would clearly enhance the evidence regarding the longitudinal trends of physical fitness among children and adolescents around the world.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/ supplementary material.

ETHICS STATEMENT
Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

AUTHOR CONTRIBUTIONS
BM wrote the manuscript, collected the data, and performed analyses. JG wrote the manuscript, overviewed previous studies, and discussed the results. AM performed analyses. MP discussed the results. YD revised manuscript. DS revised manuscript. SP discussed the results and revised manuscript. All authors contributed to the article and approved the submitted version.