Planning to win: Health and lifestyles associated with physical activity amongst 15,423 adults

Public Health (2006) 120, 206–212

ORIGINAL RESEARCH

Planning to win: Health and lifestyles associated with physical activity amongst 15,423 adults R.A. Harrisona,b,*, P. McElduffb, R. Edwardsb a

Directorate of Public Health, Bolton Primary Care Trust, St Peters House, Silverwell Street, Bolton, BL1 1PP UK b Evidence for Population Health Unit, Division of Epidemiology and Health Sciences, University of Manchester, Manchester, UK Received 6 December 2004; received in revised form 30 March 2005; accepted 19 August 2005 Available online 9 December 2005

KEYWORDS Physical activity; Sedentary behaviour; Public health; Lifestyles; Prevention

Summary Study objective: To assess levels of physical activity in the general population and amongst the ‘healthy’, and to identify factors associated with this important health behaviour. Design: Population-based cross-sectional study. Setting: North-west England. Participants: Adults (aged at least 18 years) registered with a general practitioner with a residential address within two local administrative districts (local authorities). Main results: Less than one-third of adults performed adequate amounts of physical activity for health protection, and this differed little when analyses were restricted to ‘healthy’ people. Lower levels of physical activity were observed amongst women, older people, ethnic groups, those with obesity and in each increased quintile of social deprivation. Current smokers, but not previous smokers, were less likely to be physically active, as were those not eating at least five portions of fruit and vegetables per day. Lack of physical activity was associated with poor general health and a history of, or current, chronic disease. Conclusions: Concerted efforts are required by the Government, society and individuals to overturn our predominantly physically inactive adult population. Interventions may be needed which specifically target certain groups, especially the most socially deprived, and that consider individuals and societal barriers to becoming physically active. Evidence of the effectiveness of individual and population-based interventions remains scant and this needs to be addressed urgently. Q 2005 The Royal Institute of Public Health. Published by Elsevier Ltd. All rights reserved.

Introduction * Corresponding author. Tel.: C44 1204 907728. E-mail address: [email protected] (R.A. Harrison).

Like many developed countries,1,2 people in England are increasingly sedentary, spending less time

0033-3506/$ - see front matter Q 2005 The Royal Institute of Public Health. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.puhe.2005.08.021

Health and lifestyles associated with physical activity engaging in physical activity as part of their daily lives. Over three-quarters of adults fail to meet physical activity recommendations, with a staggering 38% leading predominantly sedentary lifestyles.3 Substantial evidence links sedentary behaviour to detrimental effects on health including cardiovascular disease, type 2 diabetes, cancer, bone health and mental health.4 This is influencing the increasing prevalence of obesity.5 Reducing sedentary behaviour is a priority of the World Health Organization. In England, this is supported in the White Paper ‘Choosing Health’.6 This follows ambitious targets in the Government’s ‘Game Plan’ aiming for 70% of adults to be undertaking 30 min of physical activity on at least 5 days per week by 2020, i.e. a relative increase of around 120% over a 20-year period.7 Evidence to support policy makers to increase physical activity is scant.8,9 So too is information on individual factors associated with sedentary behaviour. For example, the National Health Survey for England3 only reports on physical activity levels by sex, age and social class. Thus many factors are overlooked that have a potential to influence physical activity, and the survey provides little information to inform policy development. The aim of the current study was, to assess levels of physical activity in the general population and in a subgroup defined as ‘healthy’. Associations with demographic, health and other lifestyle factors were considered to provide detailed information about this important health behaviour. The information is being used to inform local policy with respect to reducing sedentary behaviour, and provides a baseline to measure change in future years.

