¹School of Health Sciences, Deakin University, 221 Burwood Hwy, Burwood, Victoria 3125, Australia

²School of Community Medicine, University of New South Wales, Sydney, New South Wales 2052, Australia

Correspondence to: Adrian Bauman, Locked Mail Bag 17, Liverpool NSW 2170, Australia a.bauman@unsw.edu.au

OBJECTIVES: To investigate the effect of physical activity on the association between television viewing and overweight (body mass index (BMI) 25 kg/m²).

DESIGN: Cross-sectional study administered by interview to adults randomly selected from the electronic white pages.

SUBJECTS: 3392 adults (64% response rate) from a representative population sample in the State of New South Wales, Australia.

MEASUREMENTS: Self-reported height and weight, two-week leisure-time physical activity recall, one-week average television viewing recall.

RESULTS: BMI and physical activity patterns were both associated with hours of television watched. Compared to those participants who reported watching less than one hour of television per day, those watching 1 to 2.5 hours were 93% more likely to be overweight (BMI25 kg/m²), those watching 2.5 to 4 hours were 183% more likely to be overweight, those watching more than 4 hours per day were four times more likely to be overweight. Physical activity was not directly associated with being overweight, but an interaction between activity and television watching was present. Respondents in the low, moderate and high physical activity categories who reported watching more than 4 hours of television per day were twice as likely to be overweight compared to those who watched less than one hour of television per day, irrespective of physical activity participation.

CONCLUSIONS: With approximately half the Australian adult population overweight or obese, these findings indicate that public health strategies to reduce overweight and prevent weight gain may need to focus on reducing sedentary behaviours such as television viewing in addition to increasing physical activity.

International Journal of Obesity (2000) 24, 600-606

leisure-time physical activity; television viewing; overweight; adults

Introduction

Physical inactivity is associated cross-sectionally with overweight, and longitudinally with weight gain.^1,2,3 Western nations have a high prevalence of overweight and current trends indicate that the incidence of overweight is increasing rapidly, becoming more common even in developing nations.⁴ In 1995, 66% of Australian men and 46% of Australian women aged over 19 years were classified as overweight or obese (Body Mass Index (BMI)>25 kg/m²).⁵ Recent increases in rates of overweight are thought to be occurring against a relatively static metabolic backdrop, at a time when total energy intake is in decline. Increasing rates of overweight have been attributed mainly to increasingly sedentary lifestyles.^6,7

Television viewing is a common sedentary behaviour. Australian adults spend an average of two hours per day watching television.⁸ Examining the association between television viewing, overweight and physical activity has become a focus of research into the determinants of overweight. The majority of television studies have involved children and although not all studies of television viewing habits and body mass index of children have found consistent associations,^9,10,11,12 many have shown a positive association.^{13,14,15,16,17,18,19} In addition, studies of the association between television viewing and physical activity for children have presented mixed findings.^{9,10,11,12,14,16,17,20}

Fewer studies have examined the relationship between overweight and obesity, television viewing, and physical activity in adults. One study of adult males²¹ and another of adult females,²² found significant positive associations between body fat percentage and hours of television viewing. A positive association between television viewing and BMI was reported among adult Pima Indians, together with a negative relationship between television viewing and physical activity.²³ A longitudinal study of adult male health professionals found positive associations between BMI and television viewing, independent of physical activity level.²⁴ A recent longitudinal study of adults participating in a weight gain prevention study found television viewing predicted weight gain in high-income women at one-year follow-up,²⁵ however at the 3-year follow-up there was no significant association.²⁶ All of these studies used non-representative adult population samples. Therefore, it may be that the associations found were particular to participants in these studies.

We used data derived from a population sample of adult Australians to examine whether physical activity modifies the association between television viewing and overweight. It was hypothesised that any positive association between television viewing and overweight would be weakened with increased levels of participation in physical activity.

Methods

Subjects

The data reported here are derived from the 1996 state Physical Activity Survey in New South Wales (NSW), Australia. This was a cross-sectional study designed to examine participation in physical activity during leisure-time and at work, among adults in NSW. The survey was administered by telephone interview to a random sample of adults from the electronic white pages. Prior to receiving a telephone call, the selected households were mailed a letter explaining the survey. An adult from each household was randomly selected using the 'next birthday' method, where the household resident with the next birthday was defined as the sampled individual. The survey was administered using a computer-assisted telephone interview (CATI) system and took a median of 16 to 17 minutes to complete. The total number of household contact calls made was 5303, of which 3392 interviews were successfully completed, representing a 64% response rate.

