Safety Culture - Statens Institut for Folkesundhed

Session 1: Safety Culture – does it provide ideas for prevention? Chairperson: Kirsten Jørgensen

Safety typologies – a way of comparing national safety cultures? Karin Melinder, Karlstad University We prevent accidents because the conglomerate wants to be among top 5 Mette Bang Dyhrberg, Technical University of Denmark Safety and the role of whistleblowers Sverre Røed-Larsen, AFI, Norge Improving safety culture - a research and development project in the Norwegian petroleum industry Jorunn Tharaldsen, University of Stavanger Organisational changes and accidents in a chemical industry company Tuula Räsänen, Finnish Institute of Occupational Health Seamenship – between techniques and practical wisdom Fabienne Knudsen, Research unit of Maritime Medicine, Esbjerg

SAFETY TYPOLOGIES - A WAY OF COMPARING NATIONAL SAFETY CULTURES? Karin Melinder, Karlstad University, PROBLEM UNDER STUDY Analyses have shown that countries in Europe do not only have different levels of injuries, but also different kind of injuries. According to which type of injury is dominating the countries in Europe form clusters. One example is that suicides and motor vehicle traffic accidents to a great deal complement each other as dominating injuries. (1) In welfare research different welfare-typologies have been found. Esping-Andersen (2) in the 1990:s found three: a liberal, a conservative and a social-democratic one. Also in alcohol research Europe countries are found to cluster, according to the dominant kind of alcohol consumed and whether there exist any alcohol policies. (3). In injury research there is a debate on safety culture. This is however connected to the group level and mostly encompasses working places. (4). OBJECTIVES The objective is to present tentative safety typologies similar to the ones in welfare and/or alcohol research. METHODS Fifteen Western European countries were compared for two periods, 1989-1991, and19971999. The studied variables were different socio-cultural factors; economy, religion and values related to this, regulations related to traffic safety, such as speed and alcohol limits, and fatal traffic accidents. RESULTS Despite increased similarity as to traffic regulations, the differences regarding deaths in motor vehicle road accidents remain in the countries studied. Being a Catholic country or not seems to be as important as being a wealthy country or not. Being a non-wealthy Catholic country leads to more traffic accidents than being a wealthy Catholic country. Being a wealthy Catholic country, however, does seem to lead to more traffic accidents than being a wealthy non-Catholic country. Three to five typologies are found. The most explicit typologies are Catholic, Protestant and mixed. The five typologies should be non-wealthy Catholic, wealthy Catholic, Anglo-Saxon, safe Protestant and continental Protestant. DISCUSSION The typologies have similarities to the ones developed in welfare and alcohol research. CONCLUSION In the safety typologies tentative found important factors, and basic ones, are how people in the country look upon questions related to safety. One basic factor is how people look upon questions related to safety. The safety typologies might be seen as the safety culture of the country.

REFERENCES 1.

Melinder K, Andersson R. The Impact of Structural Factors on the Injury Rate in Different European Countries. European Journal of Public Health 2001;11(3):301-8.

2.

Esping-Andersen, G. The Three Worlds of Welfare Capitalism. Polity Press 1990.

3.

Österberg E, Karlsson T. Alcohol policies in the EAS countries, 1950-2000. In Noström T. (Ed) Alcohol in Postwar Europé. Almqvist & Wiksell International, Stockholm 2002.

4.

Guldemund F.W. The nature of safety culture: a review of theory and research. Safety Science 2000 (34): 215-257.

WE PREVENT ACCIDENTS BECAUSE THE CONGLOMERATE WANTS TO BE AMONG TOP 5 Mette Bang Dyhrberg, Department of Manufacturing, Engineering and Management, Technical University of Denmark PROBLEM UNDER STUDY Accident prevention theory has developed a rather profound understanding of the internal organizational conditions for preventing accidents in enterprises, and in terms of structure, processes and culture a lot of different methods and recommendations for preventing accidents have been generated. But there has been little focus on how to get the enterprises to work out an accident preventing strategy. The question is: what are the motives for an enterprise to prevent accidents. Describing the actors who influence enterprises accident prevention and their roles in the prevention processes might indicate the reasons for and the results of the enterprises prevention strategy. OBJECTIVES The aim is to present an analysis of the process which leads to accident prevention activities in the enterprise. The analysis will be based on contextual theories and the results will be an overview of the key actors and their roles in the accident prevention process. METHODS The study encompasses literature on accident prevention, occupational health and safety management theories, and management theories. Management theories on the interaction between organisations and their context, called strategizing, have been in focus. Regular visits to the company which were starting and accomplishing an accident prevention program have been conducted, in order to analyse the key actors and their roles in the prevention process. Interviews and observations have been used to analyse the strategizing of the organisations and its influence on the accident prevention. RESULTS In the study four actors was identified having influence on the enterprises accident prevention project. The actors were (1) the conglomerate, (2) the Danish Working Environment Authority, (3) the Occupational Health Service and, (4) the safety manager. The theoretical analysis substantiates the strategizing of the different actors and their influence on the accident prevention project. DISCUSSION The contextual theories indicate that the enterprise had an interest in collaboration with the conglomerate which they did. Through the process the enterprise make a great effort in order to show the conglomerate that they are working hard to bring the accident rate down. The reason is the conglomerate had a goal: we want to be among the top five of the world when the subject is safety. This produces two demands to the enterprise; an accident rate under 0,5 accidents per 200.000 working hours and implementation of ISO 18001. The enterprise therefore focuses on fulfilling the demands from the conglomerate. This is done by coping different activities they had seen at other sites without evaluating the use and effect in their enterprise. This is called isomorphism.

