Leisure-time organised physical activity and dietary intake among Swedish adolescents

ABSTRACT

This study aimed to investigate associations between participation in leisure-time organised physical activity (LTOPA) and dietary intake in a large representative sample of Swedish adolescents participating in the national dietary survey Riksmaten Adolescents 2016–2017. A sample of 2807 participants aged 11–12, 14–15 and 17–18 years were included. Information about LTOPA and dietary intake were collected through questionnaires and two 24-hour recalls on the web (RiksmatenFlex). For dietary intake, overall healthy eating index, intake of fruit, vegetables, candy, sugar-sweetened beverages, and the proportion of added sugar to total energy intake were analysed. Significance-testing for associations was performed with analysis of covariance. LTOPA was associated with lower sugar-sweetened beverages intake among adolescent boys (p < 0.001, ηp2 = 0.012) and girls (p = 0.007, ηp2 = 0.005), higher fruit intake among boys (p = 0.043, ηp2 = 0.003), and lower vegetable intake among girls (p < 0.001, ηp2 = 0.009). However, LTOPA was unrelated to the overall healthy eating index, candy intake, and the proportion of added sugar to total energy intake (p > 0.05). LTOPA was only associated with some healthy eating behaviours, and there is much room for improvement in the diets of Swedish adolescents.

KEYWORDS:

• Adolescent health
• diet
• food
• physical activity
• sports

Introduction

Adolescence is a developmental life stage during which the foundations for future health behaviours are established (Sawyer et al., 2012; Sheehan et al., 2017). Investment in adolescent health therefore requires explicit attention and becomes paramount as early establishment of health behaviours may influence future health outcomes.

Among health behaviours, research shows the health benefits of regular physical activity and a balanced diet rich in fruit and vegetables and low in high-calorie, low-nutrient foods and sugar-sweetened beverage consumption (Nordic Council of Ministers, 2014; Piercy et al., 2018). Globally, however, only 20% of adolescents self-report the recommended 60 minutes or more of moderate-to-vigorous physical activity (MVPA; Guthold et al., 2020), and accelerometer data from Sweden show that 43% and 23% of adolescent boys and girls reach this recommendation (Nyberg et al., 2020). In addition, many adolescents do not eat fruit and vegetables on a daily basis (Vereecken et al., 2015). Recently, in the Swedish national dietary survey Riksmaten adolescents 2016–17, less than 10% of the adolescents reported 500 g or more of fruit and vegetables per day as is recommended in Sweden (Warensjö Lemming et al., 2018). Furthermore, one-fifth of the energy intake came from foods such as candy, cookies, snacks and sugar-sweetened beverages (Warensjö Lemming et al., 2018). Another study from Riksmaten adolescents 2016–17 found that boys and participants from families with lower parental education scored lower on overall healthy eating index (Moraeus et al., 2020).

Worldwide, many adolescents participate in leisure-time organised physical activity (LTOPA), such as sport and exercise (Aubert et al., 2018). According to results from the 2018 Report Card on Physical Activity for Children and Youth, about half of children and adolescents worldwide participate in some type of sport (Aubert et al., 2018). Data from the Swedish part of the Health Behaviour in School-aged Children show that approximately 75% of adolescents aged 11–15 years participate in sport two times or more per week (Delisle Nyström et al., 2018).

Several studies show associations between LTOPA and healthy behaviours such as more physical activity (Lee et al., 2018; Nelson et al., 2011) and better social and psychological health, including social skills, self-esteem, confidence and competence (Eime et al., 2013). For example, a recent study from Sweden showed that LTOPA was associated with more physical activity and less sedentary time among adolescents (Fröberg et al., 2020). Against this background, LTOPA might be considered a promising strategy to promote health among adolescents, both in the short- and long-term perspectives (Khan et al., 2012; Malm et al., 2019).