Measures

The interview included questions regarding sex, age, country of birth, language spoken at home, educational attainment, post-code and suburb or town of residence. Television viewing habits were assessed by asking: 'How many hours do you spend watching TV and/or videos on a typical weekday?' This question was repeated for a typical weekend day. BMI was derived from self-reported weight and height. Using nationally-representative Australian data it has been shown the correlation between measured and self-reported height is 0.93 for men and 0.92 for women, and that the correlation between measured and self-reported weight is 0.97 for men and 0.98 for women.²⁷ BMI was calculated as weight (kg)/height² (m).

To assess leisure-time physical activity, global measures of walking, moderate and vigorous physical activity were used. For example, the global walking question asked: 'In the past two weeks, how many times have you walked (whether for recreation or exercise or to get to or from places) for at least 10 minutes continuously? Following this was a question asking about total time spent walking in this way during the past two weeks. The type, frequency, duration and intensity of vigorous and moderate activity performed in the past two weeks were also assessed. For example, the vigorous activity question asked: 'Which two vigorous activities did you spend most time doing in the past two weeks?' Respondents were further prompted for other vigorous activities they may have spent time doing in the past two weeks. These questions were then repeated for moderate activities. The physical activity measures used were similar to those used in previous Australian population surveys of physical activity,^28,29 and have been shown to have good reliability and validity.^30,31

Statistical Analyses

Following the system recommended by the National Health & Medical Research Council, a categorical variable was created for BMI which classified respondents as 'not overweight' (<25 kg/m²) and 'overweight' (25 kg/m²).³²

A continuous variable of leisure-time energy expenditure (kcal/week) was created using metabolic equivalent values (Mets) for each type of physical activity.^33,34 The summed total time spent walking and in moderate activities was multiplied by 3.5 Mets, then by body weight (kg). The total time spent in vigorous activities was multiplied by 9.0 Mets, and by body weight. These figures were summed and divided by two to provide an estimate of weekly energy expenditure expressed as kcal/week.^30,31,35

Based on this continuous variable of energy expenditure (kcal/week), four levels of physical activity were created.^28,29 Respondents were classified in the 'inactive' category if their energy expenditure was <50 kcal/week, in the 'low' physical activity category if energy expenditure was between 50 and 799 kcal/week, in the 'moderate' physical activity category if energy expenditure was between 800 and 1600 kcal/week, and in the 'high' physical activity category if estimated energy expenditure was >1600 kcal/week. The moderate kcal/week cut points of physical activity are based on the amount of energy expenditure recommended by the US Surgeon General as being adequate for health benefits.³⁶

Hours of reported television viewing on a typical weekday were multiplied by five, and hours watching television on a typical weekend day were multiplied by two, these were then summed and divided by seven to estimate the average television watched per day. Four categories of daily average television viewing were created: <1 hour, 1-2.5 hours, 2.51-4 hours, and >4 hours.

Using SPSS version 8.0, Chi-square (²) tests were performed to examine differences between socio-demographic variables and BMI, physical activity, and television viewing. A series of one-way ANOVAs were performed within each level of physical activity, examining differences in BMI means for each of the four categories of television viewing. Forced entry logistic regression analyses were performed for age, sex, education, television viewing and physical activity levels predicting overweight and obesity (BMI25 kg/m²). Interaction terms were tested for in the ANOVA and logistic regression models. Global interaction terms were used, based on the categorical forms of the television viewing and physical activity variables; in the ANOVA model, only two way interactions were sought. Four separate logistic regression models explored the association between television watching and being overweight at each level of physical activity.

Results

Sample characteristics

Of the 3392 participants in the survey, 1555 (45.8%) were males, and 1837 (54.2%) were females (Table 1). Compared to Census data,³⁷ females, those with a university education, and those over 60 y were over-represented in the sample.

Demographic data (age, gender and household size) were weighted to the NSW population of 4.22 million adults aged 18 years and over.³⁷ Although the data were weighted to the prevalence estimates of the NSW population, the sample size remained at 3392 to enable statistical tests to be performed. All statistical analyses in this paper use weighted data.

