Objectively determined habitual physical activity in South African adolescents: the PAHL study

Adolescence is a period characterised by rapid physical growth as well as changes in physiological and behavioural patterns [1]. Recent studies in South Africa have shown unsatisfactory levels of physical activity (PA), overweight and obesity among children and adolescents [2‐4]. Existing reports from nationwide cross sectional studies [3, 4] and a national survey [5] indicate that only about 42% of young South Africans aged between 13–19 years participate in sufficient moderate-to vigorous physical activity (MVPA). South Africa is a country faced with the burden of a rise in non-communicable diseases of lifestyle [4], hence the low levels of adolescent PA are disturbing as there is increasing evidence that PA has an important impact on long-term health and behaviour outcomes [6, 7]. It is essential to analyse patterns of habitual PA in adolescents as they serve as important functional indicators of an individual’s health status.

The most recent PA recommendation for Americans aged 6–18years is 60 minutes or more per day of MVPA [6] while other recommendations call for 30–60 min of daily MVPA [8, 9]. The accurate measurement of PA in epidemiological studies still poses a major challenge in terms of feasibility and validity of the methods proposed [10]. When assessed by accelerometry, PA has been shown to decrease during adolescence [11]. After an extensive review of literature, Dumith (2011) indicated that among adolescents aged 13–19 years, the decrease in PA was higher in boys (7%) than in girls (6.3%) [12]. Very limited PA data exists for representative samples of adolescents, with even fewer objectively collected data [13], especially in Africa.

In South Africa, there are currently not many studies which objectively quantify PA in adolescents. With the exception of Cook (2008; 2012) who used pedometers [14] and uniaxial accelerometry [15] to objectively measure PA in rural black South African women, there is a dearth of information on other population groups such as adolescents. Most studies that contain important information about the habitual PA of South African adolescents have used self-report methods to assess PA [3‐5, 16]. The use of objective instruments to determine PA would give a much needed, clearer understanding of the habitual PA patterns of South African adolescents.

In recent years, the measurement of PA with accelerometers has become more widely used [17]. These instruments quantify body movement at 5–60 second intervals for periods of up to 21 days, thereby enabling patterns of movement or inactivity to be measured. The Actiheart® has been validated against the doubly labelled water method; and it emerged a technically reliable and valid tool for PA assessment in populations [18].

South Africa is a multi-racial country, where black Africans form 79.2% of the population and Caucasians form 8.9% of the population. Race is a factor that has been hypothesised to have an effect on PA. Using questionnaires, McVeigh and colleagues (2004) found significant differences in patterns of PA in 9 year old South African school children of different races [16]. White children were found to be physically more active, more likely to participate in physical education classes at school and watched less television than black children [16]. No comparable data on the racial disparity of physical inactivity and television watching among South African adolescents could be found, but it is probable that the same trend seen in 9 year old South African children as reported by McVeigh et al. (2004) could be observed in South African adolescents.

There is a need for the accumulation of accurate data with regard to objectively measuring the pattern and amount of habitual PA in South African adolescents [19]. Baseline data from the Physical Activity and Health Longitudinal Study (PAHL study) may provide valuable information on objective PA patterns of adolescents for future studies on intervention strategies. Therefore, the purpose of the study was to objectively determine habitual PA of adolescents in the North West Province of South Africa by race and gender.

The present study provides data on objectively measured habitual PA of adolescents from the Tlokwe Municipality of the North West Province of South Africa. PA was determined using a combined accelerometer and heart rate monitor, the Actiheart®. One of the most compelling findings of our study is its consistency with the international phenomenon of low PA in adolescents [22, 23]. Up until now, most of the studies on habitual PA in South Africa were done using questionnaires [4, 24]. However, pedometers [14] and accelerometry [15] have been used to assess PA in rural black female South Africans as well as 8–12 year old South African children from disadvantaged schools [25]. The implication of our study is of great consequence, as to our knowledge, this study is the first to report objectively measured PA over a period of seven days, using the Actiheart® in a large number of South African adolescents.