Given that research shows associations between LTOPA and healthy behaviours and positive health outcomes (Eime et al., 2013; Lee et al., 2018; Malm et al., 2019; Nelson et al., 2011), it might be hypothesised that adolescents who participate in LTOPA also are motivated to eat healthy. However, LTOPA might also generate circumstances that is less favourable in terms of dietary intake. For example, sports and exercise can introduce time constraints that might affect meal planning and preparation (Bean et al., 2014; Nelson et al., 2011), and unhealthy dietary products are often available at sport venues and sports clubs (Carter et al., 2012). Previous systematic reviews have shown somewhat mixed results for associations between LTOPA and dietary intake (Kanellopoulou et al., 2021; Nelson et al., 2011). In a systematic review from 2011, Nelson et al. found that participation in LTOPA was associated with both healthy and unhealthy dietary intake, such as more consumption of fruits and vegetables and more consumption of fast food and sugar sweetened beverages (Nelson et al., 2011). One limitation, however, was that the majority of the studies reviewed focused on elite athletes (Nelson et al., 2011). A more recent systematic review from 2021 showed that participation in extra-curricular sport-related activities was associated with increased consumption of fruits, vegetables and dairy, but that the results regarding sugar-sweetened beverages and fast-food consumption were inconsistent (Kanellopoulou et al., 2021). Among the studies reviewed, Vella et al. analysed a large nationally representative sample of Australian adolescents and found that participation in sports was associated with a greater likelihood to reach the fruit and vegetable consumption guidelines, but no association was found for consumption of sugar-sweetened beverages and high-fat foods (Vella et al., 2013). Another study that was not included in the review involved Norwegian adolescents and showed that participation in sports was associated with healthier food choices, such as decreased odds for high intake of sweets and sugar-sweetened beverages and energy drinks, as well as decreased odds for low intake of fruit, vegetables and berries (Torstveit et al., 2018).

In Sweden, however, less is known about associations between LTOPA and dietary intake among the general adolescent population. Studies investigating associations between LTOPA and dietary intake might be important because it contributes with background information that can be used when developing intervention strategies to target certain groups who might be in need to improve their dietary intake. For example, if LTOPA is associated with healthy dietary intake, future interventions might focus on those who do not participate in any sports and exercise during leisure-time.

The aim of this study was to investigate associations between participation in LTOPA and dietary intake in a large representative sample of Swedish adolescents.

Methods

This study is based on the national dietary survey Riksmaten Adolescents that was conducted by the Swedish Food Agency in 2016–2017. Statistics Sweden selected schools based on the type of municipality, school organisation (public or independent) and geographical location from which students in grade five (11–12 years), eight (14–15 years) and eleven (17–18 years) were recruited. A total of 619 schools (about 200 schools from each grade) were invited to take part via emails addressed to the principal. Of these, 131 schools agreed to take part and 1–2 classes from each school were selected to participate in the survey. All the students from these classes (n = 5145) were invited to participate, but an exclusion criterion was not being able to read and write Swedish. In total, 3477 students participated, and these were overall considered representative for the population concerning socioeconomic background, school organisation and type of municipality. A comprehensive description of the study design, methods and participation is available elsewhere (Moraeus et al., 2018).

The Regional Ethical Review Board in Uppsala (No. 2015/190) approved Riksmaten Adolescents. Prior to data collection, information letters were distributed to all students and their parents. As approved by the ethics committee, written consent was retrieved from all participants and both parents of those adolescents younger than 16 years who provided biological samples (not used in this study), and opt out consents were used among the remaining participants. Students could also opt out at any stage of the survey without providing any reason.

Data collection

Information about LTOPA and dietary intake was collected through a web questionnaire and a web-based 24-hour recall method, respectively. During school hours, trained field staff from the Swedish Food Agency informed the students about the study, introduced the web methods, and measured body height and weight.

Leisure-time organised physical activity

In the web-based questionnaire, the participants were asked whether they were active in any organisation/club (e.g., soccer, swimming, dance and scouts) with the responses being yes or no. In a follow-up question, they were asked which organisations/clubs they participated in and what kinds of activities that were included. This question had an open response alternative, and participants could provide multiple responses. These responses were manually reviewed, and activities involving physical activity in any organisation/club, such as, e.g., soccer, floorball, dance and martial arts, were considered as LTOPA. Leisure-time activities such as playing musical instruments and theatre were excluded. To assess the frequency of participation in LTOPA, the participants were asked how many times per week they participated in their activities with the response categories once per week, twice per week, three times per week, four times per week and five times or more per week.

Dietary intake

To assess dietary intake in Riksmaten Adolescents, the Swedish Food Agency developed the method RiksmatenFlexDiet. RiksmatenFlexDiet is a self-administered, web-based dietary assessment tool that could be completed on Internet connected devices, such as computers, tablets or smartphones. RiksmatenFlexDiet has shown to be as valid as information retrieved from 24-hour recall interviews (Lindroos et al., 2019).

