HEALTH PROMOTION INTERNATIONAL Vol. 19. No. 2 Oxford University Press 2004. doi: 10.

1093/heapro/dah208

All rights reserved Printed in Great Britain

The Healthy Bus project in Denmark: need for an action potential assessment
KJELD B. POULSEN
National Institute of Occupational Health, Lers Parkall 105, DK-2100 Copenhagen, Denmark
Downloaded from http://heapro.oxfordjournals.org/ at Universidade Federal do Paran on November 30, 2011

SUMMARY
Research over the last 50 years has repeatedly documented that bus drivers are exposed to several physical and psychological risk factors, which are associated with health problems in the form of heart, musculo-skeletal and stomach disease, and increased coronary mortality. So why has there been little action to improve the situation when it is so obviously indicated by such assessments? This article describes the long and complex process that has made it possible to launch almost 200 interventions among the 3500 municipal bus drivers in Copenhagen. Using a participative action research design, new evidence was gathered by broadening the traditional work environmental scope to lifestyle, health issues and private matters. Comparing this updated needs assessment with a national reference population, it was found that drivers were often still worse off. Again, simply presenting new evidence did not seem to lead to changes and further work is needed to empower the stakeholders so that they can commit to start making effective interventions. It is concluded that every needs assessment has to be supplemented with an evaluation of the action potential.

Key words: action assessment, bus drivers, workplace health promotion

INTRODUCTION Over the last 50 years, needs assessments have consistently documented bus drivers problems with both their health and their working environment. Drivers are known to be at risk for ischaemic heart morbidity and mortality, hypertension, prolapsed vertebral discs and cancer (Morris et al., 1953; Heady et al., 1961; Michaels and Zoloth, 1991; Rosengren et al., 1991; Jensen et al., 1996; Hansen et al., 1998; Ragland et al., 1998). Danish bus drivers, for instance, have double the risk of being hospitalized for heart disease compared with the rest of the workforce (Tchsen and Endahl, 1999). Key workplace risk factors include the physical ones of ergonomic strain and exposure to carcinogenic substances, and others that are related to psychosocial strain (Netterstrm and Juel, 1988; Krause et al., 1998; Meijman and Kompier, 1998). Although specific efforts have been made in the past to test workplace interventions capable of reversing the trend (Johanning et al., 1996; Rydstedt et al., 1998; Netterstrm, 1999; Kompier et al., 2000), there have been no major improvements in drivers health in general. This seems to suggest that most of the ill health seriously affecting bus drivers might be caused by an interrelationship between factors at work and in their private life. It is surprising, therefore, that there has been no previous examination of a holistic approach to improving drivers health, where change occurs simultaneously in the work environment, private life circumstances and lifestyle. This may be because the bus trade has been under pressure in the past decade to become economically more profitable through tendering, outsourcing, privatization and more efficient use
205

206

K. B. Poulsen

of the workforce. These processes may downplay employee health and well-being. It might also be that the kind of needs assessments used in this period was not sufficient to bring about substantial change and that what was required was a needs assessment supplemented with a wider ranging operationalized action assessment (Greenwood and Levin, 1998). The first of the two purposes of this paper was, for the first time, to present a multi-arena needs assessment and to compare bus drivers with a national reference population. The second purpose was to present an example of a successful case of how such an assessment of the potential for action was carried out in the Danish municipal bus trade, as a component of a large-scale intervention study.

