Predicting Health Behaviour Population-Based Studies of Knowledge and Behaviour Related to Cardiovascular Diseases

Transcription

1 Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 186 Predicting Health Behaviour Population-Based Studies of Knowledge and Behaviour Related to Cardiovascular Diseases PER ANDERSSON ACTA UNIVERSITATIS UPSALIENSIS UPPSALA 2006 ISSN ISBN urn:nbn:se:uu:diva-7200

2

3 List of papers This thesis is based on the following papers, referred to by their Roman numerals: I Andersson P, Leppert J. Men of low socio-economic and educational level possess pronounced deficient knowledge about the risk factors related to coronary heart disease. J Cardiovascular Risk Dec;8(6): II Andersson P, Sjöberg RL, Krysa M, Sidorowicz W, Öhrvik J, Leppert J. Lags in behavioral change: A population based comparison of Cardiovascular risk behavior in Poland and Sweden. Cent Eur J Public Health Jun;14(2):82-5. III IV Andersson P, Sjöberg RL, Öhrvik J, Leppert J. Knowledge about cardiovascular risk factors among obese individuals. Eur J Cardiovasc Nurs Apr 6; (Epub ahead of print) Andersson P, Sjöberg RL, Öhrvik J, Leppert J. Effects of family history and personal experience of illness on inclination to change health related behavior. Submitted Papers I, II and III are reprinted with permission of the publishers.

4

5 Contents Preface...11 Introduction...12 Risk factors for cardiovascular diseases...12 Prevention of cardiovascular diseases...13 Health models...15 Factors related to knowledge about risk factors...16 Obesity related to knowledge and risk behaviour...17 Risk behaviour in relation to family history and personal experience of illness...18 Influence of different social systems on behaviour and health status Aims of the thesis...20 Methods...21 Design...21 Settings...21 Population...22 Procedure...23 Measurements...24 Screening...24 Questionnaire...25 Statistical analyses...26 Ethics...28 Results...29 Study I...29 Study II...30 Study III...32 Study IV...36 Discussion...38 Methodological considerations...41 Conclusions...44 Epilogue and implications...45

6 Acknowledgements...46 Summary in Swedish; svensk sammanfattning...48 References...51 Appendix...68

7 Glossary This glossary is primarily based on definitions taken from the Scientific Dictionary of Public Health (1). Bias Cross-sectional study Confidence interval (CI) Confounding Factor analysis Generalise Systematic distortion of the estimated intervention effect away from the thruth, caused by inadequacies in the design, conduct, or analysis of a trial (2). A study that examines the relationship between diseases, or other health-related characteristics, and other variables of interest, as they exist in a defined population at one particular time A measure of the precision of an estimated value. The interval represents the range of values, consistent with the data that is believed to encompass the thrue value with high probability (usually 95 %) (2). Mixing of more than one effect. A confounding factor is one that is associated with the studied exposure, while at the same time being a risk factor for the dependent outcome variable. A statistical approach to reduce the number of variables in an analysis on the basis of their associations to each other. Can be used to study the extent to which one, or several, variables are expressions of the same underlying phenomena. To make conclusions from one or several separate phenomena to something general by distinguishing their special characteristics and identifying what they have in common.

8 Health behaviour Interaction Knowledge Logistic regression Any activity undertaken by those believing themselves to be healthy for the purpose of preventing or detecting disease at an asymptomatic stage (3). Interplay. Refers to the joint operation of two or more causes to produce a different effect than just the sum of the separate effect. In this thesis, knowledge refers primarily to awareness of the factors that increase or decrease the risk of cardiovascular disease. A regression analysis where the association between two or more quantitative variables is a function of the included determinants. Obesity BMI kg/m 2 (4) Odds ratio The ratio between two odds. The odds of a particular exposure among persons with specific characteristics divided by the corresponding odds of exposure among persons without the same specific characteristics. Overweight BMI kg/m 2 (4) Predictor Prevention Risk behaviour A variable that can foretell something about future events. Measures to prevent occurrence of disease or to arrest its progress and reduce its consequences once it is established. Primary prevention seeks to prevent the initial occurrence of a disease. Secondary prevention seeks to arrest or retard existing disease through early detection. Specific forms of behaviour which are proven to be associated with increased susceptibility to a specific injury, disease or ill health.

9 Risk factor Validity An individual s characteristics indicating an increased risk of ill health, disease, injury or adverse health behaviour. This characteristic can be related to any aspect of an individual (hereditary, social, economic or environmental). Risk factors precede the phenomenon that they are indicating. A risk factor can be either congenital or acquired. The ability of the study to capture what was stated in the aims

10 Abbreviations CVD CHD SOC HBM SES SEI BMI Cardiovascular Diseases Coronary Heart Diseases Stage Of Change Health Belief Model Socio-Economic Status Socio-Economic Index Body Mass Index

11 Preface He lives a dangerous life, is 48 years old, overweight, stressed and untrained. The fact that he is on medication for high blood pressure and feels a pressure over his chest when he goes up and down the stairs does not make the situation better. There are obvious signs that something is wrong and that he should see a doctor. But it is difficult to break old habits and it is not the easiest thing to convince him that he has a problem. What should the health professional do to make him understand how serious his situation is? One traditional way is to let him read about risk factors and illness. Another possibility would be to have a discussion with him and to learn more about his own understanding of the situation. One goal would be to make him understand that his lifestyle is unhealthy and that he needs to do something radical to avoid the risk of having a heart attack. But if he listens to you, what happens then? What does it mean to understand? While working as a nurse in both the medical emergency and primary health care settings I was frequently confronted with the health promotion challenges these patients pose. As a lecturer in nursing education, I have also had the opportunity to deal with the very same issues from a theoretical point of view. These experiences made me appreciate the need for further exploration of the nature and dynamics of predicting the health behaviors of these kinds of patients and writing this thesis has allowed me to do so. 11

12 Introduction Risk factors for cardiovascular diseases Cardiovascular disease (CVD) is responsible for a considerable proportion of the mortality and morbidity of the general population. Myocardial infarction is the most common cause of death in Sweden and stroke is the third most common. This is much the same for most other countries in Europe (5). In the present thesis CHD (coronary heart disease) is a part of CVD. About 40,000 Swedes will have a myocardial infarction annually, and about 35,000 will have a stroke. According to a current report from the Swedish National Board of Health and Welfare the death rate for CVD continues to fall (6, 7). The survival of more people from CVD means that there will be an increased demand for health care after the event. From a national perspective, there has been a favourable trend regarding CVD in many of the world s industrial countries, but there are still considerable geographical differences. In some of the Eastern European countries, CVD deaths are still at a very high level (8-10). Development of primary and secondary prevention must thus be considered vital to most European countries. The most important risk factors for CVD are age, gender, heredity, high blood pressure, high blood lipids, tobacco smoking and physical inactivity (11-16). Hereditary factors constitute important predictors of CVD (11, 17). Close blood relatives are in a high risk group where a high proportion will have the same genetic defects as the patient and, thus, also run the risk of developing the same illness (18-20). Studies have also shown that stress at work and within the family is associated with an increased risk for CVD (21-25). Low education and low socioeconomic status have a correlation to CVD in both men and women (26-33). Other risk factors for CVD are obesity, diabetes, social vulnerabilities, and insufficiencies in the social network (34-39). Obesity is the risk factor that has shown the greatest increase in prevalence during the latest decades (40-45). However, at least in Sweden, this increase appears to have tailed off between 2002 and Despite this, around 10 % of the population in the year age group, have a BMIvalue over 30 (6). Fifteen percent of the men in the age group are extremely overweight. A higher proportion of those with a BMI over 30 are 12

13 found among those with low level of education and low socioeconomic status. Thus, socioeconomic inequality may increase health consequences of obesity in many countries (46-51). Prevention of cardiovascular diseases The aims of taking measures against CVD are to prevent illness from occurring, primary prevention, and to prevent relapse, secondary prevention. Considerable progress has been made in the field of medicine-based prevention (52, 53), but the development has not been as beneficial in relation to people s lifestyles and habits. The European guidelines on cardiovascular disease prevention in clinical practice focus on both behavioural and biological-related risk factors (54). Living habits is an important factor in population-focused prevention of CVD. People can feel well, even though they may have developed preliminary stages of illness such as diabetes, dyslipidemia, or increased blood pressure. Another problem concerning prevention that becomes apparent in the screening for risk factors is that the motive to change one s lifestyle can be weak in an individual who does not have any symptoms of illness (55). Screening involves an attempt to identify risk factors (56). Traditionally such screening is most often combined with health information and counselling, based on the results found. Screening is usually of the opportunistic type, which means that the opportunity is taken to screen people who attend a health centre for a different reason (57). Another, more systematic, way is to invite all people from various age groups, for example, all 40 and 50 year olds, in order to identify possible risk factors for CVD. In addition, mass screening is a strategy for screening when a whole population is invited for testing (58). It is still unclear which strategy is the most effective (59-63). It is also important to be aware that populations show different reactions when different strategies are applied (64). A large number of preventive CVD programmes which include screening and health information, have been carried out in different parts of the world. Several of these programmes have combined measures directed at whole populations with opportunistic attempts to discover and treat people with a high risk of becoming ill with CVD. To analyse effects of CVD-programmes with population based strategies it is essential to examine the designs used and how the programmes were implemented (65, 66). When effects of different population-focused projects have been evaluated, differences between intervention and control groups have generally been small with no clear advantages towards the population-focused programmes compared to high-risk strategy programmes (67-71). In North Karelia, Finland, one of the most famous CVD programmes started in 1972 (72). The main purpose of the program in North Karelia was 13

14 to reduce morbidity and mortality levels caused by CVD, with a specific focus on middle-aged men. There was a reduce in deaths from CVD in North Karelia, but this did not differ from the average reduce in deaths in the total population of Finland. The decreasing trend had already begun before the North Karelia project started and it is therefore unclear whether any reliable effects can be attributed to this project (73). In the USA, a project called the Stanford Three-Community Study was carried out in the 1970s (74, 75). During the two years the study was performed, risk factor levels decreased in both communities where the programme was being carried out, while changes were more limited in the control community. Intervention in the Stanford Three-Community was, primarily, a mass media campaign, where the aim was to increase awareness of the causes of CVD. In the beginning of the 1980s new programmes, focusing on population strategies rather than risk-oriented components, were started in the USA. A common factor for these programmes was the development of intervention methods with more well-organised ways to spread knowledge. A unique effort was the Stanford Five-City Project where a multilingual venture was used to reach subgroups where language difficulties could be a barrier against understanding health information (70). The Minnesota Heart Health Program combined a large number of strategies though no randomisation was done, which affected the possibilities of making adjustments for the effects of gender and level of education (71). The Pawtucket Heart Health Program included towns with relatively high unemployment and with a large proportion of inhabitants who were low educated. One strategy in the Pawtucket Heart Health Program that contributed to positive effects on the risk factor levels was that health information was adapted for target groups. One example was that people with inadequate reading skills received specially designed programmes (69). A common factor for the American programmes was that consideration was given to the conditions of the individual, partly through the interpretation of results, and partly through the choice of intervention methods. Since it s outset in the 1980s, the Swedish Norsjö Project has been regarded as a demonstration project for Public Health Care involvement on a population level. The municipality of Norsjö, with a mainly rural population, had an excessive death rate from CVD (76). During the first two years of the project, 2,400 people participated, from a total of 5,400 inhabitants in the municipality. One of the aims of the Norsjö Project was to make people take personal responsibility for their health (76-78). A conclusion from the Norsjö Project was that a long-term, municipality-based, CVD programme that combines population and individual strategies can prevent CVD among high-risk individuals. Another Swedish project is the Malmö Preventive Project that involved the screening of risk factors in 33,346 people. One quarter of those who 14

15 participated were offered preventive measures for the risks related to CVD and alcohol abuse. The main results of the project was that benefits of screening and prevention on mortality risk could only be shown in certain subgroups of younger men and women, not in the screened cohort, as compared with a non screened reference population (67). Within the County Council of Västmanland in Sweden, 34,304 health checks were carried out during the years in a CVD programme aimed at 40- and 50-year olds. The evaluations showed that the proportion of smokers had decreased, in particular, among 40-year old men. However, the proportion of 40- and 50-year olds who seldom, or never, exercised had increased among both men and women. Another problem that emerged during the 10 years of the project was a continual increase in body weight among both the men and the women who were taking part in the health checks (79). Many other projects containing screening for CVD have been carried out in Sweden, and have shown positive results pertaining to the level of risk factors (68, 80-82). Naturally, the question still arises of how projects and programmes can be more effective, and how greater attention can be given to the requirements of individuals. An evaluation of a large number of CVD programmes (65), sought to find the extent to which new knowledge, communication, information distribution, and social learning in Public Health Care work, can be used. Some questions emphasised in this report consider what the effects of social groups on society are, and what kinds of interventions that are most beneficial. One other important question is whether different intervention methods increase or decrease social differences? The population-focused strategies result in high risk and low risk individuals being treated in a similar way and a large number of them being subject to unnecessary preventive measures (83). Empirical experience furthermore indicates that even though risk factor levels in populations are declining it is specific subgroups that receive the greatest benefit from preventive efforts (67, 84). It has thus been argued that the population-focused programme approach needs to develop new methods taking knowledge about behavioural and social processes into account (15, 61, 85). From a cognitive perspective, the individual s knowledge and experience of the threat of illness constitute factors that can possibly explain risk behaviour. Another factor that may contribute to the understanding of an individual s risk behaviour is the social environment. Health models Primary prevention of CVD implies, among other things, the bringing about of behaviour change. Reaching such a goal requires an awareness of how cognitive factors affect behaviour. Knowledge is a cognitive factor and a 15