Methods The survey design has been described previously.10,11 It utilized a 50-item self-completion postal questionnaire, sent to 5% of adults residing in two administrative boroughs in the north-west of England. The sample was selected systematically from the boroughs’ primary care registers, checking for recent death notifications before sending the questionnaire. The administrative methods for the survey were evidence based to encourage high response rates.12,13 The questionnaire was posted with a covering letter explaining the purpose of the survey, and a pre-paid return addressed envelope. A reminder postcard was sent to non-responders after 7 days. A second copy of the questionnaire and a reminder letter were sent to persistent non-responders 14 days after the initial mail-out. An introduction to the survey was given in Urdu and Gujarati, the two

207 prominent second languages in this area. A full translation was not possible but information was given about a free translation service. The questionnaire included items relating to general and specific health, use of health services, lifestyle behaviours strongly related to health, and participants’ views about their neighbourhood. Individual question items were developed from national surveys (including the National Health Survey for England) and, where possible, incorporated previously validated tools (e.g. General Health Questionnaire 12).14 Physical activity was assessed using the Godin and Shephard instrument.15 This instrument has been validated in epidemiological studies and can distinguish between different levels of habitual physical activity. Participants were asked to identify how many times in the past week they had engaged in light, moderate or vigorous activity for at least 15 min per session. The Godin and Shephard instrument defines moderate activity as including brisk walking, table tennis, easy cycling, golf, dancing, cleaning windows and mopping floors. Vigorous activity includes running, football, cardiovascular gym workouts and aerobics. The survey was approved for research governance and given a favourable opinion by the local research ethics committees.

Analysis Being physically active was defined as participating in at least five sessions per week of moderate or vigorous activity, with each session lasting for at least 15 min, as defined by the physical activity scale. Univariate associations with individual health and lifestyle characteristics were assessed using cross-tabulations. Relative differences (prevalence rate ratios) in physical activity between groups, defined by health and lifestyle characteristics, were calculated after adjusting for sex, age, ethnicity and socio-economic status (Townsend deprivation quintiles at enumeration district level) using a modified Poisson regression approach. 16 This involves fitting generalized linear models to the data with a log link and Poisson error term. The outcome variable in these models was being physically active, and the predictor variables were health and lifestyle characteristics. The robust variance estimator was used to adjust for the misspecification of the error term. Two main analyses were carried out. The first included all participants completing the questionnaire. The second was restricted to those participants who did not report any adverse health

208

Table 1

Factors associated with being physically active in all participants and those classified as ‘healthy’. All participants

Category Sex Age group

Ethnicity Body mass Index

Home tenure Townsend deprivation score Enumeration district quintilesf

‘Healthy’ participantsa

Variable Male Female 18–44 years 45–64 years 65–74 years 75Cyears White Non-white !20 20–25 25–30 O30 Home owner Renting 1

% (n) 27.6% (1927) 26.7% (2266) 37.8% (2409) 25.2% (1406) 14.9% (296) 5.4% (82) 27.5% (4004) 24.1% (166) 30.1% (285) 30.5% (1972) 26.3% (1363) 19.6% (422) 28.7% (3410) 22.7% (701) 30.4% (1066)

RRb (95% CI) 1.00 (Ref) 0.96 (0.92, 1.01) 1.00 (Ref) 0.66 (0.63, 0.70) 0.39 (0.35, 0.44) 0.14 (0.11, 0.18) 1.00 (Ref) 0.70 (0.61, 0.79) 0.89 (0.81, 0.99) 1.00 (Ref) 0.95 (0.90, 1.01) 0.73 (0.67, 0.80) 1.00 (Ref) 0.80 (0.75, 0.85) 1.00 (Ref)

RRc (95% CI) 1.00 (Ref) 0.97 (0.92, 1.01) 1.00 (Ref) 0.65 (0.62, 0.69) 0.40 (0.35, 0.44) 0.14 (0.11, 0.18) 1.00 (Ref) 0.75 (0.66, 0.86) 0.92 (0.83, 1.02) 1.00 (Ref) 0.94 (0.89, 1.00) 0.74 (0.67, 0.81) 1.00 (Ref) 0.85 (0.79, 0.92) 1.00 (Ref)d

% (n) 35.7 (1009) 33.7 (1088) 40.2 (1501) 29.4 (499) 21.3 (79) 6.9 (18) 35.1 (1984) 33.7 (106) 40.8 (188) 36.9 (1103) 32.7 (584) 26.8 (129) 35.4 (1711) 34.0 (339) 34.9 (529)

