Lersø Parkallé 105 2100 København Ø

Phone: (+45) 39 16 52 00 fax: (+45) 39 16 52 01

E-mail: [email protected] www.ami.dk

Kjeld B Poulsen, Kai Drewes, Sisse Grøn, Peter T Petersen og Elsa Bach

Reflections on interventions

81 123

Experience based on more than 200 interventions initiated in order to improve the health and work environment of 3500 bus drivers

001 571 9

12

4

49 9

354989 565 12 4

001 571 9

Reflections on interventions Kjeld B Poulsen Kai Drewes Sisse Grøn Peter T Petersen Elsa Bach Layout: Topp AD/grafik Text: Gitten Bugge and Signe Bonnén

National Institute of Occupational Health Lersø Parkallé 105 DK-2100 København Ø Phone: +45 39 16 52 00 Fax: +45 39 16 52 01 E-mail: [email protected] www.ami.dk ISBN no. 87-7904-146-9 Copenhagen 2005

22/08/05

11:26

Side 3

Table of contents

04

Introduction

05

The infant level – from idea to project

11

Establishing baseline – what is the start situation?

17

The intervention phase – finally some action

23

Follow-up – did anything actually take place?

29

Results – what happened?

33

Project evaluation – to take a look at oneself in the mirror

45

What has HealthyBus contributed in the way of new knowledge

49

From practice to theory

53

Recommended reading

58

12 4 49 9

20 98 354989 565

Contents

Summary

001 571 9

Reflektioner 15-12

Reflektioner 15-12

22/08/05

11:26

Side 4

Summary 4

Summary There are no easy solutions when conducting complex

where we look for specific causal relations. In HealthyBus, it was

interventions at workplaces under normal operational con-

not unambiguous what the cause of reduced health was, and

ditions, especially not in a financially pressed sector such as

therefore we had no solution at the beginning.

the bus industry. Many companies have experienced that traditional methods are not adequate. Instead of giving up,

In this book of experience, we are passing our experience on as

read this example of how it is possible to reach some

work environment professionals, and we hope it might be of

results, if all stakeholders are willing. However, it takes a

help to our colleagues, whether they are researchers or consul-

lot of trust, commitment, joint visions and hard work.

tants. But it is also our hope that in other professions the book may find use in discussion on what can be contributed to the cross-disciplinary hotchpotch that results from change. The book

There are no easy solutions when conducting complex interven-

can be used for educational purposes, and we hope it might cre-

tions at workplaces under normal operational conditions, especial-

ate respect and interest for the necessity and usefulness of beco-

ly not in a financially pressed sector such as the bus industry. Many

ming involved with professional traditions other than one’s own.

companies have experienced that traditional methods are not adequate. Instead of giving up, read this example of how it is possible

This book has fulfilled its purpose if you say, “that I knew” or

to reach some results, if all stakeholders are willing. However, it

“that is what I usually do too” at the same time that you discover

takes a lot of trust, commitment, joint visions and hard work.

that some of the things we write provide you with a new angle of insight. It is when you dare doing what you usually do, but in

This book differs from what work environment professionals usually write. It is an attempt to write about the phases that are not

another way that you obtain the kind of new awareness which leads you to changes where change has not taken place before.

reported so often. It is also about how important the start-up phase is before an idea can be defined as a research project. It is

The book was written by the project manager of HealthyBus,

about the doubts, and all the things we would have done diffe-

senior researcher Kjeld B. Poulsen, conferring with that part of

rently if we had been granted more time and more manpower,

the project group who are not employed in the bus industry.

and if the experience had been available beforehand which we

Their comments were incorporated as loyally as possible, because

had afterwards. It is about daring to administrate the insecurity in

each person has his/her own experience of the project.

a dynamic world, as it is a constant threat to the research process. If the task is to succeed, it is also about finding support in a

I would like to thank everybody who was involved in Project

cross-disciplinary team of researchers and practicians of equal

HealthyBus: the drivers who may not be aware that the changes

standing. It is about dealing with the stakeholders who wanted to

they experience are results of the project, the fiery souls emplo-

initiate hundreds of new interventions that have eliminated the

yed at the garages of the haulage contractors, The Greater

traditional measuring methods. It is not a cookery book, as there

Copenhagen Authority (Danish abbreviation: HUR), various other

is no right recipe that can be used for all intervention projects.

organisations, and the Danish National Institute of Occupational

YOU must find a method that will work in your specific context.

Health. I would especially like to thank representatives of the

The book tells you about our experience, hoping that you may

industry who were a part of the project group: Lotte Rasmussen,

find useful advice here to help in building up your own project.

Eva Holm, Karin Lehmann, Per Vendelboe, Henning Rasmussen, Brian Hansen and Jesper Lindeberg.

The purpose of HealthyBus (“SundBus” in Danish) is to improve the health and well-being of bus drivers. But what is health, and how is it affected - both positively and negatively? The story is about a situation which is similar to the challenges companies

Elsa Bach

face daily and which is far from the classic intervention research

Director of research

Reflektioner 15-12

22/08/05

11:26

Side 5

Introduction 5

Introduction

“For every complex problem, there is a solution that is simple, neat, and wrong” Albert Szent-Györguy, Nobel Prize winner 1937

HealthyBus was a four-year intervention project which included almost every bus driver in the Greater Copenhagen area. The purpose of the project was to find out how to improve bus drivers’ health and well-being on a large scale, because this occupation is known for a significantly elevated prevalence of morbidity. Methodologically, we wanted to operationalise the health promotion concept as an action-based intervention model. The project was established as a daily cooperation between the stakeholders within a democratic and cross-disciplinary framework. We developed a tailored questionnaire, the answers of which formed the basis for implementation of hundreds of interventions simultaneously. The project shows that the goal, to achieve significantly better health for our bus drivers, was a realistic goal. However, we could only achieve that goal if we simultaneously created changes at the workplace, in the drivers’

The introduction The purpose of this publication is to pass on our experience from a comprehensive Danish intervention study conducted within the bus industry to qualify work environment professionals and researchers to work with complex interventions. We tell how project HealthyBus was established and how it builds on a democratic partnership within practical intervention research.

private lives, and in health and lifestyle. HealthyBus used the workplace as basis for such a general approach. In order to find credible and specific explanation models with which to intervene, it was crucial that we were able to handle a combined quantitative and qualitative effort. The uncontrolled, dynamic sequence design and the huge amount of interventions that ensued put a spoke in the wheel of traditional statistical effect measures. Here we had to combine different methods for evaluation to hold the project as close as possible to the reality which we so strongly wanted to be able to influence. We think that the new experience gained in HealthyBus will be useful for other researchers and work environment professionals who are concerned with complex interventions in reality. The angle of this story is primarily based on the experience of the project manager with that specific goal in view.

We go through the healthand work environment-related reasons for dealing with the health of the bus drivers.

Reflektioner 15-12

22/08/05

11:26

Side 6

Introduction 6

The National Institute of Occupational Health, Denmark (NIOH) has been involved in research on the relationship between health and exposure in bus drivers’ work environment for many years. Our publication of the figures below was a political eye-opener, as it shows an increasing number of hospital admissions of bus drivers with cardiovascular diseases. It certainly pressured the entire industry to consider how to improve the bus drivers’ work environment. NIOH decided to establish a co-operation program with the industry in which to initiate the necessary interventions. And the introduction of health promotion at the workplace was a good opportunity to develop a new method for conducting intervention research. That was the birth of the HealthyBus idea, and this book tells the story of what happened as time moved on. Health promotion is a method which makes it possible for people to exert an influence that can change their living and working conditions. That makes health promotion much more than just a change of lifestyle. It is important to recognize that there is no simple or authorised health promotion method. However, the workplace is a good basis for practising health promotion – especially as the new qualifications achieved in such a program are useful in other connections as well. HealthyBus is an example of how we chose to design and understand health promotion.

Hospital Admission due to cardiovascular disease SHR 200 Bus drivers 180

160

140 Unskilled labourer 120 Skilled labourer Other salaried employees

100

Leading salaried employees Managing directors 80 Senior salaried employees, public sector 60 1982

1984

1986

1988

1990

1992

1994

Reflektioner 15-12

22/08/05

11:26

Side 7

Introduction 7

How can you use the HealthyBus method? The HealthyBus model is demanding and should only be used to its full extent when handling important, extensive and complex health problems related to the work environment – problems that have existed for a long time without being solved by means of the methods that are generally applied. A considerable part of the challenge is to handle the extensive variation in the causes for work environment and health problems at various workplaces. It is also important to be aware from the beginning that it takes a long time – often years – to initiate, accomplish and anchor effective solutions. It is the rule rather than the exception that cross-disciplinary methods should be used when intervening in several areas at the same time. And it is only by cooperating and establishing an equal partnership that such a project will succeed. A partnership is an equal, democratic cooperation in which everybody is heard and takes responsibility for the task solving effort. The HealthyBus model can not be used where injunctions and legislation are the logical consequences, for example in connection with rules for shielding and safety equipment. You can say that the HealthyBus model is a supplement to what is usually practiced, but it is not a replacement for existing experience. The model is useful in cases where various stakeholders perceive causality and conduct of action differently, and the differences are significant. The project is an example that shows how researchers can become better intervention researchers by learning from and with practicians. The workplaces can also use this method to do a better job at transforming research results into action. In both cases the result depends on how well researchers, consultants and users interact.

The health problems of bus drivers It is not our intention to provide you with a complete reading of the scientific literature, but we do want to give a brief summary of the situation at the start of our HealthyBus project. This book should give you an idea of the wide spectrum of problems you would encounter if you were in our “health promotion shoes”. It is not sufficient only to study the work environment. You must also study the interaction between work conditions and a person’s life outside of work. Already 50 years ago, Morris and his employees discovered that bus drivers had an increased risk of cardiovascular diseases in comparison to conductors. Since then, many studies have confirmed this elevated risk for cardiovascular disease. In Denmark, bus drivers are admitted to hospital for treatment of cardiovascular disease twice as often as other wage earners, and the heart disease mortality has increased. Drivers also have a higher incidence of lung cancer and other diseases of the respiratory tract, hypertension, prolapsed disk, and a number of stomach and indigestion symptoms, as well as muscle and back pain.

Work environment of the bus drivers Drivers suffer exposure to carcinogenic agents (primarily from exhaust gas) and ergonomic exposures such as inconvenient design of the driver’s seat and humps in the road.

This publication covers the period from 1998 to 2002. From the moment the idea was born through development of a questionnaire, completing the baseline study in 1999 and initiation of interventions, to the follow up in 2001. Then came the evaluation and the first report of effect measures. The last follow-up took place at the end of 2003, and then we looked at how the changes were anchored. Those results are not included in this book, but they are available on www.ami.dk

Reflektioner 15-12

22/08/05

11:26

Side 8

Introduction 8

However, the stress causing factors have caught our primary attention as the possible cause of cardiovascular disease. On the other hand, no research has been conducted on the effects that management and cooperation, outsourcing, education etc. have on the drivers’ work environment.

Intervention research in the bus industry During the last three decades, Danish research has been at the forefront when it comes to studying the work environment of bus drivers, and there are several good examples of intervention projects. Just as it is the case in international research, intervention research does not play an important role in implementing changes in practice. The intervention projects have focused on specific problems such as muscle symptoms and stress, and relatively few drivers have participated under the controlled, laboratorylike conditions. Research has mostly contributed descriptive and causal studies which were done to gain information on the health damaging character of the work environment. So it has been up to other players to react to that information. Although only a few studies have been published in scientific journals, it is not so that nothing has happened. A part of the explanation to that discrepancy is that many intervention projects take place as an integrated part of the daily safety and environmental work of a company.

Health promotion from the holistic point of view There is no doubt about it – work has a health damaging potential. But at the same time, bus drivers have a high frequency of many of the diseases related to lifestyle. It can not be precluded that there may be explanations of their disease frequency other than the work alone, and this point is seldom included in scientific research studies. In a significant percentage of the lung cancer diagnoses for bus drivers the cause was not attributed to air pollution, but to the fact that drivers smoke more than people do in other occupations. We just want to point out, there may be multiple solutions to one and the same problem, and we would also like to stress the need for examining whether or not a combined general effort would be more useful than anything else. We do the drivers a disservice if we make them believe that they could possibly compensate the effects of other environments on their health by changing the work environment. Then there is the matter of social lopsidedness. Drivers belong to the group of unskilled workers who suffer in general from health and work environmental problems. That is another factor which makes it difficult to solve the task of raising the drivers’ health to the same level of other wage earners. It has to be done! If the goal is best possible improvement of health, then there is no doubt about getting full value for our money if we bet on other environmental factors than the work environment. Many work environment professionals find that way of thinking wrong, because it is the interaction between work and other environments which is most beneficial to health in general. The health promotion method centres on the workplace as a good place to start, where the holistic approach is desired. If it is to succeed, however, researchers and other work environment professionals must prioritize and communicate the core elements of this approach.