16 logical prerequisite for the intention of bringing about health related behaviour. How type of knowledge and level of knowledge are related to behaviour is important for the understanding of health promotion (86). A number of cognitive factors within health promotion have been developed into some general dimensions including, for example, perceptions of disease threat, knowledge about disease, social network characteristics and demographic factors (87). These dimensions have, thereafter, been included in a number of theoretical models where the aim is to predict health behaviour and give a structure for how interventions can be carried out (88). The Health Belief Model (HBM) is one of the most applied models in these contexts. The HBM is based on the assumption that health behaviour is dependent on individuals understanding of their own health (89). Perceived vulnerability and perceived obstacles are other factors influencing the ability of individuals to implement action that leads to behaviour change. According to the HBM, the factors contributing to behaviour change are motivation, degree of social support, and knowledge about risk factors. Family history of illness or the need for medical treatment further strengthens the reason for trying to change behaviour (90-93). Stage of Change (SOC) is another influential model often used to explain health related behaviour. According to this theory health related change is a process with different qualitative stages (94, 95). Prochaska et al identified five of these stages: pre-contemplation, contemplation, preparation, action and maintenance. The individual is, according to the model, more or less susceptible to different types of influence depending on which stage they are in (96). SOC has clear support regarding behaviour change for the individual and the group level, where specific focus has been on the structure for making people quit smoking (97-100), change exercise habits ( ) or eating habits ( ). Factors related to knowledge about risk factors One way of taking cognitive factors into consideration is to adapt the health information to specific risk groups. For example, earlier studies have shown that risk classification is important (108) and has an impact upon patients readiness to change modifiable cardiovascular risk behaviours (109). Evaluations of the effect that an educational program had on CVD prevention knowledge showed that those with a low level of education only marginally increased their CVD knowledge during a 10-year period, while those with a high level of education increased their knowledge (110, 111). Another important aspect to take into consideration when regarding individual conditions is the discrepancy between the perceived risk of illness and the actual risk of illness ( ). A longitudinal study (6 year) among 1024 women have shown that the awareness of CVD increases, but at the 16

17 same time, the discrepancy remains between perceived and actual risk of illness (119). In an American study, the knowledge about risk factors was less among individuals with increased risk behaviour. Low income and little formal education were other factors that correlated to insufficient knowledge (120). Insufficient information about diseases and a wrong perception of the condition of one s illness clearly reduced the chances for a healthy behaviour ( ). Knowledge and age are sometimes referred to as positive associated phenomena but not when it comes to knowledge about risk factors for CVD. On the contrary, particularly older men have less knowledge about risk factors for CVD (124, 125). As mentioned above there are both theoretical and practical reasons to believe that experience of illness as well as symptoms of illness may make you more aware of illness related risk factors. In line with this is has been shown that patients who believes themselves to be in generally poor health feel more vulnerable to getting any illness (126), compared to others where the illness symptoms are not as obvious (127). A study of smoking/non smoking CVD patients with elevated blood glucose levels or high blood pressure showed that general knowledge about CVD is not sufficient enough to change risk behaviour (128). The awareness of the individual concerning their state of health may furthermore be an important prerequisite for behaviour change and it is also a tool that can motivate behaviour change. Obesity related to knowledge and risk behaviour Obesity is an important risk factor for CVD. It may be influenced by behaviour and it increases the risk for developing a number of other well-known CVD risk factors such as raised blood pressure and elevated cholesterol (34, 38, 43). The fact that obesity is influenced by behaviour may suggest the hypothesis that insufficient knowledge about risk factors for CVD may contribute to obesity. An opposite hypothesis would be that obese individuals because of their condition are even more aware of risk factors for CVD than others. The Metabolic syndrome is a disease including conditions of obesity, insulin resistance, dyslipidemia, and hypertension (129, 130). According to current theories, metabolic syndrome is caused by poor diet and low physical activity (131, 132). There is a similar correlation between low education and the prevalence of risk factors for CVD and low education and prevalence of metabolic syndrome (133). This raises the question whether inadequate knowledge may be a contributing factor to obesity. The relation between level of education, obesity, and knowledge about risk factors has, as far as we are aware, only been studied to a limited degree. 17

18 Risk behaviour in relation to family history and personal experience of illness. As noted above, it may from a theoretical point of view be assumed that the experience of being given a diagnosis should lead to the patient developing an interest in obtaining knowledge about the illness and possible ways to avoid complications. Furthermore, doctors and nurses have an interest in providing the patient with information aimed at increasing awareness and changing any elements of risk behaviour. Information and personal interest as well as the actual experience of illness ought to be beneficial factors in the bringing about of more healthy behaviour. Women with personal experience of illness show higher awareness of the risk factors related to the illness than women without experience of illness (124). Other studies among women have also shown that personal experience of illness leads to more healthy behaviour (91). At the same time, there are opposite results from studies of patients with heart disease where both men and women had difficulties assessing the risk for recurring illness. There are doubts about the extent to which the actual experience of illness predicts behaviour among patients with CVD (134). Factors pertaining to family history of CVD have been well studied and constitute clear indications of increased risk of illness (18-20). Therefore, this ought to be a tool within primary care prevention for motivating healthy behaviour (135). When a close blood relative is stricken by illness, the personal perception of contracting the illness increases. However, a longitudinal study of 3,590 younger adults (18-30 years old), surprisingly showed opposite results. The study examined whether behaviour changed when a close relative was hit by heart disease. The result showed that this incident did not motivate long-lasting healthy behaviour (136). Relatives of patients with CVD are aware that there is a family history of disease. However, they have a low level of knowledge about the consequences of this risk factor regarding the development of CVD ( ). Essential questions in studying this further are whether these factors predict more healthy behaviour and whether family history and personal experience of illness influence behaviour to various extents. Are some of these factors stronger predictors than others? Influence of different social systems on behaviour and health status. The influence of social factors on people s health continues to be a problem in many countries where there is a difference in health between unequal social groups in society ( ). Another aspect that requires attention is the 18

19 health discrepancies between different countries. Concerns are often expressed about differences between rich, developed countries and poorly developed countries ( ). In a study of a number of countries in the European Union (EU), where the aim was to describe how perceptions of, and attitudes about, risk factors influence health, distinct differences between countries emerged, despite the fact that social and economic conditions were fairly similar (146). Another example of variation in Europe is the difference in life expectancy between countries in Eastern Europe and Western Europe, where the average life span is significantly lower in Eastern Europe (147, 148). Despite this difference in life span, health has improved in Eastern Europe during the last decades, above all through the reduction in the death rate from CVD ( ). The reduced death rate can be explained by improved diet and a higher quality of medical care, especially concerning high blood pressure ( ). Another measure taken to improve health conditions in the population of Eastern Europe is the introduction of modern primary health care. The transformation from communism to democracy has also contributed to the improvements in health conditions for the population (152, 153). In Western European countries, with Sweden as an example, health campaigns and different preventive projects have been carried out since the 1970s (76), while similar projects in Poland started about 20 years later ( ). This time-difference in ongoing preventive projects can be an explanation for the differences in health. Current studies show that gender, age, education, and income, even in Eastern Europe, constitute predictors for risk behaviour that are related to CVD ( ). Applying the research to the relation between health and social factors in special social contexts may help us better understand the causal mechanisms. For example, comparing general health in countries with different social systems increases the possibility to better understand why health follows a social gradient (143, 161, 162). An increased understanding of causal mechanisms may in turn lead to more effective health promotion (163). 19

20 Aims of the thesis The overall aim of the thesis was to study factors that influence healthrelated behaviour. Specific aims were: - to determine if certain background factors (e.g. gender, education and socioeconomic status) were associated with an individual s awareness and knowledge of the risk factors associated with CVD (Paper I). - to investigate whether knowledge about cardiovascular risk factors, motivation to control risk-related behaviour and actual risk behaviour differed between Poland and Sweden on the basis of information obtained during screening programme (Paper II). - to investigate whether obese individuals differ from individuals with normal weight regarding knowledge about risk factors for CVD. A further aim was to replicate previous findings that obese individuals are at higher risk of developing other biological risk factors for CVD (Paper III). - to examine how personal experience of illness and family history of cardiovascular disease (CVD), adjusted for gender, education and nationality, affect risk behaviour (Paper IV). 20

21 Methods Design The design for the different studies in this thesis consists mainly of population-based cross-sectional studies. In Study I, a population-based crosssectional study was conducted that focused on 50-year old men and women who participated in a screening of risk factors for CVD within the County Council of Västmanland, Sweden. Participants in this screening answered a questionnaire. In Study II, a comparison study was made of two populations. The participants in Study I formed one population that was compared with a Polish population. The Polish population was recruited from the Wroclaw area and took part in a cardiovascular screening that had the same aim, target group, and organisation, as the Swedish one. The gathering of data in Study II was done by a questionnaire combined with the results from the screening examination. In Study III, a selection of participants was taken from Study I. The BMI-values of the participants were the basis for a comparison study between those with a normal weight (BMI < 25 kg/m 2 ) and people with obesity (BMI 30 kg/m 2 ). The measurement tool in Study III consisted of questionnaires and screening results from the health examination. Study IV consisted of a pooling of populations from Study II whereby analyses were made within as well as between the subjects. Subgroups comprised of people with, and without, previous experience of illness or a family history of CVD. Settings Studies included in this thesis were carried out in connection with the project Hjärtvänliga Västmanland in the County Council of Västmanland (79). This programme was started in 1990 as a 10-year programme to reduce the death rate for CVD. During this 10-year period, the County Council of Västmanland began to cooperate with the Health Care system in Wroclaw, Poland. In 1998, the cardiovascular programme began in Wroclaw and is currently a permanent part of primary health care. 21

22 Data gathering for the studies included in this thesis took place partly in health centres in the County Council of Västmanland, Sweden, and partly in health centres in Wroclaw, Poland. The County Council of Västmanland, with a population of around 260,000 inhabitants, is mainly industrialised and has a certain extent of forestry and farming. Västmanland does not differ demographically from the other counties in this part of central Sweden (164). Wroclaw is a town in Poland with around 640,000 inhabitants situated in the South-Western part of the country, in a region called Lower Silesia. The area around Wroclaw is industrialised and one of the country s biggest universities is situated in the actual town. Population Study I A total of 36 Health Centres within the County Council of Västmanland were invited to participate in the health examination, and 34 Health Centres accepted the invitation. During the year that the data was gathered, there were 3,345 men and women of age 50 in the population register of Västmanland. Of these 3,345 people, a total of 1,981 took part in the health examination, distributed over 34 Health Centres. Data was gathered between May 1997 and April During this time 1,129 people were examined, and the remainders were invited before or after this time period. One thousand eleven of these 1,129 people completed questionnaires (554 women and 457 men). In Study I, individuals from various professional backgrounds were classified according to a socio-economic index (SEI) (165). Participants with university degree were allocated to SEI I (high socioeconomic status (SES) (n = 314, 31 %)), lower and middle civil servants were classified into SEI II (middle SES) (n = 167, 17 %) and workers were assigned to SEI III (low SES) (n = 433, 43 %); house worker, pensioners, self-employed and farmer (n =89, 7 %). Education was divided into three categories: low education (compulsory school/primary school)( n = 552, 54 %), medium education (completion of gymnasium/secondary school)( n = 229, 23 %) and high education (university studies)( n = 230, 23 %). Study II In Study II, the Swedish population from Study I was included with the Polish population from Wroclaw in Poland. In Wroclaw there are 89 Health Centres of differing sizes with varying types of organisation. Data from Wroclaw was gathered during year 2000 and 10 Health Centres were involved in the health examinations. Two thousand two hundred five men and 22

23 women, age 50, were registered at these health centres. Data was gathered from October to December 2000, and during this time 1,043 persons participated in health examinations. One thousand twenty of these 1,043 completed the questionnaire. Of these 1020 people 76 did not answered questions about educational level. See Table I. Table I. Demographic data. Poland n=1020 Sweden n=1011 Gender n % n % Male Female Level of education Highly educated (>12 years) Low education ( 12 years) Study III Study III comprised of a selection of two subgroups from the Swedish population in Study I. The BMI-values registered at the screening constituted the inclusion criteria. One group with 385 people comprised of those who had a normal weight (BMI < 25 kg/m 2 ). They were compared with a group of 159 people who were obese (BMI 30 kg/m 2 ). Study IV The population of Study IV consisted of both Swedish (Study I) and Polish (Study II) participants, all of them underwent screening for CVD. A total of 2,054 men and women of 50-years of age were included in Study IV, with 1,011 people from the Swedish population and 1,043 from the Polish population. Procedure Within the cardiovascular programme in the County Council of Västmanland all 50-year olds that were registered at a family doctor were invited to participate in a screening of risk factors for CVD. A District Nurse sent information and a questionnaire to those who accepted the invitation. The infor- 23

24 mation applied to the health examination, and the questionnaire contained questions about living habits (e.g. eating habits, exercise and smoking). During the screening examination a number of tests were carried out to identify risk factors. This model for screening was developed in the County Council of Västmanland, Sweden, and followed the same pattern during the year the programme lasted. The procedures for the invitation, screening examination, and health talk were administered in a similar way in Wroclaw, Poland. In Wroclaw, persons belong to the Health Centre that is geographically closest to their residential address. Before the screening examination and health talk at the Health Centre, all the study participants were asked to complete a questionnaire pertaining to knowledge about risk factors for CVD. Measurements Screening A standardised form was used for the screening of risk factors (see Appendix). The form contained questions about age, gender, number of people in the household, years of education, and employment. One part of the form contained questions about earlier illness such as diabetes, high blood pressure, myocardial infarction, angina pectoris and stroke. Two questions focused on family history whereby people stated whether their parents had become ill or died from CVD before the age of 55. Self reported living habits pertained to smoking and exercise. Smoking habits were classified into five groups (have never smoked, smoked but have stopped, smoke 1-15 cig/day, smoke cig/day, smoke > 25 cig/day). Exercise habits were assessed on the basis of a self-assessed, five-grade scale (exercising daily, exercising 3-4 times/week, exercising 1-2 times/week, exercising 1-2 times/month, exercising rarely or never). Exercising was considered to be some form of physically strenuous activity such as walking or gardening for at least 30 minutes. The physiological measurements taken were blood pressure, total cholesterol and fasted blood glucose in capillary blood, weight, height, and waist and hip circumferences. Weight and height measurements became the starting point for the calculation of BMI. These measurements were taken by nurses or laboratory personnel at the respective health centres. Total blood cholesterol concentrations were measured by the enzymatic dry method. For this purpose, we used the Reflotron system (Roche Diagnostic, Switzerland), standardised according to the manufacturer's instructions and carefully calibrated to the method used at the laboratory in the Central Hospital, Västerås (166). The individuals were recommended to fast and avoid smoking for at 24

