SEARCH

SEARCH BY CITATION

Keywords:

  • Adolescent;
  • Body mass index;
  • Child;
  • Physical activity;
  • Sedentary lifestyle

Abstract

  1. Top of page
  2. Abstract
  3. Key notes
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Funding
  10. Conflict of interest
  11. References
  12. Supporting Information

Background:  Both reduced moderate-to-vigorous physical activity (MVPA) and increased screen time have been implicated in the aetiology of childhood overweight/obesity. This study aimed to determine which behaviour had the stronger association with overweight/obesity.

Method:  2200 randomly selected 9- to 16-year-old Australians provided four 24-h use-of-time recalls. Participants were classified into weight status categories and as high or low physical active, and high or low screen time according to Australian guidelines (≥60 min MVPA; ≤120 min recreational screen time daily). Multivariate logistic regression was used to calculate the odds ratios (OR) for overweight/obesity for each screen time and MVPA category.

Results:  Increased likelihood of overweight or obese was often associated with high screen time (ORs, 2.13–2.55 for boys and 1.47–1.72 for girls), but only sometimes and less strongly associated with low MVPA (ORs, 0.49–2.55 for boys and 1.06–1.47 for girls). Analyses conducted for combined screen time and MVPA categories showed screen time to be a stronger indicator of weight status than physical activity, especially in boys.

Conclusion:  Overweight and obesity were more strongly associated with screen time than physical activity. Screen time may be an important target for interventions aimed at reducing childhood overweight and obesity.


Key notes

  1. Top of page
  2. Abstract
  3. Key notes
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Funding
  10. Conflict of interest
  11. References
  12. Supporting Information
  •  It is unclear whether increase screen time or reduced moderate-to-vigorous physical activity (MVPA) has a stronger association with overweight and obesity in adolescents.
  •  Results showed that high screen time was more consistently and more strongly associated with overweight and obesity, and particularly so in boys.
  •  This suggests that screen time may be particularly important target behaviour for interventions aimed at reducing overweight and obesity amongst youth, and particularly boys.

Background

  1. Top of page
  2. Abstract
  3. Key notes
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Funding
  10. Conflict of interest
  11. References
  12. Supporting Information

Given the alarmingly high rates of paediatric obesity,(1) there is considerable attention on low levels of physical activity (typically captured as time spent in moderate-to-vigorous physical activity, MVPA) (2) and high levels of sedentary behaviour (typically determined as television viewing or total screen time) (3). Although often seen as being synonymous, these two behaviours work independently and through different causal mechanisms (4). However, they are also likely to be inter-related. Adequate levels of MVPA are possible only if time is taken from other activities, and screen time, which constitutes about 25% of waking time in young people (5), is likely to act as a ‘time-buffer’. In key time-periods, such as the after-school hours, MVPA and screen time are in direct competition. However, evidence examining the relationship between MVPA and screen time is equivocal (for review see Taylor and Sallis) (6).

Few studies have examined the joint association of screen time and MVPA on risk of overweight. In 7- to 12-year-old US children, Laurson et al.(7) found that children not meeting the physical activity and screen guidelines were 3–4 times more likely to be overweight than those complying with guidelines. Eisenmann et al. (8) similarly found high screen time coupled with low physical activity was associated with greatest risk of overweight among 14- to 18-year-olds from the US 2001 Youth Risk Behavior Survey. To date, the associations between physical activity, screen time and fatness have only been examined on the basis of overweight/obesity combined without considering whether these relationships vary differently for overweight and obesity. Furthermore, the joint relationships have only been studied in samples of young people from the US.

Therefore, the aim of the present study was to quantify the relative strengths of the cross-sectional associations between MVPA and screen time and the joint association of MVPA and screen time on the likelihood of being overweight and obese in 9- to 16-year-old Australians.

Methods

  1. Top of page
  2. Abstract
  3. Key notes
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Funding
  10. Conflict of interest
  11. References
  12. Supporting Information

Participants

The participants for this study were 2200 randomly selected Australians aged between 9 and 16 years, who took part in the Australian National Children’s Nutrition and Physical Activity Survey (methods described in detail in 9). The University of South Australia’s Human Research Ethics Committee approved the study, and participants provided written informed consent.

Outcome measures

Demographic data were gathered during a computer-assisted face-to-face interview. Height and body mass were measured by trained research personnel, from which body mass index (BMI, kg/m2) and weight status category (normal weight, overweight or obese) were determined using the International Obesity Task Force criteria (10).

