The European Journal of Public Health Advance Access first published online on December 14, 2006
This version published online on December 14, 2006
The European Journal of Public Health, doi:10.1093/eurpub/ckl262
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Diet and exercise intervention in a general population—mediators of participation and adherence: the Inter99 study
Ulla N. Toft, Lis H. Kristoffersen, Mette Aadahl, Lisa von Huth Smith, Charlotta Pisinger and Torben JørgensenResearch Centre for Prevention and Health, Copenhagen, Denmark
Correspondence: Ulla N. Toft, Research Centre for Prevention and Health, Building 84/85, Glostrup University Hospital, DK-2600 Glostrup, Denmark, tel: +45 43233251, fax: +45 43233977, e-mail: ulto{at}glostruphosp.kbhamt.dk
Received March 10, 2006, accepted November 1, 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
Background: Drop-out rates are high in many lifestyle programmes. To promote adherence, the aim of this study was to identify mediators of participation in a diet and exercise (DE) intervention in a general population. Methods: Data were baseline data from a randomized non-pharmacological clinical trial in Copenhagen during 1999–2001. The participation rate was 53.3%. Participants at high risk of ischaemic heart disease (IHD) and who were offered participation in a DE counselling group intervention were included (N = 2022). Clinical characteristics, and demographic, psychosocial and lifestyle factors were measured. Results: Mediators of acceptance of participation were awareness of an unhealthy lifestyle or a bad health, low self-rated care of own health, perceived susceptibility of cardiovascular disease (CVD; overall and associated with lifestyle), high degree of motivation towards dietary changes and low self-efficacy about increasing physical activity. Overweight and impaired glucose tolerance (IGT)/screen-detected diabetes predicted acceptance whereas an absolute risk score for IHD was inverse associated with acceptance. Mediators of high adherence were low self-efficacy about changing dietary habits and perceived susceptibility of CVD and furthermore screen-detected diabetes and overweight predicted high adherence. Conclusion: Awareness of unhealthy lifestyle, perceived susceptibility of disease and motivation towards lifestyle changes were important mediators of participation. Screen-detected diabetes/IGT predicted participation and adherence whereas overweight individuals were more likely to accept but also to drop out of the course. The use of an absolute risk score in health promotion should be further evaluated.
Keywords: community participation, diet, exercise, intervention study, mediators
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
Ischaemic heart disease (IHD) remains a leading cause of morbidity and mortality in industrialized countries.1 Inappropriate diet and physical inactivity are well-documented modifiable risk factors of IHD and dietary modifications and increased physical activity level seem to be effective in modifying the risk of IHD when compliance is high.2–9 Some studies have indicated that adherence is an important factor when attempting to promote lifestyle changes in an organized lifestyle program.10 Nevertheless, drop-out rates appears to be large in many lifestyle programmes,11–13 and little is known about the predictors of participation and compliance in such programmes, especially in general populations. Better knowledge of the mediating variables of participation and adherence is vital to promote participation and to create a better foundation for developing a tailored and effective lifestyle intervention approach.8,14
Mediators can be defined as ‘causal or intermediate variables that can directly affect outcomes’15 or mechanisms through which an intervention might achieve its effects.16 In our study, we used the model of mediating and moderating variables developed by Green and kreuter15 to examine mediating variables of participation and adherence in a lifestyle programme (figure 1). Several health behaviour theories have identified important behavioural predictors but as single models they fail to fully explain nutrition and physical activity behaviours. We therefore measured potential mediating variables related to both the Transtheoretical Theory,17 the Social Learning Theory18 and the Health Belief Model19 and investigated the effect of these on participation and adherence to a diet and exercise (DE) group-counselling program in a general population aiming to reduce the risk of IHD.
|
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
Study population
Subjects were participants in the Inter99 study, investigating the effect of non-pharmacological lifestyle intervention on the incidence of IHD in a general population. The study has earlier been described in detail.20 The study population of the present study consisted of those subjects in Inter99 who were randomized to high-intensive intervention20 and comprised an age- and sex-stratified random sample of 11708 residents of the south-western part of Copenhagen County drawn by the Civil Registration System. Overall, of the subjects sampled, 72 were non-eligible, as they had died or could not be traced, leaving 11636 for invitation. The participation rate was 53.3% (N = 6206). From these, a total of 115 were excluded because of alcoholism, drug abuse or linguistic barriers.