RRb (95% CI) 1.00 (Ref) 0.94 (0.88, 1.00) 1.00 (Ref) 0.73 (0.67, 0.80) 0.55 (0.45, 0.68) 0.17 (0.11, 0.27) 1.00 (Ref) 0.84 (0.72, 0.99) 1.03 (0.91, 1.16) 1.00 (Ref) 0.91 (0.84, 0.99) 0.77 (0.66, 0.90) 1.00 (Ref) 0.93 (0.85, 1.02) 1.00 (Ref)

RRc (95% CI) 1.00 (Ref) 0.93 (0.87, 1.0) 1.00 (Ref) 0.73 (0.67, 0.79) 0.55 (0.45, 0.67) 0.17 (0.11, 0.28) 1.00 (Ref) 0.86 (0.73, 1.01) 1.04 (0.92, 1.18) 1.00 (Ref) 0.90 (0.83, 0.98) 0.77 (0.66, 0.89) 1.00 (Ref) 0.95 (0.86, 1.06) 1.00 (Ref)e

2

28.1% (853)

0.92 (0.86, 0.99)

0.92 (0.85, 0.99)

35.5 (755)

0.99 (0.89, 1.09)

0.98 (0.90, 1.09)

3 4 5

27.8% (759) 26.1% (734) 22.9% (752)

0.91 (0.84, 0.97) 0.84 (0.78, 0.91) 0.75 (0.69, 0.81)

0.90 (0.84, 0.97) 0.85 (0.78, 0.91) 0.77 (0.72, 0.84)

35.5 (398) 34.3 (369) 32.7 (366)

0.99 (0.89, 1.10) 0.94 (0.85, 1.05) 0.90 (0.81, 1.00)

1.0 (0.90, 1.10) 0.94 (0.85, 1.05) 0.92 (0.83, 1.03)

R.A. Harrison et al.

RR, relative risk; CI, confidence intervals. Note: denominator does not always add up to 15,461 because of missing responses to specific questions. a Excluding those reporting sciatica, arthritis, bronchitis, asthma or angina in the past year, or who had ever had high blood pressure diabetes, a heart attack or a stroke. b Adjusted for age (age in years) and sex. c Adjusted for age (age in years) sex, ethnicity (White/non-White) and Townsend deprivation scores for enumeration districts. d P value for trend across groups as an ordinal variable, RR 0.94 (95% CI 0.93, 0.96), PZ0.001. e P value for trend across groups as an ordinal variable, RR 0.98 (95% CI 0.96, 1.00), PZ0.11. f Townsend deprivation score, quintiles by enumeration district.

Health and lifestyles associated with physical activity Table 2

209

The association of physical activity with smoking and fruit and vegetable intake.

Category

Variable

% (n)

RRa (95% CI)

RRb (95% CI)

Smoking history

Never smoked Past smoker Current smoker Less than five portions per day At least five portions per day

29.4% (1750) 25.2% (1230) 27.2% (1160) 26.3% (3510)

1.00 (Ref) 1.0 (0.94, 1.07) 0.86 (0.81, 0.92) 1.00 (Ref)

1.00 0.99 0.88 1.00

36.1% (658)

1.49 (1.40, 1.59)

1.46 (1.37, 1.56)

Fruit/vegetable intake

(Ref) (0.93, 1.04) (0.82, 0.93) (Ref)

RR, relative risk; CI, confidence intervals. Note: total number of responders does not always add up to 15,461 due to item nonresponse. a Adjusted for age (age in years) and sex. b Adjusted for age (age in years) sex, ethnicity (White/non-White) and Townsend deprivation scores for enumeration districts.

conditions, therefore, excluding those reporting sciatica, arthritis, bronchitis, asthma or angina in the past year, and those who had ever had high blood pressure, diabetes, a heart attack or a stroke. This provided an indicator of physical activity amongst ’healthy’ people. Analyses were carried out in Stata version 8.2 (StataCorp, College Station, Texas, USA).