Reflektioner 15-12

22/08/05

11:26

Side 9

Introduction 9

Outsourcing bus service in the Greater Copenhagen area Outsourcing is often said to be the villain behind the bus drivers’ poor health. However, that can not be the entire explanation, because they were proven to have comparatively poorer health than people in other walks of life already decades before outsourcing was invented. Until 1974, public transport in this area was owned by the municipalities and gathered under the name The Copenhagen Tramway Society (“Københavns Sporveje” in Danish). Then the Copenhagen City and Regional Transport (Danish abbreviation: HT) took over, and they operated the bus service with standard contracts until 1990. In 1989, the Danish Government passed the so-called HT Bill, which formed the basis for outsourcing in this sector as we know it today. The bus service was divided into an A-division which takes care of administration and the planning of daily operations (still known as HT) and a B-division responsible for the publicly owned bus service. The daily operation has mostly been put out to tender for periods of four to six years, although shorter periods have also been seen. But especially with the use of bigger and more expensive busses in the last few years, the contract period has been extended to 8 years. The purpose of outsourcing is to reduce costs, and the work of haulage contractors paid well before outsourcing was introduced. In the beginning, focus was therefore on reducing the price per driving hour, and there is no doubt that it was a success. The profit in the bus industry is now far below what characterises a financially healthy business. The costs have been reduced by utilizing the driver capacity more effectively. In the old days, it was normal that a shift consisted of several hours of rest, and each day drivers in scores had shifts where they were paid for just being at disposal. That is not the case anymore, and we have several examples where legislation and collective agreements are interpreted as narrowly as they can go. When a bus route is out for tender, the drivers are protected by the law on business take-overs, which stipulates that the employees must be offered a job with the new employer. The job is secured, but it may continue under other conditions. For example, it can be a problem to transfer seniority from a previous employer to the new one. A company can also be taken over during the contract period. That has been seen several times. Outsourcing is a complicated affair with extremely detailed product descriptions which are hard to see through. It became a part of the HealthyBus research obligation to study what role work environment plays in outsourcing. The project also tried to determine if outsourcing could be used as a means of improving health.

Reflektioner 15-12

22/08/05

11:26

Side 10

This chapter in a nutshell: This phase of the project dealt with the clarification of roles and early organisational decisions. A thorough interpretation of possibilities and limitations in the early stages helped to build a common vision. Main topics were: – The health promotion concept defined as a method to influence one’s own life – How the purpose of the project turned into a model with general goals – Considering the design – Six principles for cooperation – Establishing the project organisation with a task force and a central project group. It was important that the representatives of the industry were employed directly in the project.

Reflektioner 15-12

22/08/05

11:26

Side 11

The infant level 11

The infant level: From idea to project We have always been willing to participate in research projects but each project has only drawn attention to the negative aspects and never illustrated when we did something that was good. Eigil Koop, managing director, Bus Denmark

When the project was still at its infant level the project manager paid visits to the stakeholders – the two unions, the employers’ association, Copenhagen Transport, and the six haulage contractors. These dialogues were far more fruitful than a joint meeting would have been in the beginning, because these visits gave us a differentiated insight of how the industry saw itself from the start. As the political landscape was hidebound after ten years of outsourcing, it was important for us to win their confidence. It was necessary to consider whether one justification of the project was in fact that we would be giving the stakeholders an alternative to the much politicized social conventions. That was done when NIOH agreed to participate as a neutral and equal partner but also as an active fellow player in preparing the future needs for research and intervention. It was also important for the researchers and other consultants to have the capabilities, the desire, and the luck required to play the game. This phase was mostly concentrated on negotiations in which the basis for a common vision and possession of the project to be established.

Is it possible to create a common vision? We had to find out if it was going to be possible to create a common vision. Generally, the stakeholders were tired of the stereotype, negative image that most people had of the industry. They all wanted action, and not more studies. At this point it is necessary to emphasize that scientific, evidence-based knowledge traditionally builds on studies conducted in the past. Many of the problems illuminated earlier had been solved by this time. For example, the defroster intake had been moved from street level to the roof of the bus, and it does not create credibility if one refuses to budge from one's scientific demands to a study when the industry can not recognise the need. The needs of the industry will always point forward, and as a matter of fact, the problems also turned out to be different from what was in the evidence base. However, everybody agreed that more problems did await solutions.

Stakeholders A stakeholder is a person, group or organisation that has an interest in a particular project and its results. As partners they will always be able to influence the progress of a project. The primary partners or stakeholders of a project are a dominating coalition whose word is always decisive.

Reflektioner 15-12

22/08/05

11:26

Side 12

The infant level 12

It is crucial that you never end up in a situation where one or several of the primary partners feel that their interests are being neglected. During the project start-up it is important to find the points of intersection where the stakeholders agree instead of searching for more disagreement. In HealthyBus, agreement was reached whenever poor health was mentioned. However, the stakeholders did not agree on what the reasons were for the poor health record. Simplistically described, the unions thought it was caused by poor work environment conditions and a financial starving out of the industry because of outsourcing, whereas the haulage contractors pointed out the effect of the financially determined need for being more efficient, and they also believed that an unhealthy lifestyle could cause poor health. The transport authority HT pointed out lifestyle and the haulage contractors’ lack of ability to make things more efficient without damaging the work environment. So at least, as a starting point, they matched by two and two. All agreed to emphasize action to replace what seemed to be common knowledge about “the bus drivers’ poor health and work environment” with something better.

Establishing industry possession of the project in an overall perspective We based our work on the extended, theoretical health concept, meaning that if we studied the psychosocial conditions of the work we also had to include private life exposures. And we had to be able to relate how these conditions and exposures were integrated. The workplace remained the basis for action, even if it was hoped that experience and results generated here would go much further. If the project was to have a lasting effect, it was necessary for the industry to take full responsibility for the action to be taken. From the beginning it was important to make clear that this was not the researchers’ job. It all boiled down to an establishment of the industry’ possession of the project.

The Luxembourg Declaration

Health promotion The concept of health promotion at the workplace was used for the new signals it represented. In practice it was an empty concept, as it had not been used before, and we lacked experience on how the concept could be taken into operation. The European definition of health promotion had been published the year before in the Luxembourg

The Luxembourg Declaration defines health promotion at the workplace as the combined effect of the initiatives taken by the employer, the wage earner and society to improve health and wellbeing among the working population.

Declaration. But in Denmark, many people still used the term “health promotion” as a broader synonym for lifestyle. HealthyBus had to develop the concept.

This can be done as a combination of: - improved organisation of the work and the work environment - improved support to the employees’ personal development - promoting the employees active participation

Reflektioner 15-12

22/08/05

11:26

Side 13

The infant level 13

Defining the purpose

Good advice

The purpose was defined to contain all the main interests of the stakeholders. The purpo-

In the start-up phase it is important for the project manager to be aware of the necessity - to be open, - to respect the participants’ points of view, - to see and create alliances, - to find solutions instead of problems, - never to present a take-it-or-leave-it solution to the stakeholders.

ses of HealthyBus were: •

to integrate health promotion in the daily operations and quickly obtain a general participation by the majority of the drivers,

•

to continuously improve health and well-being,

•

to reduce and eliminate health risks,

•

to increase job satisfaction,

•

to restore professional pride and regain a good reputation,

•

to make the job more attractive and retain more of the drivers,

•

to develop methods to control the health promotion efforts.

The first two bullets show that they all agreed on the fact that health improvement was essential. But since they did not agree on the causes and what to do about the problem, the formulation was made as general as possible and thereby it covered the stakeholders’ models of explanation. Some may say that it also made the purpose non-binding, but we would not agree with that opinion. We agreed on making the interventions as broad as possible, and that created sufficient trust to make it happen. We all expected many drivers to be unsatisfied with their jobs and have no pride in the work, considering the negative reputation both internally among the drivers and externally among the population in general. That it would be difficult to find manpower was in the cards. So again, we agreed on the purpose but not on which way to go. There were practical and theoretical wishes for operationalising and evaluating health promotion. Please notice that the word “work environment” is not mentioned once in the objectives. All the same, there was no doubt that it was a work environment project.

Design considerations From the beginning of the project there was a clear wish for a controlled study design in which the companies were divided into an intervention group exposed to well-described processes of change, and control groups which were not exposed. The four most important reasons for not giving in to the wish for a controlled design were as follows: •

All garages would be exposed to significant changes which would affect the design (for example, outsourcing).

•

We did not think that we could find sufficient general interventions which would be both effective and realistic. Something indicated that local conditions would carry more weight.

•

The changeable organisational conditions meant that it was difficult to see which persons had been fully exposed or only partly exposed to the intervention throughout the entire period.

•

The project was not causal, and the purpose was not to study which intervention was the most effective. The purpose was to see if it was possible to intervene sufficiently at all.

Reflektioner 15-12

22/08/05

11:26

Side 14

The infant level 14

We decided to use a case study design, although unfortunately that has unpleasant associ-

The case method

ations among some researchers. Many researchers find that the method is better suited for

The case method makes it possible to study meaningful characteristics in reality. Case studies are divided into - exploratory studies, - descriptive studies, - explanatory studies.

the exploratory stage, which is often conducted by the youngest researcher. However, that judgment could not be more wrong. It is an advanced design which requires that the most experienced stakeholders – both researchers and practicians – participate. It is a follow-up design which measures before, during and after, and measures the process as well as the results. But as these interventions were effected under realistic everyday conditions it could be difficult to manage information about how many drivers and who participated in what

The design may contain several scientific methods at the same time. That makes it possible to get answers on who, what, where, how many, how much, how and why.

intervention for how long. This was also complicated by interventions being initiated in several different connections. In contrast to the controlled experiment, the case method seldom provides statistical evidence of the effects of a particular intervention. Instead, one talks about the probability that a certain initiative has the wanted effect. The more evidence, the more confidence that a specific initiative has been effective. We recommended the method, as it was a continuation of the stakeholders’ working methods. Naturally, the case method also requires systematic data collection. In HealthyBus, we used the below mentioned sources of information to gather local knowledge: •

scientific standard methods such as epidemiological data from questionnaire studies and qualitative in-depth analyses based on interviews, focus groups and observations,

•

a logbook of relevant activities at the workplaces,

•

a diary which project employees kept of their own observations and informal information,

•

relevant, accessible published material from newspaper articles, industry journals, garage magazines etc.

Establishing the project organisation

Six principles of cooperation

A task force first had to be established, with representatives from the four main stakehol-

Six basic principles for cooperation were listed:

ders. They came from the trade unions, the employers’ association, the Copenhagen City and Regional Transport, and the National Institute of Occupational Health. The functions

•

of this task force were “….to support the project, motivate the stakeholders and handle

•

the overall decision-making in accordance with de facto conditions of the industry.”

•

Further, “the task force must be involved and actively participate in solving organisational and problems and policy issues.” We deliberately aimed at maintaining the functions of the task force on a general political level.

Participative involvement is a must

Business as usual ∑ Multiple interventions take place ∑ • Focus on the stakeholders main needs ∑ • The industry is the experts on how to run a business ∑ • Therefore the industry is responsible for action

Reflektioner 15-12

22/08/05

11:26

Side 15

The infant level 15

We hired five representatives from the industry to work half-time on the project. This is not at all common within work environment research, but we wanted their input. Firstly, we thought it would not be possible for the researchers to get as close to the industry as would be needed to obtain enough inside knowledge for evaluating the basis for action. Secondly, we thought that it would increase the credibility of the project within the industry. That was necessary for the industry to feel that this was their project and to take responsibility for carrying out the interventions. Thirdly, it was a condition for accomplishing other goals set for the project group, namely: •

to represent the stakeholder interests and report to the task force,

•

to secure mutual learning between researchers and practicians,

•

to work for industry commitment and their possession of the project,

•

to be the daily link between the researchers and the industry,

•

to make an effort in a partnership model.

The industry representatives in the project group were not only appointed by the task force, but the task force also insisted on paying their wages, which showed the depth of their commitment. There is no doubt that this type of cooperation was the one factor with the most influence on progress within the project.

Project manager’s meta-reflection If I were to start afresh, I would use the same method again in the start-up phase. The ping-pong game is necessary, as the freedom of action is largest at the beginning of a project. However, it takes time – a luxury which is often not available in great quantities. This phase lasted a year, in which it was uncertain if the project would continue. But it was all about building up trust in the project – and trust takes time.

Reflektioner 15-12

22/08/05

11:26

Side 16

This chapter in a nutshell: To be able to initiate effective interventions, the stakeholders needed to know the points of view of each other and what kind of responses to expect. For that purpose, the project employed a modified version of the brainstorm seminar. We developed a special questionnaire tailored for the bus industry, to encircle our possibilities for action. In the holistic approach, it was necessary to include many new topics. The stakeholders’ overwhelming commitment led to a high response rate that provided truly representative findings. These results indicated that there were still important problems to be solved within the stakeholder areas of interest. Based on that, the key stakeholders decided how to proceed. They found a method to interpret the results on their own and to describe realistic, prioritized suggestions for effective interventions which could be initiated immediately.