25 least two hours before the screening. Fasting blood glucose concentration was analysed with a factory-calibrated photometer from the Swedish company, Hemocue (Hemocue AB, Ängelholm, Sweden). In Study II, smoking habits were classified into three categories (never smoked, smoked but have stopped, and smoker), and exercising habits were classified according to the original five-grade scale. In Study III, a selection was made from the population on the basis of BMI-value. Only those with a BMI < 25 kg/m 2 (n=385) or a BMI 30 kg/m 2 (n=159) were included in Study III. In the examination group for Study III, smoking habits, exercise habits, blood pressure (systolic and diastolic pressure), total cholesterol, and blood glucose, were analysed. For smoking behaviour, classification was made for smokers and non-smokers, and exercise habits were dichotomised into those who exercised at most 1-2 times per week and those who exercised 3-4 times per week or more. Study IV included questions about personal experience of illness and family history of CVD. Illness experience was based on questions of whether the individuals had diabetes, were being treated for hypertension, angina pectoris or were receiving hospital treatment for cardiac infarction or stroke. Questions concerning family history stated whether their parents had been ill or died with CVD before the age of 55. The risk behaviours analysed in Study IV were smoking habits, exercise habits, and BMI and the risk behaviour was dichotomised into smoker/non-smoker, exercise seldom or never/exercise at least 1-2 times/week and BMI 30kg/m 2 / BMI < 30kg/m 2. Questionnaire The questionnaire included in Studies I, II and III was based on a previously used questionnaire from the Danish RISIKALK-project in Vejle, Denmark (167). It contains background data questions about gender, employment, and education, as well as a number of knowledge questions about risk factors for CVD (See Appendix). In Study I, knowledge questions were divided into three main groups: - behaviour questions (e.g. smoking, exercise, diet, alcohol, stress, obesity), asking how important these types of behaviour are in avoiding the risk for CVD. The questions had four response alternatives: very important, important, insignificant, don t know. - factors (27 items) and measures (18 items) that increase or decrease the risk for developing CVD. There were three response alternatives: yes, no, don t know. - questions to assess whether the participants understanding of knowledge about causes of CVD was adequate (with three response alternatives) and a rank-ordering of the five most significant factors regarding the risk of developing CVD. 25

26 In Study II, 18 items were included that relate to knowledge about risk factors. Of the original 27 item list of factors that increase or decrease the risk, 8 items were excluded because they were considered to overlap some of the remaining items. In addition, questions were included regarding how important it is to change behaviour to reduce the risk for CVD (smoking, exercise, diet, alcohol, stress, overweight, working environment) with the response alternatives: very important, important, insignificant, don t know. The rest of the questions in the questionnaire were excluded. The Swedish questionnaire was translated to Polish with the help of people whose native language was Polish. After that, the questionnaire was proofread by district nurses in Wroclaw, Poland who were involved in the screening program. Education level, in Study II, consisted of two categories: low educated (primary school and secondary school) and high educated (college or university education). Study III included the background variables gender and education. Education was categorised in a similar way as in Study II (low and high educated). Knowledge questions were the same as in Study II, namely 18 items about factors that increase respectively decrease the risk for CVD. Statistical analyses All the calculations were performed with the Statistical Package for Social Science 12.0 (SPSS) except one analysis in study II where STATXACT (version 4.01) was used. A p-value of < 0.05 was considered as significant, except for interaction analyses where the significance level was set to p < Study I In study I chi-square test was used for comparison of non-parametric data between groups with different educational level and socio-economic status. The Spearman statistic for ordinal data was used to study the correlation on the ranking of risk factors. The unpaired t-test (two-tailed) was used to assess the ranking of CVD-associated risk factors. A factorial analysis was used to aggregate factors and interventions, which either increased or decreased the risk of acquiring CVD. Four different dimensions were established in the following way: - the dimension "diet and blood pressure" including the following factors which increase or decrease the risk for CVD: saturated fat in food, unsaturated fat in food, sugary food, small amount sugary food, a lot of cholesterol in food, small amount cholesterol in food, a lot of cholesterol in blood, small amount of cholesterol in blood, hypertension and low blood pressure. 26

27 - the dimension "exercise" including the following interventions that increase or decrease the risk for CVD: do more, or do less, exercise. - the dimensions "diet"" including the following interventions that increase or decrease the risk for CVD: restrict cholesterol diet, restrict eating of eggs and restrict salt in diet. - the dimension "coffee" including the following interventions that increase or decrease the risk for CVD: high coffee intake and low coffee intake. For each dimension the count of correct answered items were summarized, and the mean value for each participant was calculated. Multivariate analysis (multiple linear- and logistic regression) was used to assess the adjusted relation of the independent variables (gender, SES, and level of education) to the dependent variables diet and blood pressure and self-perceived need of knowledge. Study II In study II, multiple logistic regression was used for comparisons of dichotomy responses between groups from different countries (Poland = 1, Sweden = 0), gender (men = 1, woman = 0), level of education (low education = 1, high education = 0). Ordinal logistic regression was used for analyses of ordinal responses (smoking: 0 = never smoked, 1 = ex-smokers; exercise: 0 = daily, 1 = 3-4 times/week, 2 = 1-2 times/week, 3 = 1-2 times/month, 4 = never). In figure 4, smoking habits and exercise habits in relation to country are presented as proportion. Study III In study III, the chi-square test was used to test the relationship between gender, level of education, and body weight. The dimension of the variables concerning knowledge about risk factors was reduced by a factor analysis. Differences in the profiles of these dimensions for high and low educated, with respect to BMI categories (< 25 kg/m 2, 30 kg/m 2 ), were tested by a non parametric test for interactions based on aligned ranks (168). The scores that are produced have a mean of 0 and standard deviations equal to the squared multiple correlation between the estimated factor scores and the true factor values. In case of similar profiles in both categories, a further comparison using the stratified Wilcoxon rank sum test, as adopted by STATXACT (version 4.01) was run. If the null hypothesis was rejected, a Wilcoxon (Mann-Whitney) rank sum test was run within each education level and BMI category separately. Multiple logistic regression was used for comparisons of dichotomy responses (smoking and exercise) between groups with different BMI and level of education, and to assess possible interactions between these factors. 27

28 Study IV In study IV the experience of illness and family history of CVD in relation to country, gender, and level of education were presented as proportions. Multiple logistic regression was used to study the relationship between dichotomous responses and following factors: smoking habits (non smokers = 1; current smokers = 0), exercise habits (exercise regularly = 1, exercise seldom/never = 0) and actual BMI-level (BMI < 30 kg/m 2 = 1, BMI 30 kg/m 2 = 0). The analyses were adjusted for country, gender and level of education if applicable (p < 0.05). First order interactions terms were also tested for possible inclusion (p < 0.10). Ethics The Swedish part of the studies was approved by the regional human ethics committee (Uppsala University, Sweden). The head of the health authorities in the City of Wroclaw conducted the Polish part of study II and IV. 28

29 Results Study I Women, to a greater extent than men, considered it very important to change living habits to avoid becoming ill with CVD. There was a significant difference between women and men concerning the significance of changing diet habits (61 % versus 47 % p < 0.05). Likewise, women, to a higher degree than men, considered it important to try to change the work environment to reduce the risk for CVD (36 % versus 27 %, p < 0.05). More women than men reported that it was very important to avoid high alcohol intake in order to avoid CVD (47 % versus 28 %, p < 0.05). See Figure 1. % 100 Men (n=457) Women (n=554) Smoking Exercise Diet (p<0.001) Alcohol (p<0.001) Stress Overweight Work environment (p=0.002) Figure 1. The importance of changing habits that could potentially reduce the risk of CVD; differences between men and women. People with high SES, to a greater extent than those with a low SES, considered it very important to change exercise habits to reduce the risk of CVD (SES I 63 % versus SES III 51 %, p < 0.05). 29

30 In analysis of the factors that increase/decrease the risk for CVD, significant differences between men and women emerged. Regarding factor analysis, the dimension, diet and blood pressure, showed that more women than men considered factors that were included in this dimension to increase/decrease the risk for CVD. The dimension, diet and blood pressure, includes the factors: saturated fat in food, sugary food, high cholesterol levels in food, high cholesterol levels in blood and hypertension. A comparison of groups with different SES levels showed that people with high SES had a better knowledge of the factors concerning diet and blood pressure. For this same dimension, it was shown that the high educated had a significantly better knowledge than the low educated. Significant results between the high and the low educated emerged when participants stated which measures reduce the risk for CVD. The high educated, to a greater extent, referred to measures that relate to the diet dimension. Factors included in the diet dimension were: restrict cholesterol diet, restrict eating of eggs, and restrict salt in diet. Moreover, women with high education and high SES had significantly better knowledge about the risk factors related to CVD in comparison to women with low education and low SES (p < 0.05). Similar results emerged for men with respect to education level and SES (p < 0.05). The continuous dependent variable diet and blood pressure was tested with multiple linear regression. The analysis showed significant relation for each independent variable (gender, SES), controlled for the effects of the other (p 0.05). When participants ranked risk factors, smoking was viewed as the most important risk factor regarding an increased risk in the development of CVD for both men and women. Other factors showing significant differences were blood pressure (more important for women) and exercise (more important for men). A very high correlation was found between those with different SES and level of education (high SES versus low SES r s = 0.96; high educated versus low educated r s = 0.99) on the ranking of risk factors for CVD. Study II Study II, where a Swedish population was compared to a Polish population, showed that more Swedes considered it very important to change dietary habits to reduce the risk for CVD (Sweden 54 % versus Poland 23 %; OR = 0.25; 95 % CI = ; p < 0.001). On the other hand, Poles, to a higher degree than Swedes, considered it important to influence stress in order to reduce the risk for CVD (Sweden 60 % versus Poland 72 %; OR 1.7, CI ; p < 0.001). In the Swedish population the work environment was also emphasised as an important factor to reduce the risk of developing CVD (Sweden 32 % versus Poland 27 %; OR 0.8; 95 % CI = ; p = 0.033). 30

31 These differences between Sweden and Poland also remained when the data was adjusted for gender and education level. See Figure 2. % 100 Sweden (n=1011) Poland (n=1020) Smoking Stress (p<0.001) Overweight Exercise Alcohol (p=0.014) Work environment (p=0.033) Diet (p<0.001) Figur 2. The importance of changing habits that could potentially reduce the risk of CVD; differences between Sweden and Poland. With adjustments made for nation and education level, the men regarded diet habits, stress factors, alcohol habits, and working environment to be less important factors than the women did. When adjusted for nation and gender, the high educated, to a greater degree than the low educated, considered it important to change diet habits, while the low educated considered it more important to change alcohol habits. The Polish participants (adjusted for gender and level of education) stressed significantly (p < 0.05) more often than the Swedish participants that smoking, a stressful life, high blood pressure, high levels of cholesterol in food, high blood cholesterol, high alcohol intake, saturated fat in food, old age, diabetes, sugary food, and unemployment were factors which increased the risk of developing CVD. On the other hand, the Swedish participants reported, to a significantly greater extent, that heredity increased the risk and that plenty of fibre in food reduced the risk (p < 0.05). When comparing the Swedish and Polish study populations, altered smoking behaviour was more common among Swedes than among Poles. Thirty-nine percent of Swedes were ex-smokers compared to 23 % of the Poles (OR = 0.4; 95 % CI = ; p < 0.001). In the examination group, 31

32 more Poles were smokers than Swedes (33 % versus 25 %; OR = 1.7; 95 % CI = ; p < 0.001). These differences were also significant after adjusting for gender and education level. As Figure 4 shows, Swedes exercised more regularly than Poles, even when the data was adjusted for gender and education level (OR = 0.2; 95 % CI = ). See Figure % Smoking habits Poland (n=1041) Sweden (n=1011) Exercise habits Never smoked Exsmokers Smokers Exercise 3-4 times/ week Daily exercise Exercise 1-2 times/ week Exercise 1-2 times/ month Never exercise Figure 3 Distribution of smoking and exercise habits among Polish and Swedish participants. Study III Regarding comparisons of education level and BMI, a significantly higher proportion of the low educated were obese compared to the high educated (33 % versus 18 %, p < 0.002). In line with the results in Study III, there was no difference between obese people and those with a normal weight regarding knowledge about the risk factors for CVD. With reference to analysis of the interaction effect between education level and BMI, education level was the strongest factor when knowledge of the risk factors was tested. Within the group of normal weighted people, the high educated showed a significantly better knowledge than the low educated about the fact that the risk is increased by saturated fat in food, high blood cholesterol, and cholesterol-rich food. Both obese and normal weight people who were high educated had greater knowledge than the low educated about how risk factors such as diabetes, hypertension, diet, family history, and alcohol habits, relate to CVD (Table 2). A significant interaction existed between education level and BMI for systolic and diastolic blood pressure, as well as for total blood cholesterol. Low educated people with obesity had significantly higher systolic blood 32