CONCLUSION The conglomerate was the actor witch had the biggest influence on the project. The enterprise had an interest in collaboration with the conglomerate. This results in a strategizing based on isomorphism, they copy accident prevention activities without considering the suitability with their needs.

SAFETY – AND THE ROLE OF WHISTLEBLOWERS Sverre Røed-Larsen project manager. AFI, Norway Key words: Safety, whistle blowing, freedom of speech, obligation of loyalty, sanctions PROBLEM UNDER STUDY: As employees, many people may have observed or have even participated in acts that may be evaluated as incorrect, illegal or against the objectives of the company or public authorities. Quite a few of them choose to report or speak publicly about it. Very often, such critical voices are met with formal or informal sanctions from line managers or people in top leader positions. As a consequence, many criticisable conditions continue to exist – in silence. Regarding safety, the lack of individual courage, institutionalised procedures, protecting guidelines, leadership management etc. may set people’s life and health at risk, cause material or environmental damage, contribute to production losses, and undermine reputation etc. Risks that could and should be eliminated or reduced, continue to occur. OBJECTIVES: The research project’s main objective was to investigate the obstacles and challenges nurses in the public sector experience when reporting criticisable aspects of workplace praxis, or trying to do so. METHODS: The study used several methods, e.g. a representative survey among public sector nurses, qualitative interviews of union officials, former union officials and some leaders, a process oriented case-description, discussion of reticence and the law, and a study of the national and international research literature on whistle blowing. The research project was organised as teamwork. RESULTS: Although Norwegian laws and regulations oblige health personnel to report any condition that may set the patients life and health in danger, the survey revealed that: • 44 % of the nurses report that they do not know when they shall report, and • 47 % do not know to whom they shall report. • 50 % of the respondents have personally experienced negative reactions after having reported criticisable conditions, or they have colleagues that have had such experiences, • 19 % feared retaliations by internal remarks, and • 13 % state that limitations in the freedom of expression have had negative consequences for patients’ health. In general, lack of or weaknesses in systematic reporting procedures and routines, and an insufficient safety culture at the working place, were some of many barriers against improvement of the quality assurance or safety standard. DISCUSSION: Freedom of speech is essential to democracy and thereby protected by law. On the other hand, many employers demand an obligation of loyalty from their employees and limit them from uttering opinions regarding criticisable aspects of workplace praxis. This dilemma can hamper

the necessary protection of life and health of patients, prevent efficient safety reporting and undermine the development of an open safety culture. CONCLUSION: Statutory protection for whistle-blowers, as proposed by the Norwegian Government Spring 2005 in its proposal to the revision of The Work Environment Act, may be a necessary measure, among others, to strengthen the protection of “whistle blowers” in working life.

IMPROVING SAFETY CULTURE – A RESEARCH AND DEVELOPMENT PROJECT IN THE NORWEGIAN PETROLEUM INDUSTRY Jorunn Tharaldsen, Rogaland Research, Working life and business development, Research scientist and PhD student at the Faculty of Social Sciences, University of Stavanger, Norway. PROBLEM UNDER STUDY This project is a research and development project involving Rogaland Research and Vetco Aibel (2003-2005), a contractor company in the petroleum industry. We have used both qualitative and quantitative data in order to map the safety culture. The results have been reported back to the company and a training program has been developed. The training is being implemented in 2005 and involves 3500 employees located in Norway and their subcontractors. The aim is to strengthen the safety culture and increase the employees’ ability to solve everyday dilemmas. The biggest challenge will be to investigate how “dilemma reflection training” (DRT), can be combined with and intertwined with a broader learning process in the organization? In this project we connect the concept of safety culture to the construction of meaning, symbols and the actual formation of working communities and their collective identity (Geertz 1973, Alvesson 2002). Culture is understood as a system of common symbols and common meaning connected to these symbols. We also understand culture as in constant development, often fragmented, diversified and split into different subcultures (Frost et al 1985 and 1991, Gherardi & Nicolini 2000). OBJECTIVES The project have four main objectives: 1. Identify the attitudes, values and norms related to the safety culture in the company, 2. Develop methods to improve safety culture, and 3. Implement and evaluate the methods. METHODS During the project, different types of data have been collected. The statistical data are collected from a large survey involving all offshore personnel, funded by the Norwegian Petroleum Directorate (2001 and 2003). The survey measures subjective risk perception. The qualitative data are: Interviews, fieldwork, work shops, informal conversations, videotapes and camera. A 4 hour training program is developed in close dialogue with the company, called dilemma reflection training (DRT). The DRT is based on different cases which illustrate dilemmas concerning health, safety and environment (HSE). The cases reflect findings from our data. Improvement suggestions from the DRT will be considered implemented in the company. The DRT method is continuously being evaluated.