Reflektioner 15-12

22/08/05

11:26

Side 17

Establishing baseline 17

Establishing baseline – what is the situation HealthyBus must not end up as just another report! Reiner Burgwald, union representative, Trafikfunktionaerernes Fagforbund

Even though we already had gained much useful information from the baseline study, it was necessary to obtain a more detailed picture of the general outlook of the individual stakeholders and their impressions of each other. As mentioned, there were many reasons why the stakeholders had such different views on the reasons for the bus drivers’ poor health. That is important to consider when creating a common vision for future action. Some people call these basic positions the party's “programme theory”. However, in this book we call them the “model of explanation”. To obtain detailed knowledge about variations in the bus drivers’ perceptions of their jobs and their own health conditions, it was suggested that all of the drivers should be asked, and not only their organisations. It was obvious that we should conduct a questionnaire study which we could use again later to follow up on the drivers’ perceptions the intervention effects. But it was not obvious which questions we should ask in order to solve the task. The explanation models presented by the stakeholders had to be explicit so that we could be sure of asking the right questions in the right way.

Brainstorm seminar – a way to talk about outlooks on life The first task of the project group was to describe how the stakeholders saw the problems, and also what we thought was most important to study in HealthyBus. The method was a modified brainstorm seminar. The process was divided into four phases: •

All words referring to bus drivers were listed and categorised - words like, for example, running times, health, cooperation, management,…

•

The next phase was to write only positive things. In both phases it was not allowed to discuss the content, but it was permissible to ask clarifying questions.

•

The next step was the meta-position in which we described how we generally saw ourselves and each other. The many myths about this industry were brought to light at this point, like the “shitty job” myth, the “we need more studies” myth, and the “healthy in, sick out” myth.

•

Finally, the participants gave their spontaneous and realistic reactions to what they thought could be done to improve the conditions on the basis of the knowledge we had shared.

Reflektioner 15-12

22/08/05

11:26

Side 18

Establishing baseline 18

Developing a questionnaire intended for the bus drivers The more than 600 items contributed in the brainstorm seminar were collected in twenty chapters which formed the basis for developing a questionnaire tailored to this specific industry. Many of the subjects had never been studied before, but the stakeholders said they were important in the scope of the holistic approach chosen by the project team. The subjects were, for example, the experience of colleagues, the shop steward function, the company's management, and private life. In HealthyBus, it was important that we could study all angles of a subject. Problems at work are often related to private problems, which then starts a vicious circle. Of necessity, both factors had to be included when evaluating the potential for action. The most significant paradigmatic challenge was that the practicians insisted that several of the questions in the scientific standard questionnaires were too abstract. They argued that if a question was difficult for the bus drivers to understand, it would not be possible to get a clear answer. We knew from the brainstorm seminar that there were enough problems here for the stakeholders to demand action. The purpose of the questionnaire was not so much to study if there were any problems, as it was to find out where the problems were, how far-reaching they were, and more specifically what was causing the problems. Therefore, it was decided to tailor a questionnaire for this specific project. It was a questionnaire focusing on problems in the bus industry - and it allowed us to solve two problems at once. We would qualify the problem analysis and at the same time we would find out how to proceed. An example was our hope of attaining some general information about the relationship of drivers to management. Instead of asking drivers what they thought about their management, we chose to ask six specific questions, as follows: •

Do you get a negative reaction anytime something goes wrong?

•

Does the management here treat you well?

•

Is it possible to communicate with the management?

•

Does it lead to an even more unpleasant situation if you discuss problems with someone in management?

•

Is it always the same drivers whose voices are heard?

•

Does your manager give you a feeling of insecurity?

During the project we conducted more or less structured interviews with the stakeholders to test some of the most important ideas, to go into details with some subjects or to amplify others. We became more and more aware that it was important for group members to distance themselves from their practical routines and take time for meta-reflections. For instance, it was necessary to make the extended health model more detailed, so that it also provided explanations. At a theory of recognition level we had moved away from our starting point.

Questionnaires intended for action When it comes to obtaining specific knowledge to act on, the best thing to do is to direct a questionnaire at the target group. Many problems and solutions are related to a specific context. Furthermore, it is necessary to: - focus on solutions, - be loyal to main interests of the parties, - use scientifically correct questions taken from good questionnaires that are available, - include new areas, - demand up-to-date information.

Reflektioner 15-12

22/08/05

11:26

Side 19

Establishing baseline 19

Practical organisation of the baseline study The task force commented on the questionnaire, and then it was tested in a pilot study. Then local HealthyBus organisations were established at company and garage levels, with representatives from the local managements, and shop and safety stewards. We arranged a kick-off conference at the National Institute of Occupational Health in September 1999 to present the plan for the questionnaire study that had been tailored for this industry. The next step was to work for a high and representative response rate. The entire project organisation invested much energy in an extensive publicity campaign with newsletters, folders for the drivers, posters, garage visits, oral presentations and meetings. The key stakeholders donated four prizes that could be drawn among those who filled out the questionnaire. The questionnaires were handed out at the garages by the project group and the local HealthyBus groups. Exposure was highly prioritized - a questionnaire was given to every bus driver on the payrolls as of October 1st. If a driver had not returned the questionnaire within 14 days, a new one was sent to that person’s home address with a friendly request to please send it back. The last reminder offered a phone interview or a visit at home by the interviewer. Trust in the project, evidenced by the high response rate (76%), beyond doubt that the National Institute of Occupational Health was accepted as a trustworthy organisation that could handle confidential information. Should the HealthyBus method be used as a matter of routine, the assistance of advisers and researchers as data managers and analysts will be needed.

What did the needs analysis show? The immediate conclusion was that, "the total amount of exposures in the work environment, organisation of the work and lifestyle is so great that it must be expected that the morbidity of the drivers will stay at the present high level unless a sufficiently effective, widerange and long-lasting health promotion effort is initiated.” The problems were divided into four areas - organisation of the work, qualifications and education, physical/psychosocial work environment, and lifestyle. This model was called the “four-leaf clover” model. The needs analysis also confirmed that there are still considerable problems. However, as something new and conclusive we could provide documentation which supported all the stakeholders’ models of explanation. As we also focused on the positive elements, it became possible to point out where to find the energy to do something about the problems.

Action yes, but when? A credibility problem can easily occur in relation to how you really create change for the better when it takes two years after the idea is born for the first intervention to come. Our advice is to keep a low profile during that period of time.

Reflektioner 15-12

22/08/05

11:26

Side 20

Establishing baseline 20

We graphically presented the results for three target groups: •

Copenhagen City and Regional Transport,

•

haulage contractors,

•

each garage.

It was important that not only the researchers worked out the conclusions, prioritized the work, and told the industry what they needed to do, because: •

This industry has expertise in handling its daily routines.

•

The industry can interpret the results based on a local context.

•

The industry can come up with realistic solutions.

•

The industry feels a proprietary responsibility for action.

Our role was to act as sparring partner and anchormen. We were not just objective observers, but participated actively in the discussions, also contributing ideas. How to convert results to interventions At a task force meeting in April 2000 it was decided to act on the basis of existing knowledge, and agreements were reached concerning the future of the project. However, increasing lifestyle problems among the drivers were a troubling issue, and some members feared that some of the haulage contractors would only work with that issue. This is when the four-leaf clover model proved its merit. The task force emphasized that all of the four issues had to be dealt with. If a garage had at least one intervention within each of the four leaves, it could call itself a spearhead garage. Lifestyle issues were not allowed to take up more than a fourth of the intervention activities. The first results were published, and the drivers who had filled out a questionnaire now received a newsletter reporting the main results, and it was sent to their home address. The members of the project group also participated in discussion meetings at the garages. A month later the local HealthyBus groups met at a working conference held at the National Institute of Occupational Health. There we presented a suggestion for systematic procedures they could use to go through their own data and prioritize their own need for interventions. All HealthyBus interventions should be noted in an activity plan (can be found in the Inspiration Book written by the project group). Each quarter year, a project group member met with the local contact person so they could evaluate whether the milestones of the activity plan had been reached and what changes they needed to make. The results were gathered and briefly summarised in a joint quarterly report which was sent to the local HealthyBus groups. At the following quarterly meeting the groups could discuss their experiences with one another.

Project manager's meta-reflections - It is productive to employ a partnership model when the parties are involved in defining problems and suggesting interventions. However, I wish that I had had more time to go into the processes more deeply in order to describe the intervention processes in greater detail. - As it is faster and less resource demanding to use CATI interviews, I will use that method in the future. However, my colleagues say that I must not underestimate the importance of handing out questionnaires personally.

Reflektioner 15-12

22/08/05

11:26

Side 21

Establishing baseline 21

The industry publicly commits itself to solving the problem Finally, we were ready to show that the time had come for something to happen. It was time to show that HealthyBus was more that just another study. The official event in connection with the intervention phase was a conference held in September 2000. The Ministers of Labour and Traffic participated in the conference. The industry presented a temporary intervention catalogue with more than 100 concrete interventions, that they had committed to carry through. The interventions covered a wide spectrum - from top level activities of the Copenhagen City and Regional Transport administration for all drivers to initiatives that came at the driver level. Six garages committed their groups as spearhead garages. And the Minister of Labour announced that he would be following the work of this industry so closely that it would feel him breathing down its neck. The action potential was established and we were on course.

Reflektioner 15-12

22/08/05

11:26

Side 22

This chapter in a nutshell: In the HealthyBus project, the companies organised themselves very differently. But that only reflects the differences in company culture. We found that most of the interventions were based on contextual analysis locally anchored. This meant that a certain answer in the questionnaire had many different explanations and therefore many solution possibilities. Often there was not one solution which was best for everybody. How the quarterly evaluation went is described here. A description is also given here of the many very extensive changes conditional upon external events which were of great significance to the project conditions. The project had gained a foothold by this time, because so many fiery souls believed in it.

Reflektioner 15-12

22/08/05

11:26

Side 23

The intervention phase 23

The intervention phase – finally some action The combined efforts of the four areas correspond closely to the government’s declared policy, and it is pointless to discuss which elements come first when all are necessary in order to achieve the goal of better health for bus drivers. Ove Hygum, Minister of Labour

The way the companies chose to approach the practical organisation of the project varied. The largest companies chose a very visible structure. One contractor chose to manage the environmental and operational areas centrally. Another chose a decentralised organisation model. Others chose looser and at times even more hesitant methods. The smallest haulage contractors, who at the beginning of the project employed less than 100 drivers all together, were usually organised in a pater familia structure. Therefore, many things depended on the company culture within the haulage contractor firms. We found that it was important to support the existing cultures if they were constructive, instead of forcing an externally defined model on them. Our principle was that we would support existing structures as much as we possibly could, just as it was a principle to keep things at a simple level when introducing new administrative and organisational methods.

Interventions are not only interventions The questionnaire study pointed out several areas in which changes were necessary. For example, more than one third of the bus drivers drove a bus with defects at least once a week. However, that average was distributed among the garages at rates ranging from 7% to 55%. Naturally, some garages saw it as a severe problem which became highly prioritized, whereas other garages did not find it necessary to do anything more about that particular problem, because they had registered other problems which had higher priority. Now we saw that it would be a problem if we, on the basis of averaged results, were to claim, for example, that defective busses were a general problem and thus introduced that as a standard intervention at all garages. We encouraged the garages to consider a local interpretation of what their statements about defective busses meant. One garage meant defects reported to the mechanics, another garage meant there were not enough mechanics, and a third garage said that it was not possible to maintenance busses during the morning and afternoon as they were out on the streets and roads during the daytime hours. At a fourth garage, they said that

Reflektioner 15-12

22/08/05

11:26

Side 24

The intervention phase 24

they lacked information on how the system worked when a bus needed to be repaired. We also heard various combinations of these definitions. But the most interesting point is that we did not get this information from the questionnaire! It goes without saying that the interventions here were directed toward the causes which the local experts pointed out. One place bought new busses, another garage hired more mechanics, and some of them work night shifts now. A third garage offered courses to teach the drivers how to handle defects when they occurred. Now the drivers also learned how the system works so they have an understanding of how things connect from dot-to-dot. Even though the intervention complexes depended on the context, it was possible to measure the effect of intervention against defective busses with the questionnaires. The effect was an expression of a meta-perception. The test was to see if the garages had been able to make their own analyses and act upon them. As you will see later, it was a success.

Does anything really ever happen? Evaluation of the process in practice Of course, most garages experienced variations during the evaluations. So what we present here is an ideal model. The project group established regular contact with the garages participating in HealthyBus – the best was to pay them a visit. But it could also just be a phone call. Each quarter the activity plans were evaluated and updated. If significant changes were made, such as changes in the timetable, they had to be noted in the revised plan. When all the plans had been collected, the project group wrote a brief summary in each garage’s quarterly report. The summaries included information on purpose, activities, status and contact persons. The examples below represent each of the four clover leaves: •

The garage wants to improve training in new routes. New, individual training programmes have been initiated. The intervention has been carried through. NN is the contact person.

