1

Essential, primary preventive,

occupational safety and health interventions

for low and middle income countries

Evidence Report

Jos Verbeek

Finnish Institute of Occupational Health

Cochrane Occupational Safety and Health Review Group

PO Box 310

70701 Kuopio

Finland

e-mail: jos.verbeek@ttl.fi

tel: +358-46-8108709

2

Contents

1. Introduction .................................................................................................................................................................. 4

2. Methods ......................................................................................................................................................................... 4

From burden of occupational diseases and injuries to exposure .................................................................................. 6 Work-related cancer and its related exposure ............................................................................................................. 7 Pneumoconiosis and related exposures ....................................................................................................................... 8 Chronic Obstructive Pulmonary Disease and related exposures................................................................................. 8 Occupational Asthma and related exposures ............................................................................................................... 9 Noise induced hearing loss and related noise exposure .............................................................................................. 9 Back Pain and related biomechanical exposure .......................................................................................................... 9 Injuries and related exposure to hazardous situations .............................................................................................. 10 From exposure to preventive interventions ................................................................................................................ 11

3. Evidence of effectiveness of occupational health interventions .............................................................................. 14

3.1 Interventions for decreasing inhalation exposure ................................................................................................. 14 Environmental interventions: substitution ....................................................................................................... 14 Environmental interventions: other control measures ...................................................................................... 15 Environmental measures: regulation and other incentives ............................................................................... 18 Behavioural: respiratory protection for preventing inhalation exposure .......................................................... 18 Clinical: pre-employment examinations and drugs .......................................................................................... 18

3.2 Interventions for decreasing exposure to noise .................................................................................................... 19 Environmental interventions ............................................................................................................................ 20 Behavioural interventions: promotion of hearing protection ........................................................................... 20 Clinical interventions ....................................................................................................................................... 21

3.3 Interventions for decreasing biomechanical exposure ......................................................................................... 21 Environmental interventions: load reduction and ergonomics ......................................................................... 22 Behavioural interventions: education and training .......................................................................................... 22 Clinical interventions: pre-employment examinations .................................................................................... 23

3.4 Interventions for prevention of injuries ................................................................................................................ 23 Environmental interventions ............................................................................................................................ 23 Prevention of injuries in agriculture and construction industry ....................................................................... 25 Behavioural interventions: Occupational Safety Training ............................................................................... 26 Behavioural safety interventions: feedback and rewards ................................................................................. 27 Clinical: Pre-employment examinations for preventing injuries ..................................................................... 28

3.5. Approaches to Small Enterprises ........................................................................................................................ 28

4. Conclusions and discussion ........................................................................................................................................ 29

8. References .................................................................................................................................................................... 35

9. Appendices .................................................................................................................................................................. 40

3

Executive Summary

There is still a considerable global burden of occupational diseases and injuries that leads to many fatalities

each year. It is not well known which interventions can effectively reduce the exposures at work which

eventually cause these occupational diseases and injuries. The objective of this report is to provide the

available evidence from systematic reviews of essential preventive interventions that can reduce the global

burden of occupational diseases and injuries.

Essential interventions are interventions that reduce most of the global burden of occupational disease and

injuries. These interventions should therefore aim to reduce the incidence of work-related cancer, dust-

related diseases, occupational asthma, COPD, noise-induced hearing loss, back pain and occupational

injuries. Primary prevention of those diseases occurs through reduction of the exposures that lead to these

diseases. These interventions are categorised as environmental, behavioural and clinical. The literature was

searched to locate systematic reviews of interventions that can reduce each of these exposures through any of

these interventions.

The evidence available in these systematic reviews shows that there are many technical measures to reduce

exposure available that can have a major impact on the global burden of work-related cancer, dust-related

diseases, asthma, COPD, noise and injuries. However, to effectuate this potential, better implementation is

needed. This can be realised by better regulation, reinforcement or incentives for employers. Feedback and

rewards for workers probably help in reducing occupational injuries. However, the available systematic

reviews do not provide evidence that back pain can be prevented. Personal protective equipment also has

technical potential to reduce exposure but without proper use and instruction this can not be realised. On the

contrary, there is no evidence in the available reviews that education and training reduce occupational

disease and injuries. Clinical interventions such as drugs and health examinations have little to offer for

primary prevention of occupational diseases and injuries.

More and better systematic reviews are needed to enable a better overview of the evidence especially in the

area of injury prevention.

Acknowledgements

I am grateful to Irja Laamanen and Leena Isotalo, information specialists at FIOH for helping me with the

searching of systematic reviews

4

1. Introduction

Experts estimate that less than 15% of the global workforce has some coverage with occupational health

services.1 The 60th World Health Assembly in 2007 urged the 193 Member States of WHO to work towards

full coverage with essential interventions and basic occupational health services, particularly in agriculture,

small and medium size enterprises, informal economy, and migrant workers. WHO was requested to provide

guidance to countries on basic packages, tools, working methods and models of good practices for

occupational health services and to stimulate international efforts for capacity building as part of the Global

Plan of Action on Workers' Health 2008‐2017.2

The range of the interventions to prevent occupational and work‐related diseases and injuries may include

both clinical (e.g. health examinations) and non‐clinical interventions (e.g. workplace risk assessment). The

interventions can be categorised as primary preventive interventions and treatment interventions. Primary

preventive interventions aim at preventing disease or injury outcomes before the disease or injury process

has started. In occupational health, this means usually that efforts are directed at decreasing exposure known

to be hazardous to health. Providers of such interventions may include health practitioners in clinical settings

such as primary care centres, non‐clinical providers such as workers' representatives and employees

responsible for health and safety in the enterprise or occupational hygiene and safety experts in specialized

occupational health services.

Many countries have already in place some form of basic occupational health services to deliver essential

interventions for the prevention of occupational and work‐related diseases and injuries 1. However, it is not

well known what evidence exists for the effectiveness of these interventions. To be better able to develop

guidance on essential primary preventive occupational health and safety interventions this evidence is

needed. WHO is especially interested in evidence for essential interventions in basic occupational health

services targeted at underserved working populations with constrained resources and integrated in primary

health care. The first step in this process of guidance development is to locate systematic reviews that have

synthesised the evidence available from primary studies.

Based on available systematic reviews, I report here what is the evidence for the effectiveness of the most

essential occupational health interventions for primary prevention of work‐related diseases and injuries in

agriculture, small and medium sized enterprises and the informal economy across WHO regions.

2. Methods

5

To answer this policy and research question I will first further define essential occupational health

interventions for primary prevention of occupational diseases and injuries. These essential interventions

should prevent occupational diseases and injuries in underserved populations with constrained resources. The

mechanism behind such interventions is that they cut the causal chain between exposure at work and the

resulting occupational disease or injury (figure 1).3 Basically, occupational health interventions can be

categorised into three major classes: environmental, behavioural and clinical. Environmental interventions

aim at changing the working environment in order to eliminate the exposure in a technical sense.

Behavioural interventions focus on individual workers' behaviour to eliminate exposure such as increasing

the use of personal protective equipment. Clinical interventions use a clinical method to prevent disease such

as vaccination. Even though other preventive methods are available such as screening or early diagnosis, I

will not take these into consideration because I will focus entirely on primary prevention.

I consider those interventions as essential that aim at eliminating exposures with the biggest impact on the

target population. The WHO has examined the global burden of occupational disease in a recent project.4 I

took the diseases and injuries that are mentioned in this report as the point of departure. Next, I determined

which exposure lead to these diseases and injuries. This resulted in a limited list of exposures that should be

addressed by essential interventions.

Risk factor at work

Worker Health

Behaviour

Disease, Disability, Injury

Environmental interventions Behavioural Interventions Clinical Interventions

Interventions:

E.g Work-site visits.

Organisational

changes. Regulation.

Technical changes:

substitution,

ventilation.

Interventions:

E.g. Health

promotion.

Education. Reward-

punishment.

Personal protective

equipment

Interventions:

E.g.Treatment

Counselling

Vaccination

Verbeek, Scand J WEH 2004

figure 1: Model of occupational health interventions

6

In the next step, I looked for evidence of effectiveness of essential interventions as generated by evaluation

research and reported in the literature. Since it would be impossible to look for all primary studies on all

exposures and interventions I searched only for evidence at an aggregated level in the form of systematic

reviews. I defined systematic reviews as reviews of the literature that had a clearly formulated question and

searched electronic databases. I combined search strategies for finding occupational safety and health

intervention studies with search strategies for finding systematic reviews. For locating occupational safety

and health intervention studies, I used the search strategy developed by the Cochrane Occupational Health

Field 1 and for locating systematic reviews I used the search strategy developed by the Perosh Systematic

Clearing House Working Group (www.perosh.eu)2. In principle, I did not search for primary studies when

there were no systematic reviews available. I searched first in Medline through Pubmed and then in Embase

through OvidSP to see if any additional systematic reviews could be found.

I explicitly did not review the literature on secondary prevention, treatment or rehabilitation. Even though

there are many effective and essential interventions that address rehabilitation issues and that could be

carried out in a low-resource setting, these are not included because this project only addresses primary

prevention.

From burden of occupational diseases and injuries to exposure

Table 1 provides an overview of the essential occupational diseases and injuries with a big impact on the

target population derived from the global burden of disease project.4;5

The target group of the essential

occupational health interventions consists of workers in small businesses in low and middle income countries

especially in agriculture and the so-called informal sector.

Work-related disorder to be prevented Risk factors to be addressed

1. Work-related Cancer

Arsenic, asbestos, berylium, cadmium, chromium, diesel

exhaust, nickel, silica, benzene, ionizing radiation, ethelyne

oxide

2. Asthma Biological agents: grains, flours, plants, gums, fur, feathers,

insects, fungi, drugs, woods

chemical agents: chlorofluorocarbons, alcohols, metals,

welding fumes

3. COPD Non-specific dusts and fumes

4. Pneumoconiosis Silica-containing dusts

5. Noise-induced Hearing Loss Sound levels above 80 dB(A)

1 ((effect*[tw] OR control[tw] OR controls*[tw] OR controla*[tw] OR controle*[tw] OR controli*[tw] OR

controll*[tw] OR evaluation*[tw] OR program*[tw] OR prevention*[tw] OR protect*[tw]) AND (work[tw]

OR works*[tw] OR work'*[tw] OR worka*[tw] OR worke*[tw] OR workg*[tw] OR worki*[tw] OR

workl*[tw] OR workp*[tw] OR occupation*[tw]) NOT animals[mh]) 2 (meta-analysis[mh] OR meta-analysis[pt] OR meta-analysis[tiab] OR review[pt] OR review[tiab]) NOT

(letter[pt] OR editorial[pt] OR comment[pt]) NOT ((animals[Mesh:noexp]) NOT (humans[Mesh]))

7

6. Back Pain Ergonomic risk factors: manual material handling, bending

and twisting, heavy physical load, static work posture,

repetitive movements, whole-body vibration, stress-related

risk factors

7. Injury Prevention Hazardous situations at work

Table 1 Overview of risk factors to be addressed by essential occupational health interventions

Work-related cancer and its related exposure

For work-related cancer, there are two major cancer types that are associated with the exposures at work.