Results The survey was carried out in June 2001. Of the 21, 923 adults in the sample, 15,465 (70.1%) returned a useable questionnaire. Four people returning the questionnaire did not answer any questions on physical activity. The mean age of responders was 49 years and 45.2% were men. The mean age of responders was 8.3 years greater than that of nonresponders. The ethnic mix based on self-defined ethnicity was 95.5% White, 3.5% Asian and 1.0% Black, Chinese or other. This was similar to census statistics and was regrouped in the analysis as ‘White’ and ‘non-White’. The percentage of adults participating in at least five sessions of moderate/vigorous physical activity per week, with each session lasting for at least 15 min, was 27.1% (4193/15461). There was little difference amongst men and women, but physical activity was strongly related to age (Table 1). People defined as ‘White’ were more likely to be physically active than ‘non-Whites’, as were people not classed as obese based on body mass index. Physical activity was more likely amongst people who owned rather than rented their own home, and decreased across each Townsend quintile of social deprivation. In terms of sedentary behaviour, 44% (6814/15461) of participants did not report any moderate or vigorous physical activity lasting for at least 15 min over the previous 7 days.

Of all participants, 39% (6058/15,461) were defined as ‘healthy’ in that they did not report at least one of a number of health conditions. Of the ‘healthy’, 62% (3730/6058) were aged between 18 and 44 years and 46.6% (2826/6058) were men. The percentage of physically active subjects increased by 8.1% from 27.6% of all participants to 35.7% among those defined as ‘healthy’. A similar relationship between sex, age and body mass index was observed to the analysis including all participants. The effect of ethnicity bordered on statistical significance, and the relationship between home tenure and Townsend deprivation score on physical activity was attenuated when restricted to ‘healthy’ people (Table 1). Amongst all the participants to the survey, physical activity was more likely among people who did not smoke and those eating the recommended five portions of fruit and vegetables per day (Table 2). There was a clear relationship between physical activity and perception of one’s own health during the past year and with respect to specific chronic health problems, mental health, mobility and pain (Table 3). Stroke, mobility problems and severe pain in the past 4 weeks were the strongest predictors of not being physically active.

Discussion This survey confirmed worryingly low levels of regular physical activity in the general adult population, even among people with no reported health problems. The proportion of people who were at least moderately physically active was similar to that in the Health Survey for England (1998),3 with the same relationships overall in terms of age and social deprivation. Unlike the Health Survey for England, this study did not find that men were significantly more likely to be physically active than women. A slightly reduced probability of physical

210 Table 3

R.A. Harrison et al. Health-related factors associated with physical activity.

Category

Variable

% (n)

RRa (95% CI)

RRb (95% CI)

General health in past year

Poor

7.9% (127)

0.27 (0.23, 0.32)

0.28 (0.23, 0.33)

Fair Good Very good Excellent

16.1% (559) 29.2% (1302) 36.9% (1537) 41.1% (648)

0.54 0.84 0.96 1.00

(0.49, 0.59) (0.78, 0.91) (0.89, 1.03) (Ref)

0.55 0.85 0.96 1.00

(0.50, 0.61) (0.79, 0.92) (0.90, 1.04) (Ref)

Has No Has No Has No Has No

21.8% (1348) 33.2% (2599) 14.5% (127) 31.9% (3450) 18.6% (427) 32.2% (3252) 13.0% (152) 32.1% (3433)

0.85 1.00 0.62 1.00 0.67 1.00 0.54 1.00

(0.80, (Ref) (0.52, (Ref) (0.62, (Ref) (0.46, (Ref)

0.85 1.00 0.64 1.00 0.69 1.00 0.56 1.00

(0.80, (Ref) (0.54, (Ref) (0.63, (Ref) (0.48, (Ref)

Has this condition-yes No Yes No

13.3% (89) 27.8% (4104) 18.1% (738) 30.3% (3455)

0.65 1.00 0.81 1.00

(0.53, 0.78) (Ref) (0.75, 0.87) (Ref)

0.68 1.00 0.81 1.00

(0.56, 0.82) (Ref) (0.76, 0.88) (Ref)

Yes No Yes No Yes No Yes

11.7% (79) 27.8% (4114) 7.9% (38) 27.7% (4155) 17.7% (831) 31.2% (3362) 12.6% (478)

0.68 1.00 0.47 1.00 0.79 1.00 0.48

(0.55, (Ref) (0.35, (Ref) (0.73, (Ref) (0.43,

0.52)