•

The purpose is to work out realistic timetables. The garage and The Greater Copenhagen Authority have made new rush-hour timetables for two of the busiest lines. The activity has not yet been fully implemented, but it is on schedule. NN is the contact person.

•

The garage wants to increase the drivers’ trust in the bus radio system, which also functions as an alarm system – for example, in the case of assaults. The garage runs a radio check on all of the radio devices at least once a week. The result is noted and defects are followed up. NN is the contact person.

•

The garage wants to increase the drivers’ well-being by informing them about healthy nutrition. The garage offers a course on healthy nutrition and lifestyle to the drivers and their families. The plan has not yet been completed, but is proceeding satisfactorily. The manager is the contact person.

The quarterly report was sent out to the local HealthyBus groups, the task force and

Reflektioner 15-12

22/08/05

11:26

Side 25

The intervention phase 25

other stakeholders. Then a quarterly meeting was held - scheduled to take up half a day to which everybody was invited. The purpose was to exchange ideas across companies and organisations. The idea was to inspire participants and to discuss various subjects. Many participants came, but mainly those who were already the most active people and those whose companies could let their employees participate during the working hours. It was always the same persons who had not been able to come, so we felt it was a big pat on the shoulder when a group of drivers who had never participated before actually took a day off to be at this quarterly meeting.

When reality changes during the project period We were right! The industry goes through many changes, and it would have been difficult to lock things for a period of four years. To give an impression of the externally induced changes, we can mention that six haulage contractors were attached to the project in the beginning - two large, two middle-sized, and two small haulage contractors with all together 20 garages. During this time, Bus Denmark was taken over by Arriva International and the board of directors had been replaced. Linjebus was taken over by Connex, which became the other large foreign player in the Danish market sector of bus transportation. Technically, Combus went bankrupt a couple of times and the management was replaced, so this company experienced a very long period in which it was not known what its future would be. That insecurity came to an end when the company sold the Greater Copenhagen part of their business to Connex. City Traffic and the other two small companies (Fjordbus and De Hvide Busser) did not go through organisational change of this kind. There were strikes, both local and within the entire industry. Each year a part of the routes were outsourced, and routes and drivers were transferred to new employers. Completing the Copenhagen Subway took longer than expected, and it meant that some bus routes could continue a while longer than had been scheduled, and then they were shut down when the subway was opened. To this many internal changes can be added in that management, employees and procedures rotated or were replaced. This is a market which sells for approx. two billion Danish Kroner ( 270 million) per year so the project had to work under the conditions of business as usual. The project was a success and an experience which work environment professionals accept more easily than researchers do. We think there are several reasons why HealthyBus caught hold in the industry. The most significant element were the many fiery souls at all the right places on all levels – from the shop steward to the board of directors, from the haulage contractor to the traffic company, and from the industry to the organisations.

Success in intervention projects It is worth considering whether one should hope for three small, visible successes rather than risking one huge failure. Many researchers have laughed at the intervention which tried to offer free fruit to the drivers. Beside the fact that it is a good and health promoting idea, the process of the intervention quickly led to mutual trust which was used as the basis for success in other, more complex interventions

Reflektioner 15-12

22/08/05

11:26

Side 26

The intervention phase 26

Stability was a precondition for the "fiery soul" leaders to get a foothold and create results. That is why the companies had to be committed to the project for the entire project period. The local HealthyBus groups strongly depended on continuity in the composition of their groups and teams. It was obvious that a void was left if the project's "fiery soul" transferred to another garage, if they were not substituted by another leader with a fiery soul. At Combus the project had a very impressive start. However, uncertainty over the future of the company made it difficult for the "fiery soul" leaders there to be heard. Obviously, it was necessary to have a good climate for cooperation in order to obtain good results, and people had to be totally committed to the project. It was also important that a visible organisational structure became established at all levels. Much had to be built up from scratch. These companies did not have the same traditions and experience that are available to the streamlined industries which are often referred to as examples of modern business organisation. In other words, a committed and solid organisation had to be created, if the bus drivers’ conditions were to be improved despite the external changes. Such an organisation depended totally upon a small group of initiative takers. It was because of cooperation with people like these that it was possible for the project to gain latitude. There were no easy solutions. Trees do not grow as tall as the sky and there were many problems to face during the project period. However, only few problems were so severe that they were a threat to the project. One of the reasons was that the participants had committed themselves to stick to certain rules already at the beginning of the project. However, we certainly had problems and everything could have gone wrong in the extreme, because it often was impossible to foresee problems that were right around the corner. Political factors could affect the project. One unfortunate factor was the unintended focus on lifestyle, which the project manager was not able to avoid. Conflicts between the stakeholders could burst out all of a sudden, maybe in connection with negative comments on HealthyBus or a critical situation which did not even directly involve HealthyBus. The National Institute of Occupational Health received formal criticism from some researchers who disapproved of the project concept and the methods used. We even had a situation where an anonymous driver threatened the employees who were attached to the project. It was important that the task force could meet and find solutions to these problem issues. The project suffered at times from the loss of project members, and in some cases it took a while before replacements were found. At the end of the project it turned out that it was impossible to find a replacement for one of the members. There were also problems with internal and external conditions at the various garages. So we had our hands full.

Project manager’s meta-reflections - The quarterly meetings did not work exactly as intended, and I could have involved the participants more than I did. - Seen from a researcher’s point of view it was very demanding to work with people of so many different professions, and we used a relatively large amount of resources on participation in meetings in various connections. On one hand, it was what other researchers recommended. On the other, it took too long before we produced any scientific results. I wish more senior researchers had been attached to the project from the early phases on. This kind of research is fairly expensive if you want to take a holistic approach and work with synergy, not just addition. If I had to start all over again I would wish: - for more structured interviews, - to include the stakeholders more directly in the research. - But it was all about doing the possible, and not attempting the impossible. It made a difference too, that we were reaching out to 3500 drivers, and not only 350.

Reflektioner 15-12

22/08/05

11:26

Side 27

The intervention phase 27

Reflektioner 15-12

22/08/05

11:26

Side 28

This chapter in a nutshell: – A brief description of how the follow-up study was conducted – By combining qualitative and quantitative measures we found answers to the following four questions: what, how much, why, and how. – Researchers and advisers must be careful when interpreting their own models of explanation as a foundation for reason, action and effect. – The preliminary results play an important role in projects which are based on reality. The reason is that it is often necessary to act before the results from the scientific studies are available. – The industry issued a press release which indicated that HealthyBus was a success.

Reflektioner 15-12

22/08/05

11:26

Side 29

Follow up 29

Follow-up – did anything actually take place? Health promotion at the workplace creates optimism, enthusiasm, involvement and new thinking, and it prevents employers and employees from being defensive at the workplace – and in general. Freely rewritten after Tom Tiller

The interventions had turned into routine functions and they worked well. The next step was to find out if we could measure these changes in the parameters which had triggered the action. The idea was to use the same method as in the baseline study. However, we decided to change some parts of the procedure: •

We revised the questionnaire on the basis of our experience from the first round.

•

There were no prizes to win.

•

We would take local conditions into consideration, for example, how the questionnaires were handed out.

The follow-up study was conducted in autumn 2001, exactly the same way as the baseline study. Again the response rate was quite high, namely 72%.

Obtaining knowledge of what the figures do not tell us At the request of the task force we also conducted a number of focus group interviews to find out what the figures did not tell us. We wanted to find out if the results derived from the questionnaires corresponded with the drivers’ real life experience. The task force thought it would increase the credibility of the project if we got the drivers’ own words for their experiences, and the seven focus group interviews indicated that the questionnaire results did correspond with the drivers’ own words. But it turned out all the same that the drivers did not associate their improved work environment with the HealthyBus project. That may be explained by the fact that the project was incorporated in their daily routines, so it was not seen as an external “intrusion”. We recommend for that reason that you consider the possible need for external branding in similar projects. It is essential that you actively confront your models of explanation with reality. It is a way to challenge the established approaches which characterize this type of project. It creates a dynamic process, and the qualitative methods are better suited than the quantitative. In HealthyBus we never did choose a specific hypothesis. Instead we adjusted the project to the reality in which the project existed, as this was where the interventions should be

Qualitative or quantitative? In connection with complex problems in the work with interventions it is often a good idea to choose a combined qualitative and quantitative approach. It is not a question of either/or and both/and. The project will show that synergy generates new knowledge. By choosing a combined qualitative and quantitative approach both methods become qualified. The result is a better output than if the methods are used separately.

Reflektioner 15-12

22/08/05

11:26

Side 30

Follow up 30

tested. It is interesting that the method makes it possible to test and adjust the models of explanation during the project, and within reasonable time it provided us with results to be used in the intervention process. The results indicated that two thirds of the interviewed bus drivers thought that their timetables did not work in the rush hour, but were perfectly adapted to the traffic situation outside of rush hours. By conducting the interviews it was possible to render visible explanations for the figures. It could be that the drivers were not given enough time (however that was most often not the explanation). Another possible explanation was that traffic problems led to large variations in how long it took to get from one point to another, and that was impossible to incorporate in the timetables. The qualitative method made it possible to translate the available knowledge within that specific area to scientific concepts that could be included in the analysis. One of the interventions was therefore to introduce interval timetables for two large routes in the rush hours.

Ability to inspire - As a work environment professional or a researcher it is important that you are able to inspire other people. That you are able to make the stakeholders understand when it “kicks ass” as we said in HealthyBus. That was part of the coaching work. The researcher has a problem if the stakeholders involved do not want to accept his/her active influence in the project. In action research it is a necessity that you are actively involved.

Thereafter it was possible to study the quantitative representativeness by using the questionnaire to compare all the drivers who worked on the interval timetable with all the other drivers. The advantage of the combined quantitative-qualitative method increases with the complexity of problems, for example when you study the reasons for poor health. Many researchers and advisers will probably say that they already knew that. However, it is crucial that you apply the means of a scientific method to render the stakeholders capable of finding the association in their accepted models of explanation, and not to allow it to come as the input of external persons (who seldom are experts in bus service). It is important that the stakeholders involved believe in the model of explanation if the purpose is for them to operationalise interventions themselves. Our task as external researchers was to give advice on what we know something about, i.e. how to measure whether they achieved the wanted results during the wanted process.

Data analysis and feedback We made a graphic presentation of the results for each haulage contractor and for each garage, and also compared 1999 with 2001. This time we had no “interpretation by the researcher”. The thought was that it was only if the stakeholders had to study the material themselves that they would develop a feeling of possession and responsibility for the process. Instead, it resulted in internal disagreement in the project group and external disagreement with the task force. However, it would have been the wrong methodology to tell the stakeholders how we, the researchers, thought they should interpret the material. It was the truth with a few modifications. Long before the data material was collected, preliminary comparisons had been presented, and they turned out to be right. We used data which was not standard in order to give the process a kick - based on meta-knowledge from the first years of the project - because we did not think that the industry should wait for the data set to be cleaned and scientifically analysed before they could take action. When the stakeholders saw a difference that made sense, the statistical significance level became less important. Surprisingly, they were very willing to change their models of explanation on the basis of the preliminary results, even though the data did not support their original opinions.

Meta-knowledge - Meta-knowledge is knowledge of knowledge, meaning knowledge about the assumptions of creating knowledge. It is about creating meaning which can be used reflexively. Meta-knowledge is therefore strongly attached to the context.

Reflektioner 15-12

22/08/05

11:26

Side 31

Follow up 31

Local use of the results As the resources were sparse, the work at this point was no longer conducted like we did it at baseline. It was not possible to invite people to a full-day seminar at the National Institute of Occupational Health to work with the same tool that we used in the first round. In practice we were using the same preliminary results - this time in a local version. Besides the already mentioned details, there were three conditions of importance here: 1) the results had to make sense in the local context, 2) they had to be easy to communicate in the organisation, and 3) they had to a big great impact on the industry. Originally, the project group wanted the activity plans to specify percentages for the improvements that were targeted. Of course, not all questions were suited for that - or somebody was reluctant to come up with a percentage estimate. But when percentages were indicated, the project group often found that it was too optimistic to place those expectations on the results.

What does the industry say about HealthyBus After all, the task force agreed that the project was already a success and it sent out the following press release:

The HealthyBus project shows that hard work aimed at improving the work environment pays off At more than 20 garages in the Greater Copenhagen area, problems have been seriously dealt with and good cooperation has been established between the managements and shop stewards. A new culture has developed and there is confidence that within this cooperation and task delegation a number of well known work environment problems can be solved. It can be said in addition that the same garages not only document a convincing development within the work environment area, but they also present better operating results and enjoy a better reputation, and they are seeing more satisfied customers these days. This is reflected in high quality figures from passenger opinion polls. The efforts made to achieve a better work environment have paid off. The HealthyBus project proves that a good work environment and credible cooperation are all closely associated with customer satisfaction. There is no doubt that both the employers and the shop stewards at the garages which did not participate actively in the project have much to learn from the best garages. The industry has accepted this conclusion, and the traffic company, the employers and the employees will commit themselves to ensuring that experience from the four garages spreads throughout Greater Copenhagen and on to the rest of the country. The various organisations were involved in spreading their experience from the garages that had achieved improvements at a seminar for all garages in the Greater Copenhagen area. The garages received guidance on effective initiatives and how to implement them. There was no doubt that employee development via quality courses and other training which had taken place should be included as an important lever in the industry.