Cancer of the trachea, bronchus or lung is associated with inhalatory exposure to arsenic, asbestos,

beryllium, cadmium, chromium, diesel exhaust, nickel and silica and radon. Leukaemia is associated with

exposure to benzene, ionizing radiation and ethelyne oxide. Except for arsenic and nickel the effects of all

these agents come through inhalation. Arsenic's route of exposure is through ingestion and for nickel it can

be both ingestion and skin absorption.6

Exposure to these agents in low and middle income countries is not self-evident. Beryllium is only used in a

small and specific industry. The same holds for benzene, ionizing radiation and ethelyne oxide which are

mainly used in chemical and other industries and in small amounts in health care. These compounds are thus

outside the scope of this study.

Arsenic is an important component of pesticides and herbicides and used in timber treatment and pigments.

However, exposure mainly occurs during the manufacturing of these products. It is therefore not expected

that this exposure will be very prevalent among our target group and thus it falls outside the scope of this

study. Arsenic occurs also naturally in the soil and leads to arsenic contaminated ground water. A main

source of exposure to arsenic is through the use of contaminated ground water for cooking. There are several

solutions to prevent uptake of arsenic from contaminated groundwater but this is outside the scope of this

occupational health study.

Both cadmium and nickel are ingredients of batteries that have a ubiquitous presence but this does not lead to

occupational exposure of our target population. Cadmium, chromium and nickel are present in welding

fumes. Welding is a ubiquitous activity that is very prevalent in all parts of the world. Therefore I included

interventions for reducing the risk of inhaling welding fumes.

Exposure to diesel exhaust is prevalent around the world but it is only to a limited extent occupational.

Occupational exposure mainly occurs in traffic controllers, railroad workers and truck/car drivers which are

outside the scope of this study.

Asbestos and silica are widely spread around the world and lead to health problems through inhalation. Both

substances lead also to pneumoconiosis. Exposure to asbestos occurs in the asbestos mining, ship building

8

and construction industry. The latter seems most relevant for our study. The same holds for silica, to which

workers are exposed during mining of coal and in the construction industry during all kinds of activities that

involve cutting, drilling or blasting of stone that contains silica. These activities occur frequently throughout

the world and should be the focus of preventive activities and are thus included.

After considering the prevalence of exposure in LMI-countries, I further elaborated preventive activities to

lower exposure to asbestos, silica and welding fumes because these are the main exposures that cause work-

related cancer.

Virtually all exposure to asbestos, silica and welding fumes occurs through inhalation. Preventive efforts

should thus focus on prevention of inhalation of these substances. Technically, there are two approaches to

prevent inhalation: source-oriented and exposure-oriented. Source-oriented solutions are directed at changing

the source in the production process by substituting for example asbestos with another material so that no

fibres will be released anymore. Exposure-oriented solutions focus on taking away the exposure without

actually eliminating or reducing the source. Examples of exposure-oriented solution are local exhaust

ventilation, personal protective equipment or dust control measures. Especially exposure to asbestos and

silica occurs as a dust in which tiny solid particles are carried by air currents and are capable of remaining in

suspension for a period of time. In a technical sense, dust control is well studied and there is a wealth of

information on control measures both source-oriented and exposure-oriented.7

Pneumoconiosis and related exposures

Pneumoconiosis is the term used for all dust damage done to the alveolar part of the lung including the parts

of the lung that do not have a mucociliary lining, but it does not include bronchitis, asthma or cancers by

convention.6 Pneumoconiosis can thus be caused by a number of inorganic dusts the toxicity of which

depends on particle size, shape and chemical composition of the dust. The most toxic are asbestos fibres and

silica dust. The most commonly occurring is coal dust. Ultimately the disease leads to fibrosis and

emphysema of the lung. There is no cure and the only way to prevent the disease is to lower dust levels or

eliminate exposure totally. Exposure to asbestos, silica and coal dust is a relevant exposure for low and

middle income countries.

The cause of pneumoconiosis is inhaled dust and therefore dust control and prevention of inhalation are also

here the hallmark of prevention.

Chronic Obstructive Pulmonary Disease and related exposures

Chronic obstructive pulmonary disease (COPD), defined as non-reversible chronic airflow limitations, is

related to dust exposure which is a comparable route of exposure as in pneumoconiosis or occupational

asthma.

9

Primary prevention thus uses the same pathways and I will not further elaborate the prevention of COPD

only. I assume that by preventing pneumoconiosis and occupational asthma also COPD will be prevented.

Occupational Asthma and related exposures

Also the primary prevention of occupational asthma is based on the prevention of inhalation of substances

known to cause asthma. These can be divided into high molecular weight substances, usually organic dusts

such as grain and wood dusts, and low molecular weight substances such as diisocyanates, metals and

welding fumes.

Primary prevention of occupational asthma is thus similar to primary prevention of inhalation exposure and

will require the same means as for other inhalation exposure.8

Noise induced hearing loss and related noise exposure

Noise induced hearing loss is caused by long term exposure to noise at levels greater than 80 dB(A). The

dB(A) stands for decibels for which the sound level has been corrected according to the hearing abilities of

the human ear. Once hearing ability has been lost, there are no means to recover the lost capacity.

Therefore, decreasing exposure to noise levels greater than 80 dB(A) is the only means of prevention for

noise induced hearing loss.

However, it is clear that not every worker is equally sensitive to the damaging effects of noise. Even after

forty years of exposure to 100 dB(A) a substantial proportion of workers will still have normal hearing for

their age.9 Given these differences in sensitivity, early detection of workers who are most susceptibility to

noise and focussing our preventive efforts on these workers would be worthwhile. Currently, such a test is

however not available and we are not able to accurately predict who will sustain the greatest hearing loss.10

This leaves us with exposure reduction as the main means of prevention. Along the same lines as with

inhalation exposure prevention, there are two general approaches for exposure reduction: technical

engineering controls and hearing protection. I will look for evidence for these two approaches.

Back Pain and related biomechanical exposure

For back pain, there is no clear cut physical model that explains how exposure at work causes back pain

comparable to exposure to silica dust and silicosis. This makes the primary prevention of back pain more

complicated than inhalation exposure prevention. The cardinal symptom of back pain is subjective and has

various dimensions that are difficult to measure. Even though attempts to standardise approaches to

measuring back pain have been undertaken, reporting of back pain as an outcome still varies widely across

studies 11

. In general, two different models of occupational back pain causation are in use. One model is that

back pain occurs because biomechanical overload during a considerable time period has led to osteoarthritis

10

of the vertebrae and the intervertebral discs in the lumbar spine. These anatomical and physiological

changes, in turn, lead to pain sensation. Psychosocial stress is thought of as an intermediary factor in this

process. Stress can either lead to different positions leading to greater wear and tear of the lumbar spine or it

can lead to an increased pain sensation. Another model used mainly in the North American workers'

compensation insurance context is the idea of a back injury that results from biomechanical overload at

work. The idea of an injury implies that there is an immediate connection between the overload and the

resulting pain. Thus, only back pain that occurs in immediate connection to the biomechanical overload is

recognised as occupational back pain. For our search for relevant exposures, both models would lead to the

same inference that exposure to biomechanical overload of the lumbar spine is the main cause of back pain.

Primary preventive interventions should therefore lead to a decrease of biomechanical exposure.

Even though back pain as such is an unpleasant experience, more importantly back pain leads to incapacity

for work with resulting increase in sick leave and long term occupational disability. Because this is a major

problem both for the afflicted individual and society at large, many preventive efforts thus focus on

preventing disability. More psychosocial factors, both at individual, company and society level, influence the

pathway from back pain to back disability. The primary cause will still be the origin of back pain and

primary prevention would still aim at taking away these causes.

Biomechanical overload of the lumbar spine is defined as a load that exceeds the tolerability of the structures

of the spine. There is no consensus on how to calculate this 12

. Nevertheless, consensus exits on activities

that increase the load on the spine and that should be minimised. These activities are manual material

handling such as lifting, frequent bending and twisting of the lumbar spine and exposure to whole-body

vibration 13

.

Biomechanical overload seems a very common exposure in LMIC where work is less mechanised and

sophisticated manual material handling aids are less available. Therefore, I searched for evidence of

effectiveness of interventions that can decrease biomechanical exposure at work especially through lifting,

bending and twisting and whole-body vibration.

Injuries and related exposure to hazardous situations

The etiology of injuries is more complicated than that of occupational diseases and research has long been

hampered by a lack of a conceptual framework. Even though it is not always easy to discern injuries from

illnesses, I will define injuries as sudden-onset occupational injuries where energy exchange produces

immediately discernable tissue damage 14

.

In the public health arena, Haddon was one of the first to model the etiology of injuries. He argued that an

injury results from the interaction between a host, an agent and the environment. In this model the host is the

person who receives the energy and the agent is the vector that transfers the energy. In addition, Haddon

11

conceptualised three temporal phases that determine the likelihood of injury. First, there is the pre-event

phase that includes the activities and the conditions of the host, the agent and the environment. Then there is

the event during which the energy is transferred to the host. In the post-event phase first-aid and medical care

can still add to survival and recovery. Together the temporal and energy dimensions make up the so-called

Haddon matrix that can be used as an analytical tool for injury prevention. Preventive approaches are further

divided into active and passive approaches in which the passive approach is, counter-intuitively, considered

the most effective. An active approach means that a person has to actively take countermeasures to avoid

injury whereas the passive approach does not require human interaction 15-18

.

Another approach comes from psychology where the orientation has been on behavioural theories to explain

occupational injuries. Based on a meta-analysis of personal and environmental risk factors for injuries and

accidents, Christian and Wallace (2009) put forward a conceptual model of 'workplace safety'.19

In the

model, situation-related factors such as safety climate and leadership interact with person-related factors

such as personality characteristics, job attitudes and safety knowledge and motivation. Together, the

situational and personal factors shape safety performance such as safety compliance and safety participation

which ultimately leads to the prevention of injuries.

Some sectors are notoriously more dangerous than others. For example agriculture and construction industry

are among the sectors with the highest injury rates 20;21

. In addition, certain groups of workers sustain more

injuries such as immigrant workers.22

It seems, that the conceptual model of injury etiology fits well with the approach of primary prevention that I

have taken in this report. I looked for evidence of effectiveness of environmental and behavioural

interventions that address hazardous situations in the pre-event phase.

From exposure to preventive interventions

For exposure reduction, the general approach is, similar to other risk reduction strategies, a three step

approach in which risks first have to be identified, then evaluated and finally controlled or managed. For the

management of especially chemical health hazards this approach has been best developed but the strategy is

applicable to all kinds of risk problems.

In practice many approaches have been developed to facilitate risk reduction. For example hazard

identification checklists have been developed that list potential risks in specific businesses. Such check lists

enable workers or occupational health professionals to immediately focus on problems at hand in their

specific business.23

Especially the evaluation of chemical risks is greatly facilitated by the lists of Occupational Exposure Limits

(OEL). An OEL list states for a number of chemicals what the level of exposure is below which there is no

12

appreciable health risk. Most of these lists are available through the Internet nowadays.24

In practice, this

means that one measures the exposure in a specific workplace and compares this with the appropriate OEL.

If the exposure is below the OEL, no control measures are needed.