Self-administered questionnaire
Before attendance at the centre the participants completed a comprehensive questionnaire including questions about educational level, employment, mental vulnerability, smoking, dietary intake and physical activity level.
Smoking habits were self-reported as: never smokers, ex-smokers, occasional smokers and daily smokers. Some participants were offered both smoking and DE counselling groups. To take into account both smoking status and the eventual acceptance of smoking intervention, a combined variable was constructed: (i) never smokers, (ii) occasional smokers, (iii) ex-smokers, (iv) daily smokers and declined smoking cessation, (v): daily smokers and accepted smoking cessation.
To measure physical activity level in leisure time, a previously developed and validated measure was used categorizing participants into one of four categories: sedentary, moderate activity, regular exercise and regular hard exercise.21 Because of low numbers in the highest level of physical activity, the two highest classes were merged in the analyses.
Mental vulnerability, defined as ‘a tendency to experience psychosomatic symptoms or inadequate interpersonal interactions’ was measured using a 12-item questionnaire, categorizing individuals in three classes: not vulnerable, moderately vulnerable and highly vulnerable.22
Dietary habits were measured using a validated dietary quality score developed from a 48-item food frequency questionnaire. In short, the dietary score was developed as a crude index of the overall quality of the dietary habits. The score was based on questions regarding the intake of fruits, boiled vegetables, raw vegetables, vegetable or vegetarian dishes, fish for lunch, fish for dinner, fat spread and fat used at preparation. A three-point scoring system was developed in respect to create two smaller groups with prudent and non-prudent dietary habits respectively and a larger group with average dietary habits.23
Furthermore, questions were included to assess psychosocial mediating variables of lifestyle changes. Participants were asked about their self-rated degree of healthy eating and physical fitness; motivation towards changes in dietary and exercise habits (‘Are you prepared for/minded on eating more healthy/exercising more?’ and ‘Do you think that it is necessary for your health that you change eating/exercise habits?’); susceptibility of cardiovascular disease (CVD), susceptibility of CVD associated with DE habits (‘Do you think your dietary/exercise habits increase your risk of CVD?’); and self-rated care of own health (‘How well do you think that you take care of your own health’). As a proxy measurement of self-efficacy, the participants were asked if they believed to be able to eat more healthy/exercise more. The questions regarding motivation and self-efficacy were added to the questionnaire on 01 February 2000 and therefore only answered by approximately half of the participants at baseline.
Physical examinations and laboratory measurements
Blood pressure was measured twice after 5 min of rest while lying down. Height was measured without shoes to the nearest 0.5 cm and weight was measured without shoes and overcoat to the nearest 0.1 kg. Body mass index (BMI) was calculated (kg/m2). Waist circumference was measured midway between the lower rib margin and iliac crest.
Fasting blood samples were drawn and total cholesterol was determined with enzymatic techniques (Boeringer, Mannheim, Germany).
Diabetes and impaired glucose tolerance (IGT) were defined according to the WHO criteria (WHO, 1999) on the basis of an OGTT, and individuals were categorized into four classes: known diabetes mellitus, screen-detected diabetes mellitus, IGT or normal glucose tolerance.
The study was approved by the local ethical committee.
Intervention
The subjects were invited for a health-screening programme at the Research Centre for Prevention and Health during 1999–2001.
Data from each individual were entered into a computerized program (PRECARD®) and the absolute risk of IHD at 60 years was estimated by the Copenhagen Risk Score.24 On the basis of the personal risk estimate, each individual had a lifestyle counselling talk focussing on smoking, physical activity, diet and alcohol. Motivational interviewing, conducted by trained nurses, dieticians and doctors, was used as a method.25 The Transtheoretical Model17 was also used as a central instrument to target the counselling according to the degree of motivation of the participant, using questions regarding the participants motivation towards changes in lifestyle and targeting pros, cons and self-efficacy to produce stage progress.26 The participants were furthermore offered participation in a DE group-counselling course over 6 months if they were at high risk of IHD. High risk was defined as either an absolute risk in the upper quintile of the distribution, according to the Copenhagen Risk Score, or at least one of the following isolated risk factors: systolic blood pressure 
160 mmHg, total cholesterol 7.5 mmol/l, BMI (30 kg/m2, diabetes or IGT.