Project manager's meta-reflections - Could I do something over again I would know enough the next time to insist that: 1) everyone who is responsible for interventions should also be very active in the systematic collation of results, 2) the project group should interact more in the field (home work + coaching) , and 3) they should agree on a systematic interpretation of the results at the time when the new experience base is operationalised.

Reflektioner 15-12

22/08/05

11:26

Side 32

This chapter in a nutshell: Here we can offer a wide range of examples on how to inspire work environment professionals to find out where research is needed and help companies with their problems. The chapter reveals new types of skill which we find that advisers must possess if the workplace is to function as a dynamo for general improvement of public health. After the introduction we present 11 principle case situations:

– Work environment is like a four-leaf clover – status before intervention. – Bus drivers’ health – how poor is it? – There are more things in life than the career path. – Lifestyle – a taboo where drivers are the only losers. – A national comparison – are bus drivers worse off than others? – Is this the most stressful job in the world? – A good work environment is profitable for the company. – From “old Danish” to “new Danish” means cross-cultural cooperation. – Healthy in and sick out – a myth with modifications. – One must know what good health is before it can be promoted. – The health promotion method promotes a better work environment

Reflektioner 15-12

22/08/05

11:26

Side 33

Results 33

Results – what happened? If you are on time and do your job then you can stay employed for 100 years, but if you screw up you are fired. But that is just like anywhere else and not only in the bus industry. Bus driver with more than 30 years experience

Definitely – it's the management! We have a new management we can discuss things with. They treat us well. A small group of drivers

It is important to remember that the purpose of HealthyBus was not to test if one intervention was better than the other. We were supposed to study if it was possible – in practice - to improve the drivers’ health conditions by intervening. From the beginning, we knew it would be impossible to measure greater health effects within such a short period of time. So it all depended on the stakeholders’ convictions that the initiated interventions were a significant step in the right direction. It was also necessary that measures derived from the questionnaire and linked with qualitative knowledge could be used to inspire a feeling of trust in the efforts that would lead to improved health among the drivers. With 208 implemented interventions at the time of follow-up, we had to accept that it was impossible to do a classical epidemiological follow-up of the effects of intervention. We could not limit the scope to initiation of only a few interventions, because that would not be in accordance with our original agreement. Furthermore, evaluations would be problematic when they suggested implementing interventions that were already in planning beforehand. So it was decided that as long as it was convincing that an intervention had been necessary to create measurable change, it could be held to be a HealthyBus intervention. Therefore, the epidemiological analyses consisted of what we agreed on as important for inclusion, depending on the response variable. It could be characterized as circumstantial evidence. In the following you can read about some of our results. These results were selected before the third and final round of the questionnaire study in 2003. The subjects illustrate some of the conditions which we found to be of principle character. They reveal how our knowledge of earlier conditions has developed or they can point out new areas which have not been studied as yet. The first part deals with the situation at the beginning of the project. The rest are examples of our analyses of the effect of changes that came about.

Reflektioner 15-12

22/08/05

11:26

Side 34

Results 34

The work environment as a four-leaf clover aspects before intervention In general it can be said that bus drivers have very different opinions about their work environment – both as colleagues and at the garage level. General conditions Two out of three drivers are satisfied with their position as bus driver. But only one out of twenty thinks that the industry has a good reputation. Half of the drivers expect that the reputation will get even worse within the next two years. It turned out that expectations were strong predictors in general. Approximately three out of four drivers would recom-

Selected models of explanation - The poor reputation leads to the experience of a poor work environment. - Shift work is the cause of poor health. - Old-fashioned busses with poor driver’s seats cause the trouble. - Health is damaged by inhalation of exhaust gas via the defroster. - Lack of influence on general problems is detrimental. - It’s just a “shitty job”.

mend the job to others, but only 15% would do so without hesitation. Furthermore, 20% do not feel a commitment to their job as a driver, where 1/3 are highly committed. Organisation of the work

Are the models of explanation right?

Compared to previous studies the amount of drivers who feel pressed to comply with the

- Half of the drivers state that the job has a poor reputation and often a poor work environment. But it is new to us that this includes many conditions other than lack of influence, poor timetables etc. Special attention should be paid to managerial conditions. - Many new busses have been included in the fleet. The defrosters have been moved and shift work has been reduced to a third. The drivers’ conditions have not improved corresponding to the changes, but still they see it as a good job. More than three fourths of the drivers are satisfied with their jobs.

timetable has increased considerably. While 62% often find it difficult to comply with the timetable, 50% say that breaks and adjustment breaks are not long enough. A quite significant change in the organisation of the work is that the rotation shift schedule has been replaced by regular working hours for a majority of the drivers. Among the drivers, 57% have been through an outsourcing process. Half of these drivers say that the working conditions are worse after being outsourced, 33% say that conditions are unchanged, and the rest say that conditions are better. There is a popular demand for individually adapted types of employment, including part-time jobs. And 75% of the drivers would like to have more influence on planning the shifts. Approximately half of the drivers say they get some kind of feedback when something goes wrong. In all 29% say that their management creates insecurity among the employees, and 34% say their management does not treat employees well. Their satisfaction with the public transportation service, the unions and the local shop and safety stewards corresponds at about the same levels. Qualifications and training A total of 40% of the drivers do not find the route training and the selection of courses sufficient, and 23% do not know the job demands in detail. One out of three thinks that incorrect information on what to do when mechanical defects occur and how to report them is at the bottom of many operating problems. One fourth of the drivers say that poor qualifications explain the operating problems. Physical and psychological work environment One third say that the driver’s seat is not well designed, even though the fleet of busses is relatively new. More than half of the drivers say they are always or often exposed to trouble with impassable roads, humps in the road, poor roads all together, thermal exposure, exhaust gas, draught, vibrations and noise. One third of them drive a defective bus at least once a week. Almost nine out of ten are comfortable with working alone, although one out of four often think about the exposure to violence or threats of violence which 5% have experienced.

Selected models of explanation - Bus drivers have a high morbidity rate caused by their work as drivers.

Reflektioner 15-12

22/08/05

11:26

Side 35

Results 35

The bus drivers’ health – how poor is it? In the baseline study, 41% state that their health is excellent or very good, whereas 10% say it is not so good. That is better than in the results of a large study done 20 years previously, in which 30% said their health was not so good or extremely poor. The rate of absence due to illness reported by the drivers is relatively low, and it is lower than previous figures. A majority of the reported sick leave is characterized as long-term absence. And against all expectations, we found that a job considered to be among the ten most dangerous had one of the lowest rates of absence due to sickness. Qualitative studies indicate that possible explanations could be an aggressive policy against absenteeism, screening on recruitment of new employees, and the fact that the rate of their sick leave actually is low. It is not likely that health promotion can reduce the rate further.

Are the models of explanation right? - There is an association with working conditions here which supports previous studies. However, it is new that HealthyBus includes alternative explanations, and that is of significance when planning the optimal prevention strategy. It is not always rational only to study work environment factors. Often, it is more interesting to find out how the project can be most effectively built into an alternative model – that is a challenge. The results actually indicate that this is also about health lifestyle issues in the unskilled job sector.

Bus drivers complain of fatigue, headache, palpitation and musculoskeletal problems twice as often as the population in general. They suffer three times as often from stomach pains, respiratory problems, nervousness, and melancholy - and four times as often they suffer digestive problems. However, the conditions have improved and the drivers are generally less affected by disease than they were shown to be in previous Danish studies. If you take a look at the symptoms and diseases which the drivers suffer from it is evident that they name the same symptoms that characterise the so-called lifestyle morbidity. On one hand you can say they are very common symptoms, and on the other hand, there may be many other causes for them other than the job as a bus driver. Therefore, we think that we would do the drivers a disservice by only studying their work as a causative factor. We think that a holistic effort is better suited to improve bus driver health. There were many job-related risk factors which, at bi-variate level were associated with poor health in general, even though they were weak associations with odds ratios below 2 (excess risk > 100%). The same applies for personal situations and lifestyle. Positive stress indicators were the only significant risk factors in the job. When statistically adjusting for these conditions, only stress, overweight, negative expectations for future health, sick leave, and personal problems could be related to poor health. The same applied for changes in general health during the two years till follow-up in the HealthyBus project. It turned out that the effects of job-related factors on general health, especially in the description of health symptoms and diseases, decreased in the statistics for that period. The point is that a holistic approach reduced the significance of job factors, although the job still had some effects.

There is more to life than work We think it is an advantage for the adviser to be able to use another model to explain the causes apart from the risk factor model. If stress damages health, then less stress must be good. However, total lack of stress is also damaging, so what is the preferable stress level? Not having a social network is damaging, so a network must have a positive effect. But does that theory also apply if you do not like the people in your network? Problems at work are damaging, but if you are not challenged, then it is also a problem.

Selected models of explanation - Work environment professionals should only give advice on the work environment. - Removing sufficient risk factors from the work environment makes the employee healthier.

Reflektioner 15-12

22/08/05

11:26

Side 36

Results 36

And private problems will follow you at work. When you sit behind the steering wheel

Are the models of explanation right?

and it is stressful, you need to rest when you get home. But it is better to go for a walk

Work and private life affect each other, and that must be taken into account when promoting health. Maybe it is politically correct to stick to a scope in which our own qualifications cover the work environment only, but when the work environment concept changes and it becomes harder to delimit it from the rest of life, we think that work environment professionals will have to update their qualifications accordingly.

than sit in front of the TV. If you work in a company with no canteen, you will often buy fast food to eat. Another way of thinking is to encourage drivers to bring a lunch box from home. It might have some value in itself. Health is more than the reduction of sick leave and risk factors. We think that work environment professionals are responsible for encouraging integration and not for dividing the world into little boxes. Only 12% of the drivers surveyed say they are only 'sometimes' or 'never' satisfied with the quality of their-off duty hours. That is more than the 9% which the project registered as being unsatisfied at work. Two out three think that having more influence on their work conditions would ease their private life. One third say they do not have enough time to be with family and friends. Still, 50% report having no problems. But many have severe problems both at work and at home, and these problems may affect their health. The study shows that problems have a way of accumulating, and furthermore that drivers with problems at work also have problems at home and vice versa. The challenge for the work environment professionals is to become better at supporting holistic efforts where they are required. But they must also be better at finding out how the workplace can contribute to giving employees new qualifications which make them better at handling changes. Not only when it comes to working conditions but in all other areas as well. Such a transfer effect of the work environmental efforts is called empowerment. And there may be a larger potential to start with among the bus drivers than among other occupational groups. It is an unskilled culture, but only 40% have no education, and 55% of the skilled drivers have more than three years of occupational training/education. Why not use these resources, and see who of the drivers could be pioneers? Who could be involved in developmental work, and who can increase commitment in the fleet'?

Lifestyle – a taboo where the drivers are the losers The work environment professionals do not all agree on lifestyle being within the scope of their counselling. Everybody probably agrees that if there are work environmental conditions which lead to an unhealthy lifestyle, then it is important to eliminate these damaging exposures. But afterwards, when the exposure has been eliminated, then it is a private matter. In total, 54% of the drivers smoke, and 69% of these are heavy smokers, which is approximately twice as high as the figures for the general population. There is

Selected models of explanation:

no association between how many years the person has been a bus driver and how

- Drivers smoke because they have a stressing job - The drivers can only find energy to stop smoking if the stress level is reduced - Lack of canteen facilities makes it difficult to eat healthy food - The workload makes the drivers to tired to exercise when the come home form work

much he/she smokes. But it is evident that a bus driver does not start smoking just because of the stressful job, simply because it is impossible to obtain a driver’s licence for busses until the age of 21, and most of the drivers started smoking before that age. Just as many believe that there is no association between work and smoking, 39% of the smokers say that they would definitely take an employer’s offer for a stop-smoking course, and another 39% say they might take up the offer.