Use-of-time data were collected using the Multimedia Activity Recall for Children and Adults (MARCA) (11). This software allows people to report everything they did on the previous day from wake-up to bedtime, in time intervals as short as 5 min. Young people chose from a list of about 250 activities grouped under seven main rubrics (Inactivity, Transport, Sport and Play, School, Self-Care, Chores and Other). The MARCA has a same-day test–retest reliability of r = 0.84–0.92 for major outcome variables [MVPA, physical activity level (PAL) and screen time] and convergent validity with reference to pedometer of rho = 0.54 for PAL (12). The MARCA was administered on two occasions. On each occasion, young people recalled their activities over the two previous days. On the first occasion, the MARCA was administered in a face-to-face interview, and on the second occasion, typically 1–3 weeks later, via a telephone interview. Mean daily minutes of MVPA were calculated as the number of minutes the participant reported being involved in activities expected to elicit at least 3 METs. Screen time was the number of minutes the participant reported watching television, playing videogames or using a computer.

Area-level socio-economic status was quantified using the Australian Bureau of Statistics’ Socio-economic indicators for areas (SEIFA) Index of Relative Disadvantage, which takes into account factors such as income and education.

Statistical analyses

Young people were divided into being high and low active, and high and low screen users by splitting them based on meeting or not meeting the Australian physical activity and screen time guidelines (<60 min/day for MVPA, and >120 min/day for screen time).(13) Each participant was allocated to one of four categories:

  •  Low Screen-High MVPA (reference category);
  •  Low Screen-Low MVPA;
  •  High Screen-High MVPA;
  •  High Screen-Low MVPA.

Logistic regression was used to calculate the odds ratios for being overweight alone, being obese alone and being overweight/obese combined for each of the screen time-MVPA categories for boys and for girls separately. Age and SEIFA were entered as independent variables for all logistic regression analyses. Alpha was set at 0.05. Analyses were undertaken using Statview for Windows Version 5.0.1.

Results

  1. Top of page
  2. Abstract
  3. Key notes
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Funding
  10. Conflict of interest
  11. References
  12. Supporting Information

Subject characteristics are shown in Table 1. Approximately 25% of the sample was overweight (19.0%) or obese (6.5%), with slightly higher percentages in females. About 80% of youth did not meet screen time recommendations, whereas about 20% did not meet the recommendations for physical activity. MVPA was negatively correlated with screen time in males (r = −0.39, p < 0.0001) and in females (r = −0.31, p < 0.0001).

Table 1.   Subject characteristics. Values are shown as percentages or means (SDs) or percentages
 BoysGirlsTotal
  1. BMI, Body Mass Index; SEIFA, Socio-economic indicators for areas (see text for a description of this metric).

n108911112200
Age (years)13.5 (2.2)13.4 (2.2)13.4 (2.2)
BMI (kg/m2)20.4 (3.9)21.1 (4.2)20.7 (4.1)
% obese5.87.16.5
% overweight17.520.519.0
Socioeconomic status (SEIFA; SD)1004 (66)1000 (63)1002 (65)
Mean screen time min/day (SD)255.4 (119.1)204.3 (71.0)229.5 (114.5)
Mean MVPA min/day (SD)136.8 (77.3)107.5 (71.0)122.0 (75.6)
% meeting screen time guidelines15.726.221.0
% meeting MVPA guidelines85.773.279.4

Amongst boys, the level of MVPA (adjusted for age and SEIFA, but not for screen time) was not significantly associated with the likelihood of being overweight [OR = 0.90 (95% CI = 0.55–1.45), p = 0.65] but was associated with the likelihood of being obese [OR = 2.35 (1.25–4.44) p = 0.008]. Furthermore, the level of screen time was associated with the likelihood of being overweight [OR = 2.54 (1.45–4.44) p = 0.0001] but not of being obese [OR = 1.75 (0.78–3.94) p = 0.18].

Amongst girls, the level of MVPA was not significantly associated with the likelihood of being overweight [OR = 0.76 (0.52–1.11) p = 0.16], nor the likelihood of being obese [OR = 1.56 (0.92–2.65) p = 0.10]. The level of screen time was significantly associated with the likelihood of being overweight [OR = 1.56 (1.08–2.23) p = 0.02] and the likelihood of being obese [OR = 1.90 (1.02–3.52) p = 0.04)].