based on a bringing together of communication theory (Rogers, 1995) and action research (Elliott, 1991; Greenwood and Levin, 1998). Obtaining an initial commitment Although there is a general acceptance in Denmark that bus drivers health is poor, it does not follow that there is agreement among stakeholders about either the causes or where to make changes. This project realized that only if stakeholders could be persuaded that the workplace was a key location for promoting health changes, the trade would benefit and make it possible for interventions to be introduced. The first task for the project, therefore, was to find out if stakeholders wished to have their current situation surveyed. The initial stage of the project comprised personal visits to the main stakeholders to gather their formal and informal perceptions. The results, crudely represented, were as follows. Employers saw economic constraints as the main barrier to the introduction of any programme to improve working conditions, although they also felt drivers bad lifestyles were the main cause of their ill health. The drivers unions felt it was the economic constraints brought about by privatization, together with stressful working conditions, that caused drivers ill-health. Copenhagen Transport were of the view that companies were unable to adjust sufficiently to new economic conditions, that disagreements between the two unions spoiled many opportunities to improve working conditions, and that drivers in general had a lifestyle problem. Looking at these comments, there seemed to be several causal arguments for action that could be bracketed together, but importantly there was a general agreement on the need to put an end to the trades bad reputation. The last stakeholder, who was not interviewed, was the Danish National Institute for Occupational Health (NIOH), which produced the previous surveys of drivers health. Despite the scientific basis of these earlier needs assessments, they had not resulted in real improved conditions for the bus drivers health and working conditions. The next step was to introduce health promotion as a way of tackling the problems identified. It turned out that all stakeholders regarded health promotion primarily in terms of promoting change in individual lifestyle. Since lifestyle has always been regarded as a private

Downloaded from http://heapro.oxfordjournals.org/ at Universidade Federal do Paran on November 30, 2011

METHODS Accounting for 97% of all drivers working in the Greater Copenhagen area, 3505 bus drivers made up the study base for the needs assessment. They were employed by two large, two mediumsized and two smaller haulage contractors. The overall manager of public bus traffic in the area is Copenhagen Traffic, whose responsibility it is to decide the level of service contractors should provide. A tendering process (through which companies bid competitively for routes) commenced in 1989 and was reaching its final stage when the SundBus (Healthy Bus) Project started. During the tendering period, the employers were united in one organization while the drivers were organized in two separate unions, each with its own specific concerns and agendas. These key stakeholders were the targets for the action assessment. The overall theoretical framework adopted by the SundBus Project was conceptualized in the Ottawa Charter and the European Luxembourg Declaration of Workplace Health Promotion [World Health Organization (WHO), 1986; WHO, 1998; Workplace Health Promotion Network, EU, 1998]. Since occupational stress theory relies on the job strain model, a key focus for the project was on how to increase stakeholders influence on job functions and how to help companies provide the kinds of support that would reduce stress (Karasek and Theorell, 1993). Innovations were encouraged and disseminated through a process of active participation (Toulmin and Gustavsen, 1996)

Extensive, large scale WHP effort for bus drivers

207

matter, even if influenced by working life, the concept was a hindrance to progress at first. Only when the project brought a vision of health promotion as an active, dynamic process that encompassed the physical, psychosocial, organizational working environment together with lifestyle, was it seen as an innovative way of freeing the situation from the deadlock in which it found itself. Finally, stakeholders played a crucial part in the final formulation of the project purpose, where the intention was to include tacit knowledge used before in research interventions. At this point, consent was given to start the project properly and to carry it out in two main stages; the first would investigate whether there were realistic openings for interventions, if there was still a need for them and if it was thought that they might lead to sufficient improvements in health and trade reputation.

The collaborative framework The collaboration was founded on seven main principles, agreed by all stakeholders.
G

Creating the organization A steering group was formed with representatives from the key stakeholders. Their task was to support the project, clear political issues and take decisions on the continuation of the project at critical stages. They appointed five trade members to a project group, consisting of two drivers, two manager representatives and a project associate from Copenhagen Traffic. The task of this group was to be the practical link with the trade and to gain competences from each other. Each of the appointed members of the project group was engaged half-time for the whole project period of 4 years. The stakeholders insisted that the project group had to be paid by their organizations because this would increase their commitment. Local project groups were formed at each garage, consisting of a manager and the shop and safety stewards. In addition, the large haulage contractors created a central project organization mirroring the local ones. NIOH researchers affiliated to the project represented epidemiology, occupational health, health philosophy, organization and management theory, communication and anthropology.