The adverse effects of physical inactivity in adolescents have been reported to be related to the development of cardiovascular risk factors which lead to mortality [4, 6]. Consistent with the accelerometry determined PA findings of Americans aged 12–15years [26] and Thai adolescents aged 13–19years [27], our study reported that over 60% of South African adolescents are not sufficiently active. This means that less than 40% of adolescents from the North West province of South Africa meet the recommendation of 60 minutes of MVPA per day and are at risk of obesity, hypertension and diabetes. Despite our sample being less representative than the 2012 South African National Health and Nutrition Examination Survey (SANHANES) [4], the low level of PA among South African adolescents from the Tlokwe Municipality of the North West Province is remains evident.

Our findings indicate that on average, South African adolescents are more active on weekdays than weekends irrespective of gender and race. Our findings are in agreement with Compte (2013), who used accelerometry to determine MVPA in 12–15 year old Canadian adolescents and reported that similarly, they spent more minutes/day in MVPA on weekdays than on weekends irrespective of gender [28]. Likewise, using accelerometry, Konharn (2012) also reported more activity on weekdays than on weekends in 13–18 year old Thai adolescents [27]. With reference to South African adolescents from the Tlokwe Municipality of the North West Province, it is possible that walking to school (Non Exercise Activity Thermogenesis–NEAT transport domain) [29] or school related physical education/scheduled sport is the reason behind higher weekday MVPA. Conversely, in 2008, Mamabolo and colleagues revealed that ±14½ year old township adolescents residing in the North West Province of South Africa were more active on weekends than on weekdays [24]. This disparity is likely due to the fact that Mamabolo and colleagues used questionnaires and not objective means to asses PA. The homogeneity of our findings with those of Konharn [27] and Compte [28] highlights the need to take into account weekday and weekend differences when developing PA interventions for adolescents.

Habitual PA influences a large portion of energy expenditure. In addition to being represented in terms of minutes/day of MVPA, physical activity and can also be articulated as a calculated PAL value or measured AEE value. The higher AEE observed among girls in the present study may be an indirect derivative of their higher FM which has been shown to also influence energy expenditure. Our findings are in agreement with Livingstone (1992) who, using indirect calorimetry, reported higher PAL values in 15 year old girls than boys from Northern Ireland [30]. Several years later, using 15 year old Swedish adolescents, Bratteby (1998) gave disparate findings as boys recorded higher PAL values than girls [31]. Since PAL is calculated by dividing TEE by RMR, it is evident that RMR is a critical factor in the determination of PAL. The peculiarity of the results of the current study may be attributed to the dissimilar ratio of measured TEE/RMR between the girls and the boys. It is clear that contemporary studies investigating the PAL and AEE of adolescents need to be done as reference data is essential for a better understanding of the effect of PA on adolescent growth, health and development.

It is apparent that the recommendation of 60 minutes/day of MVPA is currently not being met and this has potentially negative implications related to the development of cardiovascular risk factors. Our findings concur with those of Collings (2014), who using accelerometry, determined that United Kingdom adolescents (15 years) spent only 41.1 minutes/day in MVPA [22]. Our results, however, provide more relevant race specific data and give a better profile of habitual PA of black and white adolescents from the Tlokwe Municipality of the North West Province as physical inactivity in South Africa is often associated with poor, disadvantaged, black township communities [25]. A 2005 study by Miller et al. suggested that girls are more likely than boys to participate in moderate activity only and not in vigorous activity [32]. However, the main limitations of Miller’s report are that PA was assessed by questionnaire and not objectively, the second is that an American college-aged cohort was used.

PA varies by both gender and race in adolescents from the North West Province of South Africa. Objectively determined data from our study indicates that girls habitually spend more time in MVPA per day than boys, and that white adolescents habitually engage in more MVPA than black adolescents. Given the fact that the study sample did not meet the recommended daily PA guideline, urgent strategies to inculcate the culture of regular PA as a preventative measure of chronic diseases of life style are needed. Adolescents should be the foremost targets of interventions aimed at enhancing habitual PA.