The participants recorded all foods and beverages retrospectively, as a 24-hour recall, on two non-consecutive days. The first dietary day was registered during the school visit and the second day was automatically generated by the system, 4–10 days after the school visit, to get an even distribution of days over the week at group level.

In RiksmatenFlexDiet, the participants could provide information on the time of the eating/drinking occasion, the type of meal and place of the eating/drinking occasion and specify foods and drinks, and portion size (Lindroos et al., 2019). Using a built-in search engine, the participants could search for foods and chose from a returned list of foods. The food list, linked to the Swedish food composition database, was adapted for adolescents and contained 778 foods, dishes and beverages. Thereafter, the participants specified the amount consumed through pieces, portion pictures or household measurements. Automatic prompts to register beverages and condiments appeared throughout the registration period. Energy and nutrient intake was calculated automatically from the registered food intake using the Swedish food composition database, version Riksmaten Adolescents.

The information from RiksmatenFlexDiet was used to construct the Swedish Healthy Eating Index for Adolescents 2015 (SHEIA15; Moraeus et al., 2020). The individual components of SHEIA15 were based on the key advices of the 2015 Swedish Food Agency’s food-based dietary guidelines and the 2012 Nordic Nutrition Recommendations. The calculations were constructed as the ratio between the reported intake and the recommended intake of vegetables and fruit, fibre, wholemeal, fish, polyunsaturated fatty acids, monounsaturated fatty acids, saturated fatty acids, red and processed meat and added sugar. SHEIA15 takes all amounts of reported intake into account and not only those that reach the cut-offs for the food-based dietary guidelines and nutrition recommendation. For example, the food-based dietary guidelines recommend at least 500 g vegetables and fruit per day and was calculated as intake/500. The components were assigned a score between zero and one with values below zero and above one recoded to zero and one, respectively. A high score for vegetables and fruit, fibre, wholemeal, fish, monounsaturated fatty acids and polyunsaturated fatty acids reflects a high intake, whereas a high score for saturated fatty acids, red and processed meat and added sugar reflects low intake. Total SHEIA15 was assigned a score between zero and nine where a higher score reflects healthier eating.

The information from RiksmatenFlexDiet was also used to obtain information on intake of fruit, vegetable, candy, including chocolate and sugar-sweetened beverages and the proportion of added sugar to total energy intake (E%).

Parental education, country of birth and municipalities

Parents reported their educational attainment in the online questionnaire. The highest level of education attained by either of the parents was used, and the responses were dichotomised into low parental education (≤12 years of schooling) and high parental education (>12 years of schooling). The participants reported their country of birth as Sweden or outside Sweden. The municipalities of the participating schools were categorised into one of following five groups based on the Swedish municipality classification 2011: 1) metropolitan municipalities; 2) suburbs surrounding metropolitan areas; 3) larger cities and surrounding suburbs; 4) densely populated municipalities and 5) other municipalities (commuter municipalities, tourism and travel industry municipalities, manufacturing municipalities, sparsely populated municipalities and municipalities in sparsely populated regions). Groups 1, 2 and 3 were categorised as urban areas, and groups 4 and 5 as rural areas.

Anthropometry

Trained staff from the Swedish Food Agency measured the body height (0.1 cm) and weight (0.1 kg) using portable stadiometers (SECA 213) and digital weighing scales (SECA 862 or 899), respectively. The body mass index (BMI) was calculated (kg/m²), the body weight status was defined according to recommendations (Cole & Lobstein, 2012) and the BMI standard deviation (BMI-sds) score was calculated according to a Swedish reference standard (Karlberg et al., 2001).

Data analyses

In this paper, a sample of 2807 participants who provided information on LTOPA and dietary intake (from 2 days) were included in the analyses. Two days do not reflect habitual dietary intake, and to account for the day-to-day variation, the statistical method Multiple Source Method (MSM) was used to transform the intake from current to habitual intake (Harttig et al., 2011; Haubrock et al., 2011). Descriptive statistics are presented as means, standard deviations (SD) and proportions. Significance testing for associations between participation in LTOPA and dietary intake was performed with analysis of variance (unadjusted) and analysis of covariance (adjusted for parental education). Associations between the weekly frequency of participation in LTOPA (including those who did not participate) and dietary intake were also investigated with analysis of covariance (adjusted for parental education). If significant differences between categories of weekly frequency of participation in LTOPA were observed, pairwise comparisons were made, with a Bonferroni adjustment applied to the level of statistical significance. Effect sizes were reported as partial eta squared (ηp2) and interpreted as small (0.01), moderate (0.06) and large (0.14). An intra-class correlation of 0.095 indicates that 9.5% of the total variation of the SHEIA15 score was explained by adolescents attending the same schools. A linear mixed model approach was also tested with schools as a random intercept, but schools had a negligible effect on the analyses.