Diet and exercise group-counselling course
Each DE counselling group consisted of 15–20 subjects who were scheduled for six 2 h meetings during a 6-month period. The staff (nurses and dieticians) were all trained in lifestyle counselling based on the Health Belief Model,19 the Social Cognitive Theory18 and the Transtheoretical Model.17 The staff focussed on the individuals’ perceived health risk, their readiness to change behaviour and their self-efficacy. They used goal setting and specific behavioural advice. The meetings included education in healthy DE, motivational support and group dynamics.
Overall, 2022 subjects were offered participation in a DE group-counselling course at baseline and 977 (48%) individuals accepted. Attendance rates for the DE counselling group sessions were incomplete for 80 subjects, leaving 897 subjects for the analyses including attendance rate.
Statistical methods
Statistical analyses were carried out using the SAS statistical software package (version 9.1, 2002–2003, SAS Institute, Inc., Cary, NC, USA).
The individuals who were offered participation in a DE group were distributed into four groups, which: (i) declined participation; (ii) accepted participation, but never attended (drop-out); (iii) accepted participation, but attended less than four sessions (low adherence); (iv) accepted the participation and attended at least four sessions (high adherence). Descriptive statistical analyses were made of these four groups of participants. Means and 95% confidence intervals (CI), adjusted for age and sex, were calculated for the continuous variables. For categorical variables, an indirect age- and sex-standardization was made.
In order to identify mediators of acceptance of participation, logistic regression analyses were applied with acceptance of DE group counselling as the dependent variable (no/yes). The crude model consisted of the dependent variable and each of the measured potential cognitive mediating variables as an independent variable, adjusting for sex and age. Interaction between the independent variable and both sex and age were explored. Education, employment, mental vulnerability, sex, age, smoking habits, physical activity and dietary habits were considered as moderators of the effect of the mediators and were adjusted for in the corrected model. Furthermore, weight status was considered as a potential moderator and BMI was therefore in a third model, added to the earlier described corrected model.
To identify mediators of high adherence, logistic regression analyses similar to the analyses described previously were used, but with adherence as the dependent variable (high adherence/low adherence and drop-out).
The mediating variables were measured before attendance at the centre. However, the results from the health-screening programme could potentially have changed these variables and thereby affect participation and adherence. Therefore we investigated if risk factors measured at baseline (the Copenhagen Risk Score, systolic blood pressure, cholesterol, waist circumference, BMI and diabetes) predicted participation and adherence, adjusted for sex, age, education, employment, lifestyle and psychological vulnerability.
Odds ratio and 95% CI were calculated. The continuous variables were tested for linear association with the response variable, testing for a cubic, quadratic or a linear spline association.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
Table 1 describes characteristics of the participants according to their participation rate at the DE group-counselling course.
|
Predictors for acceptance of the DE course
Perceiving oneself as susceptible of developing CVD or perceiving an increased susceptibility of CVD associated with personal dietary and exercise habits were mediators of acceptance of DE group counselling. Furthermore, awareness of unhealthy dietary habits or bad physical fitness and bad self-rated health and self-rated care of own health significantly predicted acceptance. Participants with low self-efficacy towards increasing physical activity level and with a high motivation towards changing dietary habits were significantly more likely to accept participation. There was no effect modification for age and sex was observed. The relationship between acceptance and both self-rated health/care of health, susceptibility of CVD (overall and associated with exercise habits) and awareness of physical fitness were attenuated when adjusting for BMI.