RESULTS AND DISCUSSION In the primary study we concluded with the following characteristics of Vetco Aibel’s safety culture: Flexibility, energy, good team spirit, justice-seeking, nomadic existence, high focus on safety, procedure-driven, and high focus on the reporting of accidents and injuries. The statistics show an improvement concerning the employees subjective risk perception from 2001 to 2003. We have no clear explanation of this improvement, but the trend is a generally higher focus on safety. So far we have carried out a pilot in Vetco Aibel and tested the DRT on the company’s subcontractors. The training in the company starts after Easter. The results from this and a preliminary evaluation will be integrated in the presentation at the conference. ----------------------------------------------------------------------------------------------------LITERATURE Alvesson M. (2002): Understanding Organizational Culture. Sage Publications. Frost et al (1985): Organizational Culture, Newbury Park, Sage. Frost et.al. (1991): Reframing Organizational Culture. Sage Publications, California. Gherardi and Niccolini (2000a): The Organizational Learning and Safety in Communities of Practice. Journal of Management Inquiry, Vol.9, No.1, March 2000, pages 7-18.

ORGANISATIONAL CHANGES AND ACCIDENTS IN A CHEMICAL INDUSTRY COMPANY Räsänen Tuula, Finnish Institute of Occupational Health INTRODUCTION Many Finnish companies have experienced remarkable changes during the last twenty years. For example the ownerships have changed, subcontracting has increased, and employment has become shorter and more insecure than before. These changes can have a direct effect to safety management strategies and to safety practices as e.g. safety instructions, risk assessment, training of employees and the safety communication. In this study the factors mentioned above have bee examined at the time period 1994-2004 in a chemical industry company. Also the changes in accident statistics have been examined at the same time period. OBJECTIVES The objective of this study is to examine how the organisational changes have influenced the occupational accidents in a chemical industry company. MATERIAL AND METHODS The accident statistics of the company during the time period 1994 - 2004 have been examined. The members of the company’s EHS group (Environment, Health and Safety) have been interviewed. Interventions and other reported factors effecting the occupational health and safety have been analysed by annual and other reports. RESULTS According to company’s own accident statistics during the time period 1994-2004 the accident frequency (accidents that have caused absence per million working hours) have varied from 11.2 (1994) to 7.5 (2004). There have also been many organisational changes in this company. The results will be examined in more detail in the paper. DISCUSSION The organisational changes during the time period 1994-2004 have been quite remarkable in the company. EHS management strategies and practices have followed these changes. All these factors can be supposed to have affected the safety level in the company. This level can be verified by the statistics. CONCLUSION The organisational changes during the time period 1994-2004 in the chemical industry company have been as follows: the ownership of the company has been changed. It has been a part of large Finnish energy concern. The amounts of employees who are working to some subcontractors have been increased. The age distribution has also changed. There are now younger workers than before. During the time period 1994-2004 the accident frequency have varied from 11.2 to 7.5.

SEAMANSHIP - BETWEEN TECHNIQUES AND PRACTICAL WISDOM Fabienne Knudsen, anthropologist. Research Unit of Maritime Medicine, Denmark PROBLEM UNDER STUDY Seafaring is a risky profession with a high rate of accidents. Efforts to change this state have recently leaded to a proliferation of measures such as workplace assessments and checklists. Unfortunately, the demand for written procedures is perceived by many seafarers as counteracting the use of common sense, experience and professionalism epitomized in the concept of seamanship, and even as counteracting safety. One of the reasons to take the practitioners' resistance seriously is that a decrease in the number of occupational accidents and deaths at sea has not yet been observed. OBJECTIVES To understand and analyse the reasons for the resistance to written procedures and to qualify a discussion on what can be done to promote a situation where written procedures are perceived as complementary, and not alternative, to common sense. METHOD Two already published ethnographic studies on compliance with safety rules and on cooperation on board Danish vessels provide the empirical material of this presentation. Some key statements on the conflicting relation between seamanship and requirements of written procedures are analysed in the light of theories of knowledge, mainly Dreyfus and Dreyfus' novice-expert model, and Aristotle's distinction between techne and phronesis (practical wisdom). RESULTS Dreyfus and Dreyfus' model of skill acquisition explain why seafarers perceive systematic rule-following as a regression. While novices steadily follow context-independent rules, context and experience play a growing role during the acquisition of expertise. Contextindependency is necessary at the beginning, but with growing experience, rules become superfluous and even impeding. In contrast to the novice, the expert's behaviour goes beyond analytical rationality, and is situational, experience-based, and intuitive. The expert's qualities conform precisely to the concept of 'seamanship' as delineated by the respondents. However, the expert may become overconfident and blind to new solutions. The Aristotelian concept of phronesis brings us further by adding a reflexive, a social and an ethical dimension to the knowledge implied in the expert model. Phronesis is, like expertise, experience based knowledge which cannot be abstracted from the context where it is practiced. But it is also value- and action- oriented. It combines the capacity to recognize what ought to be done with the ability to do it. While phronesis rests on value rationality, 'techne' (know how’), another Aristotelian concept, is based on instrumental rationality and is production-oriented. Phronesis is, inter alia, choosing the proper techne for the circumstances. However, the seamen fear a development where techne predominates at the expanse of phronesis.