0.68 1.00 0.49 1.00 0.79 1.00 0.49

(0.55, (Ref) (0.36, (Ref) (0.74, (Ref) (0.44,

No Severe/very severe Moderate None/mild Possible/probable Unlikely

34.5% (3625) 11.8% (219) 20.1% (637) 32.3% (3316) 20.6% (675) 29.8% (3467)

1.00 0.47 0.75 1.00 0.70 1.00

(Ref) (0.41, 0.53) (0.70, 0.81) (Ref) (0.65, 0.75) (Ref)

1.00 0.48 0.76 1.00 0.71 1.00

(Ref) (0.42, 0.55) (0.71, 0.82) (Ref) (0.66, 0.76) (Ref)

In the past year, suffered from: Arthritis or sciatica Bronchitis Asthma Angina Has/ever had: Diabetes High blood pressure Told by a doctor that: Had a heart attack Had a stroke Has or at risk of CHDc At least one mobility problemd Pain in past 4 weeks

Psychiatric casenesse

this condition-yes this condition-yes this condition-yes this condition-yes

0.90) 0.72) 0.73) 0.63)

0.84) 0.64) 0.84)

0.91) 0.75) 0.75) 0.65)

0.84) 0.66) 0.85) 0.53)

RR, relative risk; CI, confidence intervals. Note: total number of responders does not always add up to 15,461 due to item nonresponse. a Adjusted for age (age in years) and sex. b Adjusted for age (age in years) sex, ethnicity (White/non-White) and Townsend deprivation scores for enumeration districts. c Has or has ever had diabetes or high blood pressure, ever been told by a doctor that has had a stroke or a heart attack. d Mobility problems were: difficulty walking one-quarter of a mile on a level, great difficulty walking up/downstairs, difficulty going outside the house/garden without help, difficulty bending down/straightening up without holding on to something, falling or having great difficult keeping balance, difficulty using arms to reach/stretch for things, great difficulty holding/gripping/turning things. e General Health Questionnaire, score of R4 indicating caseness.

activity was observed among ‘non-Whites’ compared with ‘Whites’. The majority of ‘nonWhite’ participants were South Asian. Previous population-based studies in the UK have suggested much lower rates of physical activity in South Asian adults.17 The fact that people with obesity were less likely to be physically active is not surprising, as physical activity is a key component of preventing and treating obesity.18 Other studies have reported

similar relationships.19 The present study also found that current smokers, but not past smokers, and those not eating the recommended amounts of fruit and vegetables per day were less likely to be participating in regular physical activity, therefore, suggesting a relationship with physical activity and other health-promoting behaviours. This study provided an opportunity for additional insight into the characteristics associated with

Health and lifestyles associated with physical activity physical activity, including the relationship with individual health; a factor not reported previously in any detail. Levels of physical activity among people not reporting certain health-related problems, compared with those that did, were not found to alter markedly. The prevalence of physical activity was only 8% higher among ‘healthy’ participants, as categorized in this study. Thus, regardless of existing health, few adults in our sample, representative of the local areas, were likely to be performing adequate amounts of physical activity to benefit their health. As such, the majority of adults were not participating in physical activity in order to reduce the risk of future health problems, despite strong evidence of a protective effect.20 Similar patterns of physical activity were noted among people with and without existing health conditions, with respect to age, sex, obesity and ethnicity. However, the association with social deprivation, represented here by home ownership and Townsend deprivation score, was not statistically significant amongst ‘healthy’ participants. Hence, the impact of deprivation may only be associated with lower levels of physical activity among people already experiencing problems with their health. This is a particular concern given that systematic reviews have not specifically considered the effectiveness of interventions to increase physical activity in disadvantaged groups.8 The presence of each self-reported health condition examined in the survey increased the association with sedentary behaviour. Perhaps lower levels of physical activity compared with the ‘healthy’ population could be expected among people with musculoskeletal and mobility problems. However, low levels of physical activity were also noted among people with risk factors for coronary heart/vascular disease, and among those who had already suffered a stroke or a myocardial infarction. This suggests that primary prevention amongst high-risk groups and secondary prevention rehabilitation strategies are not leading to sustained increases in physical activity despite established benefits.21,22 This study has a number of strengths. The large sample size provided statistical power to examine associations within subgroups. Random sample selection and the high response rate enabled generalization to the entire reference population. The questionnaire included a wide range of subjects to examine the relationship between various individual factors and levels of physical activity. It has been recognized that no instrument currently exists that can effectively quantify true levels of individual physical activity.23 The method to assess physical activity has the advantage of including domestic as