Furthermore, diabetes or IGT and overweight (waist circumference and BMI) predicted acceptance of participation. The Copenhagen Risk Score estimate did not have a linear association with the response variable. Rather, the association could be described as a linear spline (figure 2). The linear spline function was set with knots at the 10, 50 and 90 percentiles. The analysis showed a significant negative association with accepting participation (P = 0.0088) for those with a medium risk according to the Copenhagen Risk Score. For those with the very lowest and highest risk, there was a tendency (non-significantly) of a positive association with accepting participation (data not shown). Changing the knots did not change the association found.
|
Predictors for high adherence to the DE course
Participants with a low self-efficacy towards changing dietary habits, low degree of motivation towards increasing physical activity and perceived susceptibility of CVD were significantly more likely to have high adherence to the course. Screen-detected diabetes or IGT and lower BMI furthermore predicted acceptance. No effect modifications for age and sex were observed and the Copenhagen Risk Score estimate did not significantly predict high adherence (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
The present study identified mediators of acceptance and adherence to a DE group counselling course. The findings indicate that variables related both to the Transtheoretical Model, the Social Cognitive Theory and especially the Health Belief Model significantly predicted acceptance of participation and adherence to a life-style course. Furthermore, single risk factors measured at the health screening, including diabetes and overweight, predicted participation whereas the Copenhagen Risk Score, showed an inverse association with acceptance of participation.
The focus in this study was on mediators of acceptance and adherence to a lifestyle course whereas most other studies in this area have focussed on mediators of changing lifestyle or adherence to an exercise programme in individuals, who already accepted to participate.12,13,27–29 However, to be able to develop a more targeted intervention and to promote higher participation in a general population, it is essential to identify characteristics of those who accept and adhere to the intervention and potential mediators of participation and adherence. It is, however, important to bear in mind that participation is not equivalent to behavioural changes; and, individuals who decline participation can also be participants who choose to make changes on their own.
The potential mediators were measured before attendance at the centre and hence might have been modified by the results from the health screening. However, the fact that several of these mediators significantly predicted acceptance of participation at the DE course gives a good foundation for promoting participation in future intervention. Thus, the staff will be able to identify the participants’ motivation, perceived susceptibility, etc. from a simple questionnaire completed before attendance and try to modify these in order to promote participation.
Motivational variables towards lifestyle changes were used as indicators of the stages of change of the participants both in the lifestyle counselling and in the analyses in this study. These questions are useful to differentiate between participants in the pre-contemplation stage and participants in the contemplation or preparation stage. Higher degree of motivation towards eating more healthy significantly predicted acceptance of participation at the DE course. This is in accordance with the finding of other studies regarding changes in dietary habits.30,31 However, in contrast with our expectations, participants with a high degree of motivation towards increased physical activity were more likely to drop out. One reason for this could be that the course at baseline focussed more on dietary habits than on exercise, and the course did not include exercise sessions. Furthermore, the Transtheoretical Model predicts success in behaviour change, not acceptance and adherence to a lifestyle programme.
Acceptance of participation can partly be compared with the first three stages in the Transtheoretical Model, whereas adherence can be compared with the action and maintenance stage. In the last stages, self-efficacy, from social learning theory,18 has shown to become increasingly important, which is consistent with the fact that self-efficacy towards dietary changes predicted high adherence in this study. However, it seems to be an important mediator of acceptance also since self-efficacy about increasing physical activity was significantly associated with acceptance. In contrast with our expectation, it was lower self-efficacy that predicted acceptance and high adherence. A hypothesis for this finding could be that individuals with a high self-efficacy towards lifestyle changes will try to perform changes on their own whereas those with low efficacy are more likely to seek help by participating at the course. It is also essential that the measurements were not specific on participation and adherence but measured self-efficacy about changing dietary and exercise habits in general, which also might be a reason why these measures overall were weak predictors. In earlier studies, self-efficacy has shown to be a strong predictor of changes in both dietary32–34 and exercise habits.28,35 Similar to the finding for self-efficacy, participants where more likely to accept participation with decreasing degree of self-rated care of health.
