Injury Prevention and Safety Promotion

Lund, Johan; Di Giannantonio, Paolo; Mannocci, Alice

doi:10.1007/978-3-319-13620-2_10

Johan Lund⁴,
Paolo Di Giannantonio⁵ &
Alice Mannocci⁶

1874 Accesses

Abstract

Injuries (outcomes of accidents, assaults, and intentional self-harm) place heavy burdens on societies worldwide in terms of human suffering, health expenses, and compensation costs. This burden is predicted to increase.

Across the European countries, there is a great variation in injury mortality rates, from less than 25 per 100,000 in the western part to more than 180 in the eastern part. In EU-27, each year more than 230,000 persons will be killed, about 1 million will be permanently impaired, about 6 million will be admitted to hospitals, and about 34 million will be treated as outpatients.

Dangers are the origins of accidents and injuries. To avoid them, dangers should be removed or modified/diminished, or people should be trained to master them. Injury surveillance is a tool for obtaining knowledge of the epidemiology of the injuries. A two-step surveillance system is recommended. The first step should be the collection of data: (1) for estimation of injury incidence rates and (2) for establishing trends. The second step should be the collection of more detailed data for understanding factors contributing to the accidents and injuries. Accident prevention measures can be divided into three main categories: modification of (1) attitudes, (2) behaviours, and (3) structural conditions. A fourth category occurs when measures from two or all three categories are utilized (orchestration). It is shown through literature studies that interventions from categories 3 and 4 have more significant positive results than interventions from category 1 and 2.

Many experiences on best practices and evidence-based strategies on prevention of accidental injuries are referred to, and divided into traffic, occupational, home, sports, child, and elderly accidents and injuries.

Access provided by Autonomous University of Puebla. Download chapter PDF

Preventing injuries must be a priority to prevent disease in the twenty-first century

Article 29 August 2019

Injury Prevention and Safety

Injury Prevention

Keywords

Introduction

Injuries (outcomes of accidents, assaults, and intentional self-harm) place heavy burdens on societies worldwide in terms of human suffering, health expenses, and compensation costs. In 1990, injuries were responsible for 10 % of world mortality, and this rate is predicted to increase to 12 % by 2020 [1]. It is also estimated that injuries account for 14 % of global life years lost when using the measure years of life lost [2].

There is an increasing global effort to control the injury problem and to establish prevention strategies [3]. A prerequisite for planning, implementation, and evaluation of injury prevention strategies is an understanding of the epidemiology of injuries , i.e. the occurrence of injuries in terms of time, place, person, and factors contributing to injuries in the population [4]. Here, we focus on the epidemiology and prevention of accidental injuries (also called unintentional injuries), as outcomes of traffic, occupational, home, and leisure accidents. Assault (interpersonal violence) and intentional self-harm (self-inflicted violence) are viewed as separate phenomena in terms of the preventive strategies that are employed. As aspects of the general injury problem, however, they are included in discussions of treatment and registration in the medical sector.

Definitions and Concepts

Some important concepts in accident analysis and prevention and a model of the relations between them are given in Fig. 10.1 [5].

Dangers Are the Origins of Accidents and Injuries

Accidents might happen wherever humans are present, and at all hours. Each day, we are exposed to, or expose ourselves to, dangers: external sources of energy, or something that causes or is likely to cause harm [6]. The dangers are mostly “normal” and are coped with in the daily life (Fig. 10.1). However, when a person loses control over the situation, or observes a danger too late, an accident occurs (Fig. 10.1). An accident is defined as an unintentional event, characterised by the sudden release of an external force or impact which can manifest itself as a body injury [7], if the threshold of the human tolerance is exceeded [6, 8]. Humans with low mastering abilities, e.g. children and the elderly, have more accidents than humans with high mastering abilities.