Body Mass Index

The mean BMI for males was 25.6 kg/m² (s.d.=3.8), and for females was 24.1 kg/m² (s.d.=1.4). There were more overweight and obese (BMI25 kg/m²) men (51.6%) than women (33.8%) in the sample. The proportions of the sample overweight or obese, by socio-demographic characteristics, are presented in Table 2. Overweight and obesity was more common in those aged over 45 y, those with less education, and those not in paid employment. Overweight prevalence increased with increasing age (up to 45-59 y), and decreased with increasing education.

Physical activity

The percentage distribution of leisure-time physical activity levels by selected sociodemographic characteristics is presented in Table 3. Physical inactivity was more common in women, those aged over 60 y, those with less education, those with no paid work, and those born in non-English speaking countries. Physical activity category was not related to BMI (²_trend=1.76, P=0.18).

There was a significant difference between physical activity groups and BMI categories (²=11.4, df=2, P<0.01). A high proportion of those in the inactive group were overweight (47%) compared to those in the low (39%), moderate (44%), and high (41%) physical activity categories (data not shown).

Television viewing

The average time spent watching television by respondents was 2.4 hours per day (s.d.=1.4). There was no significant difference in the time spent watching television between males and females. The hours of daily television viewing at the 10th percentile was 0.75 (or 45 minutes), and at the 90th percentile was 4.1. The distribution of television viewing by sociodemographic characteristics is presented in Table 4.

Respondents reporting watching an average of more than 4 hours television per day were more likely to be female, over 60 y, in the low or inactive physical activity categories, overweight or obese, less educated, and with no paid employment.

Overweight, television viewing and physical activity

The relationship between BMI and television viewing across the four levels of physical activity is displayed in Figure 1, where BMI increased with increasing television watching. For example, the prevalence of overweight was 29.7% among those who watch <1 hour of TV per day, increasing to 50.4% of those who watched at least four hours of TV per day (²_trend=41.3, P£0.001). This is also illustrated in Figure 1. One-way ANOVAs revealed significant differences in mean BMI with increased hours of television viewing for those in the high level of physical activity category (P<0.001), moderate (P<0.001), and low (P<0.01), but not for those in the inactive category. For those in the high physical activity category, the mean BMI (±s.d.) was 23.1 (±3.6) for those watching <1 hour of television per day compared to 26.3 (±3.9) for those watching >4 hours per day. The mean BMI of respondents in the moderate physical activity category who reported watching less than one-hour of television per day was 23.7 (±3.9) compared to 26.1 (±5.7) for those watching more than 4 hours. For those in the low physical activity level watching less than one hour of television per day, mean BMI was 23.6 (±3.9) compared to 25.0 (±4.3) for those watching more than 4 hours per day. An overall model was fitted to these data, including an interaction term between activity and TV watching; this two-way interaction term was nearly significant (F₉=1.74, P=0.08).

Logistic regression modelling, adjusting for age, sex, education, employment status, physical activity level, and television viewing, was performed to assess the likelihood of being in the overweight or obese BMI category (25 kg/m²) (Table 5). Age, sex and education were all significantly associated with overweight. Odds Ratios for overweight decreased with increasing education. The likelihood of being overweight increased with increased hours of television viewing. Analyses were conducted for each category of physical activity, and the sedentary level of activity was associated with increased BMI compared to the most active category. An interaction term was fitted, representing the interaction between hours of television viewing and physical activity. Since the interaction was significant separate models were fitted (Table 6).

The first model tested the relationship between BMI and television viewing for the inactive respondents, the second model tested the relationship between BMI and television viewing for respondents in the low activity category, and so on.

Apart from the respondents in the inactive category, increased hours of television viewing was significantly related to overweight within each level of physical activity. For those in the low physical activity category (Model 2), the likelihood of overweight increased significantly with each category of television viewing >1 hour per day. Those watching between 1-2.5 hours of television per day were 57% more likely to be overweight; those watching 2.5-4 hours were 91% more likely to be overweight; and those watching >4 hours per day were two times more likely to be overweight. For respondents in the moderate physical activity category the likelihood of overweight increased for those watching >2.5 hours of television viewing per day. Those in the high physical activity category who reported watching between 2.5-4 hours per day were 85% more likely, and those watching >4 hours of television per day were two times more likely to be overweight compared to those watching <1 hour per day. The trends are striking in Models 2-4. At all levels of physical activity, except for the inactive category (Model 1), odds ratios for overweight increased with increasing television viewing time.