However, this approach, of first measuring risks and than evaluating, has been criticised especially for small

businesses because it focuses very much on measurement strategies. Measurement of chemicals often

requires specific professional expertise and financial resources that is not available in small businesses and

thus, valuable resources are spent on activities that do not necessary lead to risk control. Therefore,

Ellenbecker argued that risk control strategies in small businesses much better immediately focus on

engineering controls than on measuring and evaluating.25

Engineering controls or interventions are part of the so-called hierarchy of controls. The meaning of this

hierarchy is that is postulated that some control measures should always take priority over other control

measures because they are valued higher. Basically, this theoretical approach lists two approaches of

exposure reduction: engineering controls and administrative controls. Engineering or technical controls

comprise exposure reduction through substitution of the hazardous agent or process changes that eliminate

exposure and isolation or ventilation of the source of exposure. Administrative controls comprise personal

protective equipment, worker education and training and scheduling work such that the duration of exposure

is reduced. Because engineering controls solve exposure problems in a more fundamental way, they are

always to be preferred over the administrative controls. For example, substitution of the source should be

preferred over the other solutions that are supposed to solve exposure problems less well. Since this is a

theoretical approach it is unclear how this works out in practice.

Applying engineering controls without an intermediate step of carefully evaluating the exposure levels is

also in line with the idea that many exposures do not really have a level below which there are no health

effects. Even though there are OELs for ionizing radiation, the strategy here is to get the exposure As Low

As Reasonably Achievable (ALARA) because there is no real threshold for an effect on health and with each

increase in exposure there is a concurrent increase in risk to a worker. The OEL is based on what is

considered an acceptable risk compared to other activities in life but this does not guarantee absolute

protection.26

Another related concept that has been developed in the control of hazardous chemical exposures at work in

the past decade is the so-called control-banding.27

"The concept is to put chemicals (and other substances and

processes) into one of a number of “bands” depending on its hazard. The risk is a combination of the

inherent toxicity of the substance and the likely exposure. The exposure is assessed qualitatively by

determining the form in which the chemical exists (particle size, gas, vapour), the quantity used, and the

processes in which it is used. The approach recommends control strategies to be chosen prior to and possibly

in lieu of exposure measurements taken in the workplace. Control banding may be useful in situations where

13

there are no OELs and/or when quantitative exposure assessments would be difficult to obtain (e.g., small

industries that lack the technical expertise or in developing nations)." 28

Thus, it follows that for primary prevention of occupational diseases and injuries, specifically environmental

interventions should be the primary choice. If these are not available, behavioural or clinical interventions

can be used such as education and training or health examinations.

In terms of environmental interventions, it is important to make a distinction between interventions that are

technically possible and that work under laboratory conditions and the effect of these interventions under

field conditions. Dust control measures, for example, are effective provided that they are carried out

appropriately. However, the implementation of these technical measures in work places is an entirely

different matter. How can we make an employer or a worker take measures and change their working

routines so that no dust will be stirred up? For an effective occupational health intervention, we need a

control measure that is technically working but as such this is not sufficient. In addition, we also need

interventions that convince employers and workers to change their behaviour and implement the technical

control measures. Niewöhner and Cox have nicely illustrated this in what they call the Mental Models

Approach 29;30

. They showed that it is highly unlikely that safety information as provided by Material Safety

Data Sheets will be understood and used by managers and workers in its current technical format. If

information is taken up and acted upon, depends on the mental model that managers and workers have of the

health hazards at work and the possibilities of their control. It is therefore important to take their beliefs and

their concepts of a safe workplace into account. Thus interventions become less technical and much more

behavioural.

Most engineering measures are evaluated in uncontrolled before-after experiments over a short-time period.

For example technical changes on machinery to make it less noisy can be implemented. A noise

measurement before the implementation and after the implementation would convince most of us of the

effectiveness of the technical measures. In such a technical case study, it is usually clear that the technical

improvements are the only factors that have changed over the relatively short time period of follow-up such

as hours or days. These studies usually do convince that technical measures are feasible and do lead to lower

exposure. However, experiments intended to demonstrate that the measures can be implemented need to be

well-controlled and need a longer follow-up time to be convincing. In addition to the intervention, there are

always many other factors that can lead to workplace changes in companies that can easily confound the

effects of an intervention. Interventions will attract the most motivated firms to participate in the research

project and thus the results of an experiment can be easily distorted as a result of this selection bias. There

are only very few well randomised experiments of exposure reduction available and the results have been

disappointing probably because the behavioural change component of the intervention has been

underestimated.31

Also in less well-controlled implementation projects, researchers report that the results of

workplace changes are only modest.32

As part of this report, I especially looked at evidence of the

14

effectiveness of implementation of technical measures that reduce the exposure to occupational carcinogens

in workplaces.

Pre-employment examinations can also be considered as primary preventive interventions as their objective

is to prevent that susceptible workers will be exposed. Thus this can also be called an exposure prevention

intervention. It is important to take into consideration that the benefits of pre-employment examinations

usually are not realised without the draw-back of denying employment to the prospective employee.

There is a recent systematic review of the effect of pre-employment health examinations that lists all

controlled studies that have been performed.33

I will not search for the effect of pre-employment health

examination for each exposure but I will use the results of this review.

3. Evidence of effectiveness of occupational health interventions

3.1 Interventions for decreasing inhalation exposure

Environmental interventions: substitution

Roelofs et al performed a systematic search of the occupational hygiene literature to locate articles

describing, what they call, control or prevention strategies for chemical hazards in actual workplaces.34

They

were disappointed to find, among the located 92 articles, only two that explicitly discussed the hierarchy of

controls and only four that actually compared different control strategies. They concluded that “despite their

theoretical primacy, primary prevention strategies—those focusing on reduction of hazards at the source—

are not commonly considered in practice; they ranked third in frequency of citation in the literature. And

although industrial hygienists sometimes do consider these strategies, they are not often implemented. Only

11% of the articles described the actual implementation of primary prevention strategies.

In line with their findings I could locate only one systematic review that provided evidence of effectiveness

of substitution as an intervention to decrease inhalation exposure. The effect of substitution on the very

specific exposure to latex was studied in one systematic review. The authors included eight studies of

varying methodological quality. They concluded that there is 'adequate evidence' that substitution of

powdered latex gloves with low protein powder-free natural rubber latex gloves or latex-free gloves leads to

a reduction of occupational asthma.35

Cullinan et al, in a review of asthma prevention, argue that there few

examples of the evaluation of primary prevention of asthma but those that are described as case studies

provide evidence that this is a feasible and highly effective strategy. Examples of primary prevention are the

change from the use of powder to granulated proteases in the detergent powder industry, laboratory animal

allergy and also latex allergy measures in health care.8

15

Environmental interventions: other control measures

The effectiveness of control measures for inhalation exposure has been studied in the occupational hygiene

literature in various ways. Researchers have tried to show the effectiveness of control measures as a generic

approach. Some researchers have tried to summarize the effectiveness of what they called -risk management

measures- based on studies that compared the exposure with and without control measures.36

Others

provided an overview of measures to control dusts especially.7 Then, there are researchers that reviewed

studies that reported trends in exposure over time and tried to relate these trends to specific interventions.37-40

However, none of them used formal systematic review techniques or techniques for meta-analysis and it is

therefore difficult to judge the value of these reviews. They neither make a distinction between studies that

measure the technical outcome versus studies that have evaluated the implementation of these technical

measures in companies. The results of these studies are summarised underneath.

Fransman et al provide an overview of the efficacy of seven broad but well defined categories of control

measures for inhalation exposure excluding substitution. They calculated with what percentage the exposure

could be reduced by various control measures: 36

- enclosure (50%, 95% Confidence Interval 4% to 74%)

o complete or partial encapsulation or encasing of the source such as lids or screens

- local exhaust ventilation (82%, 95% CI 78% to 84%)

o any kind of exhaust ventilation system close to the source with or without enclosure

- specialized ventilation systems (87%, 95% CI 73% to 94%)

o mechanical ventilation systems specifically designed to displace air in small designated

areas such as a walk-in booth or clean zone

- general ventilation (43%, 95% CI 17% to 61%)

o natural or mechanical ventilation of whole work areas such as opening windows or doors

- suppression techniques (83%, 95% CI 77% to 88%)

o the addition of an additive to an activity or process to suppress emissions such as water or oil

- segregation of sources (no studies)

o total isolation of source from the work environment such as in a special room

- separation of the worker (87% 95% CI 71% to 94%)

o worker is in a personal enclosure within the work environment

They included any kind of study that compared the exposure with and without the control measure. This

could be a cross-sectional study, a laboratory experiment or a field study to evaluate the implementation.

Efficacy was calculated as the percentage reduction of an exposure: exposure without controls minus

exposure with controls divided by exposure without controls times 100. For example the exposure without

controls is 10 mg/m3 and with controls it is 1 mg/m3. This results in a efficacy value of 90%. The results

were averaged over the various control measures to give an indication of the overall efficacy per category

and adjusted for study design, sampling strategy and measurement type.

16

In total they included 433 comparisons derived from 90 articles. Most comparisons (N=177) were derived

from intervention field studies and from local exhaust ventilation (N=280). The average efficacy values are

given in the table with the categories above. The results indicate that most control measures are highly

effective in reducing exposure between 80% and 90% except for enclosure and general ventilation that

reduce the exposure to a lesser degree of between 40% and 50%. The authors did not find a difference for

dusts, fumes, vapours or mists in the efficacy of the control measures.

For practical reasons the authors did not include studies on measures that changed work processes such as

substitution and neither had they included studies on respiratory protection. The authors acknowledge that

the preferred study design to evaluate the effectiveness of the control measures would be an intervention

study aimed at evaluating the implementation of the control measures. Thus the included studies can only

provide limited evidence of effectiveness. In addition, it is easily conceivable that there will be considerable

publication bias with studies of this type. Because many of the included studies are actually case studies,

authors would be reluctant to publish studies that do not show a decrease in exposure. A review of published

studies will thus mainly include studies with positive effects.

Creely et al summarized what they called trends in inhalation exposure with the aim to summarise factors

that may have been responsible for the observed changes.39

To be included papers had to describe inhalation

exposure over some time span but the authors did not specify the length of the time span. From all papers the

authors extracted or calculated the percentage decrease in exposure per year. The results were reported

according to exposures to aerosols, gases and vapours and fibers. The suggestions on the causes of the

changes by the authors of the primary papers were also reported. The results were summarised in a narrative

way and no attempt to meta-analysis was made. They included 25 papers that reported on exposure trends

and these papers included 87 separate exposure trends. One paper that reported on exposures in the American

pulp and paper industry over the time period between 1979 and 1994 contributed 36 exposures trends.41

Of

the 87 reported exposure trends, 61 were measured in the US. The authors summarise the evidence available

in these 87 trends as in the majority of instances there were significant reductions in exposure. Across studies

the average yearly decrease in exposure was - 7% ± 5% for aerosols (N=38), - 8% ± 7% for gases and

vapours (N=39) and - 15% ± 11% for fibres (N=10). From another not totally overlapping 21 papers the

authors summarised the reasons for decline or increase of exposure. Regulatory changes were mentioned

most often as the reason for decline of exposure. Other reasons were implementation of occupational health

programmes, changes in production equipment or methods and installation of control measures such as

ventilation or elimination or substitution of the source. It was not possible to make any general conclusions

about the causes of the decline.

A third paper reviewed the trends in exposure to metal working fluids. Even though this exposure is outside

the scope of this study, it provides better evidence of trends in exposure over time. The review included 48

papers on fluid aerosol measurements of metalworking-fluids over a time period ranging from 1949 to 2007.