There are two main strategies for avoiding accidents and injuries:

1.
Remove or modify/diminish the dangers, mostly connected to injury prevention : environmental changes, hindrances, lack of freedom, rules, and regulations. Examples are: drain swimming pools of season, do not permit elderly people to use stairs, enforce speed limits on roads, make bathtubs less slippery, place guards on dangerous machines, enforce safety regulation on playground equipment, etc. Availability and use of safety equipment might influence the outcome of the accident (Fig. 10.1).
2.
Master the dangers, mostly connected to safety promotion . Examples are: train to master dangers (very relevant in child development), traffic training (bicycling, etc.), but only when the child is able/ripe to fulfil the requirements, physical and mental training, exercises in all ages, tai chi, dancing for the elderly, etc.

To master dangers is an essential part of building self-esteem. Not all dangers should be eliminated. The child needs risk-taking, adventures, and challenges. It is essential to find the balance between development and protection. If a child is falling from a high climbing frame, she/he should fall on a soft surface. When voluntary risks are coped with, the risk-taker might be rewarded with joy, happiness, and realisation of their potentialities. The downside of risk-taking, however, is injury, death, and tragedy. This ambiguity in risk-taking is a challenge to planning and implementing injury prevention strategies [9].

Injury Prevention

A model for accident prevention is proposed, the two lower shaded boxes in Fig. 10.1 [10]. The measures used for prevention of accidents and injuries directly and indirectly influence the personal and contextual factors. Here, they are divided into three main categories :

1.
Attitude modification measures: attitudes are changed by means of persuasive messages in mass media campaigns, leaflets, booklets, films, posters, or direct mail. Also included in this category are one-way counselling schemes, such as counselling on car safety to mothers of newborn children. Health education normally utilises measures from this category.
2.
Behaviour modification measures: behaviour is changed through more direct approaches, without assuming that attitudes have an intermediary function, for instance by skills training with or without a rewards system.
3.
Structural modification measures: contextual factors are changed through legislation, regulation, enforcement, organisation, and economy. This also refers to changes in the physical environment and to modification and the availability of products.

The model shows connections between these three types of preventive measures and two risk factors : behaviour and physical/organisational environment, and two process factors: attitudes/beliefs and social norms/(safety) culture .

A classic model for injury prevention has three different phases [11]:

Primary prevention: preventing new accidents and injuries
Secondary prevention: reducing the severity of injuries
Tertiary prevention: decreasing the frequency and severity of disability after injury

Important distinctions between prevention measures are made between:

Active measures that require conscious action by individuals to prevent or minimize the risk of accident or injury, such as using safety equipment.
Passive measures that do not require an individual to act to prevent injury, and are often more effective than active measures. Examples are airbags in cars, smoke alarms in homes, and dumps in roads.

Injury Surveillance

A medically based injury registration system is a tool for gaining knowledge of the epidemiology of the injuries that are referred to the medical care system (Fig. 10.1, upper shaded box). Medically based injury surveillance systems have emerged around the world from the early 1970s. The forerunners were computerised trauma registers in the USA established in the late 1960s [12]. Surveillance comprises “ongoing and systematic collection, analysis, interpretation and dissemination of health information” [13]. The term “register” is used synonymously with the related term surveillance system. According to Frerichs (p. 265) [14], “a surveillance system goes one step further than a monitoring system by including a ‘controller’ who is a person or agency with some ability to take corrective actions”(Fig. 10.1, lower shaded box). A controller can be found at national, regional, and local levels. After receiving input from the monitoring system, the controller compares the registered data with established standards or goals. Information on factors contributing to accidents and injuries might lead to proposals for preventive actions. Action follows only “if the administration has the will and political power to act” (p. 265) [14].

The first medically based injury surveillance systems were designed by product safety authorities to identify unsafe products [5]. They are data sources additional to occupational and traffic accident registers that are generally incomplete and fragmented [15, 16]. They are also a valuable data source on home and leisure accidents, the largest accident group in most countries. More current systems are based on reports of all types of injuries: accidental, assaults, and intentional self-harm. All systems collect information on injuries treated in a more or less representative sample of hospitals and/or accident and emergency departments (AED) in the country/region to be surveyed.