DISCUSSION The case raises several issues on which conditions and contexts are appropriate for rules and procedures to be perceived as a tool serving the practitioners' safety. The practitioners' arguments against written procedures are worth an analysis. They cannot be altogether rejected as mere reluctance to change; some of them stem from a genuine concern for safety which has to be taken seriously.

Session 2: Intervention - what does and what does not work? Chairperson: Sverre Røed-Larsen

Finnish Zero Accident Forum Jorma Saari, Finnish Institute of Occupational Health Vision zero – possibilities and obstacles for intervention Beate Elvebakk, Institute of Transport Economics, Norway The forgotten role of risk management in urban planning Anna Johansson, Karlstad University Fatal accidents among Danish children between 1975-2000 Astrid Gisèle Veloso, National Institute of Public Health, Denmark Railway suicide prevention; development of a checklist in support of event investigations Helena Rådbo, Karlstad University Does anti-skid devices and studded footwear increase safety during slippery outdoor activities? (Pilot study) Pauliina Juntunen, Finnish Institute of Occupational Health

FINNISH ZERO ACCIDENT FORUM Jorma Saari, Finnish Institute of Occupational Health OBJECTIVES OF THE STUDY The Finnish Government launched a national occupational accident prevention program in 2001. The primary purpose of the program was to motivate work places for better safety performance. Two main needs were identified. (a) Many Finns believe that accidents are caused by chance. Therefore, zero accident vision became an essential part of the national program. (b) There should be more work places that support employees’ safety during leisure time. METHODS A way for motivating better safety is to have more success stories. Therefore, we established a Zero Accident Forum for work places that are committed to improving there safety. These are work places, which have a high level of ambition in safety. The forum was introduced in the early 2004. In March 2005, the number of member organizations is about 105, and the number of employees in these organizations was around 100 000. The working model of the Forum is to give an arena for exchanging ideas and experiences. The forum brings together people with similar goals. Therefore, the forum is a source of social capital for its members. During the first year of existence, the forum has been seeking for forms of functioning. Joint workshops have been quite successful. Also an extranet has started positively. There is in the extranet a discussion area for members and it has activated quite nicely. Members are also sending short stories about their activities for being published in a newsletter, which is sent not only to members but also anyone who is interested. Since the Forum has existed only one year, it is too early to say anything about its effects on accidents. Clearly, it has activated the discussion about challenges in safety and we are hopeful about seeing new success stories in the coming years.

VISION ZERO - POSSIBILITIES AND OBSTACLES FOR INTERVENTION Beate Elvebakk * & Trygve Steiro ** * Institute of Transport Economics ** SINTEF Technology and Society PROBLEM Norway has an overall vision of zero killed and permanently harmed (Vision Zero) within all branches of the transport system. A vision can be defined as “power of seeing” and the “ability to view a subject imaginatively” (Oxford Advanced Learner’s Dictionary). The relative indeterminacy of the concept means that it is open to interpretation for those whose task it is to transform it into practical interventions. In this study we examine how the Vision Zero is implemented in the Norwegian road sector. The road sector is the most open sector in the transport system. It is also the transport mode with the highest number of fatalities, permanent injuries etc, and therefore the area where Vision Zero is supposed to have the greatest impact. OBJECTIVES OF THE STUDY The overall objectives of our study are to examine how Vision Zero is implemented in the road sector. What changes and interventions have ensued from the adoption of Vision Zero? We also examine the main obstacles for working towards Vision Zero both within the National Road Administration and for society as a whole. How is Vision Zero understood and interpreted by other relevant stakeholders in society? METHODS Our empirical approach is 30 in depth interviews with different stakeholders. We also use data from interviews with the committee for transportation within the Norwegian Parliament. RESULTS The focus has shifted from number of accidents to severity of accidents. Strategy documents include traffic safety better than they did before. It has become easier to implement new interventions since more resources are allocated within the National Road Administration to traffic safety. One example is the allocation of resources to perform traffic safety audits at the end of projects. All informants claim that Vision Zero has had positive effect, albeit to a varying degree. But almost all actors seem to agree that the arenas for discussions have been few and the debate has been very limited. DISCUSSIONS In our study we observe that public documents mostly ignore dilemmas associated with the realization of the vision, such as conflicting goals. The interviews indicate that the National Road Administration is more aware of the dilemmas involved with the translation of the vision into practical efforts. A lot of the improvements have come from implementing physical barriers and shifting prioritisation within the National Road Administration. This has been necessary to have a platform for further work. At the same time we see that the public debate about Vision Zero

has been very limited, leading to different interpretations. One important consequence of this is that it is up to the stakeholders to define the agenda. CONCLUSION For the Vision Zero to become reality, more people need to be involved in its process of implementation. We also argue that in order to maintain the good work and intentions, the vision needs to be communicated to a wider audience. Otherwise conflicting interpretations will be made and the outcome may become fragmented and even lead to resistance to change.