Reflektioner 15-12

22/08/05

11:26

Side 37

Results 37

Lack of exercise can be just as detrimental to health as smoking is. So it is a risk factor that

Are the models of explanation right?

at least one fifth of the bus drivers do not exercise, according to our questions about cycling,

- There does not seem to be such a strong association between stress and smoking as expected. - It is not necessary to have access to a canteen in order to eat healthy food and avoid overweight. Light exercise is more effective than a couple of hours on the couch if a driver wants to relax. The art is to find an exercise level which motivates the driver to continue exercising and fits into the daily routines. - In the HealthyBus project we found that the drivers were let down if we did not take their problems seriously enough. We also believe that it is necessary to include lifestyle, and not only where it is directly related to work, but also where qualifications obtained through work can lead to a change in lifestyle.

walking, dancing, sport and the like. At the other end of the scale, only 20% get the recommended amount of exercise, namely 3- hours per week. That is a problem when you have sedentary work. However, 25% say that they plan to do something about it. We do not believe much in fitness centres, as that approach has been tried before. The most beneficial initiative might be that the driver should get a dog, because the dog has to be walked. Of the drivers we surveyed, 43% are overweight, and further 19% suffer from extreme overweight. That is respectively 2 and 2- times higher than in the Danish population in general. There is no apparent association with their seniority as a driver. Working hours, breaks and the fact that they have no easy access to a canteen are all conditions which the drivers say influence their unhealthy eating habits. The majority of the overweight drivers are motivated to lose weight, and many say they would like to develop healthier eating habits. We do not know how many of them have tried to improve their lifestyle, but it is possible that the increased attention on the subject has made more drivers think of trying to make a change in lifestyle. But the fact is that 34% of the smokers tried to change their smoking habits during the project period, 28% tried to improve their exercise habits, and 47% tried to eat healthier food. The smokers were most successful, in that 6% stopped

Selected models of explanation

smoking and the share of daily smokers decreased by 5%.

- The driver’s seat has been improved. - Violence is an increasing problem. - The bus maintenance is not adequate. - Bus drivers are worse off than other occupational groups because of stress

A national comparison – are bus drivers worse off? How good is good enough? That was the question we had to ask ourselves when we began this project. How much must be done before we have achieved the levels targeted for bus drivers’ health and well-being? It is easy to say that they ought not to be worse

Relative excess risk (OR) Information too poor Shoulder pain Heavy smoker Low back pain Obsesity Unsatisfied with the job Qualifications not used Initiative not asked for No exercise Not involved in the job Stomach pain Experience treats and violence The job is not important Poor cooperation No possibilities for development Bad health Noise Heat Cold Draught Vibrations

46 0

1

2

4

6

8

10

12

14

Reflektioner 15-12

22/08/05

11:26

Side 38

Results 38

off than the rest of the employees in the Danish labour market. Much to our surprise we

Are the models of explanation right?

found that they had never been compared with other employees other than when regis-

- The technical standard of bus features has been raised, but it is still not good enough, and especially the lack of maintenance causes problems. - Violence and threats must be dealt with more effectively. - It was not possible to compare stress in this work with stress in the Danish Work Environment Cohort. - It should be discussed if the intensive focus on stress, influence and timetables might have displaced other important tasks. Is it because stress, work environment and cardiovascular diseases have more prestige in society and in the thinking of researchers?

ters were linked to compare diseases and mortality, which eventually became one of the purposes of HealthyBus. At the same time we were conducting HealthyBus, the Danish Work Environment Cohort was updated, so an obvious consequence was to compare the position of bus drivers with other unskilled job positions in this cohort. That was an innovation in the HealthyBus project. The figure on page 37 shows that there are 37% more heavy smokers among the drivers than among other unskilled workers, and they are exposed to noise and thermal exposures ten times as often. In general, the adjusted odds ratios are significantly higher for work-related psychosocial factors, lifestyle and several of the physical risk factors in the work environment. It is worth noticing that the risk of threats and violence is relatively high, and on the basis of a recent company visit, the Danish Work Environment Authority estimates that this factor affects the psychic work environment most of all. Beside that, many of the other conditions here may affect the stress level and the psychic work environment, for example, driving a bus under high temperatures which makes the driver tired and inattentive, problems at home, or poor health no matter whether caused by the job or other factors. Time will show us whether or not more attention will be focussed on these factors because of the HealthyBus results.

The most stressful job in the world?

The most important self-selected causes for stress

That is how the bus driver’s job has been characterised. But is it right? That was one of our questions when HealthyBus started. The reason was that it did not correspond with the participants’ experience. Of course, we found many exposures such as traffic, timetables that were impossible to keep, complaining passengers, the quality of the busses, shifts, and relations to colleagues and management. Such strain leads to stress if the driver can not handle it professionally. In this sense, a driver is better off who says “What the hell, I can't keep the timetable anyhow” than those who react in other ways.

Timetable Traffic conditions Passengers The bus Shift planning Working hours The management Job demands

Every fourth driver often or always feels that he/she is stressed at work. We measured stress by means of a psychosocial questionnaire designed by the National Institute of Occupational Health (The Copenhagen Psychosocial Questionnaire) which has been introduced as a standard tool, so it was only natural to use that reference population for a comparison to the 2,677 bus drivers in our baseline study. Our statistical analyses showed that the bus drivers were not more stressed than the reference population. Their vitality was lower but the bus drivers were not found to be at the extreme end of the scale.

Other The bus is not ready Colleagues The traffic Service Outsourcing Collective agreement Nature Stress

Recognizing this does not remove the need of finding a way to reduce stress for the drivers who suffer from it. As HealthyBus was not only directed toward the aspect of stress, we seldom focused on it directly, but often indirectly. Many interventions had stress as a positive side effect to the primary problem they concerned. Examples: A defective bus has been fixed for the next day; this involved better communication with the management, a course on handling difficult customers, certification of the drivers, interval timetables and more

The accomondations The independence Job content Physical conditions The work 0

10

20

30

40

50

60

%

Reflektioner 15-12

22/08/05

11:26

Side 39

Results 39

influence on planning shifts, holidays and days off. As a consequence, it is not possible to

Selected models of explanation

analyse specifically which interventions were most effective towards stress.

- Research has documented that a high stress level at work causes heart diseases. - The practicians say that stress is so significant that it deserves the most focus – it is one subject among many. - There is disagreement about the causes of stress.

How stress changed varied considerably among the garages. In total 12% experienced more stress, but 14% experienced a reduction of stress. Seen from a preventive point of view it is more fruitful to study what lies behind these changes than studying the fact that stress was only reduced by 2%. It is through that recognition that learning and the possibility of intervention lie, as there are often many explanations of the two directions of change. Interval timetables were introduced first as a test and then permanently on city lines. An interval timetable has no fixed times in the rush hours. We expect that it is possible to analyse if interval timetables have a stress reducing effect in the third and last follow up in 2003. By that time, this kind of timetable will have been tested by many of the drivers.

From “old Danish” to “new Danish” – a multicultural work environment One third of the bus drivers in the Greater Copenhagen area are Danes with another ethnical background (referred to as “new Danes”) and more than half of those recently employed are “new Danes”. It surprised HealthyBus that work environment researchers have not focused on that area before, especially as we found out that “new Danes” say the work environment is better than their “old Danish” colleagues say it is. Please see the illustration for more detail. Only a few factors were perceived more negatively by the “new Danes” than by “old Danes”. (See the figure on page 40, and in particular the possibilities of getting a special shift or days off. All the columns at the right show where “old Danes” have a more

Are the models of explanation right? - It looks as if the practicians’ model of explanation was the best, under the assumption that the self-rating and the questionnaire methods held water. The stress measures from The Copenhagen Psychosocial Questionnaire are popular, as they can be used to follow the effect of interventions. - The combined quantitative and qualitative results indicate that the stress factors are often smaller or local everyday incidents. Therefore, it only takes a few general interventions to reduce stress. Examples of that are a prevention of violence and the introduction of interval timetables.

negative perception of the job than the “new Danes”.) So it is natural to ask if the cultural frames of reference differ? Does the questionnaire method suit all groups or are there differences? We do not know that answer, but it should be subjected to scientific discussion. And in the meantime, work environment professionals should only use questionnaire results with caution and try to supplement with further information. From the qualitative interviews we could see that “new Danes” were more positive, and the employers stated that they were perfect employees. It is a history of success in which the stakeholders say they have experienced no problems in hiring ten times as many “new Danes” as the average figure for the rest of the Danish labour market. Still, HealthyBus found that it may be problematic in the long run that they are not represented equally at the organisational level. There was only one “new Dane” among the trade union representatives and the shop and safety stewards. They had no representatives in the managements, in the employers’ organisation or The Greater Copenhagen Authority.

Selected models of explanation - On the basis of foreign work environment studies we expected that the many “new Danes” would have found the work environment to be poor - “New Danes” isolated themselves.

Reflektioner 15-12

22/08/05

11:26

Side 40

Results 40

Excess frequency of problems related to

Are the models of explanation right? - Many of the employees are “new Danes” and they find the work environment to be better than “old Danes” do. It is correct that they isolate themselves and do not participate in the professional or organisational environment.

Not easy to get wanted rotation Unskilled Wants more influence Not easy to get a day off Can not recommend the job Unsatisfied with the job

“new Danes”

Initiative not needed

- More research in cross-cultural perception of the work environment is needed.

Other disease The driver’s seat is too poor Must make quick decisions

- It is interesting that it is not only about “new Danes” at multi-ethnic workplaces. In HealthyBus we think it is just as important to be selective when studying the work environment of “old Danes” as it is to find out what we can learn from each other. It is all about diversity.

Heavy smoker Cognitive stress Somatic stress No possibilities for development Poor management Coorporation problems Exposed to unpleasant noise Poor training Vibrations Poor reputation -100

0

100

200

300

Healthy in and sick out – a myth with modifications Many within the industry could not recognize what might easily appear as a research myth. The myth is, if you get a job as a bus driver, you have a hard job, and even though you are among the most healthy persons when employed you will end up as a disabled person, on early retirement pension, or in the worst case scenario, you die prematurely. We assumed that the work related deterioration of health happens gradually and there-

%

Selected models of explanation - Healthy people get jobs. - The high early retirement rate can be explained by the health damaging effects on the job. - The high exclusion rate is due to work-related health damage.

fore becomes worse the longer a person is employed. However, that is not the case. Is there a high health-conditional turnover? Every fourth driver had left The Greater Copenhagen Authority between the two studies. Of these, 57% changed to another job, 16% retired on some kind of pension, 13% were unemployed, 7% were on long-term sick leave, and 5% were on a job release scheme. In 41% of the cases, health conditions were contributory factors when a person left the job. The combined approach furthermore showed that many drivers quit the job after only a short period of employment. But unless the health damage occurs immediately, which is almost impossible, it is difficult to conclude anything about the extent of exclusion due to work-related health damages. It requires that we have information about the health conditions from before the driver started, and that information has not been obtained. That led us to ask whether it was only the healthiest persons who were being recruited. The argument that a compulsory medical examination before acquiring a bus driver’s

Are the models of explanation right? - We proved a negative health selection into the industry, which means that many bus drivers bring health problems with them when they get the job. As poor health is not associated with seniority and as there are competing causes of disease it is impossible to say how great the health damaging effect of the work is. We do not know if a generally high vulnerability at the time of employment increases the risk of work-related health threats. - However, the industry must get better at recruiting and retaining employees in the long as well as in the short run.

Reflektioner 15-12

22/08/05

11:26

Side 41

Results 41

license could lead to a positive selection into the job is not true, because you can easily drive a bus with treatment-controlled diabetes, you can suffer from hypertension as long as it is well controlled, and the same applies basically to musculoskeletal problems or stomach symptoms. You can even drive a bus after recovering from myocardial infarction. At baseli-

Selected models of explanation - A good work environment is lucrative for the company.

ne, one out of twelve bus drivers said they had applied for the bus driving job because of health problems in their previous work. However, in periods with economic upswing it is common that one fourth of the drivers come from the public job retraining programme. This indicates that the negative reputation of the industry can be explained by rumours that say the work environment is very poor, it is one of the ten most dangerous jobs, the industry has a selective hiring policy, and it is difficult to hold the employees. These rumours were confirmed by the studies we conducted. The high short-term turnover can be partly explained by new drivers getting the poorest shifts, and the working conditions in general; but one must be worthy of staying in the job, fit into the job and want to stay employed as a bus driver. Another explanation is that the companies do not focus on retaining their workforce. However, 75% who left the job said they were satisfied with their time in the company, and 70% would recommend the job to others.

A good work environment is lucrative for the company In a Ph.D. study we looked at outsourcing as a strategic tool and studied the question whether the work environment played an independent role commercially as well as economically. The study tried to assess whether the work environment served as an organisational concept in a strategic relation, meaning that it paid off to aim for a good work environment, as it should increase the company’s access to resources and market shares. However, there are not enough incentives for the companies to attract and retain the best employees, as the work environment has become part of competitive conditions. The strategic analysis showed that work environment has not yet been used as an organising concept in The Greater Copenhagen Authority, but it could be possible. The construction of outsourcing does not prevent it. Actually, work environment conditions have become more visible in the latest tenders, and it is possible to expand the idea further. The work environment as an organising concept for the company will probably seldom be used in the labour market of the future. The costs of a poor work environment are spread out on other areas and they do give the company a competitive advantage. Instead, one could imagine that the work environment will be actively included in other organised concepts. So the most competitive companies are those with the best work environment, as it is a condition for attracting the best employees. This type of integration has been seen before, but it can also be used in industries which are not so streamlined, including the bus industry.

Are the models of explanation right? - At a general level it is true that the funds could be allocated differently if work edified rather than shattering personal health. It is not always profitable for a company to go as far as society may wish, but neglecting to integrating the work environment within other strategic organising concepts is to neglect the potential it bears.