The use of the medical care system for acquiring statistics on injuries and injury control has necessitated development of new multiaxial classifications [7, 17]. Minimum data sets (MDSs) are developed for settings where registration resources are scarce [13, 18]. In such settings, the possibilities to identify the contributing factors leading to injuries are restricted because few details on each accident/injury can be collected. For such purposes, expanded data sets (EDSs) are collected in specially designed and funded in-depth investigations, or in accident commissions.

The most important aim of an injury surveillance system is to provide information useful for the prevention of injuries. To fulfil this aim, three different types of data need to be recorded:

1.
Data enabling estimation of injury incidence rates (for the purpose of guiding programme priorities and resource allocation)
2.
Data for establishing trends (for the purpose of evaluating the effectiveness of prevention programmes
3.
Data on factors contributing to injuries (for the purpose of developing prevention measures, strategies, and programmes)

In order to record all three types of data simultaneously, many details from each accident/injury have to be collected in a sufficiently large sample to provide high representativeness of the population to be surveyed. Such an ideal system requires large registration resources, hardly to be found anywhere in the world. As an alternative, a two-step injury surveillance system is suggested (Fig. 10.1, upper shaded box). In the first step, limited data (an MDS) on all or a representative sample of all injuries are recorded within the medical care system using routine collection procedures. This data collection includes injuries sustained by both residents and nonresidents within a defined geographical area and fulfils the need for data types 1 and 2. The second step involves periodic sampling of specific injuries, injured persons, or places for in-depth investigations from the database established in the first step, or selecting relevant injured persons seeking treatment in the medical system, in order to collect many data on a limited number of injuries. This step might provide data for developing prevention measures, strategies, and programmes (data type 3).

Such a two-step injury surveillance system was implemented in Oslo, the capital of Norway, with a population of about 500,000 [19]. During 1 year, 48,283 persons were registered using an MDS. They were treated by 17 general practitioners (GPs) in one city district, at the city’s main AED, as inpatients in four hospitals, and deceased persons were registered on death certificates (step 1). Two in-depth investigations of each of a total of 273 serious occupational injuries treated at the AED were carried out. Detailed interviews as well as on-site studies were undertaken (step 2). This two-step injury register showed potential for providing valid monitoring data and revealing factors contributing to accidents and injury.

In the European region, a lot of efforts have been made in order to establish injury surveillance systems. Both the World Health Organization (WHO) European Region (Resolution EUR/RC55/R9) and the European Council (Recommendation 2007/C164/01) have urged member states to develop injury surveillance systems, so that programmes for prevention, care, and rehabilitation can be better targeted, monitored, and evaluated. The WHO European Region evaluation of these actions shows that there has been some progress, but calls for “improved access to reliable and comparable injury surveillance information to make the extent, causes and circumstances of the problem more visible across the Region” p. 34 [20].

A Joint Action on Monitoring Injuries in the EU (JAMIE) has led to an updated methodology and format for collecting basic information in a large number of emergency departments at hospitals at almost no additional costs. At present, a substantial number of hospitals across 26 countries in the European Union (EU) are collecting data in their emergency departments in line with the harmonised methodology and classification. During these years, injury statistics in the EU has been produced based on this system, the last for the years 2008–2010 [21].

These data are currently being used for a wide range of safety promotion purposes including in helping to design better and safer consumer products. That is one of the reasons why a broad coalition of European organisations called earlier this year (2013) on the European Commission to set up a Pan European Accident and Injury Data System. These EU-level umbrella organisations in commerce, standardisation, consumer, and health and safety are convinced that such a system would contribute to fewer accidents and injuries.