For dietary intake, the following six variables were analysed: overall healthy eating index (SHEIA15) and energy-adjusted intake (g/10 MJ) of fruits, vegetables, candy, sugar-sweetened beverages and the proportion of added sugar to total energy intake (E%). The analyses were performed separately for boys and girls but included all age groups since no interaction effect was observed for age. When significant; however, the results for boys and girls are shown to be stratified by age groups. Data analyses were conducted using IBM SPSS Statistics for Windows, Version 24.0. (IBM Corp. Armonk, New York, USA), and the level of statistical significance was set at p < 0.05.

Results

Descriptive characteristics among boys and girls who did not participate in LTOPA and those who participated are presented in Table 1. Those who provided information on LTOPA and diet were more likely to be girls, older and born in Sweden (all p < 0.05). Of the 2807 participants who were included in the analyses, 1886 participated in LTOPA. The most frequently reported LTOPAs were soccer, floorball, dance, martial arts, horseback riding, handball, gymnastics and ice hockey. Among those who participated in LTOPA, 1856 participants answered the question about weekly frequency of participation. Participation rates were as follows: 15% reported once per week; 21% twice per week; 24% three times per week; 21% four times per week; and 20% five times or more per week.

Table 1. Descriptive characteristics among boys and girls who did not participate in LTOPA and those who participated

	Boys, participated in LTOPA			Girls, participated in LTOPA
	No	Yes		No	Yes
	(n = 373)	(n = 861)	p	(n = 548)	(n = 1025)	p
Age (years) (mean ± SD) (n = 2805)	15.4 (±2.6)	13.9 (±2.4)	< 0.001	15.8 (±2.4)	14.0 (±2.4)	< 0.001
High parental education (%) (n = 2807)	52	64	< 0.001	55	66	< 0.001
Country of birth, Sweden (%) (n = 2801)	86	90	0.045	85	95	< 0.001
Type of municipality, urban (%) (n = 2807)	70	71	0.798	64	66	0.277
Weight status^a (%)^b (n = 2784)			< 0.001			0.001
Underweight	8	6		8	6
Normal weight	63	76	***	68	75	**
Overweight	21	15	*	18	16
Obese	9	3	***	6	3	**

Abbreviation: BMI, Body mass index

Data were analysed with Analysis of variance and Chi-Square test (Bonferroni adjusted post-hoc)

^aDefined according to International Obesity Task Force

^bTotals may not add up due to rounding

*p < 0.05; **p < 0.01; ***p < 0.001

Table 2 shows the results of analysis of variance (unadjusted) and analysis of covariance (adjusted for parental education) for dietary intake among boys and girls who did not participate in LTOPA and those who participated, respectively. In the adjusted analyses, no difference was observed between those who did not participate and those who participated for overall healthy eating index (p = 0.109 for boys; and p = 0.450 for girls). Boys who participated had a significantly higher (small effect size) fruit intake compared to those who did not participate (p = 0.043, ηp2 = 0.003), but no differences were observed among girls (p = 0.637). Figure 1 shows fruit intake among boys who participated and those who did not, stratified by age groups. The differences among boys for fruit intake were not significant when stratified by age groups: 11–12 (p = 0.081), 14–15 (p = 0.051) and 17–18 years (p = 0.063).

Table 2. Dietary intake among boys and girls who did not participate in LTOPA and those who participated