33 pressure (140 mmhg versus 125 mmhg; p < 0.001), diastolic blood pressure (90 mmhg versus 80 mmhg; p < 0.001) and total blood cholesterol (5.8 mmol/l versus 5.5 mmol/l; p = 0.004) than low educated people with a normal weight (Table 3). People with normal weight exercised with more regularity than those who were obese (49 % versus 36 %; p = 0.002) and the high educated, irrespective of BMI level, tended to exercise more often than the low educated (50 % versus 44 %; p = 0.082). Smoking was more common among normal weighted participants than among obese participants (28 % versus 19 %; p = 0.03). See Table 3. 33

34 Table 2. Knowledge about CVD risk factors in relation to educational level and body weight expressed as median of dimension scores (Q1, Q3). Dimension (factor-loading within brackets) Dimension 1 Obesity (0.709) Fatty food (0.696) Smoking (0.693) Stressful life (0.630) Dimension 2 Saturated fat in food (0.587) High blood cholesterol level (0.722) Cholesterol rich food (0.816) Dimension 3 Diabetes (0.647) Sugary food (0.476) Hypertension (0.489) Heredity (0.418) High alcohol intake (0.547) Dimension 4 Physical inactivity (0.779) Small amount fibre in food (0.684) BMI<25 kg/m 2 n= (-0.01,0.47) 0.55 (-0.13,0.64) 0.30 (-0.33,1.03) 0.62 (-0.42,0.89) High-educated Low-educated BMI 30 kg/m 2 BMI<25 kg/m 2 n=23 p n= (-0.32,0.50) 0.55 (-0.43,0.71) 0.06 (-0.39,0.49) , (-0.08,0.50) 0.37 (-0.36,0.64) (-0.91,0.90) 0.03 (-0.69,0.82) BMI 30 kg/m 2 Interaction n=136 p p-value 0.21 (-0.03,0.54) 0.44 (-0.24,0.68) (-0.81,0.62) 0.26 (-0.74,0.76)

35 Table 3. Individual risk factors in relation to educational level and bodyweight presented as percentage or median (Q1,Q3). High-educated Low-educated Smoking habits Current smoker (%) BMI<25 kg/m 2 n=102 BMI 30 kg/m 2 n=23 p BMI<25 kg/m 2 n=283 BMI 30 kg/m 2 n=136 p Interaction p-value Exercise habits Exercise least 3-4 times/week (%) Blood pressure Systolic pressure mmhg 120 (111, 135) 130 (120, 142) (115, 140) 140 (130, 150) Diastolic pressure mmhg 80 (70, 85) 85 (80, 90) (75, 85) 90 (85, 98.7) Total blood cholesterol mmol/l 5.65 (5.0, 6.3) 5.3 (4.9, 6.1) (4.9, 6.3) 5.8 (5.3, 6.5) Fasting blood glucose mmol/l 5.3 (4.9, 5.8) 5.2 (4.7, 5.8) (4.8, 5.7) 5.4 (4.9, 6.0)

36 Study IV Of the total number in the study group, 19 % (n = 388) had personal illness experience and 21 % (n = 424) had a family history of CVD. Considerably more Poles than Swedes had personal experience of illness (Poland 27 % versus Sweden 10 %) and a family history of CVD (Poland 27 % versus Sweden 14 %). Among both Poles and Swedes, those with a low education had a greater extent of illness experience than those who were high educated (Poland low educated 29 % versus high educated 23 %; Sweden low educated 12 % versus high educated 5 %). There was a significant association between family history and personal experience of illness (x 2 = , p < 0.001). The results showed that those who had personal experience of illness more often were non-smokers than those who did not have experience of illness; adjusted for family history, gender, level of education, and nationality (OR = 1.48; 95 % CI = ) (Table 4). Family history of CVD, adjusted for illness experience, gender, level of education and country, did not lead to more positive smoking behaviour (OR = 0.84; 95% CI = ). These results indicated that experience of illness influences smoking behaviour to a higher degree than family history. Family history and personal illness experience of CVD were not predictive of more physical exercise compared to participants with no illness experience or no family history of CVD. Swedes had increased odds (OR = 2.82; 95 % CI = ) for exercising regularly compared to Poles; adjusted for illness experience, family history, gender, and level of education (Table 4). The results also showed an interaction effect for illness experience and nationality. Poles with experience of illness exercised more rarely than Swedes with experience of illness. Furthermore, the results showed that those who had illness experience and a family history of CVD were more often obese than those with no illness experience and no family history of CVD; adjusted for gender, nationality, and level of education. Poles had increased odds for obesity compared to Swedes, and more men than women were extremely overweight. The interaction between family history and nationality showed that there were increased odds that Poles with a family history of CVD were obese. In addition, for participants with family history the odds for being obese were greater among men than women. See table 4. 36

37 Table 4. The influence of adjusted experience of illness (reference = no experience of illness) and family history (reference = no family history) of CVD on smoking (event = non-smoker), exercise habits (event = exercise regularly) and obesity (event = non-obese) including first order interactions. Smoking habits p- value OR 95% CI Lower Upper Experience of illness (no = 0)* Family history (no = 0)* Exercise habits Experience of illness Family history Country (Poland = 0)* Experience of illness x Country Obesity Experience of illness Family history Country Gender (Men = 0)* Family history x Country Family history x Gender * reference category 37

38 Discussion The primary aim of this thesis was to study factors that influence healthrelated behaviour. The results showed that knowledge varies with education, gender and, to a certain degree, nationality. Moreover, knowledge was influenced by behaviour related medical conditions such as obesity. However, knowledge is not the only condition for behaviour change. The results of the present thesis support earlier theories that describe risk behaviour change as a process. This process with reference to behaviour change is thought to be facilitated, to a certain degree, by personal experience of illness, at least in the inclination of stopping smoking. Processes of change can also be discerned on a national level since Poles, despite having good knowledge about risk factors, changed their behaviour to a lesser extent than Swedes. One way of explaining this can be that public health work focused on the prevention of CVD has had a shorter history in Poland. A starting point has been the idea that public health work could be more effective if more consideration was given to people s different abilities to take on board primary preventive interventions. Prevention in the area of CVD has been described as a scientific dilemma (83). The dilemma in population-focused strategies is that high risk and low risk individuals are treated in a similar way and that a large proportion of the individuals are unnecessarily subjected to preventive measures (83, 169). This thesis indicates that unnecessary preventive work could be avoided if a greater degree of consideration was given to individual conditions such as level of knowledge and perceived threat of illness. This strategy with a focus on cognitive aspects should make a more effective preventive health project and, at the same time, fulfil internationally recommended guidelines for the prevention of CVD (54). These guidelines stress that individuals should understand the connection between behaviour, health, and illness, and also that personnel must help individuals to become more aware of the barriers that can complicate behaviour change. One result that emerged in the present thesis was that those with a low education had insufficient knowledge about the significance of risk factors for developing CVD. This result is consistent with several findings from studies with other designs (111, 120, 125, ). 38

39 One explanation for these findings may be that health information in the media and through campaigns may sometimes not reach the low educated. Another explanation can be that the information is designed in a way that the low educated can not make use of. General knowledge about the facts that increase the risk for CVD is relatively good. However, specific knowledge is more inadequate ( ). Knowledge among the low educated, above all, was insufficient concerning diet and eating habits. Other studies have also shown that people have insufficient knowledge about the content of food products and the risk of developing illness (176, 177). Insufficient knowledge among the low educated has consequences for behaviour (178). Studies indicate that the low educated eat more fatty food and less fruit and vegetables compared to the high educated (179). Insufficient time for cooking and the high cost of food were mentioned by low educated women as reasons for bad eating habits (179). Gender was an additional factor that had a correlation to knowledge about risk factors for CVD, where women had better knowledge than men. This confirms previous research (180, 181). In a study consisting only of men, this lack of knowledge was corroborated and primarily related to knowledge about diet (182). A traditional gender perspective states that women take greater responsibility for health and factors relating to health within families (183, 184). Despite the fact that women and men spend a similar amount of time at work, women are more often responsible for food and eating habits at home (179, 185). One explanation for the findings may be that women s responsibility for cooking and eating habits result in an increased involvement and interest in questions concerning diet and health. In general, women indicate that things such as the change of eating habits come about from dietary advice and recommendations. Women willingly choose to read magazines where health related issues are central themes. Gender differences in knowledge about diet also varies with level of education and socioeconomic status, where differences are more pronounced between, rather than within, socioeconomic groups (48, 186, 187). The results in Study II showed that Poles had good knowledge about the risk factors for CVD. However, Swedes had less risk behaviour. One example of this was that more Swedes than Poles had stopped smoking. The fact that the discrepancy between knowledge and behaviour was greater in Poland than in Sweden suggests that behavioural change is a gradual process. Sweden may have come further in this process of change because preventive health projects have been going on for a long time. The first preventive health projects in Poland were started 20 years after such projects started in Sweden (76, 156). Swedes indicated that it is very important to change eating habits to avoid CVD, while Poles rated stress factors as more significant. Economic 39

40 difficulties, which characterised Poland during the last decade, may naturally have contributed to a lower priority being given to good eating habits in everyday life. People in different social groups are exposed to social and economic circumstances to different extents, a fact that contributes to differences in risk behaviour and health (188, 189). According to the results in Study III, insufficient knowledge was not the primary reason for obesity. Difference in knowledge about risk factors appeared to be more pronounced among low and high educated individuals than when normal weight and obese people were compared. The results in Study III indicated that obese individuals were, to a great extent, low educated. Preventive primary work with obesity is complicated because the reasons behind obesity are both genetic and dependent on behaviour and, in addition, there are a number of other known risk factors for CVD. Current studies show that health information and counselling of individuals has decreased and this has primarily affected the low educated (190). Preventive work with those who are obese requires new strategies, whereby a multiplicity of different factors needs to be taken into account (48). Surveying both level of knowledge and self-efficacy ought to be starting points to make these strategies successful ( ). It seems to be a reasonable assumption that perceived threat of illness will predict increased health behaviour. However, contrary to this logic, Study IV suggested that personal experience of illness and a family history of CVD did not influence health behaviour to any great extent. The only exception was that personal experience of illness was associated to more healthy smoking behaviour, evidence that information about the dangers of smoking have been successful (195). One reason that threat of illness did not affect other risk behaviour may be that information did not reach out to or was badly adapted for the individual. The same condition showed in relation to family history as a risk factor, where a number of studies have indicated insufficient knowledge about family history as the reason ( ). The utilisation of personal illness experience and family history to motivate behaviour change is a strategy that ought to be used to a greater extent in preventive work (135, ). Knowledge about risk factors and the threat of illness have been central concepts in this thesis. From a cognitive theoretical perspective, knowledge of illness and perception of the threat of illness constitute predicting factors for behaviour (87, 199). According to the HBM (Health Belief Model) the information that individuals receive from mass media and health campaigns consists of signals to start behaviour change, but if this information is not adapted to the individual it may not achieve its goal. Personal experience of illness and family history, according to the HBM, also constitute the factors that, theoretically, will lead to changed behaviour. From the results of Study 40

41 IV, only smoking behaviour tended to be affected by personal experience of illness. The HBM, to a certain degree, can describe what predicts behaviour but have difficulties in describing what factors bring about behaviour change. This is a limitation in the model which emerged when the HBM was tested (92, 200, 201). In order to increase the practical benefits of the HBM, preventive work must pay more thorough attention to the cognitive factors of HBM. One example regarding individual health counselling can be to have more detailed discussions of perceptions of the threat of illness, and the possibilities and hindrances for behaviour change. The SOC model has been used previously on individual and group levels (100, 101, , 202) but the results of Study II indicates that SOC also is applicable to the national level. The finding that more Poles than Swedes were smokers and that a greater proportion of the Swedish population had stopped smoking may indicate that many Poles were in a phase of precontemplation. That is, even though they may be aware of the dangers of smoking they may not relate this to their own behaviour. Other possible phases are contemplation (contemplating risk behaviour) and preparation (preparing for behaviour change). The SOC model may thus be a helpful framework for explaining lags in behaviour change between Eastern and Western European countries. Methodological considerations In the screening programs that provided data to the present thesis some 60 % of the entire target group in Sweden participated in the screening. In Poland the corresponding figure was about 50 %. The fact that a great proportion of the target group did not participate poses a risk for systematic bias. Nonparticipants were studied in the County Council of Västmanland s cardiovascular programme a few years before the actual gathering of data. A nonparticipants analysis was done on a randomised selection of 112 people at 8 health centres. The result showed that low educated people, men, and smokers were over represented among those who did not participate in health examinations (164). No systematic differences between participants and nonparticipants emerged for other variables. The demographic differences between participants and non-participants that emerged with the non-response analysis fits very well with the results of other studies and the bias effects there were also considered marginal ( ). In the World Health Organisation s MONICA Project, concerns were expressed over respondents rates in community based surveys of cardiovascular risk factors. Declining respondents rates pose a threat to the accuracy of estimates of risk factor trends in many countries. These trends seem to follow the same pattern among populations that differ geographically and 41

