Elsevier

Safety Science

Volume 62, February 2014, Pages 16-27
Safety Science

Review
Occupational safety theories, models and metaphors in the three decades since World War II, in the United States, Britain and the Netherlands: A literature review

https://doi.org/10.1016/j.ssci.2013.07.015Get rights and content

Highlights

  • In safety science, the medical domain became dominant after WW II.

  • The theory of ‘task dynamics’, and the hazard-barrier-target model were developed.

  • Causes of accidents shifted from workers behavior to task and managerial causes.

  • Complexity in process and military industry created disasters and risk assessment tools.

  • Accident proneness remained popular in the professional safety domain.

Abstract

Objective

Which theories, models, and metaphors were developed in the period and countries under study, within what context, and if available based upon what collection of data.

Method

For the literature review, original articles were consulted, including volumes of the Dutch safety professional journal ‘De Veiligheid’ (Safety).

Results and conclusions

One theory and three models on accident causation were developed in the domain of safety science. The focus on the causes of accidents and on their prevention was gradually changing from victim behavior, via task aspects to management causes.

Willem Winsemius, a Dutch physician, is the father of ‘task dynamics theory’, explaining, predicting accidents from the reflex reactions and improvisations of workers during process disturbances. Based on a survey of 1300 accidents at the former Dutch steel works Hoogovens, his theory described human behavior as a response reaction, instead of an accident cause. British observational research conducted on more than 2000 accidents also highlighted the relationship between tasks, actions, process disturbances, and accidents. And William Haddon Jr., an American physician introduced the ‘epidemiological triangle’, and the known ‘hazard – barrier – target’ model. Finally, the ergonomics domain developed two models on disturbed information flows and inadequate decisions of workers.

The increased complexity in the military domain and the process industry, just after World War II initiated a movement to increase system reliability, leading to a number of safety techniques which were not based on systematic research, but were rather a coded collection of practical experiences.

The accident proneness theory still remained popular in the professional safety domain in the Netherlands.

Introduction

Three previous articles on the subject have provided an historical overview on knowledge development within the safety science domain. These articles described the professionalization of the field of safety science from 1870 till the first decades of the twentieth century (Swuste Gulijk and Zwaard, 2010, Gulijk et al., 2012). In the United States the ‘Safety First Movement’ was launched to reduce the extremely high incidence of fatal accidents. That was when the first safety theories arose. Crystal Eastman explained accidents from both the organizational and conditions of work points of view. Her theory was an example of the environmental hypothesis. In England the hypothesis of Greenwood and Woods explained accidents from the individual’s angle; they considered the predisposition of victims. This theory later came to be known as the accident proneness theory, and it was elaborated simultaneously in England and Germany (Burnham, 2009). According to this theory, accidents were caused by mental states of victims, and this presumption was supported by the then-popular Freudian theory on human behavior. In American literature, with more emphasis being placed on safety management, the human factor was viewed differently; the focus was more on responsibility in relation to accidents. The responsibility carried ranged from that of victims and fellow workers (Cowee, 1916, Heinrich, 1931) to the foreman and management (DeBlois, 1926, Heinrich, 1931).

This article will cover the period between the Second World War and 1974 and will discuss developments within the occupational safety domains in the United States, Britain and the Netherlands. The management approach, which was propagated in the early 1960s and 1970s (ABCM, 1964, Petersen, 1971, Bird, 1974), and built on the ideas of DeBlois and Heinrich, will be discussed in a forthcoming article. Also the development in process control, Loss Prevention, which started in the early 1960s (Lees, 1980), will be the topic of a next article. Only some safety techniques, developed in this period will be discussed at the end of this article. Like in previous articles, the main research questions were the following:

  • Which theories, models, metaphors and safety analysis techniques were developed in that period?

  • What was the basis to the theories, models, metaphors and techniques, and if known, what collection of data supported the theories?

  • Within what context did this development take place?

  • What were the consequences of this development for the professional safety domain in the Netherlands?

An extensive literature review serves as the basis for this investigation. After the war Britain, and even more so, the United States set the example in many areas, including in safety science. This was the argument when selecting the literature. Original references and sources were consulted. For consequences in the Netherlands all issues of the professional journal ‘De Veiligheid’, the national magazine for safety experts, were examined. The articles selected were those on accident causes together with the proposed safety solutions and interventions.

The research questions mention the term ‘theory’. A theory can be seen as a validated model. A theory can be used to analyze observed phenomena, in these cases concerning accidents and, importantly, with a theory predictions can be made. In common parlance, terms like ‘model’ and ‘metaphor’ are used interchangeably but this article makes a distinction. A model is a schematic, yet not validated presentation of reality. The precursor to a model is a metaphor, based upon a comparison. For example using this definition, the dominoes of Heinrich are a metaphor, since the accident process is compared with falling domino pieces (Heinrich, 1941). This also applies to the iceberg; it is a metaphor for the accident mechanism. Because of their simplicity, metaphors have strong persuasive power. The distinction between the development of safety as a science and the professional application of safety in practice is a second point linked to the research questions. The development of knowledge in the professional safety domain is not a priori synchronized with discussions continuing in the scientific domain.