Downloaded from http://heapro.oxfordjournals.org/ at Universidade Federal do Paran on November 30, 2011

First, the research project would be multidisciplinary, and include epidemiology, health philosophy, management and organization, economy and occupational medicine. Secondly, a participative approach should be initiated as early as possible, utilizing and respecting stakeholder expertise as being of equal importance to research. Thirdly, to take a positive approach where known/recognized positive factors are promoted, instead of just concentrating on reducing risk factors. The fourth element was called business as usual, meaning that interventions for general improvements were to be implemented and evaluated under the prevalent conditions, and not in a laboratory-like situation where most of the rules for running the business are suspended. Fifthly, multiple differentiated tailor-made interventions agreed by all the stakeholders were to be started simultaneously. Sixthly, to create an embedded ownership of the process, the methods were to be integrated as a part of the normal daily business. Finally, both processes and outcomes were to be evaluated.

Operationalized action assessment Huge reorganizations in the working environment had taken place during the last 10 years, therefore it was necessary to update the picture of where changes were most needed (Lauersen et al., 1980; Netterstrm and Juel, 1988). The first step for the project group was therefore to become acquainted and at the same time create a mutual understanding within the trade. It was crucial to create an open atmosphere where everything could be discussed, yet remain confidential. The methodology was that of a future research seminar, where all words or associations connected to a bus drivers work and life were written on wallpaper without discussion or comment (French and Bell, 1995). The words were then looked at again and positive associations were recorded. Only at this point were associations discussed in detail, with specific focus on getting each partner to articulate his or her point of view. A projectoriented formulation of each issue was then made collaboratively on whiteboards and keyed into a computer. If there were diverging opinions

208

K. B. Poulsen

they were recorded too. After this, a third wallpaper round took place to find out how each stakeholder conceived the other three. It became obvious that some key informants held views that might, if given too much recognition, create a biased picture (Hawe, 1996), so the future seminar was vital as a way of deciding, at an early stage, whether it was possible to create a common ground for future work. The project group agreed that it should be possible to suggest valuable changes and that several items seemed to have been tabooed or had never been part of the researchers interest, which was significant for the following step. The next step comprised the project group formulating the results of the future research seminar as a questionnaire to gain knowledge of how each driver perceived their health, working and living conditions. Many new questions had to be developed to reflect specific areas where no standard formulation existed. Additionally, several free-text questions were included. The baseline survey Project staff, together with local shop and safety stewards, handed out questionnaires to all drivers on the company payroll on 1 November 1999. The project group used this opportunity to publicize the project by wearing logo sweatshirts and distributing information folders about the project. Each driver was identified with a unique personal identification number (PIN) code to be able to follow up on an individual basis. The first reminder was done by mail, while the second and last reminders took the form of a telephone call, with an offer to interview the driver either directly over the telephone or to pay the driver a

visit at home. The project followed the national ethical guidelines for epidemiological research and if a driver didnt want to participate their decision was fully respected. Internal comparisons were done by multiple logistic regression adjusted for age and gender. The bus driver cohort was compared with the representative Danish National Work Environment Cohort Study, containing information about 6367 workers updated parallel to the baseline survey (Burr et al., 2002). Prevalence proportions and adjusted odds ratios were calculated using logistic regression analysis with Proc Logistic in SAS 6.12. RESULTS A total of 2677 (76%) out of 3505 drivers answered the questionnaire. During the period of enquiry, three drivers had died. From Table 1 it can be seen that most drivers were men, and in general had a more favourable social background, experiences or expectations about general health than the women. Operationalizing the operationalized baseline dataset To ensure the stakeholders would interpret and transform the data into practical interventions, only descriptive data were presented to them in the form of simple graphs for each question. Three datasets were produced: one for the whole area, one at company level and one for each garage. However, these crude data were first presented to the steering group at a 24-hour working seminar. As attention to lifestyle was a