42 culturally. Individuals who do not take part in health examinations are generally young, unmarried, low educated men with an unhealthy lifestyle (205, 209). Because Poland is one of the countries included in the MONICA Project, support is given to the assumption that the Polish population follows the same pattern when it comes to non-response. In the Polish population, a comparatively high number of the participants were high educated which can be explained by the fact that Wroclaw is a university town. The distribution of participants according to education levels would probably have been different if the population had been more of a mixture of town and rural populations. Earlier studies in Poland show that populations in smaller towns and in the countryside have insufficient knowledge about risk factors for CVD as well as greater risk behaviour (210, 211). Another possible bias effect could be that the gathering of information in Poland was done three years after the Swedish data collection. However, this effect would be likely to bias results against the lag in behavioural change noted among poles as compared to Swedes, rather than for it. Self-reported smoking and exercise habits were included in the screening of risk factors. The gathering of information with an instrument that involves self-reporting carries the risk of bias effects through the study participants tendency to underestimate smoking habits (212) and overestimate exercise habits. The element in the thesis that counteracts this bias effect is having a clear definition of what is meant by physical activity (213, 214). In order to strengthen the validity of self-reporting, combinations of measurement instruments are recommended. Self-reports of smoking habits, obesity, and diabetes has relatively good reliability, but physiological tests are recommended as complements to self-reports (215). In order to increase the reliability further for measuring smoking habits, proxy reporting is a method that complements self-reporting (216). Individual awareness about the presence of risk factors needs to improve before self-reporting can be used as a reliable instrument in risk factor screening (217). In this thesis a cross-section design was applied. This means that conclusions regarding the directions of cause-effect relationships must be drawn with caution. So for instance the finding of a relation between obesity and a tendency not to exercise may be caused by an influence of obesity on exercise (it may be harder to exercise when you re fat) but it may also be caused by an influence of exercise on obesity (if you don t exercise you may become fat). In cross sectional studies where all variables are measured at the same time and no variable is experimentally manipulated, definite conclusions regarding these kinds of directions cannot be drawn. The questionnaire from which data was derived was a revised version of a questionnaire that had previously been used in a cardiovascular project (Risikalk-projektet) in Vejle, Denmark (167). The knowledge-related questions were designed to be comprehensible to a lay audience (218). Questions regarding CVD knowledge were of a general character and were assessed as 42

43 being relevant for the risk of developing CVD. One reason for the choice of questions was that the questions should focus on lay knowledge, as well as starting from a logical and rational basis (218). Having this starting point and clarifying the relation between lay and expert knowledge has been put forward as an essential aspect in giving advice and information within public health work (219). One of the strengths of this thesis is that the results were population-based which makes it possible to generalise to a greater extent than is the case with random samples. The Swedish study population, with a mixture of people from the town and countryside, can probably be considered representative of the Swedish general population. The Polish material differed from the Swedish mainly in that the town of the study has one of the largest universities in Poland. One factor probably influencing the results in Study II could be a difference in attitude between residents in post-communist countries in Central Europe and residents in Western European countries. In general it is considered that post-communists has a more explicitly sceptical attitude to authorities where one easily adopts a wait and see attitude (220), which can influence the effects of health campaigns organised by society. In study I educational level and socioeconomic status were divided into three groups, and in study II, III and IV educational level was divided into two groups. In Sweden, 12 years of education comprises nine years of compulsory primary schooling and three years of secondary schooling. Thereafter the next level of schooling is university studies. Twelve years is thus a natural cut-off point when the Swedish educational system is considered. The study group in this thesis comprised of a single age group, which is a limitation. Nonetheless, such selection may be beneficial in some instances. For example, because of the standardization of age, it was easier to make comparisons between other important background factors. 43

44 Conclusions - Men with low educational level and low socio-economic status had deficient knowledge about risk factors related to CVD. - Processes of behaviour change can also be discerned on a national level since Poles, despite having good CVD knowledge about risk factors, changed their behaviour to a lesser extent than Swedes. - Obese individuals are at an increased risk of developing other risk factors for CVD, but are just as knowledgeable about risk factors for CVD as normal weighted individuals. - Individuals with a personal experience of illness may be more inclined to change smoking behaviour than the average person. 44

45 Epilogue and implications He lives a dangerous life, is 48 years old, overweight, stressed and untrained. The fact that he is on medication for high blood pressure and feels a pressure over his chest when he goes up and down the stairs does not make the situation better. There are obvious signs that something is wrong and that he should see a doctor. What should the health professional do to make him understand how serious his situation is? A basic condition for behaviour change is knowledge about risk factors. The results show that knowledge varies with education, gender and, to a certain degree, nationality. Moreover, knowledge is influenced by behaviour related medical conditions such as obesity. However, knowledge does not only consist of the conditions for behaviour change. The results in the present thesis substantiate earlier theories about risk behaviour change as being a process. This process may to a certain degree be facilitated by personal experience of illness, at least when it comes to the inclination of stopping smoking. In summary, the implications of these results in the meeting with the individual patient are that it can be important to take a series of factors into consideration. These factors steer the patient s ability to understand and to consider the consequences from the information given to her/him by the clinician. One dilemma within primary preventive work is that similar interventions often have different effects on different individuals and the ability of individuals to take information on board can vary. Predictors of risk behaviour on the individual level as well as national level may be important for health care professionals to understand and take into consideration in their everyday practise. 45

46 Acknowledgements I would like to express my sincere gratitude to all those who have supported me during the work with this thesis. In particular, I wish to thank: All those involved in the screening programme in the County of Västmanland, Sweden and in Wroclaw, Poland. Without your contribution this work would not have been possible Jerzy Leppert, my excellent main supervisor and co-author, for skilful supervision in the world of science, continuous guidance, encouragement, never-ending patience and support during the years that finally led to this thesis. Rickard L. Sjöberg, my supervisor and co-author, for encouragement and a never ceasing flow of constructive criticism. You have widened my horizons in the scientific realm and have always been available for consultations. John Öhrvik, co-author and statistic advisor, for being helpful with dataanalysis. Wladyslaw Sidorowitz, co-author, for being friendly and very helpful in giving me access to the screening programme in Wroclaw. Marzena Krysa, co-author and friend, who was mainly responsible for data collection in Wroclaw. Kent Nilsson, statistical advisor, for sharing his wisdom and giving me guidance and also for interesting discussions about the scientific world. Bengt Starrin for fruitful discussions and good advice. Marie-Louise Engström-Walker for help with data analysis and statistical advice at the beginning of this project. Birgitta Ellenius, for support and for being the link to the Health Care Centres in the County of Västmanland. 46

47 All district nurses in the County of Västmanland and Wroclaw, Poland for assistance during this project. All personnel at Centre for Clinical Research for kind and respectful support, especially to Katarina Ringström, Maria Dell Uva Karlsson, Tony Wiklund, and Gun Nyberg. Gunnel Kristiansson, Håkan Sandberg, och Roland Svensson, Head of the Department of Caring and Public Health Sciences, Mälardalen University, to whom I am grateful for the opportunity to share my time between lecturing in the nursing programme and being a doctoral student. All my colleagues at the Department of Caring and Public Health Sciences, Mälardalen University, for their support and for a pleasant working environment. Special thanks to Lillemor Fernqvist, Per Ekstrand, Gabriella Engström and Karin Weman for friendship and encouragement during this project. Henny Olsson for support and guidance at the beginning of this project. Jeanette Kliger and Shirley Blair Warg for linguistic assistance. Special thanks to Cecilia Åslund for an excellent and final linguistic revision. Peter Hedberg at Mälardalen University Library, who was always efficient and searched for references and literature without hesitation. Finally, to Bitte, my wife, you are the most important to me. Without your love, words of wisdom and support, I would not have been able to complete this project. To my beloved children, Erik and Anna, thank you for reminding me of the greatest values in life. 47

48 Summary in Swedish; svensk sammanfattning Hjärt-kärlsjukdomar är den grupp sjukdomar som orsakar flest förtida dödsfall i Sverige. Detta förhållande vid hjärt-kärlsjukdomar gäller de flesta länder i Europa. Ett stort antal riskfaktorer kan kopplas direkt till hjärtkärlsjukdom, där övervikt utgör den riskfaktor som under det senaste årtiondet visat en klar ökad prevalens. Att förebygga hjärt-kärlsjukdomar har högsta prioritet och primär prevention är ett område som kontinuerligt måste utvecklas med nya och mer effektiva strategier. De tidigare befolkningsinriktade strategierna har i flera fall medfört att hög-risk och låg-risk individer behandlats på ett likartat sätt och att ett flertal av individer i onödan utsatts för förebyggande åtgärder. Det övergripande syftet med denna avhandling var att i en population av 50-åriga män och kvinnor studera faktorer som påverkar beteende vid risk för hjärt-kärlsjukdom. I studie I testades eventuellt samband mellan kunskap om riskfaktorer och kön, utbildning och socioekonomiskt status, och i studie II studerades kunskap och riskbeteende i ett nationellt perspektiv (Sverige vs Polen). Vidare undersöktes i studie III om obesa individer (BMI 30 kg/m 2 ) skiljde sig från normalviktiga (BMI < 25 kg/m 2 ) med avseende på kunskap om riskfaktorer och i studie IV undersöktes om riskbeteende påverkades av egen sjukdomserfarenhet och hereditet för hjärt-kärlsjukdom. Studierna är populationsbaserade med tvärsnittsdesign och resultaten baseras på individens kunskap om riskfaktorer vid hjärt-kärlsjukdom. I datainsamlingen ingår även testresultat från screening, där syftet var att identifiera riskfaktorer. Studierna genomfördes i samband med hälsoundersökning av 50-åriga män och kvinnor i Västmanland, Sverige (n = 1011) och i Wroclaw, Polen (n = 1043). Den svenska populationens kunskap om riskfaktorer studerades i relation till kön, utbildning och socioekonomisk status (studie I). Wroclaw i Polen har tillsammans med Landstinget Västmanland utarbetat en gemensam modell för screening av riskfaktorer och det möjliggjorde en jämförande studie mellan en svensk och en polsk population (studie II). I denna jämförelse ingick formuläret med kunskapsfrågor tillsammans med resultat från hälsoundersökningen. Den svenska populationens BMI-värde var underlag för en jämförande studie (III) mellan normalviktiga personer (BMI < 25 kg/m 2 )(n=385) och personer med obesitas (BMI 30 kg/m 2 )(n = 159). Mätinstrument i studie III utgjordes av frågeformulär och resultat från hälsoundersökningen. I studie IV var undersökningsgruppen 2054 personer (Sverige n = 1011, Polen n = 1043) där samtliga genomgått hälsoundersökning. 48

49 Deltagarna i studie IV fick ange om de hade erfarenhet av hjärt-kärlsjukdom och/eller ärftlighet inom familjen för hjärt-kärlsjukdom, dessutom ingick uppgifter om aktuell BMI-nivå och självrapporterade rök- och motionsvanor. Resultaten från studie I visade att kvinnor hade bättre kunskap än män om riskfaktorer för hjärt-kärlsjukdom samt att låg utbildning och låg socioekonomisk status var associerade till bristfällig kunskap. Vid jämförelser som gjordes mellan män och kvinnor med olika socioekonomisk bakgrund och utbildningsnivå visades att kvinnor genomgående hade bättre kunskap om riskfaktorer relaterade till hjärt-kärlsjukdom oberoende av socioekonomisk bakgrund. De tydligaste skillnaderna i kunskap gällde faktorer som var relaterade till kostvanor. Vidare framkom i studie II att polacker hade minst lika goda kunskaper som svenskar om riskfaktorer för hjärt-kärlsjukdom, men svenskar ansåg att det är viktigare att ändra kostvanor och påverka faktorer i arbetsmiljön för att undvika risken att utveckla hjärt-kärlsjukdom. Svenskar hade också ett mindre riskbeteende än polacker och fler svenskar än polacker hade lyckats med att sluta röka. Resultaten från studie III visade att personer med obesitas skilde sig inte från normalviktiga när det gällde kunskap om riskfaktorer för hjärtkärlsjukdom, ett resultat som gällde oberoende utbildningsnivå. Obesitas och låg utbildning hade däremot tydligt samband med andra riskfaktorer relaterade till hjärt-kärlsjukdom, som exempelvis högt blodtryck och högt serumkolesterol. I studie IV visade resultaten att personlig erfarenhet av sjukdom påverkar rökbeteende i högre grad än vad ärftlighet för hjärt-kärlsjukdom gjorde. Bland obesa individer däremot, tycktes inte erfarenhet av tidigare sjukdom vara ett hot som motiverar till normalisering av vikten. Ärftlighet av hjärtkärlsjukdom hade inte någon inverkan på rökvanor, motionsvanor eller förekomst av obesitas. En grundförutsättning för en individs beteendeförändring är kunskap om riskfaktorer och där visade resultaten att kunskap varierar med utbildning, kön och i viss mån nationell tillhörighet. Dessutom påverkades individens kunskap av beteenderelaterade medicinska tillstånd som obesitas. Kunskap utgjorde dock inte den enda förutsättningen för förändring av beteende. Resultaten i avhandlingen bekräftade tidigare teorier om att förändring av riskbeteende är en process. Denna process vid beteendeförändring kunde i viss mån underlättas av egen sjukdomserfarenhet, åtminstone benägenheten att sluta röka. Resultaten i denna avhandling antyder att förändringsprocesser kan studeras på nationell nivå eftersom polacker trots goda kunskaper om riskfaktorer i mindre utsträckning än svenskar förändrade sitt beteende. En förklaring till detta kan vara att folkhälsoarbete i syfte att förebygga hjärtkärlsjukdomar har en kortare historia i Polen. Folkhälsoarbete måste i större utsträckning ta hänsyn till människors olika förutsättningar. När det gäller riskfaktorer för hjärt-kärlsjukdom visade 49

50 resultaten i denna avhandling att ett flertal faktorer har betydelse för att förstå en risksituation. Ett första steg i preventivt arbete är att förvissa sig om vilken kunskap som finns om riskfaktorer och att denna kunskap kan variera med hänsyn till kön och utbildning. Diskrepansen mellan kunskap och beteende kan variera dels beroende på riskbeteende och dels vilken fas till eventuell beteendeförändring som individen befinner sig i. Om preventiva insatser skall lyckas krävs att man kan identifiera i vilken fas individen befinner sig och sedan anpassa åtgärder efter det. Att använda sjukdomshot som motiverande faktor vid prevention kan vara en åtgärd för att åstadkomma ett mer hälsosamt beteende. 50