Current Burden of Injuries in Europe

Injuries due to accidents, assaults, and intentional self-harm are a major health problem, killing more than 230,000 persons in the EU-27 each year (annual average 2008–2010). An estimated one million persons will be permanently disabled [22]. Injuries are the third most common cause of death, after cardiovascular diseases and cancer. Below 45 years of age, it is the number one cause of death [21]. Across the European countries, there is a great variation of injury incidences, from rather high in Russia and the previous Soviet Republics, to rather low in the western countries of Europe. The most recent data show a variation from less than 25 per 100,000 to more than 180 per 100,000, a sevenfold disparity (Fig. 10.2).

Every 2 min., one EU citizen dies of an injury. For each fatal injury case, 25 people across the EU are admitted to a hospital, 145 are treated as hospital outpatients, and many more seek treatment elsewhere, e.g. by family doctors. This means that each year a staggering 5.7 million people are admitted to a hospital and 33.9 million people are treated as hospital outpatients as a result of an accident or violence-related injury [21] (Fig. 10.3).

In addition to hospital treatments, many injuries are treated by GPs and paramedical staff without being referred to a hospital. In the Netherlands for instance, about one third of all injury patients are treated in hospitals and two thirds are seeking consultation in the office of a GP [23]. In Norway in the years 2009–2011, 80 % of the injured patients sought consultation at a GP/municipality AED, Quite many of these are transferred to hospitals, with the result that in the end about 50 % of all medically treated injury patients are finally treated by GPs and 50 % at the hospitals (both as in- and outpatients) [24]. This relation will of course vary across countries depending on the organisation of the health system.

The responsibility for injury prevention is quite dispersed over a variety of policy sectors depending on the setting in which they occur and the circumstances. Figure 10.4 depicts the key figures of the main unintentional and intentional injury categories according to place of occurrence and injury outcomes in terms of severity (death, hospital admission, or outpatient treatment).

Accidental injuries are responsible for about three quarters of all injury deaths and intentional injuries for the remaining one quarter. Suicide and road injuries account for the highest number of fatalities , in both absolute and relative terms (in relation to the number of hospital-treated injuries, i.e. lethality). Most of the severe injuries in the EU are treated in hospitals making them the proper place for injury surveillance [21]. With 73 % of all hospital-treated injuries, home and leisure (including sports and school) is by far the biggest share, which is in contrast to the fact that home and leisure injury prevention programmes appear as far less resourced than programmes for road and workplace safety. In general, the tangible and intangible consequences of home, leisure, and sport injuries are also less well covered by insurance systems compared to the compensation schemes for road and work accidents [25]. The road injuries account for 10 % of all hospital-treated injuries or a total of 4.2 million victims annually. Compared to just 1.7 million injuries reported by the police (about 40 % of the hospital-treated traffic injuries), this indicates a significant underreporting of the problem in official road traffic statistics and the need for complementary information on road injuries treated in health facilities [26]. This level of underreporting in eight European countries varied between 21 and 57 % [27].

Costs of Injuries

Costs of accidents can be measured in human, social, economic, and organisational costs [28]:

Human
- The injured person being unable to return to usual work tasks, temporarily or permanently
- Poor quality of life, due to constant pain measured in QALY (quality-adjusted life years) [29]
- Emotional physical trauma
- Financial hardship
Social
- Financial burden may fall onto other family members
- Fellow workers and family may need counselling
- Economic
- Medical expenses, compensation, and rehabilitation costs to family, company, and taxpayers
Organisational
- Costs of hiring and training of replacement staff
- Loss of production while staff respond to accident or equipment needs to be shut down to be replaced or repaired
- Some workers may not wish to return to the usual job due to the severity of accident

Recently (2001–2004), within the framework of a project named EUROCOST, a uniform injury-based method to calculate medical costs of injury was developed and applied to ten EU countries. This method allowed the calculation of medical costs of injury by sex, age, external cause, and type of injury at the country level and EU level.

Home and leisure, sport, and occupational accidents combined make a major contribution (86 %) of the total hospital costs of injury in Europe. Table 10.1 shows the costs of admitted injury patients for each participating EUROCOST country.