THE FORGOTTEN ROLE OF RISK MANAGEMENT IN URBAN PLANNING - Time to rediscover the built environment’s impact on public health and safety Anna Johansson, Doctoral Candidate, BSc, BA. Prof. Inge Svedung and Prof. Ragnar Andersson Div. Public Health Sciences, Depart. Social Sciences, Karlstad University, Sweden

PROBLEM UNDER STUDY The link between public health and the built environment was established already in the early 19th century (if not before), which strongly contributed to development of an organised planning of our societies and urban areas. Growing attention is (once again) drawn to the built environment’s impact on health and safety with reference to a broad range of health and safety issues such as physical inactivity, injury, crime and socially related illness - all indicate the same thing, the built environment does matter. OBJECTIVES A broad research program has been initiated to explore the potential of municipal planning as an instrument for community safety and security improvements and to discuss how and why a more public health oriented perspective could be relevant. In this presentation we will analyse management of different risks in Swedish municipal planning and discuss the relevance of approaching these issues in a more public health oriented way. METHODS This paper draws its conclusions from three separate studies; 1) a literature review, 2) a qualitative examination of the fifty municipality comprehensive (master, structure) plan documents adopted by Swedish municipalities and 3) a survey within these municipalities and its chief officials, regarding daily work with the management of different risks. RESULTS The literature study indicates that the built environment has broad impact on humans health and safety. Findings from the plan documents study shows that risks considered in municipal planning are biased towards disastrous hazards and transport-related safety issues, while every-day injury risks in public environments attract less attention. Our preliminary findings from the chief officials survey indicate that the risk perspectives within a municipality are extensive, but explicit for each sector. DISCUSSION There are reasons to believe that a widening to a more public health oriented focus could invite to a more health and safety friendly design and planning of our built environments. Although a widening of scope maybe problematic as findings indicates that the institutional link to public health officials has largely disappeared from other municipal sectors e.g. from urban and land use-planning practices.

CONCLUSION Our preliminary findings indicate the need for local actors from several sectors to closer cooperate around risk issues regarding a broad range of health, safety and security that are to be found within the municipal arena. To promote a more public health friendly risk perspective, a research agenda that provides a deeper understanding of what will be required of the local arenas, actors, structures, methods and tools involved, seem necessary.

FATAL ACCIDENTS AMONG DANISH CHILDREN BETWEEN 1975-2000 - with special reference to fatal home and leisure accidents Astrid Gisèle Veloso stud. scient. san. publ., Research Programme Director, Researcher, MD Birthe Frimodt-Møller & Senior Researcher Bjarne Laursen Centre for Injury Research, National Institute of Public Health, Denmark. PROBLEM In the preventive work concerning fatal home and leisure accidents in Denmark one particular misses information about the accident e.g. detailed place and products that were involved in the course of accident. OBJECTIVE The aim of the present study was to obtain detailed additional information about external conditions in relation to fatal accidents besides that information given by WHO´s ICD-10 codes. METHODS The study is based on information obtained from The Danish Causes of Death Registry and includes all fatal accidents among children in the age group 0-14 years between 1975-2000 with that exception that within the period 1975-1993 fatal transport accidents were excluded. The original death certificates from 1975 to February 1997 were found on microfilm and the remaining ones on electronic files and then copied. The death certificates were perused (especially the descriptive text) and subsequently coded using the NOMESKO classification. The obtained data included ICD-10 codes, year of accident, sex, age, municipality address, place of accident, mechanism of injury and activity at the accident moment, type of sport if any and all the products that have been involved in the course of accident. Furthermore we have worked out a short résumé of each course of accident. RESULTS According to The Danish Causes of Death Registry there were 2,349 fatal accidents among children in the age group 0-14 years between 1975-2000. Of these fatal accidents, 1,101 fulfilled the criteria mentioned above (primarily home and leisure accidents). We were able to find death certificates for 1,062 accidents that subsequently were included in the study. So far our findings show that the number of death has decreased considerably during this period. The decrease in the incidence rate is mainly seen during the first half of the period and there might be a tendency showing that during recent years there has been a stagnation in bringing the number of death due to fatal accidents further down. The distribution shows further more that the most frequent causes of death next to traffic accident are suffocation and secondly death by poisoning. DISCUSSION AND CONCLUSION The first impression of the results indicates that in regard to obtain additional information on fatal accidents the largest gain was achieved in a more detailed coding of place of accident, coding of involved products and in the small description of the accident since every accident is unique.

RAILWAY SUICIDE PREVENTION; DEVELOPMENT OF A CHECKLIST IN SUPPORT OF EVENT INVESTIGATIONS Helena Rådbo, Inge Svedung, Ragnar Andersson Karlstad University,Dept. Social Sciences,Div. Public Health Sciences PROBLEM UNDER STUDY AND OBJECTIVES Suicide constitutes a major public health problem. In Sweden 1200-1500 persons commit suicide annually. About 5 % of them position them self in front of a running train on the Swedish railways. Of all deaths in man-train collisions, both intentional and unintentional, suicides constitute the major part, 75%. At Karlstad University research into suicide prevention is performed on commission by The Swedish National Rail Administration: Banverket. Preventive measures to be applied in the railway system are looked for and analysed. The idea is to influence the suicidal person and make the tracks unattractive as a mean to commit suicide. METHODS The work started by collecting and analysing data from suicides committed in the Swedish railroad system within the time period 2000-2002, to describe the pattern of when, where, and how, the collisions occurred. The main results are that railway suicides are mainly committed in densely populated areas and that the victims normally dwell on or close to the tracks for a wile, before the train arrive. Based on a general knowledge of the rail traffic system, a generic suicidal model and the findings from the epidemiological investigation a theoretical analysis of the conceivable preventive measures was performed. These measures were discussed in terms of functionality (efficiency and vulnerability) and transformed into a checklist to be used as support of data collection and analysis of future suicides in the railway system. The theoretical approach, the generic preventive measures and the preliminary checklist will be discussed.