211 well as recreational physical activity, thus reducing measurement error in terms of ‘total’ time spent in physical activity.24 A weakness of this instrument is that it does not enable one to identify participants who are meeting current UK physical activity recommendations, i.e. participating in at least 30 min of moderate physical activity on at least five occasions per week. However, it is able to identify people with varying levels of physical activity. In the present study, moderate physical activity was combined with vigorous physical activity. This makes sense from a public health policy perspective as the aim is to get more people to be at least moderately active.23 The study definition of ‘healthy’ individuals was based on information collected in the survey. As such, it may include some people with cancer and other illnesses because information on these illnesses was not obtained in the survey. This study provides stark findings about the task in hand if local and national policy makers are to meet the national target of 70% of adults being physically active by 2020.7 This is likely to require a combination of interventions targeted at populations and certain subgroups. The latter may require specific support to increase physical activity, particularly for people with disabilities.25 Structural interventions, such as transport, environmental and planning policies, seem particularly appealing given that they are implemented at the population level and are capable of influencing the lives of many people. There is also some support for their increased implementation to reduce sedentary behaviour.26– 29 Interventions directed at individuals, such as primary care physical activity referral schemes, seem unlikely to shift the population curve with respect to sedentary behaviour.30 The case for further primary research has been made8 to identify effective policies and interventions to increase physical activity in the general population and specific subgroups. These must consider the many inter-related factors operating at the individual level31 through to top-level policy contexts.32 Experience in Canada and Finland shows that population levels of physical activity can be increased, but that concerted efforts and investment are needed to meet and sustain this aim.33,34

Acknowledgements The authors would like to thank David Holt for data management and Alex Cornwell for assisting with some aspects of the analysis.

212

References 1. Manson JE, Skerrett PJ, Greenland P, Vantallie B. The escalating pandemics of obesity and sedentary lifestyle. A call to action for clinicians. Arch Intern Med 2004;164: 249–58. 2. Varo JJ, Martı´nez-Gonza ´lez, de Irala-Este ´vez J, Kearney J, Gibney M, Martı´nez JA. Distribution and determinants of sedentary lifestyles in the European union. Int J Epidemiol 2003;32:138–46. 3. Department of health. Health survey for England 1998. London: The Stationery Office; 2000. 4. World Health Organization. World health report. Geneva: World Health Organization; 2002. 5. Keim NL, Blanton CA, Kretsch MJ. America’s obesity epidemic: measuring physical activity to promote an active lifestyle. J Am Diet Assoc 2004;104:1398–409. 6. Department of health. Choosing health: making healthier choices easier. London: The Stationery Office; 2004. 7. Joint Department of culture, media and sport and strategy unit report. Game plan: a strategy for delivering government’s sport and physical activity objectives. London: Strategy Unit; 2002. 8. Hillsdon M, Foster C, Naidoo B, Crombie H. The effectiveness of interventions for increasing physical activity amongst adults: a review of reviews. London: Health Development Agency; 2004. 9. Agency for healthcare research & quality. Effectiveness of behavioral interventions to modify physical activity behaviors in general populations and cancer patients and survivors. Evidence report/technology assessment: number 102, publication no. 04-E027-1. Minnesota: Agency for Healthcare Research & Quality; 2004. 10. Harrison RA, Holt D, Pattison D, Elton PJ. Who and how many people are taking herbal supplements? A survey of 21,923 adults. Int J Vitam Nutr Res 2004;74:183–6. 11. Harrison RA, Holt D, Pattison D, Elton PJ. Are those in need taking dietary supplements? A survey of 21,923 adults. Br J Nutr 2004;91:617–23. 12. Harrison RA, Holt D, Elton P. Do postage stamps increase response rates to postal health surveys: a randomised controlled trial. Int J Epidemiol 2002;31:872–4. 13. Edwards P, Roberts I, Clarke M, DiGuiseppi C, Pratap S, Wentz R, et al. Increasing response rates to postal surveys: systematic review. Br Med J 2002;324:1183–5. 14. Goldberg D, Hillier V. A scaled version of the general health questionnaire. Psychol Med 1979;9:139–45. 15. Godin G, Shephard R. A simple method to assess exercise behavior in the community. Can J Appl Sports Sci 1985;10: 141–6. 16. Zou G. A modified Poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004;159: 702–6. 17. Fischbacher CM, Hunt S, Alexander L. How physically active are South Asians in the United Kingdom? A literature review. J Public Health 2004;26:250–8. 18. Grundy SM, Blackburn G, Higgins M, Laur R, Perri MG, Ryan D. Physican activity in the prevention and treatment of obesity and its co-morbidities. Med Sci Sports Exerc 1999;31: S02–S508.