Table 10.1 Costs per capita, incidence, and mean costs per patient for admitted injury patients (all causes) per country [30]

Full size table

In addition, the study shows that the elderly patients aged 65 years and older, especially women, consume a disproportionate share of hospital resources for trauma care, mainly caused by hip fractures and fractures of the knee/lower leg, which indicates the importance of prevention and investing in trauma care for this specific patient group (Fig. 10.5).

Prevention of Accidental Injuries: Best Practices and Evidence-Based Strategies

What Works Well and What Does Not Work So Well?

According to the model presented in Fig. 10.1 (lower part), accident prevention measures are divided into three main categories : modification of (1) attitudes, (2) behaviour, and (3) structural conditions. Prevention measures from two or three categories might be utilised in the same programme (orchestration), creating a new category of measures across categories. Through a literature review, the pathways and impacts on the incidences of accidental injuries of the various accident prevention measures in Fig. 10.1 were examined [10]. An attempt was made to identify the most effective accident prevention measures. A total of 249 interventions (166 in meta-analysis and systematic reviews and 83 separate interventions) of relatively high quality were identified in the literature. They were divided into the following main and subcategories with regard to which prevention measures they utilised:

1.
Attitude modification programmes (N = 27):
1. a.
  Information measures: mass media campaigns, leaflets, booklets, films, posters, and direct mail (N = 17)
2. b.
  Counselling and education in classrooms, small groups, or individually (N = 10)
2.
Behaviour modification programmes (N = 32):
1. a.
  Instruction, skills training, and feedback, focus on behavioural change (N = 24)
2. b.
  Rewards for desired behaviour (N = 8)
3.
Structural modification programmes (N = 156):
1. a.
  Legislation (N = 86)
2. b.
  Environmental and product modification (N = 70)
4.
Programmes combining measures across categories (orchestration; N = 34):
1. a.
  Combined prevention measures (N = 22)
2. b.
  Community-based interventions (N = 12)

All interventions were analysed with regard to the results of the interventions. Some of them had significant positive effects on reductions in accidents or injury incidence rates, or on changes in behaviours that might have an impact on injury rates, e.g. use of bicycle helmets. Others did not have any effects, or sometimes have negative effects. The percentages of interventions in each subcategory with positive results are shown in Fig. 10.6.

Based on Fig. 10.6, some of the hypothesised paths in the model (Fig. 10.1, lower part) seem to be weak: Attitude modification measures (information) → attitude → behaviour → accidents and injuries (the traditional KAP model: knowledge → attitude → practice; subcategory 1a). Others seem strong: Structural modification measures → physical and organisational environment → behaviour → accidents and injuries (subcategory 3a, b). Behaviour modification measures in workplaces have shown positive effects, while there are mixed experiences with educational and skills training programmes for children and adolescents. Such training might create unrealistic beliefs in one’s own abilities, and parents might overestimate their children’s abilities, so they may be exposed to situations more dangerous than they can master. Behaviour modification measures that concentrate on a single behaviour usually seemed to be more effective than a more general effort directed at a range of hazards [31]. The use of rewards (subcategory 2b) has consistently shown positive effects on children’s and adults’ use of safety equipment in cars and in workplaces. When attitude, behaviour, and structural modification measures were used in combination (subcategory 4a, b), the interconnections and mutual influences taking place among the personal and contextual factors in the model mostly seemed to produce stronger effects than if one category of preventive measures was used alone. The overview of community-based interventions showed significant positive preventive effects in all but one study reviewed.

Injury prevention , with its broad range of injury types and possible countermeasures, lends itself to community-based approaches. The use of multiple interventions, repeated in different forms and contexts, can lead to a culture of safety being developed within a community [11]. There is evidence that the WHO Safe Community model is effective in reducing injuries in whole populations [32].