DOES ANTI-SKID DEVICES AND STUDDED FOOTWEAR INCREASE SAFETY DURING SLIPPERY OUTDOOR ACTIVITIES? (PILOT STUDY) Grönqvist Raoul, Juntunen Pauliina Mattila Susanna, Aschan Carita, Hirvonen Mikko; Finnish Institute of Occupational Health, Departments of Occupational Safety and Physics INTRODUCTION In Scandinavian countries the ground is covered with ice and snow for 4 to 6 months a year. Slip-related falls are increasing every year because of changes in climate and weather, and due to insufficient maintenance of roads and pavements. The most frequent injuries that have occurred due to falls are diagnosed as sprains, contusions and fractures. Health care costs of femur fracture treatments are around 15 000 Euros per individual. The risk of injury due to falls is increasing significantly from the age of 45 for both women and men. The fear of falling also increases with age and as a consequence the willingness to participate in outdoor activities as part of daily basis among the elderly. Safety while moving should especially be taken in consideration among persons working outdoors in professions such as mail deliverers, construction workers, excavation workers, etc. The aim of this study was to investigate safety and usability features of different anti-skid devices and studded footwear gained from the users’ experiences. In Finland anti-skid devices are classified as personal protectors whereas studded footwear is not. The biggest difference between anti-skid devices and studded footwear is that the anti-skid devices can be removed when going indoors, which may increase safety. METHODS The seven Cobbler’s Shops (Taitavat Suutarit ry) in seven towns around Finland recruited 95 voluntary participants; 61 females and 34 males. A group of 73 participants received 7 different trademarks of snit-skid devices, while 22 participants received permanent studding in their own footwear. The participants made their own choice of a suitable anti-skid device. The studs for use in their own footwear were chosen and mounted by the cobbler. The participants are free to choose how often and in what weather conditions they wish to wear the anti-skid devices or the studded footwear. Two follow-up periods each lasting 45 days (15.1 – 28.2.2005 and 1.3 – 14.4.2005). All reported slip and fall injury accidents and “near accidents” will be investigated at the end of each period by using phone interviews and questionnaires. Some studded footwear will be observed and/or tested in the laboratory using standardized physical tests and evaluations to find out the footwear’s suitability and quality for studding, wear effects on the studs ad the footwear soles with relation to time of usage. Similar tests are also applied to evaluate the performance of different anti-skid devices. RESULTS Questionnaires and phone interviews are used to gather all requested information such as the following: age, gender, self-evaluated health status and diseases which may cause poor balance, type of ground used, performed work, tasks and hobbies, usability indoors on hard

surfaces, walking mode and safety, ability to control balance, benefits and disadvantages of using the aids. The results will be report when the second follow-up period is finished. The final paper will present the results and their implications for safety and usability.

Session 3: Communicating Safety Chairperson: Jorma Saari

Risk perception in preventive safety work Siri Wiig, University of Stavanger Cultivating Safe Exploration Practices Espen Olsen, Rogaland Research, Working life and business development, Norway An Intervention method for increased safety activity Christina Stave, the National institute for Working Life West, Sweden Improving Risk Communication on the Human Machine Interface in high risk industries Stefán Einarsson, Technical University of Iceland Child injury reduction following “Ekspedition skadefri” Bjarne Laursen, National Institute of Public Health, Denmark Respect for speed Preben Hoffmann Rosenberg, Vejle Amt, Denmark

RISK PERCEPTION IN PREVENTIVE SAFETY WORK. A COMPARATIVE STUDY WITHIN HEALTH CARE INSTITUTIONS AND MUNICIPALITIES Siri Wiig, University of Stavanger PROBLEM UNDER STUDY This study compares risk perception among employees within health care institutions and within municipalities’ safety related work. It applies risk perception and interaction between regulatory authorities and the regulated as an approach to study safety work within health care and municipalities. Research on risk perception (Hovden 1979; Slovic 2000; Reventlow, Hvas et al. 2001; Renn 2004) has revealed different factors that increase or decrease peoples’ perception of danger and what they conceive as risks to life and health. The research question is: How does risk perception among employees have an impact on preventive safety work within their organizations? OBJECTIVES There exist perceptions and understandings of risk among employees within different arenas such as regulatory authorities, municipalities, hospitals and hospital clinics. Risk perception within these different arenas is not necessarily corresponding and the study explores important elements in forming employees’ risk perception within health care and municipalities. Further, the study explores how belonging regulatory authorities affect employees’ risk perception and compares whether their risk perception has an impact on health care institutions’ and municipalities’ preventive safety work. METHODS The study is based on comparison of two multiple case studies within health care and municipalities. Data is collected through a triangulation of qualitative interviews, observation and document analysis. RESULTS Municipalities that have experienced unexpected events or accidents direct more attention to their preventive safety work, independent of the interaction with the regulatory authority. Municipalities located in geographical areas with natural hazards present also put more effort in their preventive safety work. Municipalities without similar experience or localization were more dependent on the interaction with the regulatory authority to form their risk perception and carry out preventive safety work. Data collection in the health care institution is still ongoing. Preliminary results show that hospital clinics experience a higher number of medical error and fatalities each year, than most municipalities. Errors and mistakes are more common and a part of the daily hospital work, which means that hospital employees face situations with a potential for gaining experience with handling risks. Because of a familiarity with risk situations hospital employees perceive them as known and controllable, but still involving a high risk level.