The Transtheoretical Model17 also includes the self-rated lifestyle and health variables. For examples, individuals rating their dietary habits as healthy are classified as being at the maintenance or action stage for changing dietary habits. However, two studies from the Netherlands36,37 have illustrated a low degree of agreement between the self-rated subjective intake and the objective assessment of dietary intake. According to Weinstein,38 awareness of own behaviour is one of the key issues in motivating people to move from pre-contemplation to a higher stage of change. Lechner et al.39 have proposed an alternative classification method of stages of change, where individuals, that wrongly rate their lifestyle as healthy, are considered pre-contemplators. The results from this study also emphasize the importance of awareness of unhealthy dietary or exercise habits or a bad health in mediating accept and thereby the decision to change behaviour and move to a higher stage of change. Thus, in the lifestyle counselling, it is relevant to try to identify individuals with a misconception of their own lifestyle as healthy and try to promote a realistic perception of their lifestyle to be able to promote positive lifestyle changes.
The Health Belief Model stresses the importance of the individual's perception of his or her susceptibility to a particular condition/illness for the decision to pursue a particular health-related behaviour19 and several mediators related to this theory appeared essential when predicting acceptance of participation in the DE course. Thus, both the overall perceived susceptibility of CVD and also the susceptibility of CVD related to lifestyle positively predicted acceptance of participation. Furthermore, single risk factors predicted acceptance and adherence.
Thus, overweight and diabetes diagnosed at the health-screening programme were strong predictors of accepting participation at the DE group-counselling course. In contrast, the Copenhagen Risk Score estimate showed a negative association with acceptance of participation at the course for the majority of the population. Only for those with the very lowest or highest risk there seemed to be tendency of a positive association. This could indicate that concrete and well-known risk factors, such as diabetes and overweight are effective in motivating individuals for lifestyle changes. However, it seems that individuals at moderate risk find it difficult to relate to a rather abstract estimate of their risk of developing CVD in the future. An absolute risk estimate might therefore, in general, not be a good educational tool in individuals with a moderate risk estimate. The inverse association might illustrate that individuals with an overall higher risk are more difficult to persuade to participate in an intervention like this; probably because they also differentiate from those at lower risk in other ways. More studies, especially qualitative studies, are needed in order to explore this area.
Adjustment for BMI attenuated the relationship between acceptance and susceptibility of CVD (overall and associated with exercise habits), self-rated physical fitness and care of own health. The explanation for this finding could be that the overweight in general are aware of the health hazards of being overweight and therefore BMI is closely related to the perception of susceptibility and own health status and therefore the mediating effect of these variables were primarily explained by BMI.
Only very few predictors were identified for adherence to the DE course. This is probably due to the fact that adherence over 6 months is a much more complex behaviour than accepting participation or not at the counselling session. To be able to predict adherence better, its probably would have been relevant to include variables on social support and network, for example from "Rogers Diffusion of innovation" theory.40
A limitation of this study is the cross-sectional design. A relevant future study would therefore be to explore if changes in the identified mediators will actually influence the degree of participation at a lifestyle course.
|
Conclusions |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
Mediators of accepting participation at a DE group counselling course in a general population were awareness of an unhealthy lifestyle and bad health, low self-rated care of own health, perceived susceptibility of CVD, high motivation of dietary changes and low self-efficacy of increasing physical activity. Mediators of high adherence were low self-efficacy about dietary changes, low degree of motivation towards increasing physical activity and perceived susceptibility of CVD.
Furthermore, screen-detected diabetes and IGT predicted both acceptance and adherence. Overweight participants were more likely to accept participation but were also more likely to drop out. The Copenhagen Risk Score, showed an inverse association with acceptance of participation.
|
|
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
We acknowledge the steering group for the Inter99 study and all members of the Inter99 team.
The study was initiated and analysed by the investigators in this project, but was supported economically by: The Danish Medical Research Council, The Danish Centre for Evaluation and Health Technology Assessment, Novo Nordisk, Copenhagen County, The Danish Heart Foundation, The Danish Pharmaceutical Association, Augustinus foundation, Ib Henriksen foundation and Becket foundation.