51 References 1. Janlert U, Folkhälsoinstitutet. Folkhälsovetenskapligt lexikon. Stockholm: Natur och kultur i samarbete med Folkhälsoinstitutet; Altman DG, Schulz KF, Moher D, Egger M, Davidoff F, Elbourne D, et al. The revised CONSORT statement for reporting randomized trials: explanation and elaboration. Ann Intern Med Apr 17;134(8): Kasl SV, Cobb S. Health behavior, illness behavior, and sick role behavior. I. Health and illness behavior. Arch Environ Health Feb;12(2): Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults--The Evidence Report. National Institutes of Health. Obes Res Sep;6 Suppl 2:51S-209S. 5. WHO. Largely preventable chronic diseases cause 86% of deaths in Europe: 53 WHO European Member States map a strategy to curb the epidemic: W0rld Health Organization; Socialstyrelsen. Folkhälsorapport. Stockholm: Socialstyrelsen; Persson G, Danielsson M, Rosen M, Alexanderson K, Lundberg O, Lundgren B, et al. Health in Sweden: the National Public Health Report Scand J Public Health Suppl Jun;67: Truelsen T, Mahonen M, Tolonen H, Asplund K, Bonita R, Vanuzzo D. Trends in stroke and coronary heart disease in the WHO MONICA Project. Stroke Jun;34(6): Levi F, Lucchini F, Negri E, La Vecchia C. Trends in mortality from cardiovascular and cerebrovascular diseases in Europe and other areas of the world. Heart Aug;88(2): Sans S, Kesteloot H, Kromhout D. The burden of cardiovascular diseases mortality in Europe. Task Force of the European Society of Cardiology on Cardiovascular Mortality and Morbidity Statistics in Europe. Eur Heart J Aug;18(8): Leander K, Hallqvist J, Reuterwall C, Ahlbom A, de Faire U. Family history of coronary heart disease, a strong risk factor for myocardial infarction interacting with other cardiovascular risk factors: results from the Stockholm Heart Epidemiology Program (SHEEP). Epidemiology Mar;12(2):

52 12. Ho JE, Paultre F, Mosca L. The gender gap in coronary heart disease mortality: is there a difference between blacks and whites? J Womens Health (Larchmt) Mar;14(2): Hu G, Tuomilehto J, Silventoinen K, Barengo N, Jousilahti P. Joint effects of physical activity, body mass index, waist circumference and waist-to-hip ratio with the risk of cardiovascular disease among middle-aged Finnish men and women. Eur Heart J Dec;25(24): Greenland P, Knoll MD, Stamler J, Neaton JD, Dyer AR, Garside DB, et al. Major risk factors as antecedents of fatal and nonfatal coronary heart disease events. Jama Aug 20;290(7): Sigurdsson EL, Thorgeirsson G. Primary prevention of cardiovascular diseases. Scand J Prim Health Care Jun;21(2): Stamler J, Stamler R, Neaton JD, Wentworth D, Daviglus ML, Garside D, et al. Low risk-factor profile and long-term cardiovascular and noncardiovascular mortality and life expectancy: findings for 5 large cohorts of young adult and middle-aged men and women. Jama Dec 1;282(21): Yarnell J, Yu S, Patterson C, Cambien F, Arveiler D, Amouyel P, et al. Family history, longevity, and risk of coronary heart disease: the PRIME Study. Int J Epidemiol Feb;32(1): Myers RH, Kiely DK, Cupples LA, Kannel WB. Parental history is an independent risk factor for coronary artery disease: the Framingham Study. Am Heart J Oct;120(4): Nilsson PM, Nilsson JA, Berglund G. Family burden of cardiovascular mortality: risk implications for offspring in a national register linkage study based upon the Malmo Preventive Project. J Intern Med Feb;255(2): Scheuner MT, Wang SJ, Raffel LJ, Larabell SK, Rotter JI. Family history: a comprehensive genetic risk assessment method for the chronic conditions of adulthood. Am J Med Genet Aug 22;71(3): Rose G, Kumlin L, Dimberg L, Bengtsson C, Orth-Gomer K, Cai X. Work-related life events, psychological well-being and cardiovascular risk factors in male Swedish automotive workers. Occup Med (Lond) Sep;56(6): Brezinka V, Kittel F. Psychosocial factors of coronary heart disease in women: a review. Soc Sci Med May;42(10): Kuper H, Marmot M. Job strain, job demands, decision latitude, and risk of coronary heart disease within the Whitehall II study. J Epidemiol Community Health Feb;57(2): Hallman T, Burell G, Setterlind S, Oden A, Lisspers J. Psychosocial risk factors for coronary heart disease, their importance compared with other risk factors and gender differences in sensitivity. J Cardiovasc Risk Feb;8(1):

53 25. Wamala SP, Mittleman MA, Schenck-Gustafsson K, Orth-Gomer K. Potential explanations for the educational gradient in coronary heart disease: a population-based case-control study of Swedish women. Am J Public Health Mar;89(3): Hoeymans N, Smit HA, Verkleij H, Kromhout D. Cardiovascular risk factors in relation to educational level in men and women in The Netherlands. Eur Heart J Apr;17(4): Jacobsen BK, Thelle DS. Risk factors for coronary heart disease and level of education. The Tromso Heart Study. Am J Epidemiol May;127(5): Kilander L, Berglund L, Boberg M, Vessby B, Lithell H. Education, lifestyle factors and mortality from cardiovascular disease and cancer. A 25-year follow-up of Swedish 50-year-old men. Int J Epidemiol Oct;30(5): Mayer O, Jr., Simon J, Heidrich J, Cokkinos DV, De Bacquer D. Educational level and risk profile of cardiac patients in the EUROASPIRE II substudy. J Epidemiol Community Health Jan;58(1): Moller L, Kristensen TS, Hollnagel H. Social class and cardiovascular risk factors in Danish men. Scand J Soc Med Jun;19(2): Strand BH, Tverdal A. Can cardiovascular risk factors and lifestyle explain the educational inequalities in mortality from ischaemic heart disease and from other heart diseases? 26 year follow up of 50,000 Norwegian men and women. J Epidemiol Community Health Aug;58(8): Yarnell J, Yu S, McCrum E, Arveiler D, Hass B, Dallongeville J, et al. Education, socioeconomic and lifestyle factors, and risk of coronary heart disease: the PRIME Study. Int J Epidemiol Apr;34(2): Bosma H, Van Jaarsveld CH, Tuinstra J, Sanderman R, Ranchor AV, Van Eijk JT, et al. Low control beliefs, classical coronary risk factors, and socio-economic differences in heart disease in older persons. Soc Sci Med Feb;60(4): Kannel WB, Wilson PW, Nam BH, D'Agostino RB. Risk stratification of obesity as a coronary risk factor. Am J Cardiol Oct 1;90(7): Rosengren A, Orth-Gomer K, Wilhelmsen L. Socioeconomic differences in health indices, social networks and mortality among Swedish men. A study of men born in Scand J Soc Med Dec;26(4): Wamala SP, Wolk A, Orth-Gomer K. Determinants of obesity in relation to socioeconomic status among middle-aged Swedish women. Prev Med Sep-Oct;26(5 Pt 1):

54 37. Kuper H, Marmot M, Hemingway H. Systematic review of prospective cohort studies of psychosocial factors in the etiology and prognosis of coronary heart disease. Semin Vasc Med Aug;2(3): Jonsson S, Hedblad B, Engstrom G, Nilsson P, Berglund G, Janzon L. Influence of obesity on cardiovascular risk. Twenty-three-year followup of 22,025 men from an urban Swedish population. Int J Obes Relat Metab Disord Aug;26(8): Wang HX, Mittleman MA, Leineweber C, Orth-Gomer K. Depressive symptoms, social isolation, and progression of coronary artery atherosclerosis: the Stockholm Female Coronary Angiography Study. Psychother Psychosom. 2006;75(2): Adams KF, Schatzkin A, Harris TB, Kipnis V, Mouw T, Ballard- Barbash R, et al. Overweight, obesity, and mortality in a large prospective cohort of persons 50 to 71 years old. N Engl J Med Aug 24;355(8): Wannamethee SG, Shaper AG, Walker M. Overweight and obesity and weight change in middle aged men: impact on cardiovascular disease and diabetes. J Epidemiol Community Health Feb;59(2): Evans A, Tolonen H, Hense HW, Ferrario M, Sans S, Kuulasmaa K. Trends in coronary risk factors in the WHO MONICA project. Int J Epidemiol Oct;30 Suppl 1:S De Bacquer D, De Backer G, Cokkinos D, Keil U, Montaye M, Ostor E, et al. Overweight and obesity in patients with established coronary heart disease: are we meeting the challenge? Eur Heart J Jan;25(2): Berg C, Rosengren A, Aires N, Lappas G, Toren K, Thelle D, et al. Trends in overweight and obesity from 1985 to 2002 in Goteborg, West Sweden. Int J Obes (Lond) Aug;29(8): Lindstrom M, Isacsson SO, Merlo J. Increasing prevalence of overweight, obesity and physical inactivity: two population-based studies 1986 and Eur J Public Health Dec;13(4): Lahmann PH, Lissner L, Gullberg B, Berglund G. Sociodemographic factors associated with long-term weight gain, current body fatness and central adiposity in Swedish women. Int J Obes Relat Metab Disord Jun;24(6): Paeratakul S, Lovejoy JC, Ryan DH, Bray GA. The relation of gender, race and socioeconomic status to obesity and obesity comorbidities in a sample of US adults. Int J Obes Relat Metab Disord Sep;26(9): Ball K, Crawford D. Socioeconomic status and weight change in adults: a review. Soc Sci Med May;60(9):

55 49. Molarius A, Seidell JC, Sans S, Tuomilehto J, Kuulasmaa K. Educational level, relative body weight, and changes in their association over 10 years: an international perspective from the WHO MONICA Project. Am J Public Health Aug;90(8): Rosmond R, Bjorntorp P. Psychosocial and socio-economic factors in women and their relationship to obesity and regional body fat distribution. Int J Obes Relat Metab Disord Feb;23(2): Rosmond R, Lapidus L, Bjorntorp P. The influence of occupational and social factors on obesity and body fat distribution in middle-aged men. Int J Obes Relat Metab Disord Jul;20(7): Tuomilehto J. Cardiovascular risk: prevention and treatment of the metabolic syndrome. Diabetes Res Clin Pract Jun;68 Suppl 2:S Zieve FJ. The metabolic syndrome: diagnosis and treatment. Clin Cornerstone. 2004;6 Suppl 3:S De Backer G, Ambrosioni E, Borch-Johnsen K, Brotons C, Cifkova R, Dallongeville J, et al. European guidelines on cardiovascular disease prevention in clinical practice. Third Joint Task Force of European and Other Societies on Cardiovascular Disease Prevention in Clinical Practice. Eur Heart J Sep;24(17): Erhardt LR. Barriers to effective implementation of guideline recommendations. Am J Med Dec;118 Suppl 12A: Bankhead CR, Brett J, Bukach C, Webster P, Stewart-Brown S, Munafo M, et al. The impact of screening on future health-promoting behaviours and health beliefs: a systematic review. Health Technol Assess. 2003;7(42): Sandbaek A. Screening in general practice--what do we (not) know? Scand J Prim Health Care Mar;23(1): Holland WW, Stewart S, European Observatory on Health Systems and Policies., Nuffield Trust for Research and Policy Studies in Health Services. Screening in disease prevention : what works? Oxford ; Seattle: Radcliffe; Marteau TM, Kinmonth AL. Screening for cardiovascular risk: public health imperative or matter for individual informed choice? Bmj Jul 13;325(7355): Getz L, Sigurdsson JA, Hetlevik I. Is opportunistic disease prevention in the consultation ethically justifiable? Bmj Aug 30;327(7413): Dobs AS, Masters RB, Rajaram L, Stillman FA, Wilder LB, Margolis S, et al. A comparison of education methods and their impact on behavioral change in patients with hyperlipidemia. Patient Educ Couns Oct;24(2):

56 62. Roderick PJ, Brennan PJ, Meade TW. Do risk factors change in men at high risk of coronary heart disease? Observations on the effect of health promotion in primary care. J Cardiovasc Risk Aug;2(4): Emberson J, Whincup P, Morris R, Walker M, Ebrahim S. Evaluating the impact of population and high-risk strategies for the primary prevention of cardiovascular disease. Eur Heart J Mar;25(6): Socialstyrelsen. Hälso- och sjukvårdsrapport. Stockholm: Socialstyrelsen; Folkhälsoinstitutet, Statens beredning för medicinsk utvärdering. Att förebygga sjukdom i hjärta och kärl genom befolkningsinriktade program : en systematisk litteraturöversikt. Stockholm: Statens beredning för utvärdering av medicinsk metodik (SBU) : Folkhälsoinstitutet; Sellers DE, Crawford SL, Bullock K, McKinlay JB. Understanding the variability in the effectiveness of community heart health programs: a meta-analysis. Soc Sci Med May;44(9): Nilsson P, Berglund G. Prevention of cardiovascular disease and diabetes: lessons from the Malmo Preventive Project. J Intern Med Dec;248(6): Wilhelmsen L, Berglund G, Elmfeldt D, Tibblin G, Wedel H, Pennert K, et al. The multifactor primary prevention trial in Goteborg, Sweden. Eur Heart J Apr;7(4): Carleton RA, Lasater TM, Assaf AR, Feldman HA, McKinlay S. The Pawtucket Heart Health Program: community changes in cardiovascular risk factors and projected disease risk. Am J Public Health Jun;85(6): Fortmann SP, Varady AN. Effects of a community-wide health education program on cardiovascular disease morbidity and mortality: the Stanford Five-City Project. Am J Epidemiol Aug 15;152(4): Luepker RV, Rastam L, Hannan PJ, Murray DM, Gray C, Baker WL, et al. Community education for cardiovascular disease prevention. Morbidity and mortality results from the Minnesota Heart Health Program. Am J Epidemiol Aug 15;144(4): Puska P, Tuomilehto J, Salonen J, Neittaanmaki L, Maki J, Virtamo J, et al. Changes in coronary risk factors during comprehensive five-year community programme to control cardiovascular diseases (North Karelia project). Br Med J Nov 10;2(6199): Tuomilehto J, Geboers J, Salonen JT, Nissinen A, Kuulasmaa K, Puska P. Decline in cardiovascular mortality in North Karelia and other parts of Finland. Br Med J (Clin Res Ed) Oct 25;293(6554):