Information measures alone (such as mass media campaigns or leaflets, Fig. 10.6, subcategory 1a) seemed to produce very little, if any, effect on safety behaviour and on the incidence of accidents and injuries, except when the target groups were highly motivated. The effects were stronger when the message was repeated, was more tailor made, and was delivered face to face in counselling schemes, or even better when this information on accident prevention was based on two-way information in small groups (subcategory 1b). Structural measures, such as regulation and enforcement, environmental and product modifications generally seemed to have a strong positive effect. Environmental changes with negative effects were marked pedestrian road crossings, which probably introduced some sort of false safety.

All 166 interventions in the meta-analyses and systematic reviews dealt with traffic accidents, as well as 51 of the 83 other interventions. The rest dealt with home (17), occupational (6), sport/leisure (5), and all accidents (4). About 80 % of the 249 interventions took place in Anglo-American cultures. Very few of the interventions took place in Africa (N = 4) and in Asia (N = 2). This bias restricts any generalisation of results to the Western World. The bias towards traffic accidents reflects the fact that evaluations of traffic accidents have a longer and broader tradition in the English-speaking world compared with evaluation research into home, occupational, and sport/leisure accidents. The bias towards traffic accident interventions, however, should not influence general conclusions, as risk and process factors are not supposed to react differently when other accident types are involved.

Although attitude change measures used alone seem to have little direct impact on behaviour, they may still have an important role in accident prevention. The model presented in Fig. 10.1 (lower part) suggests that by influencing attitudes, other factors can also be influenced, which in turn will reduce accidents and subsequent injuries. Through attitude-changing measures, we might:

Persuade more people to take both precautionary actions and to initiate passive measures that in turn will reduce the prevalence of accidents and injuries.
Contribute to the shaping of public opinion that is favourable towards the use of passive measures and legislation. If one does not have convincing support from public opinion, it is more difficult for politicians, union representatives, and others to make the decisions that are necessary to introduce various structural measures.
Mobilise social support for kinds of behaviour that will reduce the risk of accidents and subsequent injuries; for example, support for not driving under the influence of alcohol.
Contribute to changing social norms. Although individuals’ private attitudes may have limited impact on their own actions, the attitudes of others often carry more weight.
Accident-relevant aspects of culture may be influenced through a long-term use of informational and educational approaches, preferably in combination with other measures. Hence, attitude-changing measures also represent an important contribution to the development of a more safety-minded culture.

After this more general view on which preventative measures work well, and which not so well, we refer to examples of interventions which have given positive effects in some of the main accident types. We refer to handbooks, overviews of best practices, and other systematic reviews which give information about prevention of accidental injuries. One systematic review describes preventative measures in various settings and includes cost–benefit analyses (CBAs) of the measures [33]. The results of CBA depend strongly on the context to which they refer. Monetary valuations of impacts, which are a key element of CBA, vary substantially between countries. As a rule, one would therefore not expect the results of CBAs made in one country to apply directly to another country.

These CBAs are referred to in the following chapters. The country where and the year when the analysis is made are given in parentheses.

Prevention of Traffic Accidents and Injuries

Evaluations of traffic accidents have a longer and broader tradition than evaluations of other accident types. Traffic safety has been highly prioritised, probably because the field is public, there are many fatalities , and it is a high-energy and dramatic area .

The WHO has published a world report with a lot of information about road traffic injury prevention and how to tackle the situation [34]. The Handbook of Road Safety Measures [27] gives a systematic overview of current knowledge about the effects of road safety measures. This is a book of about 1000 pages. Hundreds of scientific papers and reports have been studied and analysed; 110 specific road safety measures are described with the effect on accidents, mobility, environment, costs, and also CBA for many of them.