DISCUSSION AND CONCLUSION The catastrophe potential is different within health care and municipalities. Catastrophes may happen in a municipality and the consequences can cause death for a large amount of people, while medical errors usually have consequences for only one person at a time. The number of undesired events is higher in health care and can lead to increased attention and risk perception and to the ability to handle risks. That is similar for municipalities, which have to manage and prevent present natural hazards. Regulatory authorities within health care are in possession of stronger sanctions than within municipalities, which implies a stronger impact on risk perception. To compensate for lack of sanctions, focus on dialogue and guidance between municipalities and regulatory authorities results in an increased focus on preventive safety work within municipalities. REFERENCES Hovden, J. (1979). Vurdering av ulykkesrisiko, Tapir. Renn, O. (2004). "Perception of risks." The Geneva Papers on Risk and Insurance 29(1): 102-114. Reventlow, S., A. C. Hvas, et al. (2001). ""In really great danger...". The concept of risk in general practice." Scandenavian Journal of Primary Health Care 19: 71-75. Slovic, P. (2000). The Perception of Risk, EARTHSCAN.

CULTIVATING SAFE EXPLORATION PRACTICES Idar A. Johannessen, Senior Research Scientist & Espen Olsen, Research Scientist , Rogaland Research, Working life and business development, Norway

PROBLEM UNDER STUDY In recent years we have seen an increased awareness of the impact of human resources on risk in the work place. Human resources can constitute important barriers for danger in drilling projects. This study seeks to identify, describe, refine, store and effectively share a wide range of wise practices that can improve safety in exploration drilling projects. Some of these practices are human factors in themselves, such as a Drilling Supervisor taking initiative to check the competence of all new arrivals on the rig. Each individual choice involved in safe exploration practices we have called a “safety lever”, and we have identified over 200 such levers in our material. Examples range from making policy decisions on safety to day-to-day practices as mentioned above. OBJECTIVES Firstly, we describe how a database of smart practices was developed (the technical solution). Secondly, we describe how utilization of the information means making use of the database in the leading and managing of the projects (the human side). METHODS We interviewed 25 informants who helped us make explicit some of their knowledge of how they can influence safe work practices, given their position in the system. Research interviews gave descriptive and important information of possible practices which then were developed further in a series of meeting with the people involved. RESULTS, DISCUSSION AND CONCLUSION In parallel to developing the descriptions further, the documentation has been transformed to a prototype database. This creates a tidy display of the information. It is possible to search and group by different criteria and users may pick parts of the picture that are most relevant to them. For example, people in adjacent (or interdependent) positions in the project organization can single out their best practice descriptions and quickly discover overlaps, critical times when they need to communicate, etc. We are now beginning experiments to discover how the tool can best be exploited. A pilot workshop will run in the near future as part of team building in a new well project. Current ideas include 1) training people to become familiar with the safety challenges of a position which is new to them, 2) making people in adjacent and interdependent positions aware of each others’ thinking about safe practices 3) sorting out interface problems 4) increasing the scope of safe practice ideas by comparing participants’ own views with those in the database 5) feeding back new and improved ideas to the database. Our vision is to create 1) a tool that makes safe practices accessible and easy to relate to and 2) to design learning settings that facilitate wise decision that have a real impact on safety.

AN INTERVENTION METHOD FOR INCREASED SAFETY ACTIVITY Christina Stave and Marianne Törner The National Institute for Working Life West, Sweden PROBLEM The frequency of occupational accidents in Swedish farming is high. In order to stimulate increased safety activity, new strategies and methods should be developed. Therefore, a method aimed at influencing safety attitudes, cognitions and behaviour was tested and evaluated. Objectives: The aim of the present study was 1) to evaluate effects on risk perception, perceived manageability of risks, risk acceptance, work stress and activity in safety of three different intervention approaches with different levels of structure, 2) to evaluate the intervention process in the three different intervention approaches, describing and analysing the process and evaluating feasibility from the participants and process leaders’ perspective. METHODS The intervention was based on group discussions with three different degrees of structure. 88 farmer and farm workers participated. Nine groups were created. Three process leaders were assigned three groups each. The groups were randomly divided into three different degrees of intervention structure. The first level was to create social-supportive networks implying participatory and process-oriented strategies encouraging discussions and reflection, focusing on risk manageability. The second level added structured analysis of incidents/accidents, and the third level added information on risks and their possible consequences. Each intervention group gathered on seven occasions during one year. A pre-post questionnaire was used to evaluate the effects and questionnaires, interviews and memos were used in order to evaluate the process. RESULTS AND CONCLUSIONS: Safety activity increased significantly in the entire intervention group. Work stress as well as risk acceptance were reduced, while risk perception and risk manageability had not changed after the intervention. The results showed positive effects of the basic concept used, e.g. creating networks for social support, facilitating discussions and reflections with focus on risk manageability. Those who additionally used an incident diary, and thus a structured way of analysing accident events, showed a somewhat more positive outcome. Providing risk information showed no additional positive effect. The practical arrangements functioned well and attendance was high. Only four persons left the intervention. Focus on safety was high and an increased subjectively stated safety activity was noted at the end of the intervention. The group process showed normal signs of evolution, which indicated that time is needed in order to accomplish a behavioural change. The supportive network with an honest and non-judging climate seemed to have been the most important and appreciated part of the intervention. The process and evaluations were rather similar in all of the three group types. Despite the infrequent use of the incident diaries, the structured way to analyse incidents may have influenced insight in that incidents/accident may be prevented.