Downloaded from http://heapro.oxfordjournals.org/ at Universidade Federal do Paran on November 30, 2011

Table 1: Background variables of 2677 Danish municipal bus drivers

Male (n Belongs to an ethnic minority group (%) Annual sickness absenteeism (mean days) Years working as a bus driver (mean years) Health worsened during last 5 years (%) Percentage who are unskilled (%) Living alone (%) Age (mean years) Health expected to be worse in 2 years (%) 10 years of school attendance (%)
a

2294)

Female (n 6.8 15.2 12.7 35.7 50.4 34.2 46.0 25.1 49.9

383)

p-valuea 0.0001 0.0001 0.0001 0.0002 0.0002 0.001 0.02 0.05 0.06

32.7 9.4 10.6 25.9 38.3 25.7 44.7 20.4 44.7

p-values from t-test.

Extensive, large scale WHP effort for bus drivers

209

sensitive issue for some of the stakeholders, ensuring a balance between lifestyle and interventions focussed on the work environment became a central political issue. The solution was to adapt the extended concept of health promotion to the local situation in a way that was accepted by all. In practice, data was then categorized in four different arenas symbolized as a four-leaved clover, where each leaf should be equally weighted when designing the interventions. Apart from lifestyle, the categories were psychosocial elements, physical work environment and organizational matters. Based on this format the project was allowed to hand over the data to the companies to let them assess whether they would and could come up with realistic interventions for starting the change process. Each garage was then invited to a groupworking seminar where they were coached on how to start evaluating their own results. They were provided with a simple framework tool where each result from the questionnaire was classified to prioritize the most important conditions to be changed. A second form was used to plot the interventions to visualize if all the stakeholders were responsible for actions and if all four clover leaves had been covered. Before they left, they were trained in using a 1-page project template for describing each suggestion for an intervention in a systematic way (Figure 1). Since it was such an important stage in the whole process, it was positive that 18 garages (representing 91.5% of the study base) turned up and also consented to develop an intervention catalogue. Finally, to be sure that each driver was informed about the main findings, a newsletter was sent by mail to their home address. All three stages in the data presentation were carried out in less than a month to ensure that information was as timely as technically possible without delaying the overall process. Scientifically analysed results While the partners were working with the crude dataset, the researchers made a focussed, detailed scientific analysis. Scientific analysis shows the effect of the four factorshealth, seniority, ethnicity and smokingon 18 different outcomes. It can be seen that the odds ratio for driving a defective bus at least weekly is 0.81 when assessing health as good when compared with

drivers with bad health. This was not affected by seniority or smoking status. However, only half the ethnic minorities (defined by last name) reported driving a defective bus on a weekly basis. This might suggest that having good health or belonging to an ethnic minority group reduces the chances of having to drive a defective bus! An alternative explanation would be that members of these groups have a lesser tendency to assess the bus as defective. It can also be seen in the table that drivers who had been in the job for 5 years or were heavy smokers had more negative views towards the job. The comparisons with the Danish National Work Environment Cohort Study in Figure 2 illustrate that bus drivers were worse off concerning health, lifestyle and working environment than the rest of the workforce. It appears that evaluation of the health of the general workforce is compromised in the case of bus drivers, perhaps because their health problems are a consequence of both lifestyle and a range of specific health conditions. Negative psychosocial factors featured more prominently among drivers. Finally, what is not shown in the figure is that traditional occupational risk factors like noise, vibration, draught and thermal variations were between 13 and 46 times more prevalent among bus drivers. Licence to intervene Finally, a conference was held in which the two ministers of Traffic and Work took part, to present the results of the survey and an intervention catalogue with almost 200 different suggestions for realistic first phase actions. At this conference the stakeholders publicly committed themselves to dedicate the necessary resources to the participative intervention process over the following 3 years. The next step was to transform their intervention plans into action. But only at this point, 1 year after the first meeting, was preparatory groundwork finished, fortunately with a collaborative outcome. DISCUSSION The bus-driving occupation has been overwhelmed with needs assessments crying out for follow-up action. Our baseline survey showed that compared with other scientific surveys, there seemed to have been little or no improvements in drivers health