Discussion

The results of this study indicated that even for adults who are physically active, watching more than four hours of television per day was associated with at least a two-fold increased risk of being overweight. This finding partially supports other longitudinal studies that have found a positive association between overweight and television viewing independent of physical activity.²⁴ However in this study the relationship between television viewing and overweight was modified by physical activity; a finding consistent with other cross-sectional studies that have found physical activity or physical fitness to modify this relationship.^21,22 Our study has produced mixed results, in that there was a lack of a clear relationship between overweight and television viewing for the most sedentary category but a relationship appeared to exist for other activity categories. This suggests that some active people may compensate for their participation in physical activity by increased nutritional intake or increased sedentary periods in other parts of their day. Furthermore, the study was the first to examine television viewing, physical activity and overweight using data from an adult population-based representative sample.

The association between television viewing and overweight, even for those who were physically active, may be because television viewing is a proxy 'marker' for other behaviours that lead to overweight, such as other sedentary behaviours or dietary behaviours. There is some evidence that prolonged hours of television viewing and overweight may be associated with the intake of foods high in saturated fat. A number of studies have examined the association between diet and television viewing in children and adolescents.^38,39,40,41 These have shown that television viewing is negatively associated with consumption of 'healthy foods' and positively associated with foods high in saturated fat. A recent longitudinal study of a self-selected sample of adults also found a positive association between television viewing and energy intake, as well as percentage of energy from fat.²⁵ However it is not possible to determine from these studies whether the relationships are due to the intake of particular foods while watching television, or indicative of generally unhealthy lifestyle patterns.

Although our findings indicated an association between television viewing and overweight, this relationship was not entirely independent of physical activity. For those participating in moderate levels of physical activity, only those watching >2.5 hours of television per day were at increased risk of overweight. However, in the low, moderate, and high physical activity levels, there were obvious trends of increasing odds ratios with increased television watching, which were strongest (and significant) at the higher levels of television watching. Among the most active, 9% watched at least 4 hours of television daily, and among the moderately active, 43% watched at least 2.5 hours daily; and these levels of television watching placed these individuals at risk of overweight, in spite of their 'sufficient' leisure time activity patterns. It may also be that in addition to television viewing, much of this group's leisure time was spent in other sedentary behaviours. This study suggests that inactivity may be associated with being overweight, but across activity categories television viewing time may be a marker or proxy measure of overweight risk.

Ecological studies in Britain suggest that an increase in sedentary behaviours and decrease in total energy expenditure may be contributing to the prevalence of overweight in developed countries.⁶ The total energy expenditure throughout the day is important, not only the small parts of the day spent in purposive physical activity.³⁶ Therefore, it is important to focus on incidental physical activity as an integrated part of everyday life and tasks, in addition to reductions in sedentary behaviours such as television viewing.

The strengths of our study are that it is based on a population-representative sample, it employed valid and reliable methods to assess physical activity and BMI,^27,30,31 the prevalence of physical activity and overweight in the study sample were similar to other Australian population estimates,^28,29,37 and the average time spent watching television was similar to other population data.⁸ The modest response rate of 64% was based on household level responders; if the randomly sampled adult within the household was contacted, the response rate among them was 81%. Thus the reported response rate here is a conservative estimate.

A further limitation of our study is that the data on physical activity, television viewing and BMI are cross-sectional, and it is therefore not possible to establish causal relationships. For example, greater fatness may lead to decreased physical activity and increased television viewing, as well as vice versa. Nonetheless, our study poses new questions regarding the inter-relationship among physical activity, television viewing and overweight. Even those who are active in leisure- time physical activity may be overweight if they engage in enough sedentary behaviour.

In conclusion, this study shows that adults who watch more than four hours of television daily are significantly more likely to be overweight, even if they are physically active. Thus being very active, even for recommended amounts of time per week, may not protect against obesity in the presence of prolonged sedentary behavior time, such as television watching. These findings suggest that, in order to reduce the prevalence of overweight and obesity, or at least prevent weight gain, it may be not only important to increase participation in purposive physical activity, but also to reduce the time spent in sedentary behaviours. Longitudinal studies are required to further examine the potential causal nature of these relationships between the development of overweight, television viewing and other sedentary behaviours, dietary behaviours and physical activity.

Acknowledgements

Data collection was supported by the New South Wales Health Department. David Crawford is supported by a Public Health Fellowship from the National Health and Medical Research Council.

1 Foreyt JP, Brunner RL, Goodrich GK, St Jeor ST, Miller GD. Psychological correlates of reported physical activity in normal-weight and obese adults: the Reno diet-heart study. Int J Obes Relat Metab Disord 1995 19; S69-S72.