57 74. Farquhar JW, Maccoby N, Wood PD, Alexander JK, Breitrose H, Brown BW, Jr., et al. Community education for cardiovascular health. Lancet Jun 4;1(8023): Maccoby N, Farquhar JW, Wood PD, Alexander J. Reducing the risk of cardiovascular disease: effects of a community-based campaign on knowledge and behavior. J Community Health Winter;3(2): Weinehall L. Partnership for health : on the role of primary health care in a community intervention programme. Umeå: Univ.; Weinehall L, Hellsten G, Boman K, Hallmans G. Prevention of cardiovascular disease in Sweden: the Norsjo community intervention programme--motives, methods and intervention components. Scand J Public Health Suppl. 2001;56: Weinehall L, Hellsten G, Boman K, Hallmans G, Asplund K, Wall S. Can a sustainable community intervention reduce the health gap?--10- year evaluation of a Swedish community intervention program for the prevention of cardiovascular disease. Scand J Public Health Suppl. 2001;56: Landstinget, Västmanland. Uppföljning av Landstingets Västmanlands hjärt- och kärlprogram - Hjärtvänliga Västmanland p. 33. Unpublished. 80. Bjorkelund C, Bengtsson C. Feasibility of a primary health care programme aiming at reducing cardiovascular and cerebrovascular risk factors among women in a Swedish community, Stromstad. Scand J Prim Health Care Jun;9(2): Ovhed I, Odeberg H, Troein M, Rastam L. Awareness and treatment of cardiovascular disease risk factors among middle-aged Swedish men and women. Scand J Prim Health Care Sep;16(3): Hellenius ML, Johansson J, de Faire U, Elofsson S, Krakau I. Four years experience of a cardiovascular opportunistic screening and prevention programme in the primary health care in Sollentuna, Sweden. Scand J Prim Health Care Jun;17(2): Thelle D. Prevention of cardiovascular diseases--a scientific dilemma. Scand Cardiovasc J. 2000;34(2): Hellenius ML, Nilsson P, Elofsson S, Johansson J, Krakau I. Reduction of high cholesterol levels associated with younger age and longer education in a primary health care programme for cardiovascular prevention. Scand J Prim Health Care Jun;23(2): Burke LE, Dunbar-Jacob JM, Hill MN. Compliance with cardiovascular disease prevention strategies: a review of the research. Ann Behav Med Summer;19(3): Baranowski T, Cullen KW, Nicklas T, Thompson D, Baranowski J. Are current health behavioral change models helpful in guiding prevention of weight gain efforts? Obes Res Oct;11 Suppl:23S-43S. 57

58 87. Cummings KM, Becker MH, Maile MC. Bringing the models together: an empirical approach to combining variables used to explain health actions. J Behav Med Jun;3(2): Whitehead D. A social cognitive model for health education/health promotion practice. J Adv Nurs Nov;36(3): Becker MH. The health belief model and sick role behaviour. Health Education Monographs. 1974;2: Al-Ali N, Haddad LG. The effect of the health belief model in explaining exercise participation among Jordanian myocardial infarction patients. J Transcult Nurs Apr;15(2): Ali NS. Prediction of coronary heart disease preventive behaviors in women: a test of the health belief model. Women Health. 2002;35(1): Quah SR. The health belief model and preventive health behaviour in Singapore. Soc Sci Med. 1985;21(3): King JB. Illness attributions and the health belief model. Health Educ Q Fall-Winter;10(3-4): Prochaska JO, DiClemente, Carlo C. In search of how people change. American Psychologist. 1992;47(9): Prochaska JO. Population Treatment for Addictions. Current Directions in Psychological Science. 2004;13(6): Malotte CK, Jarvis B, Fishbein M, Kamb M, Iatesta M, Hoxworth T, et al. Stage of change versus an integrated psychosocial theory as a basis for developing effective behaviour change interventions. The Project RESPECT Study Group. AIDS Care Jun;12(3): DiClemente CC, Prochaska JO, Fairhurst SK, Velicer WF, Velasquez MM, Rossi JS. The process of smoking cessation: an analysis of precontemplation, contemplation, and preparation stages of change. J Consult Clin Psychol Apr;59(2): Velicer WF, Fava JL, Prochaska JO, Abrams DB, Emmons KM, Pierce JP. Distribution of smokers by stage in three representative samples. Prev Med Jul;24(4): Etter JF, Perneger TV, Ronchi A. Distributions of smokers by stage: international comparison and association with smoking prevalence. Prev Med Jul-Aug;26(4): Etter JF, Perneger TV. Associations between the stages of change and the pros and cons of smoking in a longitudinal study of swiss smokers. Addict Behav May-Jun;24(3): Ronda G, Van Assema P, Brug J. Stages of change, psychological factors and awareness of physical activity levels in The Netherlands. Health Promot Int Dec;16(4):

59 102. Sarkin JA, Johnson SS, Prochaska JO, Prochaska JM. Applying the transtheoretical model to regular moderate exercise in an overweight population: validation of a stages of change measure. Prev Med Nov;33(5): Suminski RR, Petosa R. Stages of change among ethnically diverse college students. J Am Coll Health Aug;51(1): Drevenhorn E, Bengtson A, Allen JK, Saljo R, Kjellgren KI. Counselling on lifestyle factors in hypertension care after training on the stages of change model. Eur J Cardiovasc Nurs May Perkins-Porras L, Cappuccio FP, Rink E, Hilton S, McKay C, Steptoe A. Does the effect of behavioral counseling on fruit and vegetable intake vary with stage of readiness to change? Prev Med Mar;40(3): Laforge RG, Velicer WF, Richmond RL, Owen N. Stage distributions for five health behaviors in the United States and Australia. Prev Med Feb;28(1): Steptoe A, Kerry S, Rink E, Hilton S. The impact of behavioral counseling on stage of change in fat intake, physical activity, and cigarette smoking in adults at increased risk of coronary heart disease. Am J Public Health Feb;91(2): Schuit AJ, van Loon AJ, Tijhuis M, Ocke M. Clustering of lifestyle risk factors in a general adult population. Prev Med Sep;35(3): Doherty SC, Steptoe A, Rink E, Kendrick T, Hilton S. Readiness to change health behaviours among patients at high risk of cardiovascular disease. J Cardiovasc Risk Jun;5(3): Gans KM, Assmann SF, Sallar A, Lasater TM. Knowledge of cardiovascular disease prevention: an analysis from two New England communities. Prev Med Oct;29(4): Davis SK, Winkleby MA, Farquhar JW. Increasing disparity in knowledge of cardiovascular disease risk factors and risk-reduction strategies by socioeconomic status: implications for policymakers. Am J Prev Med Sep-Oct;11(5): Willaing I, Jorgensen T, Iversen L. How does individual smoking behaviour among hospital staff influence their knowledge of the health consequences of smoking? Scand J Public Health. 2003;31(2): Murray SA, Manktelow K, Clifford C. The interplay between social and cultural context and perceptions of cardiovascular disease. J Adv Nurs Nov;32(5): Oliveria SA, Chen RS, McCarthy BD, Davis CC, Hill MN. Hypertension knowledge, awareness, and attitudes in a hypertensive population. J Gen Intern Med Mar;20(3):

60 115. Petrella RJ, Campbell NR. Awareness and misconception of hypertension in Canada: results of a national survey. Can J Cardiol May 15;21(7): Weinstein N, Slovic P, Waters E, Gibson G. Public understanding of the illnesses caused by cigarette smoking. Nicotine Tob Res Apr;6(2): Weinstein ND, Marcus SE, Moser RP. Smokers' unrealistic optimism about their risk. Tob Control Feb;14(1): Ayanian JZ, Cleary PD. Perceived risks of heart disease and cancer among cigarette smokers. Jama Mar 17;281(11): Mosca L, Ferris A, Fabunmi R, Robertson RM. Tracking women's awareness of heart disease: an American Heart Association national study. Circulation Feb 10;109(5): Ford ES, Jones DH. Cardiovascular health knowledge in the United States: findings from the National Health Interview Survey, Prev Med Nov;20(6): Weinstein ND. What does it mean to understand a risk? Evaluating risk comprehension. J Natl Cancer Inst Monogr. 1999(25): van Steenkiste B, van der Weijden T, Timmermans D, Vaes J, Stoffers J, Grol R. Patients' ideas, fears and expectations of their coronary risk: barriers for primary prevention. Patient Educ Couns Nov;55(2): Lloyd AJ. The extent of patients' understanding of the risk of treatments. Qual Health Care Sep;10 Suppl 1:i Kirkland SA, MacLean DR, Langille DB, Joffres MR, MacPherson KM, Andreou P. Knowledge and awareness of risk factors for cardiovascular disease among Canadians 55 to 74 years of age: results from the Canadian Heart Health Surveys, Cmaj. 1999;161(8 Suppl):S Potvin L, Richard L, Edwards AC. Knowledge of cardiovascular disease risk factors among the Canadian population: relationships with indicators of socioeconomic status. Cmaj. 2000;162(9 Suppl):S Meischke H, Sellers DE, Robbins ML, Goff DC, Daya MR, Meshack A, et al. Factors that influence personal perceptions of the risk of an acute myocardial infarction. Behav Med Spring;26(1): Frijling BD, Lobo CM, Keus IM, Jenks KM, Akkermans RP, Hulscher ME, et al. Perceptions of cardiovascular risk among patients with hypertension or diabetes. Patient Educ Couns Jan;52(1): Alm-Roijer C, Stagmo M, Uden G, Erhardt L. Better knowledge improves adherence to lifestyle changes and medication in patients with coronary heart disease. Eur J Cardiovasc Nurs Dec;3(4): Isomaa B, Almgren P, Tuomi T, Forsen B, Lahti K, Nissen M, et al. Cardiovascular morbidity and mortality associated with the metabolic syndrome. Diabetes Care Apr;24(4):

61 130. Fulop T, Tessier D, Carpentier A. The metabolic syndrome. Pathol Biol (Paris) Aug Sonnenberg L, Pencina M, Kimokoti R, Quatromoni P, Nam BH, D'Agostino R, et al. Dietary patterns and the metabolic syndrome in obese and non-obese Framingham women. Obes Res Jan;13(1): Mohan V, Gokulakrishnan K, Deepa R, Shanthirani CS, Datta M. Association of physical inactivity with components of metabolic syndrome and coronary artery disease--the Chennai Urban Population Study (CUPS no. 15). Diabet Med Sep;22(9): Wamala SP, Lynch J, Horsten M, Mittleman MA, Schenck-Gustafsson K, Orth-Gomer K. Education and the metabolic syndrome in women. Diabetes Care Dec;22(12): Byrne M, Walsh J, Murphy AW. Secondary prevention of coronary heart disease: patient beliefs and health-related behaviour. J Psychosom Res May;58(5): Walter FM, Emery J. Perceptions of family history across common diseases: a qualitative study in primary care. Fam Pract Apr Kip KE, McCreath HE, Roseman JM, Hulley SB, Schreiner PJ. Absence of risk factor change in young adults after family heart attack or stroke: the CARDIA Study. Am J Prev Med May;22(4): McCusker ME, Yoon PW, Gwinn M, Malarcher AM, Neff L, Khoury MJ. Family history of heart disease and cardiovascular disease riskreducing behaviors. Genet Med May-Jun;6(3): Hunt K, Emslie C, Watt G. Lay constructions of a family history of heart disease: potential for misunderstandings in the clinical encounter? Lancet Apr 14;357(9263): Watt G, McConnachie A, Upton M, Emslie C, Hunt K. How accurately do adult sons and daughters report and perceive parental deaths from coronary disease? J Epidemiol Community Health Nov;54(11): Bobak M, Pikhart H, Rose R, Hertzman C, Marmot M. Socioeconomic factors, material inequalities, and perceived control in self-rated health: cross-sectional data from seven post-communist countries. Soc Sci Med Nov;51(9): Diez-Roux AV, Nieto FJ, Tyroler HA, Crum LD, Szklo M. Social inequalities and atherosclerosis. The atherosclerosis risk in communities study. Am J Epidemiol May 15;141(10): Lynch JW, Kaplan GA, Salonen JT. Why do poor people behave poorly? Variation in adult health behaviours and psychosocial characteristics by stages of the socioeconomic lifecourse. Soc Sci Med Mar;44(6): Marmot M, Bobak M. International comparators and poverty and health in Europe. Bmj Nov 4;321(7269):