Downloaded from http://heapro.oxfordjournals.org/ at Universidade Federal do Paran on November 30, 2011

210

K. B. Poulsen

Downloaded from http://heapro.oxfordjournals.org/ at Universidade Federal do Paran on November 30, 2011

Fig. 1: The action plan registration form used to define and follow the progress of specific interventions among 3500 Danish municipal bus drivers.

during recent decades. Even though most of the bus stock has been renewed in this period, there remains just as much need for improvement in working conditions, health and well-being as before. The job is still characterized by low driver influence on their role and a lack of opportunity

for improvement (Karasek and Theorell, 1993). It is of course inherent in the job that bus drivers have little control over the framing constrictions, such as the route, the driving schedule and the traffic. The potential for influence lies in improving drivers knowledge of the traffic system, and

Extensive, large scale WHP effort for bus drivers

211

Downloaded from http://heapro.oxfordjournals.org/ at Universidade Federal do Paran on November 30, 2011

Fig. 2: Comparison of Danish bus drivers with a Danish national representative sample of workers. The numbers after the variable show the percentage of bus drivers/workers who had the problem in question. Odds ratios are shown with 95% confidence limits.

increasing understanding of how to improve the service they give and of how to interact with the customers; these are areas where the employer is rewarded by the Copenhagen Traffic bonus system if improvement occurs. All the arguments for action are obvious, so why does nothing happen?

Our experience indicates that this is because needs assessments on their own are not sufficient. It is self-evident that we cannot do without them, but they will remain ineffectual unless the whole organization is prepared to act to implement the changes identified. The question is whether more

212

K. B. Poulsen

can be done to facilitate and accelerate such changes in the trade (Rogers, 1995)? There is no doubt that much depends on being in the right place at the right time, and with the right staff doing the right things, which cannot be read from a manual. Participative research is also a political game where everyone has to be professionally experienced, flexible, innovative and, in the end, able to compromise on the creation of a realistic joint vision (Israel et al., 1989; Yin, 1994). The activity of gaining stakeholders active support has to involve a step by step process, and at each step it is necessary to be realistic about the chances of a positive outcome (Greenwood and Levin, 1998). No single model can be used to guide you through such a process. In the projects case, different elements of different theories were found to be relevant for this specific context and were used in a piecemeal way to create a framework that was appropriate for the situation (Baker et al., 1998). We believe that the conceptual contract, specifying responsibilities and direct influence right from the beginning, created an enabling climate (Koelen et al., 2001). Such a broad approach has never been tried before in the bus trade on such a large scale (Netterstrm, 1999; Kompier et al., 2000). It is also thought that another crucial feature was the use of highly specific questions drawn from particular information supplied by the full range of stakeholders, which was tailored to include all their concerns. This replaced the usual broad-brush approach to needs assessment (Hawe, 1996). The brainstorming sessions used to create mutual understanding and a joint vision between the stakeholders were regarded as effective because they resulted in the introduction of several issues that had not been researched, but which were evident knotty problem areas for the participants. The pragmatic four-leaf conceptualization of health promotion became more a process guideline of a balanced holistic approach than a definition. Taking such things into account forced the stakeholders to discuss earlier taboo issues and therefore created new momentum. The research institution seemed to create a confident setting and provide facilitators. Another element was the focus on action, which was appealing to the trade. The acceptance by the research society of economicpolitical realities, and thus demand for interventions that were pragmatic and founded based upon local needs, was crucial. Finally, the importance of the intensive daily collaboration