R.A. Harrison et al. 19. Buchowski MS, Acra S, Majchrzak KM, Sun M, Chen KY. Patterns of physical activity in free-living adults in the Southern United States. Eur J Clin Nutr 2004;58:828–37. 20. Booth FW, Chakravarthy MV, Gordon SE, Spangenburg EE. Waging war on physical inactivity: using modern molecular ammunition against an ancient enemy. J Appl Physiol 2002; 93:3–30. 21. Gordon NF, Gulanick M, Costa F, Fletcher G, Franklin BA, Roth EJ, et al. Physical activity and exercise recommendations for stroke survivors. An American heart association scientific statement from the council on clinical cardiology, subcommittee on exercise, cardiac rehabilitation, and prevention; the council on cardiovascular nursing; the council on nutrition, physical activity, and metabolism; and the stroke council. Stroke 2004;35:11240–2130. 22. Taylor RS, Brown A, Ebrahim S, Jolliffe J, Noorani H, Rees K, et al. Exercise-based rehabilitation for patients with coronary heart disease: systematic review and meta-analysis of randomized controlled trials. Am J Med 2004;116:682–92. 23. Department of health. At least five a week. Evidence on the impact of physical activity and its relationship to health. A report from the chief medical officer. London: Department of Health; 2004. 24. Lawlor DA, Taylor M, Bedford C, Ebrahim S. Is housework good for health? Levels of physical activity and factors associated with activity in elderly women. Results from the British women’s heart and health study J Epidemiol Community Health 2002;56:473–7. 25. Rimmer JH, Rile B, Wang E, Rauworth A, Jurkowski J. Physical activity paricipation among persons with disabilities. Barriers and facilitators. Am J Prev Med 2004;26: 419–25. 26. Bauman AE, Bellew B, Owen N, Vita P. Impact of an Australian mass media campaign targeting physical activity in 1998. Am J Prev Med 2001;21:41–7. 27. Ewing R, Schmid T, Killingsworth R, Zlot A, Raudenbuh S. Relationship between urban sprawl and physical activity, obesity and morbidity. Am J Health Promot 2003;18:47–57. 28. Frank LD, Andresen MA, Schmid TL. Obesity relationships with community design, physical activity, and time spent in cars. Am J Prev Med 2004;27:87–96. 29. Saelens BE, Sallis JF, Frank LD. Environmental correlates of walking and cycling: findings from the transportation, urban design, and planning literatures. Ann Behav Med 2003;25: 80–91. 30. Harrison RA, Elton PJ. Does referral to a community based physical activity referral scheme reduce sedentary behaviour? A randomised controlled trial. J Public Health Med 2005;27:25–32. 31. Sherwood NE, Jeffery RW. The behavioral determinants of exercise: implications for physical activity interventions. Annu Rev Nutr 2000;20:21–44. 32. Cannon G. Why the Bush administration and the global sugar industry are determined to demoish the WHO global strategy on diet, physical activity and health. Public Health Nutr 2004;7:369–80. 33. Craig CL, Russell SJ, Cameron C, Bauman A. Twenty-year trends in physical activity among Canadian adults. Can J Public Health 2004;95:59–62. 34. Helkapori S, Uutela A, Prattala R, Pusk P. Heath behaviour and health among finnish population. Helsinki: National Public Health Institute; 2000.