Multivariate odds ratios were used to examine the joint relationships between MVPA and screen time and weight status. Table 2 shows the adjusted multivariate odds ratios for overweight and obesity combined and the percentage of subjects for the four MVPA-Screen time categories. Owing to the extremely small number of boys with a Low Screen-Low MVPA phenotype (n = 14), analyses were not performed for this sub-category. The prevalence of overweight or obesity was 39% (1.47/1.06 = 139%) to 113% (2.13/1) greater among those who did not meet the screen time guidelines than among those who did.

Table 2.   Percentages and multivariate odds ratios for overweight/obese combined for the four activity categories
Screen timeMVPATotal noverweight/obese combined (%)OR (95% confidence interval) p value
  1. OR, Odds Ratio.

  2. Statistically significant findings are highlighted in bold. Analyses are adjusted for age and SEIFA.

  3. *Analyses not performed owing to very small sample size.

Boys
 MeetMeet15713.41.00 (reference)
 MeetDo not meet14**
 Do not meetMeet77424.8 2.13 (1.30–3.48) p = 0.003
 Do not meetDo not meet14128.3 2.55 (1.39–4.66) p = 0.002
Girls
 MeetMeet24520.41.00 (reference)
 MeetDo not meet4719.21.06 (0.47–2.38) p = 0.88
 Do not meetMeet57231.7 1.72 (1.20–2.46) p = 0.003
 Do not meetDo not meet25027.21.47 (0.95–2.28) p = 0.08

The relationship between weight status and MVPA was less distinct; in two cases (girls with low screen time and boys with high screen time), the prevalence of overweight or obesity was 6% (1.06/1.00) to 20% (2.55/2.13) greater among adolescents who did not meet the MVPA guidelines compared to those who did, and in one case (girls with high screen time), low MVPA was actually associated with a 15% (1.47/1.72) reduction in likelihood of overweight or obese. While fewer than 20% of young people in the Low Screen-High MVPA category were overweight or obese, almost 30% of those in the High Screen-High MVPA and High Screen-Low MVPA categories were overweight or obese.

Analyses undertaken for overweight and obesity separately cast further light on the consistency, magnitude and strength of these relationships. Table 3 shows the adjusted odds ratios for overweight alone. Young people who did not meet screen time guidelines were between 31% (1.17/0.89) and 151% (2.51/1.00) more likely to be overweight than those who met the screen time guidelines. This trend was seen for both boys and girls, but was of greater magnitude amongst boys. Surprisingly, young people with low MVPA were between 11% and 30%less likely to be overweight than those with high MVPA.

Table 3.   Percentages and multivariate odds ratios for being overweight for the four activity categories
Screen timeMVPATotal nOverweight (%)OR (95% confidence interval) p value
  1. OR, Odds Ratio.

  2. Statistically significant findings are highlighted in bold. Analyses are adjusted for age and SEIFA.

  3. *Analyses not performed owing to very small sample size.

Boys
 MeetMeet1578.91.00 (Reference)
 MeetDo not meet14**
 Do not meetMeet77419.6 2.51 (1.40–4.48) p = 0.002
 Do not meetDo not meet14117.0 2.20 (1.07–4.54) p = 0.03
Girls
 MeetMeet24516.31.00 (Reference)
 MeetDo not meet4712.80.89 (0.352.27) p = 0.80
 Do not meetMeet57224.5 1.68 (1.13–2.48) p = 0.01
 Do not meetDo not meet25016.81.17 (0.711.92) p = 0.55

Table 4 shows the adjusted odds ratios for obesity alone. Young people did not meet the screen time guideline were 37% to 89% more likely to be obese than those with who met the screen time guideline, while participants who did not meet the physical activity guideline were 41–152% more likely to be obese than those who met the physical activity guideline. This pattern was consistent in both boys and girls. There were approximately two and a half times as many obese participants in the High Screen-Low MVPA category (10.7%) as in the Low Screen-High MVPA category (4.3%).

Table 4.   Percentages and multivariate odds ratios for being obese for the four activity categories
Screen timeMVPATotal nObese (%)OR (95% confidence interval) p value
  1. OR, Odds Ratio.

  2. Statistically significant findings are highlighted in bold. Analyses are adjusted for age and SEIFA.

  3. *Analyses not performed owing to very small sample size.