between trade representatives and researchers in the project group can not be overestimated (Greenwood and Levin, 1998). Scientifically, several paradigms had to be in use at the same time (Koelen et al., 2001). However, most of the scientific analysis was only available at a point when decisions had already been taken; this meant that the scientists could not give advice on their usual firm ground. As it turned out, this was a good thing as scientific advice would have prevented the stakeholders from taking ownership by making their own decisions (Greenwood and Levin, 1998). It transpired that it was important to document the fact that the trade was not as homogeneous as previously presumed (Kompier and Martino, 1995), and that it was necessary to obtain support for a highly differentiated focus for future interventions. CONCLUSION This survey shows how an action research approach played a key role in creating a conducive environment for health-promoting interventions among stakeholders in the bus trade. However, as in other occupational settings, it is not enough for a traditional needs assessment of the bus trade to invest in an intervention process. A much more complicated action assessment must be made to find out if the organizational, political, human and economic resources are available. Such a process is highly sensitive and context-dependent, particularly with respect to the people involved. A wide and pragmatic use of the current health promotion concepts helped to lend the process scientific credibility among stakeholders, and was thus of great value in practice. Time-consuming stages like initial stakeholder commitment, participative partnership, organizational set-up, operationalized action assessment, joint data analysis and final stakeholder commitment all had to be carried out before a comprehensive intervention survey could begin. In the end, however, all stakeholders did agree on launching large-scale interventions to improve the health and working environments of bus drivers. ACKNOWLEDGEMENTS Funding was provided by the National Board of Health (grant no. 9801326) and the Danish Ministry of Work (grant no. 204-0003).

Downloaded from http://heapro.oxfordjournals.org/ at Universidade Federal do Paran on November 30, 2011

Extensive, large scale WHP effort for bus drivers

Address for correspondence: Kjeld B Poulsen, MD, PhD Senior Researcher Specialist in Occupational Medicine The National Institute of Occupational Health Lers Parkall 105 DK-2100 Copenhagen O Denmark E-mail: kbpoulsen@hotmail.com

213

REFERENCES
Baker, E., Israel, B. A. and Schurman, S. (1998) The integrated model: implications for worksite health promotion and occupational health and safety practice. Health Education Quarterly, 23, 175190. Burr, H., Bach, E., Borg, V. and Villadsen, E. (2002) Arbejdsmilj i Danmark 2000 [The Danish Work Environment 2002]. Arbejdsmiljinstituttet, Copenhagen. Elliott, J. (1991) Action Research for Educational Change. Open University Press, Buckingham. French, W. L. and Bell, C. H.,Jr (1995) Comprehensive interventions. In French, W. L. and Bell, C. H.,Jr (eds) Organization Development. Prentice Hall International, London. Greenwood, D. J. and Levin, M. (1998) Introduction to Action Research. Social Research for Social Change. SAGE Publications, CA. Hansen, J., Raascou-Nielsen, O. and Olsen, J. H. (1998) Increaced risk of lung cancer among different types of professional drivers in Denmark. Occupational and Environmental Medicine, 55, 115118. Hawe, P. (1996) Needs assessment must become more change-focused. Australian and New Zealand Journal of Public Health, 20, 473478. Heady, J. A., Morris, J. N., Kagan, A. and Raffle, P. A. B. (1961) Coronary heart disease in London busmen: a progress report with particular reference to physique. British Journal of Preventive and Social Medicine, 15, 143153. Israel, B. A., Schurman, S. J. and House, J. S. (1989) Action research on occupational stress: Involving workers as researchers. International Journal of Health Service, 19, 135155. Jensen, M. V., Tchsen, F. and rhede, E. (1996) Prolapsed cervical intervertebral disc in male professional drivers in Denmark 19911990. A longitudinal study of hospitalizations. Spine, 21, 23522355. Johanning, E., Lansbergis, P., Geissler, H. and Karazmann, R. (1996) Cardiovascular risk and back-disorder intervention study of mass transit operators. International Journal of Occupational Medicine and Environmental Health, 2, 7987. Karasek, R. and Theorell, T. (1993) Healthy Work. Stress, Productivity, and the Reconstruction of Working Life. Basic Books, Inc., New York. Koelen, M. A., Vaandrager, L. and Colomer, C. (2001) Health promotion research: dilemmas and challenges. Journal of Epidemiology and Community Health, 55, 257262. Kompier, M. A. and Martino, V. D. (1995) Review of bus drivers occupational stress and stress prevention. Stress Medicine, 11, 253262.