The arithmetic mean exposure was averaged over ten year periods for 155 measurement situations with the

17

following result: prior to 1970 5.4 mg/m3, 1970-1980 2.5 mg/m3, 1980-1990 1.2 mg/m3, 1990-2000 0.5

mg/m3 and in 2000-2007 0.5 mg/m3. These figures show a ten-fold decrease or 90% decline in exposure

over time. These reductions in exposure are thought to be brought about by installation of aircleaners, partial

enclosure of machines and local exhaust ventilation based on better awareness of the toxicity of

metalworking fluids.40

Symanski et al also systematically reviewed exposure decline over time but did not include any reference to

the possible interventions that could have induced these declines.42

Their conclusions are in line with the

other reviews. Based on 119 published studies they calculated a yearly decline in exposure of 15% for

western Europe, of 17% for North Amerina, of 60% for Eastern Europe, of 47% for Japan but only 5% for

other countries.

I conclude that there is low quality evidence from four systematic reviews of low quality that control

measures to reduce inhalation exposure are effective. General ventilation and enclosure seem to be less

effective than other measures. Apparently through enforcement of regulation and implementation of

technical controls a general decrease in inhalation exposure has occurred in the US and other industrialised

countries in the past 30 years.

On a more technical level dust control as a form of specific inhalation exposure prevention has been

reviewed by Smandych as a generic measure to reduce exposure to dust.7 She reviewed dust control

measures both from the point of view of the dust sources and from the point of view of controlling dust in

general. Dust sources in the production process are usually related to storing, processing, packaging, feeding

or hauling. Enclosure of most of these processes is a feasible option. For general control strategies there are

dust collection systems such as local exhaust ventilation and dust suppression systems such as wet or oily

sprays. The authors conclude that most of the information is merely rule-of-thumb or general guidelines and

that more research is needed to provide a more consistent and systematic approach. A test of dustiness of

materials handled at workplaces has been developed that will help in evaluating the dustiness of material to

be used at work.43

This will help in selecting materials that are less dusty.

In addition, a wealth of studies have been published that reviewed techniques to reduce specific inhalation

exposure such as due to abrasive blasting, mortar removal, hand tools for concrete cutting, masonry cutting

and tuckpointing44-47

I will not review all these specific techniques as this is beyond the scope of this study.

These studies show that technical measures to control inhalation exposure are well known and well studied.

There were no systematic reviews of exposure reduction in welding. I found two studies that showed that

local exhaust ventilation can substantially reduce exposure to chromium and manganese in welding

processes applied in the construction industry.48;49

18

Environmental measures: regulation and other incentives

Torén and Sterner discussed that different regulatory strategies can lead to similar results in exposure

reduction.50

They provide some evidence from case studies in various countries that taxes, subsidies and

banning can lead to similar reduction of exposure to trichloroethylene. Elsner et al reviewed case studies of

economic incentives in various European countries and concluded that they can be an effective strategy for

improving occupational health.51

An example of successful regulation of prevention of inhalation exposure is

the case of asbestos. The use of asbestos in production processes has been successfully banned from most

industrialized countries where exposure to asbestos nowadays mainly occurs in construction workers during

demolition and renovation of buildings. Recently, the Collegium Ramazzini group has renewed their call for

a universal ban on the mining and use of all forms of asbestos.52

One of the main arguments for a ban is that

it is impossible to control the use of asbestos to ensure that no health hazards will occur.53

Even though good systematic reviews are missing in this area, it seems that regulation and economic

incentives can both lead to substantial reduction of exposure.

Behavioural: respiratory protection for preventing inhalation exposure

For respiratory protection, the same holds as for technical control measures in general. The effectiveness of

respiratory protection is the result of the interaction of proper design and manufacturing ('proper

functioning') and actual fit, maintenance and cleaning and proper use by the wearer in workplace practice

('proper use') as Brouwer et al put it.54

Thus, results of laboratory testing and technical features of the devices

never hold under practical circumstances. The protection factor of respiratory protection has been shown to

decrease with a factor 450 between simulated and real workplace testing.55

In addition, it is clear that without

proper instruction and training in its use, the exposure reduction will be even worse.56

Practical features

make the use of respiratory protection difficult. If the wearer has a beard, the protective properties of

respiratory protection will be lost.55

Brouwer et al propose that for personal protective equipment that has

been shown to be functioning properly additional criteria for its proper use should be adopted. These criteria

are: awareness of the need to use the personal protective equipment, instruction and training, ability to

perform work tasks with the protective equipment and acceptability based on discomfort rating.54

There are no systematic reviews that have looked at the effectiveness of respiratory protection under field

circumstances. In none of the reviews of exposure trends over time the effects of respiratory protection have

been taken into account. Given all the practical constraints for proper use, it seems unlikely that respiratory

protection can be an effective measure of control of inhalation exposure in low and middle income countries.

Clinical: pre-employment examinations and drugs

The effectiveness of pre-employment examinations has been debated for a long time.57

Most authors have

argued against the effectiveness of pre-employment examinations for the following reasons. Pre-employment

testing is basically a form of screening by which prospective job applicants are tested for personal risk

19

factors that make them more vulnerable to the exposure of the prospective job. For those who test positive an

intervention should be applied to eliminate the risk. Most research has focussed on the tests that are applied

in pre-employment examinations. Authors have argued that the tests are not specific enough and label many

job applicants as being at risk while in reality they are not. In epidemiological terms, the tests yield too many

false positives. Denying the job to the worker who tests positive is the most frequently applied intervention.

This means that those who are false positive are wrongfully denied access to work. Based on modelling

studies, authors have argued that the costs of wrongfully denying work to those who are false positives far

outweighs the benefits of preventing occupational diseases in those who are correctly classified as being at

risk.58-60

However, there are no studies that have conducted a proper economic evaluation. There are also no

studies that have followed the job applicants after they were rejected based on the pre-employment

examinations.

Evaluating the effectiveness of screening procedures is complicated because the outcome depends both on

the quality of the diagnostic tests involved and the effectiveness of the intervention. Screening can be

evaluated by comparing the whole screening procedure including the interventions to a control condition

without the procedure. An alternative evaluation procedure is to study the effectiveness of the interventions

only in candidates who screen positive as the interventions' effectiveness is the most important part of the

procedure. There are basically three different interventions that can be applied to job applicants that have a

risk factor: they can be denied the job they are applying for, the prospective workplace can be

accommodated to their risks or the job applicant can be trained so that they are no longer at risk.

There is only one Cochrane systematic review that has looked at controlled studies that have evaluated the

effectiveness of pre-employment examinations.33

In this review, nine studies were included. Seven evaluated

introduction of pre-employment examinations versus no pre-employment examinations. One of these seven

studies was about the prevention of occupational asthma. Incorporating a bronchial challenge test for

workers in the aluminium industry decreased the number of cases of occupational asthma over time in an

interrupted time series compared to the time period when pre-employment examinations were used without

these tests.

Prevention of work-related cancer can theoretically also be achieved through vitamins such as beta-carotene,

which could be prescribed for people at risk. There is also a Cochrane Review that concludes that there is

currently no evidence to support recommending vitamins such as alpha-tocopherol, beta-carotene or retinol,

alone or in combination, to prevent lung cancer. On the contrary, a harmful effect was found for beta-

carotene with retinol at pharmacological doses in people with risk factors for lung cancer (smoking and/or

occupational exposure to asbestos).61

3.2 Interventions for decreasing exposure to noise

20

Environmental interventions

I found one systematic review of interventions to reduce exposure to noise.62

The authors found evaluations

of the effects of the following interventions:

- the effect of the introduction of legislation on noise levels (1 study)

- the long term effect of hearing loss prevention programmes in companies on hearing loss (14 studies)

- the short term effect of hearing protection on temporary hearing loss or noise exposure (6 studies)

The one study that examined the effect of new legislation in reducing noise exposure measured the impact

during 18 consecutive years in the mining industry in the US, either side of the implementation of legislation.

It found that the median noise level decreased by 28 decibels immediately after the introduction of

legislation.

The effectiveness of the hearing loss prevention programmes in companies was evaluated by comparing

hearing loss in workers that were exposed to noise but who were protected through a hearing-loss prevention

programme with the natural hearing loss in workers that were not exposed to noise. The studies produced

mixed findings and, in some, workers who were protected from noise still had three to four times the risk of

hearing impairment compared to similar people not exposed to noise in the first place. In three

implementation studies, better quality hearing loss prevention programmes were shown to have a

significantly lower risk of hearing loss than lower quality programmes. Most of the hearing loss prevention

programmes were from the US or older studies that consisted mainly of instruction to workers to wear

hearing protection, even though some studies also included engineering controls.

The studies on the short-term effects of hearing protectors showed that these can reduce noise levels

sufficiently. An important factor is that the hearing protectors have to be worn properly. For example,

workers need instruction for the insertion of hearing plugs to make these sufficiently protective.

The authors concluded that there is low quality evidence that legislation can reduce noise levels in

workplaces but contradictory evidence that prevention programmes are effective in the long-term. Even

though case studies show that substantial noise reductions can be achieved, there is no evidence that this is

realised in practice. It is concluded that better implementation and reinforcement of legislation and hearing

loss prevention programmes is needed if people’s hearing is to be better protected.

Behavioural interventions: promotion of hearing protection

In addition to this review, there is one other Cochrane systematic review on interventions to enhance the use

of hearing protection.63

In this review, three studies were included that reported on the primary outcome

measure that was the proportion of participants wearing hearing protection. One study was a cluster

randomised trial of 17 schools for vocational training in farm work with 3 years follow-up. The participants

at the intervention schools (N=375) received a hearing test and an extensive educational intervention

whereas the participants at the control schools (N=378) received only a hearing test. The percentage of pupils

21

that reported wearing hearing protection 'at least sometimes' increased from 23% to 83% at the end of

follow-up whereas these figures for the control group were 24% and 35% respectively. Another study among

automotive workers (N= 1325) exposed to on average 90 dB(A) evaluated the effect of a computer

generated message that was tailored according to the amount of hearing loss and to the individual predictors

of the likelihood of wearing hearing protection. This tailored message was compared to general information

and to a control condition where workers saw a commercial video on hearing loss prevention. In a follow-up

of the same study the workers were again randomised to receive a booster message sent to their private

addresses after one and three months after the first message. In the third study, the same tailored intervention

was evaluated in a different population. The tailored message resulted in an increase of 6% of the self-

reported amount of time that hearing protection was worn in both studies compared to either the non-tailored

intervention or the control condition. The other studies in the review did not report on the primary outcome.

The authors conclude that there is evidence from one study that school-based interventions can be effective

and that messages tailored to the needs of the individual can be more effective than general messages.

Clinical interventions

As mentioned before, there are no reliable tests to detect those that are sensitive to noise that could be used in

a pre-employment examination. Neither did I find studies that used pre-employment examinations as an

intervention to reduce noise-induced hearing loss.

Even though magnesium has been used as a drug to prevent hearing loss among those exposed to noise, there

are no systematic studies that have reviewed its effectiveness.

A conclusion similar to that on prevention of inhalation exposure applies to reduction of noise exposure.

Technical solutions for decreasing noise levels are feasible, tested and available. There is low quality

evidence from a high quality review that legislation can help implementing these interventions, which leads

to lower noise levels. There is contradictory evidence that noise-induced hearing loss decreases as a result of

hearing loss prevention programmes. Hearing protection can adequately reduce exposure but needs

educational or behavioural interventions to be properly implemented to levels that protect workers

adequately. There is low quality evidence that educational programmes at school or messages tailored to the

needs of the individual can increase implementation but it is unclear if this leads to sufficient protection.