Section snippets

Criticism on accident proneness theory

After World War II, US doctors became interested in accidents. It was noted with some surprise that medical science had failed to substantially reduce accidents, while huge successes had been achieved in combating various contagious and non-contagious diseases. The introduction of improved water supplies, antibiotics, the pasteurization of milk, vaccination programs, improved surgical procedures, the general improvement in living conditions, and medical care had led to dramatic reductions in

Task dynamics

William Winsemius (1917–1990) was trained as a doctor and worked as a scientist at the Department of Mental Health at the Institute of Preventive Medicine in Leiden, what later came to be known as NIPG-TNO. His PhD thesis in 1951 was a psychologically tinted topic ‘The psychology of accident events’ (Winsemius, 1951). Like other scientists, Winsemius sharply criticized the statistical approach to accident causation, which formed the basis to the accident proneness theory (see, for example,

Causes and prevention of 2000 accidents

A study into the causes and prevention of 2000 accidents was published in Britain in 1971. It was based on continuous observations that had gone on for more than a year in four medium-sized manufacturing companies (100–300 workers) from the metal, assembly and distribution sectors (Powell et al., 1971). Like with Winsemius, tasks and corresponding actions were seen as the main determinants of accidents. An increase in production, resulting in an increase in actions will increase accidents

Human factors and ergonomics

The focus on ergonomics, both by Powell and colleagues, and Winsemius is a result of developments from World War II onwards. The war had led to a considerable acceleration in the technical development of military equipment, machinery and industrial processes, and had created serious control problems (for a review see e.g. McIntyre, 2000). Weaponry and machines were becoming increasingly complex, and created new demands on operators’ cognition. It is generally considered that human factors,

Safety analysis techniques

After the Second World War a number of safety analysis techniques were developed, which assumed that the accident process was sequential. The American, DeBlois (1926) was the first to postulate that an accident should be understood as a sequence of events. Later this idea was repeated in the manual by Heinrich (1931), and ultimately it culminated in the famous domino metaphor (1941). This domino metaphor was the first accident analysis technique (for a review, see Gulijk et al., 2012). In the

The Netherlands

From World War II onwards, a variety of topics have been dealt with in the professional journal for safety experts, ‘De Veiligheid’. The general impression was that in the domain of occupational safety, the Netherlands was lagging far behind. Too many accidents occurred and in the safety domain the United States, and to a lesser extent the UK, were leading the way (Fetter, 1947, Gorter, 1947a, Anonymous, 1965). Apart from many articles on classical safety techniques, the modification of

Discussion and conclusions

The first research question asked was ‘which theories, models, metaphors and safety analysis techniques were developed in that period?’ In the relevant period, one epidemiological model (Haddon and colleagues) was developed, two models from the ergonomics domain (Dunn, Hale and Hale), one accident theory was put forward on task dynamics (Winsemius) and there were four analytical tools but no metaphors. These changes in the field of safety science coincided with an increased focus on ergonomics

References (142)

  • J. Dunn

    A safety analysis technique derived from skills analysis

    Applied Ergonomics

    (1972)
  • M. Guarnieri

    Landmarks in the history of safety

    Journal of Safety Research

    (1992)
  • ABCM

    Association of British Chemical Manufacturers. Safety and Management, A Guide for the Chemical Industry

    (1964)
  • J. Andriessen

    Waarom wil men veilig werken (Why do people work safely

    De Veiligheid

    (1974)
  • Anonymous

    Contactcommissie voor bedrijfsveiligheid en –hygiëne (Contact Commission for industrial safety and hygiene – in Dutch)

    De Veiligheid

    (1941)
  • Anonymous

    Veiligheidscommissies, het oordeel van het Hoofd der Engelse Fabrieksinspectie (Safety Committees, the opinion of the Head of British Factory Inspectorate? – in Dutch)

    De Veiligheid

    (1947)
  • Anonymous

    Boterhammenpapier met veiligheidswenken (Sandwich paper with safety tips? – in Dutch)

    De Veiligheid

    (1948)
  • Anonymous, 1949. Procedures for Performing a failure mode effects and critical analysis. November 9th, United States...
  • Anonymous

    Hoe bevorderen wij het veilig werken? (How we promote safe work? – in Dutch)

    De Veiligheid

    (1953)
  • Anonymous, 1958. Home Safety Project. Final report California State Department of Public...
  • Anonymous

    Basis voor effectieve veiligheidsoverdracht door de baas (Basis for effective safety transfer by the boss? – in Dutch)

    De Veiligheid

    (1962)
  • Anonymous

    ‘n Amerikaans oordeel over Europese fabrieken (An American opinion on European factories? – in Dutch)

    De Veiligheid

    (1965)
  • Anonymous

    Grondbeginselen van de preventie in de bedrijven. (Principles of prevention in companies – in Dutch)

    De Veiligheid

    (1968)
  • B. Archer

    Safety preventing industrial accidents

    Design

    (1965)
  • ANC – Arxiu Nacional de Catalunya

    Collecció the cartels the prevenció d’laborals accidents

    (1987)
  • S. Baker et al.