Reflektioner 15-12

22/08/05

11:26

Side 42

Results 42

You must know what good health is before you can promote it One of the purposes in HealthyBus was to operationalise health promotion at the workplace, and the researcher’s or the adviser’s view on health is of importance to the intervention process. The health concept behind most work environment projects focuses on reducing the incidence of disease, and seen from that perspective health promotion is all about removing the risk factors. The consequence of that perception is that the medical profession deter-

Selected models of explanation - Health is created by identifying and removing the risk factors of disease. - Health is about well-being and life quality, and it depends on how we design our society. - Health promotion is to promote health according to the premises above.

mines what is right and wrong. Thus, health is defined as absence of disease. Conditions related to society and work will only be focused upon, if they can be related to the existence of pathogenic risk factors. The humanistic perspective views health as something attached to the way a person chooses to live his/her life and the way that we collectively organise our world. This perspective brings well-being and life quality into focus. The consequence of that kind of health perception is that there are many different players and groups with adequate knowledge of what is healthy and unhealthy. From a humanistic perspective we can analyse health promotion at the workplace as a complex practice determined by a situation with many players who see the world from different angles. That lay behind the stakeholders’ formulation of the purpose of HealthyBus, and not only the two perspectives but many different health paradigms are available for the various actors here. Partly as models of explanation of how they see the world and as elements in the conflict of interests we see. When looking for possibilities of action in an intervention project, it is important to be able to hold this differentiated view of the stakeholders as a legitimate part of the power game. It is the advisers’ task to make this view explicit – also to themselves. The HealthyBus project has transformed this insight into indicative principles for practical

Are the models of explanation right?

action. This means that a health promoting intervention project must secure a dialogue

- The industry operates with both paradigms at the same time. - Health promoting initiatives must be able to respond to challenges from the most important health paradigms at the same time, if the goal is credibility, participation and establishment of efficiency.

between the prevailing, but often complementary paradigms throughout the project process. It should be a dialogue in which each stakeholder can see his or her own role, which is a big help in becoming better fellow players. The methodological challenge is to establish the best possible forum in which the search for better ways to live and work takes place within the framework of dialogues where none of the stakeholders are experts but where everybody can contribute something – not only to the dialogue but to the truth as a whole and to good health in particular.

The health promotion method makes for a better work environment It turned out that the way in which things are organised was a significant factor for creating changes in a complex field of problems. We gained that experience during the project and by studying the process. The health promotion method is in accordance with organisation-theoretical expectations, but the applicability in relation to work environment issues must be tested further to get an indication of the overall validity. The theoretical background is that changes such as these are always based on the context. Besides the will to

Selected models of explanation - It takes a common vision, will and true partnership to change complex problems in dynamic and complex everyday routines. - It is necessary to have a systematic overall view of the problem.

Reflektioner 15-12

22/08/05

11:26

Side 43

Results 43

cooperate, the workplace must adhere to a broad, overall health concept with the purpose of improving health by giving employees, groups and the company more power to act. We set up some general political rules for establishing a democratic partnership in which to operationalise the health promotion concept. On that basis we constructed the following 4-point meta-checklist: Clarifying the needs – are there large problems to be solved? •

The stakeholders must have equal influence on what questions are asked to make sure that all areas are covered.

•

When securing that, the focus falls on those who need it the most.

Prioritized action based on the needs analysis •

Interventions are based on a systematic, general analysis.

•

This allows open prioritization.

Evaluating the process •

Both the process and the result must be evaluated systematically.

•

Conclusions must be made for each intervention process.

When can we say it is a holistic initiative (this item must be adapted

Are the models of explanation right?

to the specific workplace/industry)?

- The explanations seem to fit. It is important to create a contextualised analysis and interpretation of the problems – the question in focus. That means that there often are different interpretations of the same question, depending on where it comes up. The intervention package must be developed so that it reacts as precisely as possible to the local interpretation – otherwise it is impossible to obtain optimal effects. It is a challenge to everybody, but especially to the researcher. The health promotion method turned out to be an effective tool.

•

Initiation of at least six interventions makes it holistic in this project.

•

There is at least one intervention in three out of four clover leaves.

What we did was to ensure that the stakeholders were involved in defining what we should ask and how, in order to make sure that we focused on what we thought were the most important needs. The systematic prioritization of local problems started at a conference at the National Institute of Occupational Health, where the problems were categorised during an open process. The activity plans were used as a practical process tool for systematic follow-up and termination during the evaluation and process. The results were gathered in quarterly reports and meetings. The extent of the efforts was a post hoc assessment. Half of the drivers were employed at garages categorised as health promoting garages. And as shown in the figure, the health promoting garages outperformed the others in many areas. Some areas turned out to be difficult to change with the use of traditional work environmental measures – also in areas which everybody finds are important to health in the long as well as the short run.

Reflektioner 15-12

22/08/05

11:26

Side 44

This chapter in a nutshell: – In order to be able to evaluate a complex project, it is necessary to use several methods at the same time. Evaluation is a timeconsuming, continuous multi-disciplinary process. – We present the initiatives that were a success and those that were not. We also present the future challenges. – The evaluator gives good advice to the researchers

Reflektioner 15-12

22/08/05

11:26

Side 45

Project evaluation 45

Project evaluation – to take a look at oneself in the mirror And not only so, but we glory in tribulations also knowing that tribulation worketh patience. And patience, experience, and experience, hope. Paulus, the Epistle to the Romans, chapter

In a complex study such as HealthyBus with over one hundred interventions, it can be difficult to evaluate whether one is doing the right thing. Especially, as it was not possible to conduct a stringent, controlled analysis of which interventions worked where, on whom, and how much. Let alone the suppressed objectivity criterion which occurred because we were so deeply involved in the process. In cases like this, it is a good idea to let others evaluate whether things are going satisfactorily or not. We were lucky enough to contact a British researcher who knew all about evaluation, for him to follow what we did closely. The project was also evaluated by the stakeholders’ experience of effect and by the authorities (the Danish Working Environment Authority made inspections), and in the form of articles published in peer-reviewed scientific journals. The qualitative evaluation, which will be described in the following, came into existence over a period of twelve months. We used recorded interviews with the project group, drivers, managements, shop and safety stewards, the union and employer representatives, and The Greater Copenhagen Authority. Observations were made at garages during the project work and during seminars. The material was transcribed and followed by dilemma and discourse analyses, primarily based on action research theory and what Glaser and Strauss called grounded theory. It was tested how well the method using democratic participation and establishing possession had succeeded, and whether the method had influence on initiation of health and work environment promoting initiatives. The evaluation also showed what type of qualities an adviser has to have in order to be able to offer intervention projects to companies.

Please note that this is only the mid-evaluation. First, we describe where the method was a success – at least at some garages as follows: •

HealthyBus established a participative structure which worked well.

•

We got the main stakeholders interested in a cooperation which was a long-term and regular development process.

•

We changed the participants’ perception and conduct.

•

HealthyBus made participants believe in their ability to create change.

•

The project established better internal dialogues between the drivers and the management / shop stewards.

Reflektioner 15-12

22/08/05

11:26

Side 46

Project evaluation 46

•

Focus changed from blame to common understanding of the problems.

About evaluation

•

An effective dialogue structure was established.

•

Processes of change became institutionalized.

•

The project opened up new ways of communicating.

•

The industry was well briefed, and this was a very demanding task.

- The evaluation is an ongoing process. Different pieces must be collected as circumstantial evidence of the accomplished processes and results, as consequences of the initiated and targeted activities. - In large projects it is therefore necessary to use as many sources as possible, for example internal/external scientific epidemiological analyses and qualitative methods, all sorts of statements from the stakeholders, and third stakeholder evaluations.

Naturally there were some problems: •

The project was run via intermediaries such as shop and safety stewards. Bus drivers were surveyed, but they are not yet part of the dialogue.

•

Many drivers felt that HealthyBus was not sufficiently responsive to their arguments about lifestyle.

•

The evaluator found that the project had generated much theory, and since the research results are the most important it was crucial to present them at the end of the project.

•

As a researcher, it was sometimes difficult to avoid role conflicts.

The third and last element points out future challenges: •

We need to get better at measuring the process.

•

We need more focus on changes in the competency for taking action.

•

We need to study how the drivers can gain direct access to decision-making processes.

•

We need to focus more on the marginalised drivers.

•

We need to see what can be done for the bus drivers who said things had worsened.

•

We need to develop more subtle methods for measuring change and the consequences thereof.

•

We need to get closer to the drivers' view of these processes.

The evaluator gives good advice and guidelines, primarily to researchers who work with work environment interventions in reality: •

One must respect, describe and be able to use the many different and opposite reasons for the stakeholders to participate.

•

We should tap information sources among those who are closest to the work processes.

•

There is a dilemma between giving full ownership and maintaining control over the data which show if the project has any effect.

•

The most reliable way of presenting data is to give all available information and not only the part that can be dichotomized.

•

It is essential to mix qualitative and quantitative methods.

•

It is better to describe and analyse the dilemma of the processes and choices than to have a success/failure mentality.

Are the good processes anchored and spread out? The final question is whether the good processes will stick and be a part of future company concepts. To this end, we gradually led the stakeholders to take over during the last two years of the project, in order to see if they could anchor the processes they believed in. Whether it was a success or not is the subject of the last questionnaire two years after follow-up. The most challenging in projects such as these, which do not follow a recipe, is to become better at evaluating who, what and why, and also how the evaluation is most beneficial to all stakeholders.

The project manager's meta-reflections - By mixing several scientific paradigms, one can easily fall between two stools. Based on experience gained with HealthyBus, I would try to give the different traditions more influence on the method discussions and the entire process in future. The problem is partly that the various traditions do not respect one another, and it often ends in a discussion of which science comes closer to the truth. But as concluded at a WHO health promotion conference, there is no method which is superior to the others. There is not even a basis for a hierarchical ranking. We are left high and dry when it comes to evaluating health promotion interventions, for example in the risk of role conflicts. So we have to concentrate on making constructive contributions.

Reflektioner 15-12

22/08/05

11:26

Side 47

Project evaluation 47

Reflektioner 15-12

22/08/05

11:26

Side 48

This chapter in a nutshell: This chapter deals with a synthesis of the many sub-results. We have emphasized the results by taking a general approach. We hope that other researchers, professionals etc. will consider a use of the health promotion method when working with other complex work environment problems.

Reflektioner 15-12

22/08/05

11:26

Side 49

New knowledge 49

What has HealthyBus contributed in the way of new knowledge Let us study what makes us survive instead of only what we die from. Alfred S. Evans

The approach of multi-disciplinary action research has been the distinctive feature of HealthyBus. Together with the stakeholders we have moved the field from sheer work environment explanations of the causes to a general explanation of the bus drivers’ health problems. If such a work environment project is to accomplish its purposes for health and well-being, then it is necessary to initiate several different initiatives at the same time. Proving that stress was not only caused by the timetables, but a combination of managerial conditions, the organisation of work, maintenance of the busses, traffic obstructions, passenger conflicts, relation to the colleagues, poor well-being, personal situations and health problems, it became easier to do something about the conflict issues. Decisionmaking within many areas has become visible as part of the local, common responsibility and in that way it has moved closer to bus drivers. However, this does not excuse the responsible politicians from a superior responsibility, because price, quality, work environment and the number of routes are all closely related. HealthyBus has been very successful in operationalising a health promotion model for the

Meta-model of explanation

improvement of working and living conditions. We have shown that it is possible to con-

- Implementing sufficient and effective work environment initiatives, it is possible to raise the health condition of bus drivers to the average level of other wage earners.

vert general public health knowledge into action aimed at the large majority of the affected population. The method is a supplement to other work environment initiatives: legislation, regulations, guidelines, collective agreements etc. HealthyBus not only points out a broad-spectred view on the bus drivers’ health and work environment but also a more differentiated use of data. It is not sufficient to study the net effect of an intervention only based on an action-oriented perspective. Dealing with questions about what causes change and how one can do something about it requires knowledge about those whose situation worsens in the change, those whose conditions improve, and those whose conditions stay the same. The project shows that it is only possible to gain the required knowledge about change and evaluation by combining quantitative information with qualitative information from the models of explanation. The distorted profile of recruitment into the industry is a contributory cause to health not being as optimal as it could be expected already from the beginning. If the work environment has a damaging effect, it may especially affect drivers who are vulnerable. The companies must be better at recruiting employees who really want the job.