IMPROVING RISK COMMUNICATION ON THE HUMAN MACHINE INTERFACE IN HIGH RISK INDUSTRIES - AN EFFORT TO INCREASE HUMAN RELIABILITY AND ORGANISATIONAL RESILIENCE Dr. Stefán Einarsson, Dr. Bjarki A. Brynjarsson Technical University of Iceland PROBLEM UNDER STUDY: Human organisations, such as private companies, state owned institutions, or service enterprises experience the reality of vulnerability each year. Vulnerability may appear as an unfortunate consequence of a faulty decision that cannot be made redundant, as unwisely chosen business strategy or as a major accident followed by damage of property and casualties. In many cases managers are to blame, since they have not been able to deal properly with the sometimes contradicting goals of minimising risk and optimising financial returns. In other cases the organisational structure is the critical factor. We are inclined to believe this has often little to do with direct operator reliability, but more to do with interactions within the organisation that are not yet fully understood. OBJECTIVES: The scope of the research is to analyse the problem of human error in a holistic way, with the aim of improving safety at the human-machine interfaces and within the organisation as a whole. Serving as a model project, the most recent Scandinavian project on risk and uncertainty (www.risikoforsk.no), has been considered. In this project, important concepts like resilience, barriers and information processing pertaining to the environment in high risk workstations in industry are discussed. The main objectives are to come up with a realistic and practical model of the humanmachine interface which includes the mode of risk communication within the company and risk communication over the company’s interfaces with governmental bodies and municipalities. METHODS: This research is the first stage in a multidisciplinary project, which will draw upon resources from safety researchers from different disciplines like psychology and human factors and companies involved in high risk operation (e.g. a SEVESO company) and preferably also involve insurance companies. At later stages an intervention study is appropriate to validate results. The first stage is a theoretical study considering some important near accidents and major accidents, which may serve well to elaborate and illuminate the main issues of focus. An important basis for the study is the vulnerability concept and a new approach to vulnerability analysis as proposed by the authors. Further important foundations are the fundamental work of Reason, cognitive psychoanalysis and behaviour therapy.

RESULTS AND DISCUSSION: Different risk management strategies are used to deal with human error depending on the life system and the context wherein it takes place. Behaviour needs to be modified, models of risk need to be improved, and risk perception needs to be elaborated. Within high risk organisations the organisational resilience must be stabilised and enhanced. The impact of emotions on operator reliability and thereby organisational resilience and vulnerability needs to be analysed. Finally the safety culture needs to be analysed, so that innovations can be successfully integrated. CONCLUSIONS: First impressions of accident analysis, vulnerability and human reliability assessment show that a new modified method to assess human reliability is well worth the effort. Such a method could serve several high risk industries as a unified theoretical and practical approach for human error. Keywords: Human reliability, vulnerability, organisational accident, resilience, line of authority, vertical risk communication, horizontal risk communication.

CHILD INJURY REDUCTION FOLLOWING ”EKSPEDITION SKADEFRI” Bjarne Laursen1 and Stein Nygaard2 1

National Institute of Public Health, Denmark, and Prevention unit, Frederiksborg County, Denmark

2

PROBLEM UNDER STUDY: In the year 2002, Frederiksborg County accomplished a child safety intervention called “ekspedition skadefri” involving 2000+ schoolchildren at the age about 10 years. The intervention included one day in a military training area, where the children should accomplish several physical challenges. Further, nurses, doctors and other professionals informed them about injury prevention and treatment. Additionally, information materials were distributed in the school classes, and used in the teaching before and after the training day. OBJECTIVES The purpose of the present study was to investigate whether there was a reduction in emergency department contacts related to child injuries, following the intervention. METHODS Since personal identification numbers were not provided, the analysis was based on comparing the emergency department visits in the specific age group, compared to other counties in Denmark. ED visits for the children were followed two years before and two years after the intervention, based on data obtained from the National Patient registry in Denmark. The registrations were grouped based on diagnosis, injury mechanism, and activity at the time of the accident. RESULTS AND DISCUSSION Compared to the other counties in Denmark, there was no general reduction in the number of ED visits following the intervention. However, for specific injury types (acute overexertion resulting in dislocations, sprains and strains) there was a significant short and long-term reduction compared to other counties (p