The road safety measures in this handbook are divided in eight chapters:

Road design and road equipment: 20 measures, e.g. tracks for walking and cycling, roundabouts, black spot treatment, cross-section improvements, and road lighting
Road maintenance: nine measures, e.g. improving evenness of the road surface, bright road surface, winter maintenance, correcting erroneous traffic signs
Traffic control: 21 measures, e.g. speed limits, speed-reducing devices, pedestrian streets, yield signs at junctions
Vehicle design and protective devices: 28 measures, e.g. studded tyres, ABS and disc brakes, daytime running lights for cars, cycle helmets, and seatbelts in light vehicles
Vehicle inspections: Four measures: e.g. periodic motor vehicle inspection
Requirements for drivers, driver training, and professional driving: 13 measures, e.g. driving licence age limits, the driving test, and graduated driving licence
Road user education and information: four measures, e.g. education of pre-school children, education in schools, and road user information and campaign
Enforcement and sanctions: 11 measures, e.g. stationary speed enforcement, seatbelt enforcement, speed cameras, fines and imprisonments, and warning letters

In addition, there is a chapter on general-purpose policy instruments: 14 measures. These measures are general in nature and are used in many sectors of public policy. Thus, they are not always regarded as road safety measures. Examples of these are: safe community programmes, road pricing, motor vehicle taxation, and road traffic legislation.

Some examples from the review of Elke and Elvik [33] with benefit–cost (BC) ratio:

Alcohol-control and media campaign (NZ 2004), BC ratio 14–26
Road lighting (NO 2007), BC ratio: 1.9
Alco-lock for previous drunken drivers (NO 2007), BC ratio: 8.8
Roundabouts (NO 2007), BC ratio: 1.9–2.6
Seatbelt reminder (NO 2007), BC ratio: 16.2
Speed control (NO 2007), BC ratio 1.5

Prevention of Occupational Accidents and Injuries

In a recently published systematic review of safety intervention for the prevention of accidents at work [35], 318 relevant studies were identified . The screening process started with about 22,000 titles from the international literature from 1966 and up to now. About 6500 abstracts were assessed to be relevant. After studying them, about 600 articles were left to be read. Of the 318 articles, 162 were assessed to be of too low quality, leaving 156 studies to be studied for assessing the effectiveness of safety interventions in preventing accidents and injuries at work.

About 50 % of these studies were from the health and social sector, 10 % from industry, 10 % from construction, 5 % from agriculture, and the rest (25 %) from other sectors. The studies were divided into the same four main groups as with the study of Lund and Aaro [10] (Fig. 10.6). And the same general results were found: Group 1 and 2 had less studies with positive significant results as groups 3 and 4. In the following, some examples of these studies are listed (+ : positive effect, −: negative effect, s.: significant, ns.: not significant) .

Attitude modification programmes

Transport, professional driving, and group discussions (SE 1996): + 44 %, s.
Agriculture, one-way communication, and group discussions (DK 2003): + 3 %, ns.

Behaviour modification programmes

Transport, professional driving, and training programmes (SE 1996): + 37 %, s.
Transport, professional driving, and rewards (SE 1996): + 26 %, s.
Supermarkets, cut injuries, knives, and training programmes (USA 1997): + 9 %, ns.
Post workers, back injuries, and educational programmes (USA 1998): − 11 %, s.

Structural modification programmes

Forestry, technical changes in machinery (USA 2002): + 64 %, s.
Hospital, regulation against violence (USA 2009): + 48 %, s.
Hospital, technical change of syringe (USA 2001): + 46 %, s.
Mining, change of regulation (USA 2002): + 45 %, s.

Programmes combining measures across categories (orchestration)

Hospital, muscle-skeletal injuries (USA 2011): + 61 %, s.
Hospital, needle sticks (AUS 2008): + 51 %, s.
Construction, nail-gun injuries (USA 2008): + 37 %, s.
Construction, all injuries (DK 2002): + 25 %, s.

In conclusion, the authors recommend to develop and build strategies based on integrated measures and structural measures (p. 19) [10]. “It is therefore not a viable option for focusing exclusively on attitudinal measures through campaigns and the like, as they cannot stand alone in prevention efforts nor provide the expected pay-off when they do.”