62 144. Szaflarski M, Cubbins LA. Self-reported health in Poland and the United States: a comparative analysis of demographic, family and socioeconomic influences. Health (London) Jan;8(1): Laaksonen M, McAlister AL, Laatikainen T, Drygas W, Morava E, Nussel E, et al. Do health behaviour and psychosocial risk factors explain the European east-west gap in health status? Eur J Public Health Mar;11(1): Margetts BM, Rogers E, Widhal K, Remaut de Winter AM, Zunft HJ. Relationship between attitudes to health, body weight and physical activity and level of physical activity in a nationally representative sample in the European Union. Public Health Nutr Mar;2(1A): Bobak M, Marmot M. East-West mortality divide and its potential explanations: proposed research agenda. Bmj Feb 17;312(7028): WHO. Highlights on health in Poland [cited ]; Available from: Zatonski WA, McMichael AJ, Powles JW. Ecological study of reasons for sharp decline in mortality from ischaemic heart disease in Poland since Bmj Apr 4;316(7137): Rywik SP, W. Rywik, T. Broda, G. Szczesniewska, D. Is the Decrease of Cardiovascular Mortality in Poland associated with the reduction of global cardiovascular risk related to changes in life style? Kardiol Pol. 2003;58: Nolte E, Scholz R, Shkolnikov V, McKee M. The contribution of medical care to changing life expectancy in Germany and Poland. Soc Sci Med Dec;55(11): Carlson P. Self-perceived health in East and West Europe: another European health divide. Soc Sci Med May;46(10): Pardell H, Roure E, Drygas W, Morava E, Nussel E, Puska P, et al. East-west differences in reported preventive practices. A comparative study of six European areas of the WHO-CINDI programme. Eur J Public Health Dec;11(4): Rywik SL, Davis CE, Pajak A, Broda G, Folsom AR, Kawalec E, et al. Poland and U.S. collaborative study on cardiovascular epidemiology hypertension in the community: prevalence, awareness, treatment, and control of hypertension in the Pol-MONICA Project and the U.S. Atherosclerosis Risk in Communities Study. Ann Epidemiol Jan;8(1): Zdrojewski T, Pienkowski R, Szynkiewicz M, Krupa-Wojciechowska B. Have rapid socioeconomic changes influenced awareness of blood pressure in Poland? J Hum Hypertens Apr;15(4):

63 156. Zdrojewski T, Wyrzykowski B, Szczech R, Wierucki L, Naruszewicz M, Narkiewicz K, et al. Epidemiology and prevention of arterial hypertension in Poland. Blood Press Suppl Dec;2: Stelmach W, Kaczmarczyk-Chalas K, Bielecki W, Drygas W. The impact of income, education and health on lifestyle in a large urban population of Poland (Cindi programme). Int J Occup Med Environ Health. 2004;17(3): Stelmach W, Kaczmarczyk-Chalas K, Bielecki W, Drygas W. How education, income, control over life and life style contribute to risk factors for cardiovascular disease among adults in a post-communist country. Public Health Jun;119(6): Leinsalu M, Vagero D, Kunst AE. Estonia : enormous increase in mortality differences by education. Int J Epidemiol Dec;32(6): Nicholson A, Bobak M, Murphy M, Rose R, Marmot M. Socioeconomic influences on self-rated health in Russian men and women--a life course approach. Soc Sci Med Dec;61(11): Bobak M, Hertzman C, Skodova Z, Marmot M. Own education, current conditions, parental material circumstances, and risk of myocardial infarction in a former communist country. J Epidemiol Community Health Feb;54(2): Brajczewski C, Rogucka E. Social class differences in rates of premature mortality among adults in the city of Wroclaw, Poland. Am J Hum Biol. 1993;5: Manhem K, Dotevall A, Wilhelmsen L, Rosengren A. Social gradients in cardiovascular risk factors and symptoms of Swedish men and women: the Goteborg MONICA Study J Cardiovasc Risk Oct;7(5): EpC-Rapport. Riskfaktorer för hjärt-kärlsjukdom - regionala och sociala skillnader i Sverige (Risk factors in coronary heart diseases - regional and social disparity in Sweden); Report No.: 1997: Statistiska centralbyrån. Yrkesklassificeringar i FoB 85 enligt Nordisk yrkesklassificering (NYK) och Socioekonomisk indelning (SEI) : alfabetisk version = Occupations in population and housing cencus 1985 (FoB 85) according to Nordic standard occupational classification (Nordisk yrkesklassificering, NYK) and Swedish socio-economic classification (Socioekonomisk indelning, SEI) : alfabetical version. Stockholm: SCB; Price CP, Koller PU. A multicentre study of the new Reflotron system for the measurement of urea, glucose, triacylglycerols, cholesterol, gamma-glutamyltransferase and haemoglobin. J Clin Chem Clin Biochem Apr;26(4):

64 167. Osler M LJ, Rasmussen NK. En undersögelse af medborgarnas forebyggelse af hjerte-karsygdomme i Vejle amt 1989 (A study of citizens' prevention of coronary heart diseases in Vejle). Copenhagen: Report from Danish Institute for Clinical Epidemiology; Ohrvik J. Nonparametric methods in the two-way layout. Chiang Mai Journal of Science. 2002;29: Getz L, Kirkengen AL, Hetlevik I, Romundstad S, Sigurdsson JA. Ethical dilemmas arising from implementation of the European guidelines on cardiovascular disease prevention in clinical practice. A descriptive epidemiological study. Scand J Prim Health Care Dec;22(4): Dolecek TA, Schoenberger JA, Oman JK, Kremer BK, Sunseri AJ, Alberti JM. Cardiovascular risk factor knowledge and belief in prevention among adults in Chicago. Am J Prev Med Sep-Oct;2(5): Osler M, Lous J, Rasmussen NK. Knowledge, attitudes and cardiovascular risk factors in Danish adults. Scand J Soc Med Sep;20(3): Nourjah P, Wagener DK, Eberhardt M, Horowitz AM. Knowledge of risk factors and risk behaviors related to coronary heart disease among blue and white collar males. J Public Health Policy Winter;15(4): Carroll C, Naylor E, Marsden P, Dornan T. How do people with Type 2 diabetes perceive and respond to cardiovascular risk? Diabet Med May;20(5): Erhardt L, Hobbs FD. Public perceptions of cardiovascular risk in five European countries: the react survey. Int J Clin Pract Nov;56(9): Kjellgren KI, Svensson S, Ahlner J, Saljo R. Hypertensive patients' knowledge of high blood pressure. Scand J Prim Health Care Dec;15(4): Danielsson B, Aberg H. The public view on cardiovascular risk factors and changes in lifestyle. Scand J Prim Health Care Mar;13(1): Cumming RG, Barton GE, Fahey PP, Wilson A, Leeder SR. Cardiovascular disease-related knowledge and attitudes in a high-risk Australian population. Med J Aust May 15;150(10):551-3, Egan BM, Lackland DT, Cutler NE. Awareness, knowledge, and attitudes of older americans about high blood pressure: implications for health care policy, education, and research. Arch Intern Med Mar 24;163(6):

65 179. Inglis V, Ball K, Crawford D. Why do women of low socioeconomic status have poorer dietary behaviours than women of higher socioeconomic status? A qualitative exploration. Appetite Dec;45(3): Naslund GK. Relationships between health behavior, knowledge, and beliefs among Swedish blue-collar workers. Scand J Soc Med Jun;25(2): Avis NE, McKinlay JB, Smith KW. Is cardiovascular risk factor knowledge sufficient to influence behavior? Am J Prev Med May-Jun;6(3): Davison LE. An analysis of knowledge, beliefs and behaviour relating to the development of coronary heart disease. Health Educ J. 1980;39(4): Vlassoff C, Garcia Moreno C. Placing gender at the centre of health programming: challenges and limitations. Soc Sci Med Jun;54(11): Denton M, Walters V. Gender differences in structural and behavioral determinants of health: an analysis of the social production of health. Soc Sci Med May;48(9): Mosca L, Mochari H, Christian A, Berra K, Taubert K, Mills T, et al. National study of women's awareness, preventive action, and barriers to cardiovascular health. Circulation Jan 31;113(4): Parmenter K, Waller J, Wardle J. Demographic variation in nutrition knowledge in England. Health Educ Res Apr;15(2): Fagerli RA, Wandel M. Gender differences in opinions and practices with regard to a "healthy diet". Appetite Apr;32(2): Johansson J, Viigimaa M, Jensen-Urstad M, Krakau I, Hansson LO. Risk factors for coronary heart disease in 55- and 35-year-old men and women in Sweden and Estonia. J Intern Med Dec;252(6): Marmot MG. Status syndrome: a challenge to medicine. Jama Mar 15;295(11): Jackson JE, Doescher MP, Saver BG, Hart LG. Trends in professional advice to lose weight among obese adults, 1994 to J Gen Intern Med Sep;20(9): Rimal RN. Closing the knowledge-behavior gap in health promotion: the mediating role of self-efficacy. Health Commun. 2000;12(3): Rimal RN. Longitudinal Influences of Knowledge and Self-efficacy on Exercise Behavior: Tests of a Mutual Reinforcement Model. Journal of Health Psychology. 2001;6(1): Rimal RN. Perceived risk and self-efficacy as motivators: Understanding individuals long-term use of health information. Journal of Communication. 2002;51:

66 194. Sol BG, Graaf YV, Bijl JJ, Goessens NB, Visseren FL. Self-efficacy in patients with clinical manifestations of vascular diseases. Patient Educ Couns Jul Hunt K, Davison C, Emslie C, Ford G. Are perceptions of a family history of heart disease related to health-related attitudes and behaviour? Health Educ Res Apr;15(2): Walter FM, Emery J. 'Coming down the line'-- patients' understanding of their family history of common chronic disease. Ann Fam Med Sep-Oct;3(5): Hunt SC, Gwinn M, Adams TD. Family history assessment: strategies for prevention of cardiovascular disease. Am J Prev Med Feb;24(2): Rich EC, Burke W, Heaton CJ, Haga S, Pinsky L, Short MP, et al. Reconsidering the family history in primary care. J Gen Intern Med Mar;19(3): Conner M, Norman P. Predicting health behaviour : research and practice with social cognition models 2ed. Maidenhead Open University Press; Calnan M, Rutter DR. Do health beliefs predict health behaviour? An analysis of breast self-examination. Soc Sci Med. 1986;22(6): Berkanovic E. Behavioral science and prevention. Prev Med Mar;5(1): Siero FW, Broer J, Bemelmans WJ, Meyboom-de Jong BM. Impact of group nutrition education and surplus value of Prochaska-based stagematched information on health-related cognitions and on Mediterranean nutrition behavior. Health Educ Res Oct;15(5): Weinehall L, Hallgren CG, Westman G, Janlert U, Wall S. Reduction of selection bias in primary prevention of cardiovascular disease through involvement of primary health care. Scand J Prim Health Care Sep;16(3): Bostrom G, Hallqvist J, Haglund BJ, Romelsjo A, Svanstrom L, Diderichsen F. Socioeconomic differences in smoking in an urban Swedish population. The bias introduced by non-participation in a mailed questionnaire. Scand J Soc Med Jun;21(2): Jackson R, Chambless LE, Yang K, Byrne T, Watson R, Folsom A, et al. Differences between respondents and nonrespondents in a multicenter community-based study vary by gender ethnicity. The Atherosclerosis Risk in Communities (ARIC) Study Investigators. J Clin Epidemiol Dec;49(12): Tibblin G. A population study of 50-year-old men. An analysis of the non-participation group. Acta Med Scand Oct;178(4): Thorogood M, Coulter A, Jones L, Yudkin P, Muir J, Mant D. Factors affecting response to an invitation to attend for a health check. J Epidemiol Community Health Jun;47(3):

67 208. Waller D, Agass M, Mant D, Coulter A, Fuller A, Jones L. Health checks in general practice: another example of inverse care? Bmj Apr 28;300(6732): Tolonen H, Dobson A, Kulathinal S. Effect on trend estimates of the difference between survey respondents and non-respondents: results from 27 populations in the WHO MONICA Project. Eur J Epidemiol. 2005;20(11): Zdrojewski T, Szpakowski P, Bandosz P, Pajak A, Wiecek A, Krupa- Wojciechowska B, et al. Arterial hypertension in Poland in J Hum Hypertens Aug;18(8): Kedzierski A, Dylewicz P. Knowledge about ischaemic heart disease in healthy subjects and in patients after myocardial infarction. Polish Heart Journal. 2000;52: Patrick DL, Cheadle A, Thompson DC, Diehr P, Koepsell T, Kinne S. The validity of self-reported smoking: a review and meta-analysis. Am J Public Health Jul;84(7): Aadahl M, Jorgensen T. Validation of a new self-report instrument for measuring physical activity. Med Sci Sports Exerc Jul;35(7): Caspersen CJ, Powell KE, Christenson GM. Physical activity, exercise, and physical fitness: definitions and distinctions for health-related research. Public Health Rep Mar-Apr;100(2): Bowlin SJ, Morrill BD, Nafziger AN, Lewis C, Pearson TA. Reliability and changes in validity of self-reported cardiovascular disease risk factors using dual response: the behavioral risk factor survey. J Clin Epidemiol May;49(5): Hyland A, Cummings KM, Lynn WR, Corle D, Giffen CA. Effect of proxy-reported smoking status on population estimates of smoking prevalence. Am J Epidemiol Apr 15;145(8): Johansson J, Hellenius ML, Elofsson S, Krakau I. Self-report as a selection instrument in screening for cardiovascular disease risk. Am J Prev Med May;16(4): Coveney J. A qualitative study exploring socio-economic differences in parental lay knowledge of food and health: implications for public health nutrition. Public Health Nutr May;8(3): Angus J, Evans S, Lapum J, Rukholm E, St Onge R, Nolan R, et al. "Sneaky disease": the body and health knowledge for people at risk for coronary heart disease in Ontario, Canada. Soc Sci Med May;60(9): Mishler WR, R. Trust, distrust and skepticism: Popular evaluations of civil and political institutions in post-communist societies. Journal of Politics. 1997;59(2):

68

69 Appendix

70

71

72

73

74