Kompier, M. A., Aust, B., van den Berg, A.-M. and Siegrist, J. (2000) Stress prevention in bus drivers: evaluation of 13 natural experiments. Journal of Occupational Health Psychology, 5, 121. Krause, N., Ragland, D. R., Fisher, J. M. and Syme, S. L. (1998) 1998 Volvo award winner in clinical studies. Spine, 23, 25072516. Lauersen, P., Netterstrm, B., Pedersen, T. K. and WittaJorgensen, A. (1980) Buschauffrers Arbejdsmilj 1. En Undersgelse af Helbred og Trivsel hos Buschauffrer i Kbenhavn, Odense og rhus [in Danish]. Institut for Socialmedicin, Copenhagen. Meijman, T. F. and Kompier, M. A. (1998) Bussy business: how urban bus drivers cope with time pressure, passengers, and traffic safety. Journal of Occupational Health Psychology, 3, 109121. Michaels, D. and Zoloth, S. R. (1991) Mortality among urban bus drivers. International Journal of Epidemiology, 20, 399404. Morris, J. N., Heady, J. A., Raffle, P. A. B., Roberts, C. G. and Parks, J. W. (1953) Coronary heart-disease and physical activity of work. Lancet, 265, 10531057. Netterstrm, B. (1999) Denmark: self-rule on route 166. An intervention study among bus drivers. In Kompier, M. and Cooper, C. (eds) Preventing Stress, Improving Productivity: European Case Studies in the Workplace. Routledge, London. Netterstrm, B. and Juel, K. (1988) Impact of work-related and psychosocial factors on the development of ischemic heart disease among urban bus drivers in Denmark. Scandinavian Journal of Work, Environment and Health, 14, 231238. Ragland, D. R., Greiner, B. A., Holman, B. L. and Fisher, J. M. (1998) Hypertension and years of driving in transit vehicle operators. Scandinavian Journal of Social Medicine, 25, 271279. Rogers, E. M. (1995) Diffusion of innovations. The Free Press, New York. Rosengren, A., Anderson, K. and Wilhelmsen, L. (1991) Risk of coronary heart disease in middle-aged male bus and tram drivers compared to men in other occupations: a prospective study. International Journal of Epidemiology, 20, 8287. Rydstedt, L. W., Johansson, G. and Evans, G. W. (1998) The human side of the road: improving the working conditions of urban bus drivers. Journal of Occupational Health Psychology, 3, 161171. Toulmin, S. and Gustavsen, B. (1996) Beyond Theory: Changing Organizations Through Participation. John Benjamins Publishing Company, Amsterdam. Tchsen, F. and Endahl, L. (1999) Increasing inequality in ischaemic heart disease morbidity among employed men in Denmark 19811993: the need for a new preventive policy. International Journal of Epidemiology, 28, 640644. WHO (1986) Ottawa charter for health promotion. WHO, Geneva. WHO (1998) The Health-Promoting WorkplaceMaking it Happen. WHO, Geneva. Workplace Health Promotion Network, EU (1998) The Luxembourg Declaration on Workplace Health Promotion in the European Union. Arbejdsmiljinstituttet, Copenhagen. Yin, R. K. (1994) Case Study Research: Design and Methods. Sage Publications, Thousands Oaks, CA.

Downloaded from http://heapro.oxfordjournals.org/ at Universidade Federal do Paran on November 30, 2011