Key points
|
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
1 Tunstall-Pedoe H, Kuulasmaa K, Mähönen M, et al. (1999) Contribution of trends in survival and coronary-event rates to changes in coronary heart disease mortality: 10-year results from 37 WHO MONICA Project populations. Lancet 353:1547–57.[CrossRef][Web of Science][Medline]
2 Mensink RP and Katan MB. (1992) Effect of dietary fatty acids on serum lipids and lipoproteins. A meta-analysis of 27 trials. Arterioscler Thromb 12:911–19.
3 Wilhelmsen L, Berglund G, Elmfeldt D, et al. (1986) The Tromso Survey: the Family Intervention study–the effect of intervention on some coronary risk factors and dietary habits, a 6-year follow-up. Eur Heart J 7:279–88.
4 De Lorgeril M, Salen P, Martin JL, et al. (1999) Mediterranean diet, traditional risk factors, and the rate of cardiovascular complications after myocardial infarction: final report of the Lyon Diet Heart Study. Circulation 99:779–85.
5 Sesso HD, Paffenbarger RS Jr, Lee IM. (2000) Physical activity and coronary heart disease in men: The Harvard Alumni Health Study. Circulation 102:975–80.
6 Berlin J and Golditz G. (1990) A meta-analysis of physical activity in the prevention of coronary heart disease. Am J Epidemiol 132:612–28.
7 Sjøl A, Thomsen K, Schroll M, Andersen L. (2003) Secular trends in AMI in relation to physical activity level in the general Danish population. Scand J Med Sci Sports 13:224–30.[CrossRef][Web of Science][Medline]
8 Ammerman AS, Lindquist CH, Lohr KN, Hersey J. (2002) The efficacy of behavioral interventions to modify dietary fat and fruit and vegetable intake: a review of the evidence. Prev Med 35:25–41.[CrossRef][Web of Science][Medline]
9 Dayton S, Pearce M, Hashimoto S, et al. (2003) A controlled clinical trial of a diet high in unsaturated fat in preventing complications of atherosclerosis. Circulation 40:Suppl 2, 1–63.
10 Atkins CJ, Senn K, Rupp J, et al. (1990) Attendance at health promotion programs: baseline predictors and program outcomes. Health Educ Q 17:417–28.[Web of Science][Medline]
11 Lechner L and de Vries H. (1995) Participation in an employee fitness program: determinants of high adherence, low adherence, and dropout. J Occup Environ Med 37:429–36.[Web of Science][Medline]
12 Stiggelbout M, Hopman-Rock M, Crone M, et al. (2006) Predicting older adults’ maintenance in exercise participation using an integrated social psychological model. Health Educ Res 21:1–14.
13 Packianathan I, Sheikh M, Boniface D, Finer N. (2005) Predictors of programme adherence and weight loss in women in an obesity programme using meal replacements. Diabetes Obes Metab 7:439–47.[CrossRef][Web of Science][Medline]
14 Kumanyika SK, Van Horn L, Bowen D, et al. (2000) Maintenance of dietary behavior change. Health Psychol 19:Suppl 1, 42–56.[CrossRef][Web of Science][Medline]
15 Green L and Kreuter M. (2005) Health program planning. An educational and ecological approach 4th edn (McGraw-Hill, New York).
16 Baron R and Kenny D. (1986) The moderator–mediator variable distinction in social psychological research: conceptual strategic, and statistical considerations. J Pers Soc Psychol 51:1173–82.[CrossRef][Web of Science][Medline]
17 Prochaska J, DiClemente C, Norcross J. (1997) In search of how people change: application to addictive behaviours. In Marlatt G and van den Bos H (Eds.). Addictive behaviours: readings on etiology, prevention, and treatment(American Psychological Association, Washington DC) pp. 671–96.
18 Bandura A. (1986) Social foundation of thoughts and action: a social cognitive theory(Prentice Hall, New York).
19 Janz N and Becker M. (1984) The health belief model: a decade later. Health Educ Q 11:1–47.[Web of Science][Medline]
20 Jorgensen T, Borch-Johnsen K, Thomsen TF, et al. (2003) A randomized non-pharmacological intervention study for prevention of ischaemic heart disease: baseline results Inter99 (1). J Cardiovasc Risk 10:377–86.