    Reducing injuries and their results: a scientific approach. The Millbank memorial fund quarterly

    Health and Safety Society

    (1974)
  • F. Bird et al.

    Damage Control, A New Horizon in Accident Prevention and Cost Improvement. American Management Association

    (1966)
  • F. Bird

    Management Guide to Loss Control

    (1974)
  • G. Blaauw

    Het wekken van de veiligheidsgedachte in de bedrijven bij toezichthoudend personeel en arbeiders. Voordracht voor de Commissie van overleg. (Raising the safety concept in companies with supervisory staff and workers. Presentation for the Conciliation Committee – in Dutch)

    De Veiligheid

    (1950)
  • J. Bloemen

    Veiligheidsorganisatie in een bedrijf – NV Nederlandse Aardolie Maatschappij (Safety organization in a company – Dutch Petroleum Company NV)

    De Veiligheid

    (1967)
  • A den Boer

    Veiligheid en wetenschap (Safety and science – in Dutch)

    De Veiligheid

    (1967)
  • A. Broadhurst

    Factory legislation and safety training

    Industrial and Commercial Training

    (1971)
  • J. Burnham

    Accident Prone, A History of Technology, Psychology, and Misfits of the Machine Age

    (2009)
  • C. Buschmann

    Kunnen veiligheidsvoorschriften worden afgedwongen? (Can safety procedures be enforced? – in Dutch)

    De Veiligheid

    (1972)
  • C. Buschmann

    Hoe kan een beter veiligheidsniveau worden bereikt? (How can a better security be achieved? – in Dutch)

    De Veiligheid

    (1972)
  • G. Cowee

    Practical Safety Methods and Devices. Manufacturing and Engineering

    (1916)
  • J. Creyghton

    Dispositie tot ongevallen (Disposition to accidents – in Dutch)

    De Veiligheid

    (1949)
  • J. Creyghton

    Jan ongeluk (Accident Jan – in Dutch)

    De Veiligheid

    (1952)
  • L. DeBlois

    Industrial Safety Organization for Executives and Engineers

    (1926)
  • DeLong, T., 1970. Fault Tree Manual. Texas A&M...
  • G. Dop

    Bedrijfsongevallen oorzaken classificatie preventie (Accidents causes classification prevention)

    De Veiligheid

    (1967)
  • Dutch Health Council (Gezondheidsraad), 1965. Rapport betreffende de algemeen statistische berichtgeving van ongevallen...
  • Ericson, II., C., 1999. Fault tree analysis, a history. In: Proceedings of the 17th International System Safety...
  • C. Ericson

    A short history of system safety

    Journal of System Safety

    (2006)
  • Fetter, S., 1947. Verslag van de 25e vergadering van de Commissie van overleg inzake veiligheid en hygiëne. In: Report...
  • B. Finley et al.

    Reduction of human errors in field test programs

    Human factors

    (1974)
  • G. Fortune

    Taak en positieve van de bedrijfsgeneeskundige dienst in de bedrijven. Voordracht Commissie van overleg. (Role and benefits of the occupational health services in the companies. Nomination Committee for consultation – in Dutch)

    De Veiligheid

    (1948)
  • J. Fraser

    Psychology

    (1951)
  • W. Friedrich

    Statistische analyse van industriële ongevallen als basis voor ongevallenpreventie. Voordracht voor Commissie van overleg. (Statistical analysis of industrial accidents as a basis for accident prevention. Nomination Committee for consultation – in Dutch)

    De Veiligheid

    (1951)
  • W. Gerritsen

    De optimale samenwerking tussen bedrijfsarts en veiligheidsinspecteur. (The optimal cooperation between occupational health and safety inspector – in Dutch)

    De Veiligheid

    (1957)
  • J. Gibson

    The contribution of experimental psychology to the formulation of the problem or safety – a letter for basic research. Behavioural Approaches to Accident Research. Association for the Aid of Crippled Children, New York, 77–89

  • J. Gordon

    Epidemiology of accidents. American Journal of Public Health 39:504–515

  • R. Gorter

    Veiligheid na de oorlog. (Safety after the war – in Dutch)

    De Veiligheid

    (1947)
  • R. Gorter

    Veiligheidscommissies en veiligheidsinspecteurs. (Safety committees and safety inspectors – in Dutch)

    De Veiligheid

    (1947)
  • C. Gulijk et al.

    Heinrich’s safety models

    Safety Science

    (2012)
  • W. Haddon

    A note concerning accident theory and research with special reference to motor vehicle accidents

    Annals of the New York Academy of Science

    (1963)
  • W. Haddon

    The changing approach to the epidemiology, prevention, and amelioration or trauma: the transition to approaches etiologically based rather than descriptive

    American Journal of Public Health

    (1968)
  • W. Haddon

    On the escape or serious: an ecologic note

    American Journal of Public Health

    (1970)
  • W. Haddon

    Energy damage and the measure to counter strategies

    Human Factors

    (1973)
  • Cited by (81)

    View all citing articles on Scopus
    View full text