Reflektioner 15-12

22/08/05

11:26

Side 50

New knowledge 50

They must also do what they can to maintain people in the job. However, it requires that

Are the models of explanation right?

the companies become better at handling the situations of disease and personal pro-

- No this is not right. The meta- model is an expression of a strongly idealised conception of causality and etiological fraction. Theoretically, the model is right, but as HealthyBus points out, in the real world, more effect is obtained by including supplementary etiological models of explanation. - The two significant contributions by HealthyBus have been to extend the model of explanation beyond the drivers’ negative health and to show that this holistic knowledge, in practice, can be converted to an ability to act.

blems in a manner suited to increasing the job security. Many of the traditional physical work environment problems can be either removed or reduced with better use of technology and rational administrative solutions. Fenced, airconditioned, air-dampened driver seats in combination with more knowledge of how to handle conflicts and a removal of the task of inspecting and selling tickets ought to be technically and economically realistic suggestions that can improve this work environment. Drivers are sedentary persons applying for sedentary jobs. The inappropriate lifestyle will have been established already years before they got the job, and maybe it is easy to maintain such a lifestyle in that specific type of work. One more reason may be that lifestyle is a taboo topic, but just like all other people, bus drivers must compensate for the negative lifestyle effects caused by their job. HealthyBus has shown that it is possible to reduce smoking, to go for a walk to reduce stress, and to bring a lunchbox. We have emphasized the importance of the ethnical change of culture in this industry. There are many different opinions about the working conditions and that should be utilized more actively. It is a challenge to be better at integrating the drivers in professional and social contexts – no matter what their cultural background is. The industry ought to find out if it is possible to create more value using diversity as an asset instead of passive unification. Even though we have no causal statistical evidence, there is no doubt that HealthyBus has taken a great step in the right direction of long-lasting improvement of the bus drivers’ health. Interventions that may have been more effective than ours have been pointed out before in theory, but the problem is that they were impossible to accomplish due to political, practical or economical problems. Seen from an ethical point of view we find that it is necessary to test known methods in order to make progress. It is better than doing nothing while waiting for the perfect solution. We have shown that it is possible to get the right start, but it only continues that way if the stakeholders involved are motivated. The drivers, the haulage companies, the traffic and transport companies, the politicians and the researchers must be goal-oriented if the project concerned with bus driver health is to succeed. In our opinion, the health promoting method can be used under other circumstances where severe, extensive and complex problems have existed for a long time, and where the cooperation approach is more obvious as a model of solution than traditional work environment initiatives.

Reflektioner 15-12

22/08/05

11:26

Side 51

New knowledge 51

Reflektioner 15-12

22/08/05

11:26

Side 52

This chapter in a nutshell: We formulated the theoretical considerations which were generated in this project. The most important thing is how to qualify the start-up phase and build a bridge between the humanistic and the scientific research traditions. Finally, we formulated an Intervention Complex Model which makes it possible to conduct epidemiological analyses based on a combined quantitative and qualitative interpretation.

Reflektioner 15-12

22/08/05

11:26

Side 53

From practice to theory 53

From practice to theory A complete description of an electron can not be made only by studying wave quality or particle quality. Both qualities belong to a complete description, but in principle they can not be studied simultaneously. Niels Bohr’s principle of complementarity

As mentioned in the evaluation, HealthyBus has produced many theories. In this chapter we will try to describe the meta-knowledge of the project in a theoretical frame of understanding. A detailed description of the model will be published as a scientific article. Three general theoretical challenges crystallized during the project. Below we attempt to give some answers: How to convert knowledge to action •

Even though several studies show that bus drivers suffer from severe health problems, nothing was ever done about it. Why not?

How to handle multi-disciplinary science •

There is a need to combine epidemiological and qualitative methods in a multi-paradigmatic approach in order to go into the matter, but also to make it as broad as possible during the process.

How to evaluate the effect of multiple interventions •

HealthyBus went a step further than what is normal in feasibility studies which take place in an optimized context. That made it difficult to handle the effect measures of the many user- controlled interventions which varied according to composition, strength, conformity and extent.

How to convert knowledge to action Seen from a communication science point of view, the work environment system builds on the assumption that information makes people act. The researchers must obtain knowledge and measure effects, but it is seldom that researchers play an important role in implementing knowledge. It is only other work environment professionals and politicians who act on the basis of that knowledge. However, this assumption implies that the necessary power behind the needs analysis is mobilised, and that has not been the case for the bus drivers. The partner analysis indicated that the stakeholders’ models of explanation for the increased morbidity did not relate sufficiently to work as the foundation for some kind of action.

Reflektioner 15-12

22/08/05

11:26

Side 54

From practice to theory 54

Before the stakeholders could decide if they wanted to be involved in a large intervention project, it was necessary to update the needs analysis. Besides evaluating if there still was a health problem of a magnitude worth reacting to, the needs analysis had to include an evaluation of the validity of the key stakeholders’ models of explanation. Generally, if it is not possible to establish sufficient intersection between various models of explanation there is no foundation for continuing an intervention study based on partnership. The stakeholders did believe that it was possible to find interventions that were both effective and possible to implement in practice. However, action research helped us to understand how we could operationalise our wish to move from knowledge to action: •

It is a method which deals with practice.

•

It is a way to operationalise health promotion.

•

It is an equal partnership in which everybody participates actively.

•

This is a democratic focus on changing social inequality.

•

It involves working with reflexive, dynamic processes.

•

The researcher is not only an observer but part of the process.

•

One must argue systematically if the initiatives are to work.

Action research is a suitable method to be used in large and complex intervention studies in which constructive frames for dialectical interaction between several research traditions are created. How to handle multi-disciplinary science The foundation of action research is to convert knowledge to action which then creates knowledge. It was nice to see how easy mutual learning was created between the practicians and the researchers. As in other action research projects the problem was to make the humanistic and the epidemiological paradigms work constructively together. It was not enough that they tolerated each other, they also had to inspire one another and thereby obtain synergy. What was it we wanted? Well, our questionnaire study had been tailored to the bus drivers as a result of an extensive communication-based process. We were not only given information about the stress level but also why the drivers became stressed. This gave us new knowledge about how to intervene. Information can optimize the intervention efforts as the common models of explanation become adequate when using both qualitative and quantitative methods. The same principle turned out to be useful during the analysis work where the most interesting was not the net effect but how we could explain why so many drivers felt their situation got better or worse. The point is that it is possible to group the material on the basis of an operational angle, so that the full breadth and dynamics of answers are maintained. Outliers and extreme changes are often the most informative sources, and it is important to include them when deciding which groups are best to study qualitatively in-depth in order to upgrade

Reflektioner 15-12

22/08/05

11:26

Side 55

From practice to theory 55

the models of explanation. As qualitative analyses can be very demanding, the quantitative dialectics optimize the resources, and at the same time it is possible to go into the matter to get a wider overview of the explanations. The results can then qualify the subsequent quantitative questionnaire rounds. This systematic interactive process takes place in a constant interplay between the two paradigms and between theory and practice. We have called it the Critically Reflexive Process. How to effectively evaluate multiple interventions The technical problem when measuring interventions conducted under everyday conditions is that there is no controlling frame as in causal intervention research. Causal intervention research takes place under laboratory-like conditions. As mentioned, we were surprised that so many activities were initiated, and we could not ask the stakeholders only to select a few, broad interventions which fitted our statistical methods of measurement. So we developed an analytical interpretation - the complex model - which could handle context-dependent multiple interventions and which had two advantages. It supported existing epidemiological analysis methods, and it could hold the effects of combined quantitative and qualitative knowledge. The complex model is based on the questions that one wants to change (QaMeta), for example “having to drive a defective bus at least once a week”. The question only addresses how many drive a defective bus, and not the cause or what is done about it. Our experience was that this broad meta-question could be understood differently, depending on the person. Therefore, the interpretation of cause depends on the context. But before you get that far, it is important to recognize the importance of each actor having his/her own pre-understanding of the situation. With these individual models of explanation in mind, the stakeholders, including the advisers and the researchers, were supposed to find out if they could establish a common denominator sufficiently large to act upon. The local explication of the interpretation of causes should rest on the Critically Reflexive Process to guarantee maximum knowledge and influence, so that everybody got involved in the project. Quantitative information can be combined with qualitative information, e.g. how often should the busses be repaired or taken out due to defects, and why does it happen, which type of defects etc. At the same time, theory is confronted with practice, for example, the discovery that the technical standard of the busses is in doubt, and a lack of influence on repairing the defects are both significant stressors. In practice this means that garage A may find five good explanations of the causes for defective busses CaA(1-5). . At garage B they may find one cause CaB(1). A At garage C seven causes CaC(1-7) are determined, and so on. The total contextualised complex of causes at the three garages is CaA(1-5) … CaC(1-7). Or in general, if there are X different intervention localities CaA(1-n) … CaX(1-n). What makes the model credible is that the stakeholders can find the optimal intervention package which they think will result in significant changes of QaMeta. This is why the Critically Reflexive Process becomes the most important element of the model. In relation to the complex of causes CaA(1-5) there may be two interventions in garage A which are

Reflektioner 15-12

22/08/05

11:26

Side 56

From practice to theory 56

Problem: Meta-effect variable

Q a Meta

Pre-understanding

Proposed cause: Åa (1-n)

Å a A(1-n) :

Critical

Å a X(1-n)

Intervention package: I a (1-n)

I a A(1-n) :

Critical

I a X(1-n) Adjustment Process evaluation

Problem: effect assessment

Q a Meta(A) :

Critical

Q a Meta(X)

Q a Meta(A..X)

=

F(I A(I-n).. I X(I-n) / I ()

Reflektioner 15-12

22/08/05

11:26

Side 57

From practice to theory 57

found to be feasible and believed to give the wanted result. If there are three or four solutions that can be used by the other two garages, then the total number of interventions in the three intervention complexes are: IaA(1-2) … IaC(1-3) or in general: IaA(1-n) … IaX(1-n). The world is in a constant state of change, and this means that especially long-lasting intervention processes must be adjusted during the intervention complex. Naturally, there must be some kind of documentation for the chosen intervention complex to actually be accomplished and adjusted according to its purpose. The evaluation is thus in itself a Critically Reflexive Process and not only an objective, statistical evaluation. It summarises the stakeholders’ knowledge of the entire process to document the validity if there is a significant change in the meta-variable. We started with one common meta-question QaMeta and now we have effect variables for each locality QaMeta(A)… QaMeta(X). Therefore the total effect on QaMeta is a function of the local effects: QaMeta(A..X)=f(IaA(1-n)..IaX(1-n/Ia0), where Ia0 is the effect on situations where interventions are not necessary. By defining the intervention complex as the analysis unit, it is possible to use epidemiological standard methods, because there is a relevant interpretation behind the concept of one intervention. Also, the complex model takes into consideration and accepts that the same intervention can affect several meta-variables. General education will make the manager better at handling defective busses, and maybe teach the manager to make better duty rosters and be more attentive to the employees’ holiday wishes. The complex model is a suggestion of how to solve difficult problems when having to analyse dozens of interventions in various connections which also vary in strength, extent and similarity. In optimal situations it is possible to adapt the analysis to a controlled situation, if one knows exactly who was exposed to the entire complex. The model can be used in many widespread epidemiological causal theories, and it supports common organisation and learning theories.

Reflektioner 15-12

22/08/05

11:26

Side 58

Recommended reading 58

Recommended reading It is on purpose that we do not refer specifically to the bus drivers’ work environment. The purpose of our publication is to pass on experience and inspiration to other projects with the same type of problems to be solved. We could just as well have used other occupational groups in the project. Anyhow, we would like to pass on a small extract of references which inspired us and which are probably not known among work environment professionals. Christensen S & Kreiner K. Projektledelse i løst koblede systemer. (publication in Danish). 10 pieces of heresy advising on project management in an imperfect world. 1991. Published by the Association of Danish Lawyers and Economists. Dahler-Larsen P. At fremstille kvalitative data. (publication in Danish). Odense: Odense University Press, 2002. Dahler-Larsen P. Den rituelle reflektion. (publication in Danish). Odense: Odense University Press, 1999. Draborg EU. Effektevaluering af sundhedsinformation - med empirisk analyse af et præoperativt patientinformationsprogram og en oplysningskampagne mod narkotikamisbrug. Ph.D. thesis in Danish. 1998. Odense University. Downie RS, Tannahill C & Tannahill A. Health promotion. Models and values. Oxford: Oxford University Press, 1996. Flyvbjerg B. Casestudiet som forskningsmetode. (publication in Danish). The Department of Development and Planning, Aalborg University, 1988. Greenwood DJ & Levin M. Introduction to action research. Social research for social change. SAGE Publications Inc. California, 1998 Meredith JR & Mantel SJ. Project Management A managerial approach. New York: John Wiley & Sons Inc., 1995. Toulmin S & Gustavsen B. Beyond Theory: Changing organizations through participation. John Benjamin’s Publishing Company, 1996. Yin RK. Case study research. Design and methods. 2nd ed. SAGE Publications Inc. California, 1994.

Lersø Parkallé 105 2100 København Ø

Phone: (+45) 39 16 52 00 fax: (+45) 39 16 52 01

E-mail: [email protected] www.ami.dk

Kjeld B Poulsen, Kai Drewes, Sisse Grøn, Peter T Petersen og Elsa Bach

Reflections on interventions

81 123

Experience based on more than 200 interventions initiated in order to improve the health and work environment of 3500 bus drivers

001 571 9

12

4

49 9

354989 565 12 4

001 571 9