21 Saltin B and Grimby G. (1968) Physiological analysis of middle-aged and old former athletes: comparison with still active athletes of the same ages. Circulation 38:1104–15.
22 Eplov LF, Jorgensen T, Birket-Smith M, et al. (2006) Mental vulnerability-a risk factor for ischemic heart disease. J Psychosom Res 60:169–76.[CrossRef][Web of Science][Medline]
23 Toft U, Kristoffersen L, Lau C, et al. The Dietary Quality Score: validation and association with cardiovascular risk factors: the Inter99 study. Eur J Clin Nutr advance online publication 2006:23.
24 Thomsen T, Borch-Johnsen K, Davidsen M, Ibsen H. (1997) The ‘PRECARD’ study: identification and management of individuals at risk of developing cardiovascular disease. Can J Cardiol 13:Suppl B, 286B–7B.
25 Britt E, Hudson SM, Blampied NM. (2004) Motivational interviewing in health settings: a review. Patient Educ Couns 53:147–55.[CrossRef][Web of Science][Medline]
26 Prochaska J, DiClemente C, Norcross J. (2003) In search of how people change: applications to addictive behaviors. In Salovey P and Rothman A (Eds.). Social psychology of health: key readings(Psychology Press, New York) pp. 63–77.
27 Miller D, Trost SG, Brown W. (2002) Mediators of physical activity behavior change among women with young children. Am J Prev Med 23:98–103.[Web of Science][Medline]
28 Sallis JF Jr and Owen N. (1999) Determinants of physical activity. Physical activity and behavioral medicine(SAGE Publications, California) pp. 110–33.
29 Näslund G, Fredrikson M, Hellenius ML, de Faire U. (1996) Determinants of compliance in men enrolled in a diet and exercise intervention trial: a randomized, controlled study. Patient Educ Couns 29:247–56.[CrossRef][Web of Science][Medline]
30 de Vet E, de Nooijer J, de Vries N, Brug J. (2006) The Transtheoretical Model for fruit, vegetable and fish consumption: associations between intakes, stages of change and stage transition determinants. Int J Behav Nutr Phys Act 3: doi:10.1186/1479-5868-3-13.
31 Horwath C. (1999) Applying the Transtheoretical Model to eating behaviour change: challenges and opportunities. Nutr Res Rev 12:281–317.[CrossRef]
32 Schafer R, Keith P, Schafer E. (1995) Predicting fat in diets of marital partners using the health belief model. J Behav Med 18:419–33.[CrossRef][Web of Science][Medline]
33 Walsh M and Flynn T. (1995) A 54-month evaluation of a popular very low calorie diet program. J Fam Prac 41:231–6.[Web of Science][Medline]
34 Ma J, Betts NM, Horacek T, et al. (2002) The importance of decisional balance and self-efficacy in relation to stages of change for fruit and vegetable intakes by young adults. AJHP 16:157–66.
35 Biddle S and Mutrie N. (2003) Psychology of physical activity. Determinants, well-being and interventions(Routledge, London).
36 Brug J, van Assema P, Kok GJ, et al. (1994) Self-rated dietary fat intake: association with objective assessment of fat, psychosocial factors, and intention to change. J Nutr Educat 26:218–23.
37 Lechner L, Brug J, de Vries H. (1997) Misconception of fruit and vegetable consumption: differences between objective and subjective estimation of intake. J Nutr Educat 29:313–20.
38 Weinstein ND. (1988) The precaution adoption process. Health Psychol 7:355–86.[CrossRef][Web of Science][Medline]
39 Lechner L, Brug J, van Assema P, Mudde A. (1998) Stages of change for fruit, vegetable and fat intake: consequences of misconception. Health Educat Res 13:1–11.
40 Rogers E. (1983) Diffusion of innovations 3rd edn (The free press. A division of Mac-Millan Publishing Co. Inc., New York).
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. R. Colberg Encouraging Patients to Be Physically Active: What Busy Practitioners Need to Know Clin. Diabetes, July 1, 2008; 26(3): 123 - 127. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||