	Boys, participated in LTOPA				Girls, participated in LTOPA
	No (n = 373)	Yes (n = 861)	p	p adjusted*	No (n = 548)	Yes (n = 1025)	p	p adjusted*
Energy intake, kJ	9637.4 (2801.6)	9696.9 (2756.0)	0.729	0.854	8116.0 (2023.3)	8418.5 (1909.6)	0.003	0.006
Overall healthy eating index (SHEIA15) (mean ± SD)	5.6 (0.9)	5.7 (0.8)	0.039	0.109	5.9 (0.9)	5.9 (0.8)	0.831	0.450
Fruit, g
Fruit (g/10 MJ) (mean ± SD)	70.3 (65.8)	79.1 (61.6)	0.025	0.043	110.0 (83.5)	109.6 (76.4)	0.918	0.637
Vegetables, g
Vegetables (g/10 MJ) (mean ± SD)	160.9 (75.2)	157.8 (67.8)	0.480	0.344	197.5 (93.3)	183.2 (74.5)	0.001	< 0.001
Candy, g
Candy (g/10 MJ) (mean ± SD)	27.7 (22.4)	28.8 (23.1)	0.453	0.403	35.9 (26.0)	36.2 (24.9)	0.838	0.798
Sugar-sweetened beverages, g
Sugar-sweetened beverages (g/10 MJ) (mean ± SD)	250.4 (167.6)	215.2 (124.7)	< 0.001	< 0.001	238.7 (157.0)	215.6 (123.6)	0.001	0.007
Added sugar (E%) (mean ± SD)	10.2 (3.9)	10.0 (3.5)	0.276	0.307	11.1 (3.7)	10.8 (3.3)	0.197	0.227

*Adjusted for parental education.

Data were analysed with Analysis of variance (unadjusted) and Analysis of covariance (adjusted).

Figure 1. Dietary intake among boys and girls who did not participate in LTOPA (no) and those who participated (yes) stratified by age groups. Only those dietary variables being significant in Table 2 are shown. None of the differences remained significant when the analyses were stratified by age groups. Data were analysed with analysis of covariance and adjusted for parental education (presented as mean with 95% CI).

Moreover, no differences were observed between boys who did not participate in LTOPA and those who participated for vegetable intake (p = 0.344). However, girls who participated had a significantly lower (small effect size) vegetable intake compared to those who did not participate (p < 0.001, ηp2 = 0.009). Figure 1 shows vegetable intake among girls stratified by age groups. There were no significant differences among girls for vegetable intake when stratified by age groups: 11–12 (p = 0.081), 14–15 (p = 0.174) and 17–18 years (p = 0.248).

Candy intake did not differ between the two groups (p = 0.403 for boys and p = 0.798 for girls). However, sugar-sweetened beverage intake was lower (small effect size) among those who participated in LTOPA compared to those who did not among both boys (p < 0.001, ηp2 = 0.012) and girls (p = 0.007, ηp2 = 0.005; Figure 1). Among girls, the analyses for sugar-sweetened beverages were not significant when stratified by age groups: 11–12 (p = 0.154), 14–15 (p = 0.952) and 17–18 years (p = 0.363). Similar results were observed among boys: 11–12 (p = 0.847), 14–15 (p = 0.331) and 17–18 years (p = 0.067). The proportion of added sugar to the total energy intake did not differ between those who participated in LTOPA and those who did not among boys (p = 0.307) and girls (p = 0.227).

Moreover, Figure 2 shows dietary intake by weekly frequency of participation in LTOPA. Among boys, differences (small effect sizes) between weekly frequency of participation in LTOPA were shown for overall healthy eating index (p = 0.041, ηp2 = 0.009; differences between 0/w and 1/w: p = 0.041), sugar-sweetened beverage intake (p = 0.001, ηp2 = 0.017; differences between 0/w and 5+/w: p = 0.001), and the proportion of added sugar to total energy intake (p = 0.039, ηp2 = 0.010; differences between 2/w and 5+/w: p = 0.016). Among girls, differences (small effect sizes) were shown for vegetable intake (p = 0.011, ηp2 = 0.010; differences between 0/w and 3/w: p = 0.036). No differences were shown in the remaining analyses.

Figure 2. Dietary intake among boys (black lines) and girls (grey dashed lines) who did not participate in LTOPA (0/w) and those who participated once per week (1/w), twice per week (2/w), three times per week (3/w), four times per week (4/w), and five times or more per week (5+/w). Data were analysed with analysis of covariance (adjusted for parental education) and significant differences are indicated by asterisk (*p < 0.05; **p < 0.01) (presented as mean with 95% CI).

Discussion

This study investigated associations between participation in LTOPA and dietary intake in a large representative sample of Swedish adolescents. Overall, the main findings were that LTOPA was associated with few dietary variables. No differences were observed for overall healthy eating index, fruit (among girls), vegetables (among boys) and candy intake, as well as for the proportion of added sugar to total energy intake. When further analysing dietary intake by weekly frequency of participation in LTOPA, including those who did not participate, few significant differences were observed.