Boys
 MeetMeet1574.51.00 (Reference)
 MeetDo not meet14**
 Do not meetMeet7745.21.37 (0.603.13) p = 0.46
 Do not meetDo not meet14111.3 3.45 (1.32–9.07) p = 0.01
Girls
 MeetMeet2454.11.00 (Reference)
 MeetDo not meet476.41.78 (0.457.00) p = 0.41
 Do not meetMeet5727.21.89 (0.933.88) p = 0.08
 Do not meetDo not meet25010.4 2.67 (1.21–5.88) p = 0.02

While analyses based on recommended physical activity and screen time guidelines are useful from a policy-making/practical perspective, these cut points are somewhat arbitrary. Therefore, the data were re-examined by splitting the MVPA and screen time data into high vs. low categories on the basis of median cut points (Table S1). These analyses showed the same general patterns seen when analysed on the basis of guideline cut points, however, the magnitude of associations increased.

Discussion

  1. Top of page
  2. Abstract
  3. Key notes
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Funding
  10. Conflict of interest
  11. References
  12. Supporting Information

This study found that compliance with screen time guidelines was a stronger indicator of weight status than physical activity guideline compliance. The relationships between weight status and screen behaviour and MVPA were more consistent and of greater magnitudes in boys compared to girls. Strengths of this study include the large random sample, the wide age range and the fact that physical activity was assessed using a validated and reliable methodology that yielded very high-resolution use-of-time data.

The finding that increased weight status was more strongly and consistently associated with screen behaviour than MVPA indicates that screen time is interacting with weight status through pathways over and above displacement of physical activity. It has been suggested that screen time may lead to increased energy intake. For example, television viewing has been associated with snacking behaviour (14). Furthermore, children with high levels of screen time are exposed to more advertisements for energy-dense foods (15), which has been shown to influence food choices at other times of the day (16). In addition, evidence from studies conducted in adults suggests that sedentary time may lead to altered metabolic processes (17), with consequences for weight status. Finally, it is possible that some third factor may lead to both increased screen time and increased risk of overweight. For example, recent studies have found an association between low sleep duration and risk of overweight in children (18,19). It is likely that high levels of screen time will impact children’s sleep as well.

An interesting finding from the current study was that the relationships between weight status and screen behaviour and MVPA were more consistent, and of greater magnitude, in boys than in girls. The reasons for this are unclear. Boys participated in more MVPA and more screen time each day. Given that this study identified an important relationship between weight status and screen behaviour in particular, it is possible that boys’ greater participation in these activities accentuated the strength of relationship between weight status and activity. Alternatively, it is possible that other factors not accounted for in the current analyses (e.g. psychosocial factors) play a greater role in influencing girls’ weight status than boys’.

Our main finding that weight status was more consistently and strongly associated with screen time than physical activity contrast with those of the two US studies conducted with this age group. Laurson et al. (7) found a tendency for physical activity to be more important, whereas Eisenmann et al. (8) showed physical activity and screen time were of equal importance. One possible explanation for the discrepancy in findings amongst studies is methodological variation between the studies. Whilst both Laurson et al. (7) and the current study classified the physical activity data on the basis of meeting MVPA guidelines (which recommend 60 min a day for both US and Australian children), in our study, daily MVPA was calculated from use-of-time recalls, and 80% of participants complied with the guideline. In contrast, Laurson et al. (7) measured MVPA using pedometer step counts and categorized compliance based on a cut-off of 11,000 (girls) or 13,000 (boys) average daily steps. On this basis, only 44% of children met the physical activity guideline. It is unclear whether the large difference in the number of Australian and US children meeting the guideline is attributed to true differences between these samples, or to methodological differences. However, it is important to recognize that there are considerable cultural and climatic differences between the United States and Australia, so the relationships between weight status and activity patterns may be truly different between countries.

In the past two decades, young people’s physical activity behaviour has been the subject of scores of interventions. In comparison, interventions aimed at reducing screen time have been relatively uncommon. Our findings suggest that screen time reduction may be a more effective target behaviour for intervention studies aimed at reducing overweight and obesity amongst children. Having said this, the value of physical activity interventions should not be underestimated, given the well-recognised social and psychological benefits of sports participation and physical activity in general.

Conclusion

  1. Top of page
  2. Abstract
  3. Key notes
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Funding
  10. Conflict of interest
  11. References
  12. Supporting Information

Increased likelihood of a young person being overweight or obese was more strongly and consistently associated with high screen time than with low physical activity. Findings underscore the need for interventions targeting screen behaviour in young people.

Funding

  1. Top of page
  2. Abstract
  3. Key notes
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Funding
  10. Conflict of interest
  11. References
  12. Supporting Information

This study was supported by the Australian Commonwealth Department of Health and Ageing, the Department of Agriculture, Fisheries and Forestry and the Australian Food and Grocery Council, through the National Children’s Nutrition and Physical Activity Survey.