3.3 Interventions for decreasing biomechanical exposure

Control of biomechanical exposure can be either directed at worker behaviour such as better lifting

techniques or at the workplace such as decreasing the loads to be lifted or improving the position in which

the work has to be done. It is surprising how few studies have been conducted in this area.

22

Environmental interventions: load reduction and ergonomics

With biomechanical exposure it is not always easy to make a distinction between environmental and

behavioural interventions. Ergonomic interventions usually require some kind of human participation but it

is possible to reduce exposure without human interference for example by reducing the weight that has to be

lifted. In that sense, the American National Institute of OSH (NIOSH) has presented a valuable exposure

limit as a means to try to reduce spinal load at work. This consists of a relatively simple assessment tool that

can be used to assess a recommended weight limit. The maximum recommended weight limit is 23 kg. The

lifting conditions, such as distance of the load to the body and the degree of trunk rotation have to be

specified and entered into a formula which returns a recommended weight limit for these conditions.64

Even

though this is a practical approach that in theory should be able to reduce spinal load, there are no controlled

studies that have evaluated if using such an exposure limit prevents back pain.

The effectiveness of participatory ergonomic interventions on health outcomes has been reviewed by Rivilis

et al 65

earlier published as Cole 2005. The implementation of this type of interventions thus requires active

participation of workers. This review included seven studies that evaluated the effect of ergonomic

interventions on musculoskeletal disorder symptoms. Three of these were RCTs. The highest quality study

showed similar symptom levels in the intervention group as in the control group after 10 months follow-up.

The other two studies found little change. Of the non-randomised studies, two found a positive result and two

found a non-significant result for back symptoms. Even though the authors conclude that this is moderate

evidence that participatory interventions have a positive impact on musculoskeletal disorder symptoms, the

review does not provide evidence that back pain can be prevented by ergonomic interventions.

A more recent review of ergonomic interventions (physical and organisational) to prevent back pain in

workers included 10 RCTs. There was low to moderate quality evidence that physical and organisational

ergonomic interventions were not more effective than no ergonomic intervention on short and long term back

pain incidence/prevalence and short and long term back pain intensity.66

The same authors found a similar

result in a cluster randomised trial that they conducted as a result of their review.67

Behavioural interventions: education and training

Since long it has been assumed that there is a correct lifting posture that decreases biomechanical load on the

lumbar spine. There is however no consensus in the biomechanical literature what this correct lifting posture

is.12

Based on these assumptions, there has been considerable effort invested in interventions to make

workers change their lifting behaviour. These efforts have been especially concentrated in occupations such

as nursing in which it is difficult to change the load because this is inherent to the task of lifting patients. In

one older review on prevention of back pain in general, Van Poppel et al found 11 trials.68

There were six

RCTs that evaluated education but it was concluded that these contained no evidence of a preventive effect

of education.

23

Since then, several reviews have addressed the effectiveness of manual material handling advice to prevent

back pain.69-72

They are mostly based on the same studies and all conclude that manual handling training is

largely ineffective in reducing back pain or back injury.

Verbeek et al recently updated the review of Martimo et al of advice and training for manual material

handling to prevent back pain.73

The authors included in this updated review ten RCTs with 20.152

employees and eight cohort studies with 1176 employees. All studies focused on prevention of back pain.

Ten studies compared some kind of training to no intervention (4), a minor intervention (3), back belt use (1)

or exercise (1) and one study compared training plus lifting aids to training only or no intervention. The

intensity of training ranged from a single educational session to very extensive personal biofeedback of the

load on the lumbar spine. None of the included studies showed evidence of a preventive effect of training on

back pain. Based on seven RCTs, there was moderate quality evidence that training resulted in similar back

pain as no intervention with an odds ratio of 1.17 (95% Confidence IntervaI 0.68 to 2.02) and as minor

advice with an odds ratio of 0.93 (95% CI 0.69 to 1.25). The results of the cohort studies were similar to

those of the randomised studies. The authors conclude that there is moderate quality evidence that MMH

advice and training with or without assistive devices do not prevent back pain or back pain-related disability

when compared to no intervention or alternative interventions.

Clinical interventions: pre-employment examinations

The same Cochrane review on pre-employment examinations as mentioned before included three studies that

used a functional capacity evaluation test in workers with high physical work load to prevent

musculoskeletal injuries including back pain. The studies led to contradictory results with one study resulting

in less musculoskeletal disorders and two studies that did not lead to a decrease. All studies led to an increase

in rejected applicants. Moreover, if there are benefits, it remains to be seen if these outweigh the increase of

rejected job applicants.33

The very low quality of the evidence implies that future research could easily

change these conclusions.

3.4 Interventions for prevention of injuries

A major problem in injury prevention research is that injuries are a relatively infrequent event in most

workplaces. For evaluation research this means that studies are easily underpowered to find differences

between intervention and control group. Therefore one has to be cautious to not interpret this as interventions

not being effective.

Environmental interventions

There were no studies that systematically looked at environmental interventions such as guarding to prevent

entanglement in machines or preventing falls from roofs. Hsiao and Simeonov reviewed the literature on fall

prevention but they did so more from a theoretical point of view than as a review of prevention effectiveness

24

studies.74

Falls from roofs are one of the main causes of work-related fatalities in the construction industry

and thus an important focus for prevention. Most measures for preventing falls focus on how to mitigate the

results once the fall is happening already such as using safety belts and guards. In line with the Haddon-

matrix, the authors postulate that control of balance is the most important factor in the pre-event phase. They

reviewed which factors can improve control of balance to prevent falls from roofs. They found evidence that

visual and physical interaction with the environment, tasks such handling loads and personal factors such as

work experience and training are the main determinants of control of balance at heights. Visual interaction

can be improved for example by providing visual anchors or colour cues to improve depth perception.

Physical interaction with the environment can be improved by extending the surfaces on which a worker

stands, providing slide guards, increasing surface frictional properties, evenness and removing obstacles.

Task related factors such as load handling, fatigue and task complexity can be improved to prevent falls from

roofs. Personal factors that can be improved were training and personal protective equipment. There were no

intervention or evaluation studies that could support the suggestions for primary prevention of fall injuries.

Tompa et al reviewed the effectiveness of 'prevention incentives' used by insurance and in regulation.75

I

regarded this as implementation measures to get environmental interventions in place to prevent occupational

injuries. He discerns two approaches to induce the desired preventive behaviour: experience ratings of

insurance premiums and enforcement of occupational health regulation. Experience rating is the insurance

practice where premiums are reduced if you make fewer claims. They included all studies that evaluated one

of these interventions, were quantitative and had used a longitudinal design. Based on the quality and the

outcome of the studies they rated the available evidence as strong, moderate, limited, no or mixed.

For experience rating they found moderate evidence that both the introduction of experience rating and the

degree of experience rating led to a decrease in injuries.

There were only two studies that evaluated the introduction of occupational health regulation which led to a

conclusion of mixed evidence for its effectiveness. Enforcement of occupational health regulation was

measured in various ways. Inspections were evaluated in 18 studies and the authors concluded that there was

limited evidence that inspections were associated with a reduction in injury frequency or severity. The 11

studies that evaluated both the effect of inspections and the probability of a penalty showed mixed results.

The other seven studies that evaluated the effect of an actual penalty showed strong evidence that this

resulted in a lowering of the injury rate. The authors warn for the possibility that regulation and experience

rating can give rise to so-called perverse incentives meaning that the incentive can also induce undesired

behaviour such as not reporting injuries anymore to be able to receive a lower premium or to prevent a

penalty.

It has been shown that there is a strong relationship between safety climate in a company and the injury rate.

Safety climate is defined as a specification of organisational climate that in turn is made up of shared

perceptions among employees concerning procedures, practices and kinds of behaviours that get rewarded

and supported with regard to a specific strategic focus. When the strategic focus involves performance of

25

high risk operations, the resultant shared perceptions define safety climate.76

The topic has been extensively

reviewed by Zohar but at the moment there is no empirical evidence on how to affect safety climate such that

it would help in reducing injuries.

Prevention of injuries in agriculture and construction industry

Another approach to injury prevention evaluation is to look at a whole branch of industry. Branches of

industry that are important to our topic are agriculture and construction industry.

An older review on injury prevention in agriculture by de Roo et al concluded in 2000 that there was little

evidence that farm safety interventions were effective.77

Recently, this review has been updated in a

Cochrane Review.78

Five RCTs and three interrupted time-series (ITS) met the inclusion criteria. Five studies

evaluated educational interventions, one study financial incentives and two studies evaluated the effect of

regulation, one of regulating tractor roll-over protection structures and one banning of pesticides.

Three RCTs with 4670 adult participants evaluated safety training and education. These studies did not find

an effect on injury rates (Rate Ratio 1.02 (95% confidence interval 0.87 to 1.20)) Another two RCTs did not

find an effect of safety training among children.

Financial incentives decreased the injury level immediately after the intervention in one ITS. Legislation

requiring rollover protective structures (ROPS) on new tractors was associated with a decrease in fatal

injuries but the same requirement for existing tractors showed no effect.

Hartling et al performed a review of interventions to prevent specifically childhood farm injuries.79

They

included 23 controlled studies of which four randomized. All interventions were educational in nature. Even

though school-based programs and safety day camps appeared to be effective at increasing short-term

knowledge acquisition, other interventions showed mixed results and no studies showed an actual reduction

in injuries.79

Another Cochrane Review gathered evidence on safety interventions in the construction industry.20

Here, the

authors found five interrupted time-series studies that met their inclusion criteria. Three studies evaluated the

effect of regulations, one evaluated a safety campaign, and one a drug-free workplace program on fatal or

non-fatal injuries compared to no drug-free workplace program. The overall methodological quality was low.

The three studies that evaluated regulatory interventions did not show either an initial or sustained effect on

fatal or non-fatal injuries, with effect sizes (ES) of 0.69 (95% confidence interval (CI) -1.70 to 3.09) and 0.28

(95% CI 0.05 to 0.51).

The safety campaign, which consisted of several methods aimed at preventing injuries, reduced non-fatal

injuries significantly both immediately (ES -1.82 (95% CI -2.90 to -0.75)) and in the long run (ES -1.30

(95% CI -1.79 to -0.80)). Also the drug-free workplace program had an initial and sustained effect, reducing

non-fatal injuries compared to no intervention (ES -6.74 (95% CI -10.02 to -3.54) and -1.76 (95% CI -3.11 to

-0.41)).

26

The authors concluded that the vast majority of technical, human factors and organisational interventions

which are recommended by standard texts of safety, consultants and safety courses, have not been adequately

evaluated. There is no evidence that regulations for reducing fatal and non-fatal injuries are effective. There

is limited evidence that a multifaceted safety campaign and a multifaceted drug program can reduce non-fatal

injuries in the construction industry.