2 Tremblay A, Deprès JP, Leblanc C, Craig CL, Ferris B, Stephens T, Bouchard C. Effect of intensity of physical activity on body fatness and fat distribution. Am J Clin Nutr 1990; 51: 153-157. MEDLINE

3 Williamson DF, Madans J, Anda RF, Kleinman JC, Kahn HS, Byers T. Recreational physical activity and ten-year weight-change in a US national cohort. Int J Obes Relat Metab Disord 1993; 17: 279-286. MEDLINE

4 Popkin BM, Doak CM. The obesity epidemic is a worldwide phenomenon. Nutr Rev 1998; 56: 106-114. MEDLINE

5 Australian Bureau of Statistics. National Nutrition Survey: Selected Highlights Australia 1995. Catalogue no. 4802.0, Australian Government Printing Service: Canberra, 1997.

6 Prentice AM, Jebb SA. Obesity in Britain: gluttony or sloth? Br Med J 1995; 311: 437-439.

7 Wiseman MJ. Behaviour change in practice: Population strategies. Int J Obesity 1996; 20: S31-S333.

8 Australian Bureau of Statistics. How Australians use their time. Australian Government Publishing Service: Canberra. 1994;

9 Armstrong CA, Sallis JF, Alcaraz JE, Kolody B, McKenzie TL, Hovell MF. Children's television, body fat, and physical fitness. Am J Health Promot 1998; 12: 363-368. MEDLINE

10 Myers L, Strikiniller PK, Webber LS, Berenson GS. Physical and sedentary activity in school children grades 5-8: The Bogalusa heart study. Med Sci Sport Exerc 1996; 28: 852-859 ,

11 DuRant RH, Baranowski T, Johnson M, Thompson WO. The relationship among television watching, physical activity, and body composition of young children. Pediatrics 1994; 94: 449-455. MEDLINE

12 Wolf AM, Gortmaker SL, Cheung L, Gray HM, Herzog DB, Colditz GA. Activity, inactivity, and obesity: Racial, ethnic, and age differences among schoolgirls. Am J Public Hlth 1993; 83: 1625-1627.

13 Andersen RE, Crespo CJ, Bartlett SJ, Cheskin U, Pratt M. Relationship of physical activity and television watching with body weight and level of fatness among children. JAMA 1998; 279: 938-942. MEDLINE

14 Robinson TN, Killen JD. Ethnic and gender differences in the relationships between television viewing and obesity, physical activity, and dietary fat intake. J Hlth Educ 1995; 26: S91-S98.

15 Obarzanek E, Schreiber GB, Crawford PB, Goldman SR, Barrier BM, Frederick MM, Lakatos E. Energy intake and physical activity in relation to indexes of body fat: the National Heart, Lung, and Blood Institute Growth and Health Study. Am J Clin Nutr 1994; 60: 15-22. MEDLINE

16 Pate RR, Ross JG. Factors associated with health-related fitness. J Phys Educ Recreation Dance 1987; Nov-Dec: 93-95.

17 Tucker L. The relationship of television viewing to physical fitness and obesity. Adolescence 1986; 21: 780-806.

18 Dietz WH, Gortmaker SL. Do we fatten our children at the telvision set? Obesity and television viewing in children and adolescents. Pediatrics 1985; 75: 807-812. MEDLINE

19 Gortinaker SL, Must A, Sobol AM, Peterson K, Colditz GA, Dietz WH. Television viewing as a cause of increasing obesity among children in the United States, 1986-1990. Arch Pediatr Adolesc Med 1996; 150: 356-362. MEDLINE

20 Robinson TN, Hammer LD, Killen JD, Kraemer HC, Wilson DM, Hayward C, Barr Taylor C. Does television viewing increase obesity and reduce physical activity? Cross-sectional and longitudinal analyses among adolescent girls. Pediatrics 1993; 91: 273-280. MEDLINE

21 Tucker LA, Friedman GM. Television viewing and obesity in adult males. Am J Public Health 1989; 79: 516-518. MEDLINE

22 Tucker LA, Bagwell M. Television viewing and obesity in adult females. Am J Public Health 1991; 81: 908-911. MEDLINE

23 Fitzgerald SJ, Kriska AM, Pereira MA, deCourten MP. Associations among physical activity, television watching, and obesity in adult Puma Indians. Med Sci Sport Exerc 1997; 29: 910-915.