References

  1. Top of page
  2. Abstract
  3. Key notes
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Funding
  10. Conflict of interest
  11. References
  12. Supporting Information
  • 1
    Wang Y, Lobstein T. Worldwide trends in childhood overweight and obesity. Int J Pediatr Obes 2006; 1: 1125.
  • 2
    Janssen I, Katzmarzyk PT, Boyce WF, Vereecken C, Mulvihill C, Roberts C, et al. Comparison of overweight and obesity prevalence in school-aged youth from 34 countries and their relationships with physical activity and dietary patterns. Obes Rev 2005; 6: 12332.
  • 3
    Tremblay MS, Willms JD. Is the Canadian childhood obesity epidemic related to physical inactivity? Int J Obes Relat Metab Disord 2003; 27: 11005.
  • 4
    Biddle S, Gorely T, Marshall S, Murdey I, Cameron N. Physical activity and sedentary behaviours in youth: issues and controversies. J R Soc Promot Health 2004; 124: 2933.
  • 5
    Olds T, Ridley K, Dollman J. Screenieboppers and extreme screenies: the place of screen time in the time budgets of 10-13 year-old Australian children. Aust N Z J Public Health 2006; 30: 13742.
  • 6
    Taylor W, Sallis J. Determinants of physical activity in children. In: Simopoulos A, Pavlou K, editors. Nutrition and Fitness: Metabolic and Behavioural Aspects in Health and Disease, World Review of Nutrition and Dietetics. Basel (CHE): Karger; 1997.
  • 7
    Laurson K, Eisenmann J, Welk G, Wickel E, Gentile D, Walsh D. Combined influence of physical activity and screen time recommendations on childhood overweight. J Pediatr 2008; 153: 20914.
  • 8
    Eisenmann J, Bartee R, Smith D, Welk G, Fu Q. Combined influence of physical activity and television viewing on the risk of overweight in US youth. Int J Obes (London). 2008; 32: 6138.
  • 9
    Department of Health and Ageing. 2007 Australian National Children’s Nutrition and Physical Activity Survey - Main Findings. Canberra: Department of Health and Ageing; 2008.
  • 10
    Cole TJ, Flegal KM, Nicholls D, Jackson AA. Body mass index cut offs to define thinness in children and adolescents. BMJ 2007; 335: Epub.
  • 11
    Ridley K, Olds TS, Hill A. The Multimedia Activity Recall for Children and Adolescents (MARCA): development and evaluation. Int J Behav Nutr Phys Act 2006; 3: Epub.
  • 12
    Olds TS, Ridley K, Dollman J, Maher CA. The validity of a computerized use of time recall, the multimedia activity recall for children and adolescents. Pediatr Exerc Sci 2010; 22: 3443.
  • 13
    Department of Health and Ageing. Australia’s Physical Activity Recommendations for Children and Young People. Australian Commonwealth: Department of Health and Ageing; 2005.
  • 14
    Salmon J, Campbell K, Crawford D. Television viewing habits associated with obesity risk factors: a survey of Melbourne schoolchildren. Med J Aust 2006; 184: 647.
  • 15
    Neville L, Thomas M, Bauman A. Food advertising on Australian television: the extent of children’s exposure. Health Promot Int. 2005; 20: 10512.
  • 16
    Borzekowski D, Robinson T. The 30-second effect: an experiment revealing the impact of television commercials on food preferences of preschoolers. J Am Diet Assoc 2001; 101: 426.
  • 17
    Hamilton M, Healy G, Dunstan D, Zderic T, Owen N. Too little exercise and too much sitting: inactivity physiology and the need for new recommendations on sedentary behavior. Curr Cardiovasc Risk Rep 2008; 2: 2928.
  • 18
    Eisenmann J, Ekkekakis P, Holmes M. Sleep duration and overweight among Australian children and adolescents. Acta Paediatrica. 2006; 95: 95663.
  • 19
    Olds T, Blunden S, Dollman J, Maher C. Day type and the relationship between weight status and sleep duration in children and adolescents. Aust N Z J Public Health 2010; 34: 16571.

Supporting Information

  1. Top of page
  2. Abstract
  3. Key notes
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Funding
  10. Conflict of interest
  11. References
  12. Supporting Information

Table S1. Percentages and multivariate odds ratios for being obese and/or overweight when activity is categorised on the basis of MVPA and screen time median splits.

FilenameFormatSizeDescription
APA_2804_sm_TableS1.doc40KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.