Behavioural interventions: Occupational Safety Training

Education and training to increase knowledge and model safe behaviour have long been the mainstay of

injury prevention. An extensive review of the literature from 1980 to 1996 by NIOSH in 1998 categorised

training as more narrow instruction whereas education was defined as broader instruction.80

The authors also

categorised occupational health training and education into four different programme types with increasing

assumed effectiveness: fundamental training aimed at instruction of proper work practices and use of

personal protective equipment, recognition programmes aiming at hazard recognition and control, problem-

solving programmes and empowerment programmes that go beyond just problem-solving but use a total

quality management approach. They divided the literature further into training that aimed at reducing injury-

producing forces, toxic chemicals or materials, harmful physical agents, ergonomic stressors, biologic or

infectious agents. Here, I present only the results of training aimed at reducing injury-producing forces. All

together, they found 80 studies that evaluated some kind of training intervention. These studies used either

satisfaction, knowledge, behaviour or injuries as the outcome by which the effectiveness was judged. Of

these 80 studies, 21 studies aimed at injury prevention but only 14 used some kind of control group. The

following results were reported. Four studies reported a reduction of injuries and one study reported no

effect. Seven studies reported objective behaviour change and one no effect. The other studies measured only

knowledge or satisfaction. The authors conclude that there was "much direct and indirect evidence to show

the benefits of training in establishing safe and healthful working conditions. The intervention studies in

particular were especially supportive. Findings here were near unanimous in showing how training can attain

objectives such as increased hazard awareness among the workers at risk, knowledge of and adoption of safe

work practices, and other actions that improve workplace safety and health protection." The review did not

take into account possible biases such as publication bias and the quality of included studies. It is unsure if

the same conclusions would be drawn with current systematic review standards.

More recently Burke et al reviewed the effectiveness of occupational health training with the specific

objective to see if training that engaged workers more was more effective than training in which workers

were less engaged 81

. They used a meta-analytic approach mostly used in psychology and grouped all types

of training programmes together but looked separately at knowledge, performance and injury outcomes. To

be included in the review, studies had to compare the intervention with a control group. They found 95

studies of which 31 evaluated the effect on injuries. The authors did not mention an overall outcome of their

review but only reported that the pooled effect sizes significantly increased from least engaging to

27

moderately engaging to most engaging training programmes for knowledge, and injury outcomes but not so

for behavioural outcomes. The latter non-significant result is explained by confounding as the level of

engagement was also related to the complexity of the behaviours that had to be improved. The more complex

tasks involved more worker engagement but were at the same time more difficult to change. The authors

took methodological quality partly into account but not publication bias.

In 2010 a group of researchers from the Canadian Institute of Work and Health and NIOSH updated the

NIOSH review with better methodology and newer studies.82

They were interested only in randomised

controlled trials and found 22 of these but included in the review only 14 which were judged as of sufficient

quality. They categorized interventions as low, medium or high worker engagement and they categorised

outcomes as knowledge, attitudes and beliefs, behaviours and health-outcomes. Based on the quality and the

effect size of studies in a category, evidence for effectiveness was assessed as strong, sufficient or

insufficient. There were only four studies with six interventions which fulfilled the inclusion criteria and

which aimed at reducing safety outcomes. Only two of these measured health/injury outcomes and both had

a non-significant outcome. The authors did not look at separate outcomes but combined all studies. They

concluded that there was insufficient evidence that knowledge was increased by training, strong evidence

that behaviour improved and again insufficient evidence that health outcomes improved. This somewhat

contradictory result is due to the lack of high quality studies in the knowledge domain even though the effect

size was large. The authors concluded also that there was insufficient evidence for a greater effect of training

with higher worker engagement.

Behavioural safety interventions: feedback and rewards

Wirth and Sigurdsson provided an overview of behavioural safety research without pretending that the

overview is systematic.83

The approach is based on behavioural psychology and applied already in the 1930s.

They defined behavioural safety as "an approach designed to change safety-related behaviours directly

through the application of behavioural principles and multiple strategies such as peer observations of safe

behaviours, goal setting, performance feedback and celebrations or incentives for safety goals." They see this

as an additional approach to engineering controls. They found that evidence of effectiveness is mostly

missing but that many of the intervention ideas are useful and should be better evaluated. The authors

stressed the importance of feedback and incentives such as rewards and celebrations because they are an

important feature of behavioural psychology. The idea is that reinforcement is "a basic learning process that

occurs when a behavioural consequence increases the frequency, intensity or duration of the targeted

behaviour".

In an older review of the use of incentives and feedback to enhance workplace safety, McAfee and Winn

aptly summarised 24 studies mostly carried out in the 1970s and 1980s.84

They included studies if they had

used an incentive intervention, a before-after outcome measurement and reported quantitative data. Most

28

studies were based on the principle that rewarded behaviour is likely to be repeated. The types of

interventions used were monetary incentives, praise and feedback and team competitions. The authors found

that all studies reported positive results in terms of a reduction of injuries in eight studies or an increase of

safety behaviour in the other 14 studies at least in the short term. One study reported that a reward system

still improved safety after having been in place for 12 years. Based on the results of their review they find it

impossible to tell which incentives were the best and they argue that better studies are needed to inform

practice.

Clinical: Pre-employment examinations for preventing injuries

The Cochrane review on pre-employment examinations used earlier, included two studies that aimed at

preventing injuries. Based on the results of the pre-employment examinations, the studies divided job

applicants in a group that were judged to be at risk and in a group that were not at risk.33

In one study, the

applicants that were at risk were provided with work accommodations. After one year of follow-up, the

injury rate in the group with work accommodations was similar to the group of job applicants that were

considered to be not at risk. This was taken as evidence of effectiveness of a beneficial effect of pre-

employment examinations and work-accommodations. In another study in the military, a training programme

was offered to those that were judged to be not physically fit enough to endure military training without

sustaining injuries. At the end of follow-up the injury rate in those who received training was similar to those

who were judged not to be at risk. Also in this study, this was taken as evidence that the pre-employment

examination with the training intervention was effective in reducing injuries in those at risk.

However, the confidence intervals around the risk estimates were wide and included also a substantial higher

risk for those that were provided with workplace accommodations and training. Even though these studies

provide some evidence that work-accommodation and training can be effective interventions with pre-

employment examinations, the quality of the evidence was assessed as very low.

3.5. Approaches to Small Enterprises

Small enterprises form the majority of companies and a considerable part of workers are employed by small

companies. Small enterprises differ from bigger enterprises in that they are usually led by the owner who has

to handle all the management tasks. Halse and Limborg describe that the owner is usually suspicious about

regulation and external consultants.23

The amount of resources that small companies are able and willing to

devote to occupational safety and health is usually limited. In addition, injuries and accidents occur rather

infrequently which easily leads to an ad-hoc approach in safety matters. The authors describe two approaches

to meet the specific needs of small firms. One is to develop specific tools such as predefined checklists for

risk assessment for specific types of firms. The other one is to work with intermediary organisations that

support small enterprises such as labour unions, insurance companies or occupational health services. The

authors acknowledge that none of these methods is based on thorough evaluation studies.

29

Brooke advocates a specific hazards-scheme approach towards the occupational health problems of small

firms to control health risks from chemicals.85

Their approach is very similar to control-banding and also

based on the idea that it is much easier to work with hazard bands than to measure all exposures. They

propose to create the hazard bands based on the R-phrases that are required based on EU-regulation. An

evaluation of the hazard scheme to occupational exposure limits showed that the scheme is a potentially

powerful tool for helping SMEs to control chemical risks. However, there is no evaluation that shows if risks

in practice are more effectively controlled using the hazard-scheme.

One systematic review focussed entirely on safety interventions for businesses with less than 100 employees.

The review included five studies that covered a wide range of varying interventions. The interventions were

found to be effective in increasing safety-related attitudes and beliefs but had mixed effects on exposure and

no effect on health outcomes. It was difficult to draw conclusions that were specific for small firms from this

review.86

4. Conclusions and discussion

Table 2 summarises the evidence that I found for the effectiveness of essential primary preventive

occupational safety and health interventions. The general conclusions are that there is evidence from

systematic reviews that:

- many technical interventions for inhalation exposure reduction are effective and do not necessarily

have to be costly and can be based on simplified exposure assessments such as control-banding

- there is indirect evidence that shows that specific ventilation controls are better than general controls

- regulation and incentives for employers are probably one of the main causes of inhalation exposure

reduction in the industrialised world in the past forty years. It is therefore concluded that regulation

and incentives are effective in implementing technical exposure controls in firms.

- personal protective equipment can reduce exposure in a technical sense but there are many practical

barriers that impede its effectiveness in practice

- personal protective equipment is not a reliable tool without proper instruction and adaptation. This

holds for both respiratory and for hearing protection equipment

- that pre-employment examinations might prevent occupational asthma for specific exposures, even

though the quality of the evidence is very low

- regulation and enforcement can reduce noise levels in workplaces

- hearing loss prevention programmes that are mainly based on hearing protection are probably not

sufficiently protective even though the quality of the evidence is low

- there is no evidence in the available studies that back pain can be prevented neither by training and

education nor by ergonomic improvements nor by pre-employment examinations

- technical passive hazard controls such as roll-over protection structures on tractors can reduce fatal

injuries but for most technical controls there are no studies or no systematic reviews

- regulation and incentives for employers for reducing injuries produce mixed results but there are no

systematic reviews of measures to improve the safety climate in a company

30

- incentives such as feedback and rewards for workers improve safety behaviour and probably reduce

injuries

- education and training to prevent injuries produces mixed results with some reviews providing

evidence of effectiveness but with other reviews not providing such evidence.

- pre-employment examinations which lead to work accommodations or extra training might lead to

lower injury rates

31

Table 2: Overview of preventive occupational health interventions and the evidence for their effectiveness

Work-related

disorder to be

prevented

Risk factors to be

addressed

Types of interventions

Engineering Behavioural Clinical

Substitution / Isolation-

ventilation

PPE Education/

Training

Other behavioural

interventions

Scheduling/

Health Exam

Other

Clinical

Cancer

Asthma

COPD

Asbestos,

Silica,

Coal dust,

Welding

fumes

HMW

Biological

agents

LMW

Chemical

agents

Non-

specific

dusts and

fumes

Inhalation-

exposure

Prevention

Technical measures:

- Substitution ±

- Enclosure +

- LEV ++

- Special

ventilation ++

- General

ventilation +

- Suppression ++

- Separation ++

Implementation measures

- regulation +

- incentives ±

Respiratory protection

- Technical

properties +

- Implementation ?

OSH training:

see injuries 0/±

Worker Incentives:

see injuries ±

- pre-

employment

examinations

±

Medication

to prevent

cancer: 0

Noise-

induced

Hearing Loss

Sound levels above 80

dB(A)

Technical measures ++

Implementation measures

- Regulation +

- Incentives ?

Hearing Protection

Technical properties

- without instruction

±

- with instruction

+++

implementation

- school-based +

- work-based ±

Instruction

hearing protection

++

General see

injuries ±

Worker Incentives:

see injuries ±

Pre-

employment

exam 0

Magnesium

?

Back Pain Ergonomic risk factors:

manual material handling,

bending and twisting,

heavy physical load,

static work posture,

Technical measures ±

Implementation /

Ergonomics 0

Lifting maximum ?

Aids

Technical properties ?

Implementation 0

Instruction

manual material

handling / lifting

0

Incentives ? Pre-

employment

0

6. Injury

Prevention

Hazardous situations at

work

Technical measures

- Fall Prevention +

- Roll-over

Technical measures ?

Implementation ?

Education/training

0/±

Education

Worker Incentives

- monetary +

- praise and

Pre-

employment

±

NA

32

Work-related

disorder to be

prevented

Risk factors to be

addressed

Types of interventions

Engineering Behavioural Clinical

Substitution / Isolation-

ventilation

PPE Education/

Training

Other behavioural

interventions

Scheduling/

Health Exam

Other

Clinical

protection +

Implementation

- Safety climate ?