24 Ching PLYH, Willett WC, Rimm EB, Colditz GA, Gortmaker SL, Stampfer MJ. Activity level and risk of overweight in male health professionals. Am J Public Hlth 1996; 86: 277-280.

25 Jeffery RW, French SA. Epidemic obesity in the United States: Are fast foods and television viewing contributing? Am J Public Hlth 1998; 88: 277-280.

26 Crawford D, Jeffery RW, French SA. Television viewing, physical inactivity and obesity. Int J Obesity 1999; 23: 437-440.

27 Water AM. Assessment of self-reported height and weight and their use in the determination of Body Mass Index. Australian Institute of Health and Welfare, Canberra 1993.

28 Department of the Arts, Sport, the Environment and Territories. Pilot survey of the fitness of Australians. Australian Government Publishing Service: Canberra, 1992.

29 Bauman A, Bellew B, Booth M, Hahn A, Stoker L, Thomas M. Towards best practice for the promotion of physical activity in the areas of NSW. NSW Health Department: Centre for Disease Prevention and Health 1996.

30 Booth M, Owen N, Bauman A, Gore CJ. Relationship between a fourteen-day recall measure of leisure-time physical activity and a sub-maximal test of physical work capacity in a population sample of Australian adults. Res Q Exercise Sport 1996 67: 221-227.

31 Booth M, Owen N, Bauman A, Gore CJ. Repeatability of self-reported leisure-time physical activity measures for population surveys. Int J Epidemiol 1996; 25: 153-159. MEDLINE

32 National Health and Medical Research Council. Report of the one hundredth session. Australian Government Publishing Service: Canberra, 1985.

33 Ainsworth BE, Haskell WL, Leon AS, Jacobs DR, Montoye HJ, Sallis JF, Paffenbarger RS. Compendium of physical activities: Classification of energy costs of human physical activities. Med Sci Sport Exerc 1993; 25: 71-80.

34 Montoye HJ, Kemper HCG, Saris WHM, Washburn RA. Measuring physical activity and energy expenditure. Human Kinetics: Champaign, Illinois, 1996.

35 Booth M, Bauman A, Owen N, Gore CJ. Physical activity preferences, preferred sources of assistance, and perceived barriers to increased activity among physically inactive Australians. Prev Med 1997; 26: 131-137. MEDLINE

36 US Department of Health and Human Services. Physical Activity and Health: A Report of the Surgeon General US Dept of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion: Atlanta, GA, 1996.

37 CDATA 1996, Australian Bureau of Statistics Census. Australian Bureau of Statistics: Canberra, 1998.

38 Taras HL, Sallis JF, Patterson TL, Nader PR, Nelson JA. Television's influence on children's diet and physical activity. J Dev Behav Pediatr 1989; 10: 176-180. MEDLINE

39 Shannon B, Peacock J, Brown MJ. Body fatness, television viewing and calorie-intake of a sample of Pennsylvania sixth grade children. J Nutr Educ 1991; 23: 262-268.

40 Robinson TN, Killen JD. Ethnic and gender differences in the relationships between television viewing and obesity, physical activity, and dietary fat intake. J Hlth Educ 1995; 26: S91-S98.

41 Woodward DR, Cumming FJ, Ball PJ, Williams HM, Hornsby H, Boon JA. Does television affect teenagers' food choices? J Hum Nutr Dietet 1997; 10: 229-235.

Figure 1 Mean (±SE) body mass index (BMI) by level of television viewing (h/day), stratified by physical activity level. One-way ANOVAs: *P<0.001, # P<0.01.

Table 1 Percentage distribution of selected socio-demographic characteristics of participants in the New South Wales 1996 Physical Activity survey (n=3392) compared to the 1996 NSW Census

Table 2 Proportion (%) of sample overweight or obese (25 kg/m²) by selected socio-demographic characteristics

Table 3 Distribution of leisure-time physical activity levels by selected socio-demographic characteristics (%)

Table 4 Distribution of reported hours spent watching television per day (tv/day) by selected sociodemographic characteristics (%)

Table 5 Odds ratios and 95% confidence intervals (CI) for selected sociodemographic variables, physical activity levels and television viewing, and the likelihood of being overweight (25 kg/m²)

Table 6 Adjusted odds ratios^a and 95% confidence intervals (CI) for television viewing and the likelihood of being overweight (25 kg/m²), for respondents in the inactive, low, moderate and high physical categories