Previous studies show that participation in extra-curricular sport-related activities might be associated with increased consumption of fruits, vegetables and dairy (Kanellopoulou et al., 2021). Some large studies have also found associations between sport and overall healthier food choices, less intake of sweets and sugar-sweetened beverages and energy drinks (Dortch et al., 2014; Torstveit et al., 2018; Vella et al., 2013). This is similar to our findings that boys who participated in LTOPA had significantly higher fruit intake compared to those who did not and that sugar-sweetened beverage intake was lower among those boys and girls who participated in LTOPA. The effect sizes for these differences were small. However, the differences were not significant when stratified by age groups.

In contrast to some previous research (Dortch et al., 2014; Torstveit et al., 2018; Vella et al., 2013), we observed no association between LTOPA and overall healthy eating index among both boys and girls, fruit intake among girls, and no association for candy intake among both boys and girls. While it might be hypothesised that adolescents who participate in LTOPA are motivated to eat healthy, sports and exercise can also bring about circumstances that are less favourable in terms of dietary intake. For example, LTOPA requires time investments from both the adolescents who participate in sport and exercise and their parents in the form of transport to and from and being present during exercise sessions and matches/competitions (Bean et al., 2014). This can introduce time constraints that might affect meal planning and preparation and create situations where family meal habits are negatively affected (Bean et al., 2014; Nelson et al., 2011). Moreover, studies indicate that sports sponsorships with food and beverage companies might promote energy-dense, nutrient-poor products (Bragg et al., 2018). Although these sports sponsorships seldom explicitly target adolescents, such food and beverage marketing might affect food perceptions and preferences and send contradictory messages regarding acceptable food choices (Haubrock et al., 2011; Nelson et al., 2011). In addition, a recent systematic review showed that energy-dense and nutrient-poor foods and beverages, such as sports drinks, sweets, soft drinks and sausage sandwiches, are frequently available at sport venues and sports clubs (Carter et al., 2012). Collectively, these circumstances might lead to unhealthy dietary behaviours and explain the lack of clear association between LTOPA and a healthier dietary intake as observed in this study.

Although the effect size was small, this study showed that girls who did not participate in LTOPA had significantly higher vegetable intake compared to those who participated. This might be explained by several factors. Studies suggest a positive association between LTOPA and physical activity (Fröberg et al., 2020; Hebert et al., 2015; Marques et al., 2016). For example, accelerometer data from Riksmaten Adolescents showed that adolescents who participated in LTOPA were more physically active and more likely to reach the physical activity recommendations than those who did not participate (Fröberg et al., 2020). It is possible that girls who do not participate in LTOPA might view a healthy diet as a means to compensate for not being involved in any sports or exercise. It should also be noted that the differences among girls for vegetable intake were not significant when the analyses were stratified by age groups.

One strength of this study was the usage of a large sample of Swedish adolescents who overall were considered representative of the population concerning socioeconomic background, school organisation and type of municipality. However, the cross-sectional nature of the data precluded conclusions about causality and, as in all dietary studies, dietary intake may not be reported accurately. Energy under-reporting has previously been examined in the Riksmaten Adolescent and both under-reporting of unhealthy foods and over-reporting of healthy foods were seen. However, on the group level, energy reporting was found to be very plausible (Lemming et al., 2018). Our results must also be interpreted with some caution since we lack data on how dietary intake might change when adolescents begin to participate in LTOPA or increase their weekly frequency of participation.

In this study, dietary intake was measured with a self-administered web-based dietary assessment tool that has previously been validated against objectively measured physical activity and biomarkers. Although the web-based dietary assessment tool is comparable with estimates from 24-hour dietary recall interviews (Lindroos et al., 2019), it is likely that recall bias (misreporting) is present and this should be taken into account when interpreting the results. Another limitation is that the LTOPA questions have not been tested for reliability and validity. It is therefore possible that measurement errors exist in the LTOPA data. For example, recall bias might be present when the participants reported how many times per week they participated in their activities.

Conclusion

This study showed that LTOPA was associated with lower sugar-sweetened beverage intake among adolescent boys and girls and higher fruit intake among boys but lower vegetable intake among girls. However, an overall healthy eating index, candy intake and the proportion of added sugar to total energy intake were not associated with LTOPA. In conclusion, this study shows that LTOPA is only associated with some healthy eating behaviours and there is much room for improvement in the diets of Swedish adolescents. Places where adolescents participate in LTOPA are therefore one important setting to promote healthy dietary habits among adolescents.

Acknowledgments

The authors thank the school staff and students who participated in this study. We also thank other research team members who assisted with the data collection.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This study was funded by the Swedish Research Council for Sport Science.