Incentives:

- Regulation ±

- Experience

rating ++

- Enforcement +

- Inspections +

- Penalties ++

- Subsidies ?

agriculture 0

feedback +

- team

competitions

+

? No evidence such as systematic reviews found

0 Evidence available but no indication of effectiveness

± Evidence available and some indication of effectiveness

+ Evidence available and indication of effectiveness

++ Evidence available and strong indication of effectiveness

33

Strong aspects of this study are that I used a systematic approach to primary prevention of occupational

diseases and injuries. I followed the global burden of disease approach to determine which diseases and

injuries should be targeted to provide the highest global impact and thus could be called essential

interventions. I used a theoretical framework to underpin a general intervention approach of environmental,

behavioural and clinical interventions. This allowed for a broader view on prevention than just medical

studies. This justified our focus on exposure prevention both from the point of view of occupational hygiene

and from the point of view of occupational medicine. This also allowed considering more concrete

interventions such as local exhaust ventilation and pre-employment examination, especially compared to a

more general approach as usual in occupational medicine such as health surveillance or workplace health

surveillance. It is much more difficult to provide evidence of the latter measures because they can include

both exposure assessment as well as interventions. Moreover, these are so broadly defined that it will be

difficult to come up with concrete evidence for their effectiveness. Thanks to this framework I could focus

on concrete interventions that should impact on the global burden of disease.

Limitations of this report are that the available time and resources were restricted and I could not assess the

quality of the evidence and that I could not more concretely provide specific effect sizes for the included

interventions. I do not expect that this would have altered my conclusions but it would have made better fine-

tuning possible. Therefore, the assessment of the overall availability of evidence for specific interventions

was necessarily global and subjective. It has to be noted, however, that there are still many white spots where

there is no or little evidence available. Especially in the field of occupational injury prevention systematic

reviews of technical measures to prevent injuries are missing. This might be due to the technical nature of the

interventions and little attention paid to the implementation of these measures.

When the focus is on general interventions such as education and training or inhalation exposure prevention,

it is difficult to search the literature. The reason is that some authors use specific terms for example ‘correct

lifting postures' and others use general terms such as ‘training to prevent back pain’. When searching for

general interventions, it is easy to overlook the specific ones. That this happens in reality can be inferred

from the differences between training and education reviews and the review on manual material handling

advice to prevent back pain. Many of the studies that are included in the latter review are not included in the

former reviews. I assume that the reason for this is that the search strategies and the focus have been too

general. I tried to prevent this in my report by keeping the focus as broad as possible but it can be that I have

overlooked reviews in specific areas such as injury prevention or occupational hygiene because the

terminology used is different from that in occupational medicine. However, I believe that by thoroughly

looking at the occupational hygiene literature this bias is minimized. There were only a few studies that

evaluated the specific difficulties of small and medium sized enterprises which form the majority of the

enterprises and which employ most workers. The available studies show that specific approaches for these

enterprises are needed and that general implementation measures may not suffice.

34

There are few other studies that have reviewed the effectiveness of primary preventive occupational safety

and health interventions. Goldgruber and Ahrens discussed the effectiveness of workplace health promotion

and primary preventive workplace interventions based on 17 systematic reviews. Their focus was however

more on stress and health promotion than on chemical hazards and injuries such as in our report. For

ergonomics and training and education they used similar reviews as I did and they came to similar

conclusions. The lack of a proper theoretical framework makes their study difficult to compare to this.87

Implications for practice

There are many technical interventions available that can have a major impact on the global burden of work-

related cancer, asthma, COPD, noise and injuries. Better implementation interventions are needed to realise

this potential. Examples of such interventions are better regulation and reinforcement. However, the

available studies do not provide evidence that back pain can be prevented. Personal protective equipment has

technical potential to reduce exposure but without proper use and instruction this can not be realised. On the

contrary, there is no evidence in the available studies that education and training reduce occupational disease

and injuries. Feedback and rewards probably help in reducing occupational injuries. Clinical interventions

such as drugs and health examinations have little to offer for primary prevention of occupational diseases

and injuries

Implications for research

Better focussed questions are needed to be able to review the effectiveness of essential occupational safety

and health interventions. Better reviewing of primary studies of injury prevention is needed to get a better

overview of injury prevention effectiveness.

35

8. References

(1) Rantanen J. Basic Occupational Health Services. 1-22. 2007. Helsinki, Finnish Institute of

Occupational Health.

(2) WHO. Global plan of action on workers' health 2008-2017. 2008. Geneva, WHO.

(3) Verbeek J, Husman K, van DF, Jauhiainen M, Pasternack I, Vainio H. Building an evidence base for

occupational health interventions. Scand J Work Environ Health 2004; 30(2):164-170.

(4) Concha-Barrientos M, Nelson DI, Driscoll T, Steenland NK, Punnet L, Fingerhut MA et al. Selected

occupational risk factors. In: Ezzati M, Lopez A, Rodgers A, Murray C, editors. Comparative

quantification of health risks:global and regional burden of disease attributable to selected major risk

factors. Geneva: World Health Organisation; 2004. 1651-1802.

(5) Concha-Barrientos M, Nelson DI, Fingerhut M, Driscoll T, Leigh J. The global burden due to

occupational injury. Am J Ind Med 2005; 48(6):470-481.

(6) Hunter's diseases of occupations. 10th ed. London, UK: Hodder Arnold; 2010.

(7) Smandych RS, Thomson M, Goodfellow H. Dust control for material handling operations: a

systematic approach. Am Ind Hyg Assoc J 1998; 59(2):139-146.

(8) Cullinan P, Tarlo S, Nemery B. The prevention of occupational asthma. Eur Respir J 2003;

22(5):853-860.

(9) Malchaire J, Piette A. A comprehensive strategy for the assessment of noise exposure and risk of

hearing impairment. Ann Occup Hyg 1997; 41(4):467-484.

(10) Noise and its effects. Chichester, UK: John Wiley; 2007.

(11) Dionne CE, Dunn KM, Croft PR, Nachemson AL, Buchbinder R, Walker BF et al. A consensus

approach toward the standardization of back pain definitions for use in prevalence studies. Spine

(Phila Pa 1976 ) 2008; 33(1):95-103.

(12) van Dieen JH, Hoozemans MJ, Toussaint HM. Stoop or squat: a review of biomechanical studies on

lifting technique. Clin Biomech (Bristol , Avon ) 1999; 14(10):685-696.

(13) Lotters F, Burdorf A, Kuiper J, Miedema H. Model for the work-relatedness of low-back pain. Scand

J Work Environ Health 2003; 29(6):431-440.

(14) Keyserling WM, Smith GS. Using process control concepts to model conditions required for sudden-

onset occupational injuries. J Occup Environ Hyg 2007; 4(7):467-475.

(15) Haddon W, Jr. Advances in the epidemiology of injuries as a basis for public policy. Public Health

Rep 1980; 95(5):411-421.

(16) Haddon W, Jr. Energy damage and the ten countermeasure strategies. J Trauma 1973; 13(4):321-

331.

(17) Haddon W, Jr. Editorial: Strategy in preventive medicine: passive vs. active apprroaches to reducing

human wastage. J Trauma 1974; 14(4):353-354.

(18) Runyan CW. Using the Haddon matrix: introducing the third dimension. Inj Prev 1998; 4(4):302-

307.

36

(19) Christian MS, Bradley JC, Wallace JC, Burke MJ. Workplace safety: a meta-analysis of the roles of

person and situation factors. J Appl Psychol 2009; 94(5):1103-1127.

(20) van der Molen HF, Lehtola MM, Lappalainen J, Hoonakker PL, Hsiao H, Haslam R et al.

Interventions for preventing injuries in the construction industry. Cochrane Database Syst Rev

2007;(4):CD006251.

(21) Lehtola MM, Rautiainen RH, Day LM, Schonstein E, Suutarinen J, Salminen S et al. Effectiveness

of interventions in preventing injuries in agriculture--a systematic review and meta-analysis. Scand J

Work Environ Health 2008; 34(5):327-336.

(22) Ahonen EQ, Benavides FG, Benach J. Immigrant populations, work and health--a systematic

literature review. Scand J Work Environ Health 2007; 33(2):96-104.

(23) Hasle P, Limborg HJ. A review of the literature on preventive occupational health and safety

activities in small enterprises. Ind Health 2006; 44(1):6-12.

(24) Laamanen I, Verbeek J, Franco G, Lehtola M, Luotamo M. Finding toxicological information: An

approach for occupational health professionals. J Occup Med Toxicol 2008; 3:18.

(25) Ellenbecker MJ. Engineering controls as an intervention to reduce worker exposure. Am J Ind Med

1996; 29(4):303-307.

(26) Le Heron J., Padovani R, Smith I, Czarwinski R. Radiation protection of medical staff. Eur J Radiol

2010; 76(1):20-23.

(27) Zalk DM, Nelson DI. History and evolution of control banding: a review. J Occup Environ Hyg

2008; 5(5):330-346.

(28) Cohen H, White EM. Metalworking fluid mist occupational exposure limits: a discussion of

alternative methods. J Occup Environ Hyg 2006; 3(9):501-507.

(29) Cox P, Niewohmer J, Pidgeon N, Gerrard S, Fischhoff B, Riley D. The use of mental models in

chemical risk protection: developing a generic workplace methodology. Risk Anal 2003; 23(2):311-

324.

(30) Niewohner J, Cox P, Gerrard S, Pidgeon N. Evaluating the efficacy of a mental models approach for

improving occupational chemical risk protection. Risk Anal 2004; 24(2):349-361.

(31) Lazovich D, Parker DL, Brosseau LM, Milton FT, Dugan SK, Pan W et al. Effectiveness of a

worksite intervention to reduce an occupational exposure: the Minnesota wood dust study. Am J

Public Health 2002; 92(9):1498-1505.

(32) Meijster T, Tielemans E, Heederik D. Effect of an intervention aimed at reducing the risk of allergic

respiratory disease in bakers: change in flour dust and fungal alpha-amylase levels. Occup Environ

Med 2009; 66(8):543-549.

(33) Mahmud N, Schonstein E, Schaafsma F, Lehtola MM, Fassier JB, Reneman MF et al. Pre-

employment examinations for preventing occupational injury and disease in workers. Cochrane

Database Syst Rev 2010;(12):CD008881.

(34) Roelofs CR, Barbeau EM, Ellenbecker MJ, Moure-Eraso R. Prevention strategies in industrial

hygiene: a critical literature review. AIHA J (Fairfax , Va ) 2003; 64(1):62-67.

(35) LaMontagne AD, Radi S, Elder DS, Abramson MJ, Sim M. Primary prevention of latex related

sensitisation and occupational asthma: A systematic review. Occupational and Environmental

Medicine 2006; 63(5):359-364.

37

(36) Fransman W, Schinkel J, Meijster T, Van HJ, Tielemans E, Goede H. Development and evaluation

of an exposure control efficacy library (ECEL). Ann Occup Hyg 2008; 52(7):567-575.

(37) Vermeulen R, de HJ, Swuste P, Kromhout H. Trends in exposure to inhalable particulate and dermal

contamination in the rubber manufacturing industry: effectiveness of control measures implemented

over a nine-year period. Ann Occup Hyg 2000; 44(5):343-354.

(38) Kromhout H, Vermeulen R. Long-term trends in occupational exposure: Are they real? What causes

them? What shall we do with them? Ann Occup Hyg 2000; 44(5):325-327.

(39) Creely KS, Cowie H, Van TM, Kromhout H, Tickner J, Cherrie JW. Trends in inhalation exposure--

a review of the data in the published scientific literature. Ann Occup Hyg 2007; 51(8):665-678.

(40) Park D, Stewart PA, Coble JB. A comprehensive review of the literature on exposure to

metalworking fluids. J Occup Environ Hyg 2009; 6(9):530-541.

(41) Coble JB, Lees PS, Matanoski G. Time trends in exposure measurements from OSHA compliance

inspections of the pulp and paper industry. Appl Occup Environ Hyg 2001; 16(2):263-270.

(42) Symanski E, Kupper LL, Hertz-Picciotto I, Rappaport SM. Comprehensive evaluation of long-term

trends in occupational exposure: Part 2. Predictive models for declining exposures. Occup Environ

Med 1998; 55(5):310-316.

(43) Liden G. Dustiness testing of materials handled at workplaces. Ann Occup Hyg 2006; 50(5):437-439.

(44) Madl AK, Donovan EP, Gaffney SH, McKinley MA, Moody EC, Henshaw JL et al. State-of-the-

science review of the occupational health hazards of crystalline silica in abrasive blasting operations

and related requirements for respiratory protection. J Toxicol Environ Health B Crit Rev 2008;

11(7):548-608.

(45) Collingwood S, Heitbrink WA. Field evaluation of an engineering control for respirable crystalline

silica exposures during mortar removal. J Occup Environ Hyg 2007; 4(11):875-887.

(46) Shepherd S, Woskie SR, Holcroft C, Ellenbecker M. Reducing silica and dust exposures in

construction during use of powered concrete-cutting hand tools: efficacy of local exhaust ventilation

on hammer drills. J Occup Environ Hyg 2009; 6(1):42-51.

(47) Meeker JD, Cooper MR, Lefkowitz D, Susi P. Engineering control technologies to reduce

occupational silica exposures in masonry cutting and tuckpointing. Public Health Rep 2009; 124

Suppl 1:101-111.

(48) Meeker JD, Susi P, Flynn MR. Manganese and welding fume exposure and control in construction. J

Occup Environ Hyg 2007; 4(12):943-951.

(49) Meeker JD, Susi P, Flynn MR. Hexavalent chromium exposure and control in welding tasks. J

Occup Environ Hyg 2010; 7(11):607-615.

(50) Toren K, Sterner T. How to promote prevention--economic incentives or legal regulations or both?

Scand J Work Environ Health 2003; 29(3):239-245.

(51) Elsler D, Treutlein D, Rydlewska I, Frusteri L, Kruger H, Veerman T et al. A review of case studies

evaluating economic incentives to promote occupational safety and health. Scand J Work Environ

Health 2010; 36(4):289-298.

(52) Collegium Ramazzini. Asbestos is still with us: repeat call for a universal ban. Occ Med 60, 584-

585. 2011.

38

(53) Castleman B. "Controlled use" of asbestos. Int J Occup Environ Health 2003; 9(3):294-298.

(54) Brouwer DH, Marquart H, van Hemmen JJ. Proposal for an approach with default values for the

protection offered by PPE, under European new or existing substance regulations. Ann Occup Hyg

2001; 45(7):543-553.

(55) Howie RM. Respiratory protective equipment. Occup Environ Med 2005; 62(6):423-8, 362.

(56) Takemura Y, Kishimoto T, Takigawa T, Kojima S, Wang BL, Sakano N et al. Effects of mask

fitness and worker education on the prevention of occupational dust exposure. Acta Med Okayama

2008; 62(2):75-82.

(57) de Kort WL, Fransman LG, van Dijk FJ. Preemployment medical examinations in a large

occupational health service. Scand J Work Environ Health 1991; 17(6):392-397.

(58) de KW, van DF. Preventive effectiveness of pre-employment medical assessments. Occup Environ

Med 1997; 54(1):1-6.

(59) Serra C, Rodriguez MC, Delclos GL, Plana M, Gomez Lopez LI, Benavides FG. Criteria and

methods used for the assessment of fitness for work: a systematic review. Occup Environ Med 2007;

64(5):304-312.

(60) Palmer KT, Poole J, Rawbone RG, Coggon D. Quantifying the advantages and disadvantages of pre-

placement genetic screening. Occup Environ Med 2004; 61(5):448-453.

(61) Caraballoso M, Sacristan M, Serra C, Bonfill X. Drugs for preventing lung cancer in healthy people.

Cochrane Database Syst Rev 2003;(2):CD002141.

(62) Verbeek JH, Kateman E, Morata TC, Dreschler W, Sorgdrager B. Interventions to prevent

occupational noise induced hearing loss. Cochrane Database Syst Rev 2009;(3):CD006396.

(63) El Dib RP, Mathew JL. Interventions to promote the wearing of hearing protection. Cochrane

Database Syst Rev 2009;(4):CD005234.

(64) Dempsey PG. Usability of the revised NIOSH lifting equation. Ergonomics 2002; 45(12):817-828.

(65) Rivilis I, Van ED, Cullen K, Cole DC, Irvin E, Tyson J et al. Effectiveness of participatory

ergonomic interventions on health outcomes: a systematic review. Appl Ergon 2008; 39(3):342-358.

(66) Driessen MT, Proper KI, Van Tulder MW, Anema JR, Bongers PM, Van Der Beek AJ. The

effectiveness of physical and organisational ergonomic interventions on low back pain and neck

pain: A systematic review. Occupational and Environmental Medicine 2010; 67(4):277-285.

(67) Driessen MT, Proper KI, Anema JR, Knol DL, Bongers PM, Beek AJ. The effectiveness of

participatory ergonomics to prevent low-back and neck pain - results of a cluster randomized

controlled trial. Scand J Work Environ Health 2011.

(68) Van Poppel MN, Hooftman WE, Koes BW. An update of a systematic review of controlled clinical

trials on the primary prevention of back pain at the workplace. Occup Med (Lond) 2004; 54(5):345-

352.

(69) Martimo KP, Verbeek J, Karppinen J, Furlan AD, Kuijer PP, Viikari-Juntura E et al. Manual

material handling advice and assistive devices for preventing and treating back pain in workers.

Cochrane Database Syst Rev 2007;(3):CD005958.

(70) Clemes SA, Haslam CO, Haslam RA. What constitutes effective manual handling training? A

systematic review. Occup Med (Lond) 2010; 60(2):101-107.

39

(71) Bigos SJ, Holland J, Holland C, Webster JS, Battie M, Malmgren JA. High-quality controlled trials

on preventing episodes of back problems: systematic literature review in working-age adults. Spine J

2009; 9(2):147-168.

(72) Dawson AP, McLennan SN, Schiller SD, Jull GA, Hodges PW, Stewart S. Interventions to prevent

back pain and back injury in nurses: a systematic review. Occup Environ Med 2007; 64(10):642-650.

(73) Verbeek J, Martimo KP, Karppinen J, uijer PP, iikari-Juntura E, akala EP. Manual material handling

advice and assistive devices for preventing and treating back pain in workers. Cochrane Database

Syst Rev 2011;(7):CD005958.

(74) Hsiao H, Simeonov P. Preventing falls from roofs: a critical review. Ergonomics 2001; 44(5):537-

561.

(75) Tompa E, Trevithick S, McLeod C. Systematic review of the prevention incentives of insurance and

regulatory mechanisms for occupational health and safety. Scand J Work Environ Health 2007;

33(2):85-95.

(76) Zohar D. Thirty years of safety climate research: reflections and future directions. Accid Anal Prev

2010; 42(5):1517-1522.

(77) DeRoo LA, Rautiainen RH. A systematic review of farm safety interventions. Am J Prev Med 2000;

18(4 Suppl):51-62.

(78) Rautiainen RH, Lehtola MM, Day LM, Schonstein E, Suutarinen J, Salminen S et al. Interventions

for preventing injuries in the agricultural industry. Cochrane Database Syst Rev

2008;(1):CD006398.

(79) Hartling L, Brison RJ, Crumley ET, Klassen TP, Pickett W. A systematic review of interventions to

prevent childhood farm injuries. Pediatrics 2004; 114(4):e483-e496.

(80) Cohen A, Colligan MJ. Assessing occupational safety and health training, a literature review. 98-

145, 1-164. 1998. Cincinnati,Ohio, National Institute of Occupational Safety and Health.

(81) Burke MJ, Sarpy SA, Smith-Crowe K, Chan-Serafin S, Salvador RO, Islam G. Relative effectiveness

of worker safety and health training methods. Am J Public Health 2006; 96(2):315-324.

(82) Robson LS, Stephenson C, Schulte P, Amick B, Chan S, Bielecky A et al. A systematic review of

the effectiveness of training and education for the protection of workers. 1-140. 2010. Toronto,

Ontario, Canada, Institute for Work and Health.

(83) Wirth O, Sigurdsson SO. When workplace safety depends on behavior change: topics for behavioral

safety research. J Safety Res 2008; 39(6):589-598.

(84) McAfee RB, Winn AR. The use of incentives/feedback to enhance work place safety: a critique of

the literature. Journal of Safety Research 20, 7-19. 1989.

(85) Brooke IM. A UK scheme to help small firms control health risks from chemicals: toxicological

considerations. Ann Occup Hyg 1998; 42(6):377-390.

(86) Breslin FC, Kyle N, Bigelow P, Irvin E, Morassaei S, MacEachen E et al. Effectiveness of health

and safety in small enterprises: a systematic review of quantitative evaluations of interventions.

Journal of occupational rehabilitation 2010; 20(2):163-179.

40

(87) Goldgruber J, Ahrens D. Effectiveness of workplace health promotion and primary prevention

interventions: a review. J Pub Health 2010; 18:75-88.

9. Appendices

1. General Search strings used in Pubmed to find occupational health systematic reviews

COHF: work

((effect*[tw] OR control[tw] OR controls*[tw] OR controla*[tw] OR controle*[tw] OR controli*[tw] OR

controll*[tw] OR evaluation*[tw] OR program*[tw] OR prevention*[tw] OR protect*[tw]) AND (work[tw]

OR works*[tw] OR work'*[tw] OR worka*[tw] OR worke*[tw] OR workg*[tw] OR worki*[tw] OR

workl*[tw] OR workp*[tw] OR occupation*[tw]) NOT animals[mh])

Perosh: systematic review

(meta-analysis[mh] OR meta-analysis[pt] OR meta-analysis[tiab] OR review[pt] OR review[tiab]) NOT

(letter[pt] OR editorial[pt] OR comment[pt]) NOT ((animals[Mesh:noexp]) NOT (humans[Mesh]))

Plus specific words

2. Search for Pre-employment examinations 3.3.2011

“Pre-employment” OR "fitness for work" plus Perosh systematic review search

Resulted in 60 references

2. Inhalation exposure 26.2.2011

"inhalation exposure" AND (occupation* OR worker*) AND meta-analysis

Resulted in 256 references

3. Dust control 26.2.2011

“dust control” AND Perosh systematic review:

25 references no systematic reviews

4. Respiratory protection

Respiratory protection[mh] AND Perosh systematic review

97 references

5. Noise 2.3.2011

Noise, occupational[mh] AND Perosh systematic review

157 references searched to 1990

6. Back Pain 26.2.2011

Back Pain[MH] AND COHF strategy AND Perosh Systematic Review

356 references

7. Injury prevention

(injur* OR accident*) AND cohf strategy AND Perosh systematic review

229 references